From owner-freebsd-hardware Sun Aug 27 09:56:36 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id JAA27158 for hardware-outgoing; Sun, 27 Aug 1995 09:56:36 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id JAA27152 for ; Sun, 27 Aug 1995 09:56:35 -0700 Date: Sun, 27 Aug 1995 12:56:25 -0400 (EDT) From: -Vince- To: Brian Gottlieb cc: Gary Palmer , freebsd-hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <9508241430.AA00576@beru.wustl.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Thu, 24 Aug 1995, Brian Gottlieb wrote: > > -Vince- (-Vince-) writes: > > -Vince-> On Wed, 23 Aug 1995, Brian Gottlieb wrote: > >> > >> It all depends on what you're doing with it. In my machine at work I > >> have a 400 meg drive dedicated to swap. The circuit synthesis and > >> simulations we run here need LOTS of memory and swap. The "big" > >> machines in our group have 256M of memory and a 1 Gig swap drive. > > -Vince-> Hmmm, is there like a way to do well with a big swap and > -Vince-> like 16 megs of physical memory? How much physical memory is > -Vince-> on the machine with 400 meg swap? > > Today is a good day to answer this. Last night I got a memory > upgrade. There is now 192M of ram in the machine (I feel like a kid > in a playground..."My machine could beat up your machine" ;) > > When I originally wrote this, I had 64M RAM. But the synthesiser > swapped too much and took way to long to run on here. With more > memory, it swapped less and ran much faster. I don't remember the > numbers, but it was very significant. A Sparc 5 with 192M RAM was > keeping up with a Sparc 10 with 128M. One factor we probably didn't > consider was that the 10 may have had a faster disk on it. Hmmm, how does a FreeBSD box match up to a Sparc though? > I'm no expert on things, but I think there is a point where you may > get diminishing returns on having lots of swap. On the other hand, I > have 16 Megs RAM in my freebsd machine and 32 megs of swap, and I have > run out of memory a few times. More swap would probably help. But > perhaps if I got my swap too big, it may become less efficient since I > could run more programs, but it would spend more time swapping. I guess so but physical ram is expensive while swap is cheap so like isn't there a point where you can just put a number for the physical memory and just use swap the rest of the way and it will work fine? > I don't know. To tell the truth, the 400 Megs of swap on here > probably rarely gets filled up, since I typically ran my simulations > elsewhere. But I know that the Sparc 10 in the office upstairs was > swapping like mad when I ran stuff on it (I could almost hear the > drive down here ;). Really? I always thought SUN's were quiet ;) > Hmm...long post. So the answer is...I have no clue ;) Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Sun Aug 27 16:51:56 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id QAA11540 for hardware-outgoing; Sun, 27 Aug 1995 16:51:56 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id QAA11534 for ; Sun, 27 Aug 1995 16:51:54 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id JAA21496; Mon, 28 Aug 1995 09:23:27 +0930 From: Michael Smith Message-Id: <199508272353.JAA21496@genesis.atrad.adelaide.edu.au> Subject: Re: Upgrade to my machine To: vince@penzance.econ.yale.edu (-Vince-) Date: Mon, 28 Aug 1995 09:23:27 +0930 (CST) Cc: hardware@freebsd.org In-Reply-To: from "-Vince-" at Aug 27, 95 12:56:25 pm Content-Type: text Content-Length: 1607 Sender: hardware-owner@freebsd.org Precedence: bulk -Vince- stands accused of saying: > > When I originally wrote this, I had 64M RAM. But the synthesiser > > swapped too much and took way to long to run on here. With more > > memory, it swapped less and ran much faster. I don't remember the > > numbers, but it was very significant. A Sparc 5 with 192M RAM was > > keeping up with a Sparc 10 with 128M. One factor we probably didn't > > consider was that the 10 may have had a faster disk on it. > > Hmmm, how does a FreeBSD box match up to a Sparc though? Depends on what you're trying to do with it, and what sort of FreeBSD box you're talking about. A VLB '486DX2/66 will just about keep even with a Sparc 2 otherwise similarly configured; I don't have any data on faster Sun systems as just about everything around here is an Alpha. > I guess so but physical ram is expensive while swap is cheap so > like isn't there a point where you can just put a number for the physical > memory and just use swap the rest of the way and it will work fine? If your working set size significantly exceeds your physical memory size, it doesn't matter how much swap you have, you're still going to thrash. > -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Mon Aug 28 01:17:13 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id BAA10473 for hardware-outgoing; Mon, 28 Aug 1995 01:17:13 -0700 Received: from nietzsche (annex1s2.urc.tue.nl [131.155.12.12]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id BAA10093 for ; Mon, 28 Aug 1995 01:14:02 -0700 Received: from localhost (localhost [127.0.0.1]) by nietzsche (8.6.11/8.6.9) with ESMTP id KAA15846 for ; Mon, 28 Aug 1995 10:16:23 +0200 Message-Id: <199508280816.KAA15846@nietzsche> X-Mailer: exmh version 1.6.1 5/23/95 To: hardware@freebsd.org Subject: Quantum Fireball, any good? Date: Mon, 28 Aug 1995 10:16:22 +0200 From: Marc van Kempen Sender: hardware-owner@freebsd.org Precedence: bulk Hi, I can get a nice deal on a quantum fireball 1080S. Are they any good performancewise? Any known flaws with respect to FreeBSD (2.0.5)? It will be connected to a NCR53c825 on a Dell 90Mhz pentium. Regards, Marc. ---------------------------------------------------- Marc van Kempen wmbfmk@urc.tue.nl He's dead Jim ..., kick him if you don't believe me. From owner-freebsd-hardware Mon Aug 28 01:44:45 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id BAA12562 for hardware-outgoing; Mon, 28 Aug 1995 01:44:45 -0700 Received: from toplink1.toplink.de (toplink1.toplink.de [194.163.120.1]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id BAA12538 for ; Mon, 28 Aug 1995 01:44:33 -0700 Received: (from ck@localhost) by toplink1.toplink.de (8.6.9/8.6.9) id KAA00864; Mon, 28 Aug 1995 10:44:08 +0200 From: Christian Kratzer Message-Id: <199508280844.KAA00864@toplink1.toplink.de> Subject: Re: Quantum Fireball, any good? To: wmbfmk@urc.tue.nl (Marc van Kempen) Date: Mon, 28 Aug 1995 10:44:05 +0200 (MET DST) Cc: hardware@freebsd.org In-Reply-To: <199508280816.KAA15846@nietzsche> from "Marc van Kempen" at Aug 28, 95 10:16:22 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1293 Sender: hardware-owner@freebsd.org Precedence: bulk Hi sorry about the previous empty message. > Hi, > > I can get a nice deal on a quantum fireball 1080S. Are they any good > performancewise? Any known flaws with respect to FreeBSD (2.0.5)? > It will be connected to a NCR53c825 on a Dell 90Mhz pentium. > I have two of them in my 2.0R system. I don`t have any performance data handy but here's something else. I did have serious problems when I tried installing. After I had partioned them and booted for the second time FreeBSD kept crashing on the fsck of /dev/sd0e. Actually you could hear the spindle motor stop as if somebody had turned to power off. After going nuts over a weekend I finally managed to get the installtion going by manually running newfs and fsck. Perhaps somebody could comment on the reliability of this drive. I am running a newsfeed into a 800MB partition on one of these drives. It's taking quite a hammering when a batch of new articles comes in. Greetings Christian -- -------------------------------------------------------------------------------- TopLink GbR, Internet Services info@toplink.de Christian Kratzer Phone: +49 7452 87174 http://www.toplink.de/ Fax: +49 7452 87175 FreeBSD spoken here! -------------------------------------------------------------------------------- From owner-freebsd-hardware Mon Aug 28 02:40:19 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id CAA16267 for hardware-outgoing; Mon, 28 Aug 1995 02:40:19 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id CAA16255 for ; Mon, 28 Aug 1995 02:40:16 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id CAA00575; Mon, 28 Aug 1995 02:40:03 -0700 From: "Rodney W. Grimes" Message-Id: <199508280940.CAA00575@gndrsh.aac.dev.com> Subject: Re: Quantum Fireball, any good? To: wmbfmk@urc.tue.nl (Marc van Kempen) Date: Mon, 28 Aug 1995 02:40:03 -0700 (PDT) Cc: hardware@freebsd.org In-Reply-To: <199508280816.KAA15846@nietzsche> from "Marc van Kempen" at Aug 28, 95 10:16:22 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2109 Sender: hardware-owner@freebsd.org Precedence: bulk > > > Hi, > > I can get a nice deal on a quantum fireball 1080S. Are they any good > performancewise? Any known flaws with respect to FreeBSD (2.0.5)? > It will be connected to a NCR53c825 on a Dell 90Mhz pentium. These are Quantums low end PC class of disk drives designed to compete in the price diving PC market. They are not known for performance, or at least, not in the normal since of Quantum drives. Here is a break down of the series of Quantum disks from low performace to high performance. (For current production models, I have dropped the older Empire and ProDrive series of drives as they are no longer a production item) Maverick 14mS 3600RPM Trailblazer 14mS 4500RPM Fireball 12mS 5400RPM Lightning 11mS 4500RPM Capella 8.5mS 5400RPM Grand Prix 8.6mS 7200RPM Atlas 8mS 7200RPM Of these drives I mostly sell Capella and Atlas class drives (500MB to 4Gbytes). The DEC/Quantum DSP3xxx series of drives are really Capella class drives, but they have now dropped out of production, I really liked them, but now am again searching for drive models in the 500 and 1G sizes that meet my performance and reliablity standards :-(. I don't know what your ``nice deal'' is, but I can sell you a pair of: XX. BAS DEC3053L Dec/Quantum 535MB 3.5"x1", SCSI-II, 5400 RPM, 9.5mS $ 195.00 Or if I can still get it (Friday showed 2 in stock) the 1.07G 7200RPM 8mS atlas drive is a real little screamer for performance. A bit on the spendy side though, at $770.00. I don't have current pricing or avaliabilty status on the VP31110, that is the 1G Capella drive, should be cost effective as that is a new production series (about 3 months old now). Go visit the Quantum web page, http://www.quantum.com, you'll notice that the Maverick->Lightning class drives are under the PC section, and what I advocate (Capella->Atlas) are in the Workstation/Server pages. I do _not_ recommend the Grand Prix, it is simply there as a data point. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Mon Aug 28 02:51:45 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id CAA17217 for hardware-outgoing; Mon, 28 Aug 1995 02:51:45 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id CAA17209 for ; Mon, 28 Aug 1995 02:51:42 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id CAA00607; Mon, 28 Aug 1995 02:50:55 -0700 From: "Rodney W. Grimes" Message-Id: <199508280950.CAA00607@gndrsh.aac.dev.com> Subject: Re: Quantum Fireball, any good? To: ck@toplink.de (Christian Kratzer) Date: Mon, 28 Aug 1995 02:50:55 -0700 (PDT) Cc: wmbfmk@urc.tue.nl, hardware@freebsd.org In-Reply-To: <199508280844.KAA00864@toplink1.toplink.de> from "Christian Kratzer" at Aug 28, 95 10:44:05 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1795 Sender: hardware-owner@freebsd.org Precedence: bulk > > Hi > > sorry about the previous empty message. > > > Hi, > > > > I can get a nice deal on a quantum fireball 1080S. Are they any good > > performancewise? Any known flaws with respect to FreeBSD (2.0.5)? > > It will be connected to a NCR53c825 on a Dell 90Mhz pentium. > > > I have two of them in my 2.0R system. I don`t have any performance > data handy but here's something else. > > I did have serious problems when I tried installing. After I had partioned > them and booted for the second time FreeBSD kept crashing on the fsck > of /dev/sd0e. Actually you could hear the spindle motor stop as if somebody > had turned to power off. After going nuts over a weekend I finally managed > to get the installtion going by manually running newfs and fsck. > > Perhaps somebody could comment on the reliability of this drive. I am > running a newsfeed into a 800MB partition on one of these drives. It's > taking quite a hammering when a batch of new articles comes in. Oh, I forget to mention this in my other reply, anytime I see a drive manuafacture drop performance, price and reliabilty numbers for a new series of drives vs what they had been shipping it scares me BIG TIME about those drives. The fireball/lightning series in effect replaced the Empire/ProDrive series on the low end of Quantums scale. Along with it came a warranty reduction from 5 years to 3 years, a drop in MTBF numbers from 500k to 300K, and a huge price drop. Now you go figure it... is it Quantum wanting to break into the PC disk drive market in a bigger way, or is Quantum going to forsake it's reputation and start selling junk drives? -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Mon Aug 28 18:17:11 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id SAA03771 for hardware-outgoing; Mon, 28 Aug 1995 18:17:11 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id SAA03749 for ; Mon, 28 Aug 1995 18:16:56 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id KAA24324; Tue, 29 Aug 1995 10:48:14 +0930 From: Michael Smith Message-Id: <199508290118.KAA24324@genesis.atrad.adelaide.edu.au> Subject: Re: S.O.S -2.1Stable and ASUSP54TP4 To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Tue, 29 Aug 1995 10:48:13 +0930 (CST) Cc: hardware@freebsd.org In-Reply-To: <199508281725.KAA01867@gndrsh.aac.dev.com> from "Rodney W. Grimes" at Aug 28, 95 10:25:23 am Content-Type: text Content-Length: 2393 Sender: hardware-owner@freebsd.org Precedence: bulk Rodney W. Grimes stands accused of saying: > > Just on the 'cuda thread; I had opportunity to eyeball a pile of -4's > > tonight. Some observations for fans of big and fast disks, and > > particularly those that have met these drives before : > > By -4's do you mean ``Hawk-4'' series drives? As far as I can tell > the -4 in Barracuda-4 and Hawk-4 just means it is a 4G drive. Nope, Barracuda-4 > > - They're quiet. (Yes, sports fans, quiet) > > - They don't get very hot. (One busy unit packed in a small, > > convection-cooled case with its power supply was finger-touch > > warm - all of the drives had been running for over a week) > > That was _not_ a Barracuda drive, unless seagate did some major > changes and didn't change the model name/number. Please give exact > details as to seagate model number. I suspect you where looking at > a Hawk drive (ST15230N), which do match the above description. Unless Seagate have started sticking inch-round labels with a picture of a fish and "BARRACUDA" over the spindle motors on their Hawk family disks, I suspect that it was, indeed, a Barracuda; my notes say ST15150N (Two PWA's in the old CDC style, spindle motor sticking through one of them, SCSI connector & jumpers on the other at right angles against the back of the chamber) (Yes, I'd expect to be able to tell the difference 8) Thinking about it, it's possible that there was a _really_small_ fan in the case (it's basically the width of a 3.5" drive, twice as high as a 1/2-height drive like the 'cuda, and with about 2" of clearance behind the drive. The back panel's covered with SCSI connectors, power connector, switch & ID selector, hence my assumption that it's convection cooled. If there's a fan, it'd have to be hidden under the drive forward of the power supply... Either way, in comparison to the earlier 'cudas I've met, it was cool and quiet, which was really all I had to say 8) > Rod Grimes rgrimes@gndrsh.aac.dev.com -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Mon Aug 28 19:29:07 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id TAA06540 for hardware-outgoing; Mon, 28 Aug 1995 19:29:07 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id TAA06532 for ; Mon, 28 Aug 1995 19:29:01 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id LAA24532; Tue, 29 Aug 1995 11:55:39 +0930 From: Michael Smith Message-Id: <199508290225.LAA24532@genesis.atrad.adelaide.edu.au> Subject: Re: Upgrade to my machine To: vince@penzance.econ.yale.edu (-Vince-) Date: Tue, 29 Aug 1995 11:55:39 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: from "-Vince-" at Aug 28, 95 05:40:13 am Content-Type: text Content-Length: 1751 Sender: hardware-owner@freebsd.org Precedence: bulk -Vince- stands accused of saying: >> Depends on what you're trying to do with it, and what sort of FreeBSD >> box you're talking about. A VLB '486DX2/66 will just about keep even >> with a Sparc 2 otherwise similarly configured; I don't have any data >> on faster Sun systems as just about everything around here is an Alpha. > > I meant like use it as a server for a organization on a FreeBSD P5-90 > box, what would it be equivelent to in performance? Is the Alpha that > much better than a Pentium? Depends on what you're doing, as far as I can tell. I'll have a clearer picture when the new box (P100) comes in, and I can get some comparative numbers against the low-end (-66) alphas around here. > I thought the working size will always exceed the physical memory > since you can only expand to a certain number for physical ram while with > hd's, you can just add new ones... Huh? The "certain number" for physical ram essentially limits your maximum working set size; obviously you have a limit related to processor speed as well. (after all, you can only access so much memory in a given period). Once your working set (the pages currently in use by active processes) exceeds the available physical memory, you start thrashing, as you have to swapout/swapin on a regular basis. > -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Tue Aug 29 00:57:21 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id AAA16740 for hardware-outgoing; Tue, 29 Aug 1995 00:57:21 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id AAA16729 for ; Tue, 29 Aug 1995 00:57:19 -0700 Date: Tue, 29 Aug 1995 03:56:49 -0400 (EDT) From: -Vince- To: Michael Smith cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <199508290225.LAA24532@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Tue, 29 Aug 1995, Michael Smith wrote: > -Vince- stands accused of saying: > >> Depends on what you're trying to do with it, and what sort of FreeBSD > >> box you're talking about. A VLB '486DX2/66 will just about keep even > >> with a Sparc 2 otherwise similarly configured; I don't have any data > >> on faster Sun systems as just about everything around here is an Alpha. > > > > I meant like use it as a server for a organization on a FreeBSD P5-90 > > box, what would it be equivelent to in performance? Is the Alpha that > > much better than a Pentium? > > Depends on what you're doing, as far as I can tell. I'll have a clearer > picture when the new box (P100) comes in, and I can get some comparative > numbers against the low-end (-66) alphas around here. Hmmm, okay... I meant a Alpha 275Mhz compared to a P5-90... > > I thought the working size will always exceed the physical memory > > since you can only expand to a certain number for physical ram while with > > hd's, you can just add new ones... > > Huh? The "certain number" for physical ram essentially limits your > maximum working set size; obviously you have a limit related to > processor speed as well. (after all, you can only access so much memory in a > given period). Once your working set (the pages currently in use by active > processes) exceeds the available physical memory, you start thrashing, as > you have to swapout/swapin on a regular basis. I know what you mean but memory is still limited to 256 megs or less so there is no way you can have 1 gig of physical ram I think... Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Tue Aug 29 01:05:33 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id BAA17122 for hardware-outgoing; Tue, 29 Aug 1995 01:05:33 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id BAA17114 for ; Tue, 29 Aug 1995 01:05:29 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id RAA25296; Tue, 29 Aug 1995 17:37:39 +0930 From: Michael Smith Message-Id: <199508290807.RAA25296@genesis.atrad.adelaide.edu.au> Subject: Re: Upgrade to my machine To: vince@penzance.econ.yale.edu (-Vince-) Date: Tue, 29 Aug 1995 17:37:38 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: from "-Vince-" at Aug 29, 95 03:56:49 am Content-Type: text Content-Length: 1329 Sender: hardware-owner@freebsd.org Precedence: bulk -Vince- stands accused of saying: > > Depends on what you're doing, as far as I can tell. I'll have a clearer > > picture when the new box (P100) comes in, and I can get some comparative > > numbers against the low-end (-66) alphas around here. > > Hmmm, okay... I meant a Alpha 275Mhz compared to a P5-90... There are a few of those around here too, but it's kinda hard to get realistic number off them, as they're rather busy most of the time. > I know what you mean but memory is still limited to 256 megs or > less so there is no way you can have 1 gig of physical ram I think... On what? No reason why you can't have several GB of physical memory if you happen to want it. There may not be any Intel PCI chipsets that support it (yet), but there's no hard law-of-physics limit that applies there; there are certainly plenty of GB+ memory machines kicking around. > -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Tue Aug 29 01:39:08 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id BAA18191 for hardware-outgoing; Tue, 29 Aug 1995 01:39:08 -0700 Received: from tango.rahul.net (tango.rahul.net [192.160.13.5]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id BAA18183 ; Tue, 29 Aug 1995 01:39:04 -0700 Received: from bolero.rahul.net by tango.rahul.net with SMTP id AA20130 (5.67b8/IDA-1.5); Tue, 29 Aug 1995 01:39:02 -0700 Received: from RockyMountain.rahul.net by bolero.rahul.net with SMTP id AA15622 (5.67b8/IDA-1.5); Tue, 29 Aug 1995 01:39:00 -0700 Received: by RockyMountain.rahul.net id AA11284 (5.67b/IDA-1.5); Tue, 29 Aug 1995 01:38:56 -0700 Date: Tue, 29 Aug 1995 01:38:56 -0700 From: Pete Delaney Message-Id: <199508290838.AA11284@RockyMountain.rahul.net> To: freebsd-platforms@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: FreeBSD CD-ROM 2.0.5 - Any SPARC Porting Underway? Cc: pete@RockyMountain.rahul.net Sender: hardware-owner@FreeBSD.ORG Precedence: bulk Forwarded: ... Re-Subject: Re: FreeBSD CD-ROM 2.0.5 - Any SPARC Porting Underway? Re-Cc: freebsd-platforms, freebsd-hardware, port-sparc@NetBSD.ORG, tech-ports@NetBSD.ORG, netbsd-ports@NetBSD.ORG > For sparc, you want FreeBSD's first cousin NetBSD.. > see www.netbsd.org.. I briefly checked it out. It feels dissapointing to see the Free/Net BSD UNIX community splintered. > they have more platforms but we have a better install and cover PC > hardware better.. Sounds like the old X11R6 vs XFree disintegration. I see no reason why they should be different source trees. > you chose which you want.. I saw AT&T break up UNIX into a lot of individual releases and thought it was stupid. Same for X on the PC and everything else. This doesn't smell any different. -pete From owner-freebsd-hardware Tue Aug 29 01:48:15 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id BAA18677 for hardware-outgoing; Tue, 29 Aug 1995 01:48:15 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id BAA18670 for ; Tue, 29 Aug 1995 01:48:12 -0700 Date: Tue, 29 Aug 1995 04:47:19 -0400 (EDT) From: -Vince- To: Michael Smith cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <199508290807.RAA25296@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Tue, 29 Aug 1995, Michael Smith wrote: > -Vince- stands accused of saying: > > > Depends on what you're doing, as far as I can tell. I'll have a clearer > > > picture when the new box (P100) comes in, and I can get some comparative > > > numbers against the low-end (-66) alphas around here. > > > > Hmmm, okay... I meant a Alpha 275Mhz compared to a P5-90... > > There are a few of those around here too, but it's kinda hard to get > realistic number off them, as they're rather busy most of the time. Hmmm, can a FreeBSD machine handle as big of a load as the Alpha? > > I know what you mean but memory is still limited to 256 megs or > > less so there is no way you can have 1 gig of physical ram I think... > > On what? No reason why you can't have several GB of physical memory > if you happen to want it. There may not be any Intel PCI chipsets > that support it (yet), but there's no hard law-of-physics limit > that applies there; there are certainly plenty of GB+ memory > machines kicking around. I mean on Intel PCI Chipsets since even ftp.cdrom.com only has 128 megs of RAM and you need to use swap somehow even on servers since I haven't really seen anyone with a server with more the 256 megs of memory yet... Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Tue Aug 29 01:57:43 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id BAA19137 for hardware-outgoing; Tue, 29 Aug 1995 01:57:43 -0700 Received: from tango.rahul.net (tango.rahul.net [192.160.13.5]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id BAA19128 ; Tue, 29 Aug 1995 01:57:41 -0700 Received: from bolero.rahul.net by tango.rahul.net with SMTP id AA20513 (5.67b8/IDA-1.5); Tue, 29 Aug 1995 01:57:27 -0700 Received: from RockyMountain.rahul.net by bolero.rahul.net with SMTP id AA17052 (5.67b8/IDA-1.5); Tue, 29 Aug 1995 01:57:23 -0700 Received: by RockyMountain.rahul.net id AA11308 (5.67b/IDA-1.5); Tue, 29 Aug 1995 01:56:45 -0700 Date: Tue, 29 Aug 1995 01:56:45 -0700 From: Pete Delaney Message-Id: <199508290856.AA11308@RockyMountain.rahul.net> To: davidg@root.com Subject: Re: FreeBSD CD-ROM 2.0.5 - Any SPARC Porting Underway? Cc: hackers@FreeBSD.ORG, port-sparc@netbsd.org, tech-ports@netbsd.org, netbsd-ports@netbsd.org, davem@caip.rutgers.edu, freebsd-platforms@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Sender: hardware-owner@FreeBSD.ORG Precedence: bulk Hi Dave: > > >Anyone know if 'anyone' has raised any Legal issues with BSDI or Chris Torek > >on his SPARC Stuff in the BSD release. I was currious why it wasn't on > >the CD. With Dave Miller making big time progress on SPARC on Linux and > >Chris Torek's stuff I would think about three man months from now we could > >have FreeBSD running on the sun4c. > > That's correct. In fact, someone has already done a port of the Sparc code > to an earlier version of FreeBSD...but it would need a LOT of work to port the > port to -current. I don't suppose you know: 1. Who did it? 2. Where a copy of it is? 3. Why it wasn't integrated into the release? 4. What the problems are? 5. If some of the NetBSD guys would like to help. > I will get to it eventually (I have a Sparc sitting next to > me), but at the moment I'm too busy with other things in FreeBSD (namely, the > 2.1 release). Since the previous version didn't make it into the release, I wonder if have this be more than a one man show might help increase the odds. If we could get a few guys working on this I would think the odds greatly improved. I've got plenty of sparkstations that I can use here at RockyMountain. What have you been plaining to do? > > -DG -pete From owner-freebsd-hardware Tue Aug 29 16:00:55 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id QAA27642 for hardware-outgoing; Tue, 29 Aug 1995 16:00:55 -0700 Received: from veenet.value.net (value.net [204.188.125.5]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id QAA27628 for ; Tue, 29 Aug 1995 16:00:48 -0700 Received: (from unibrow@localhost) by veenet.value.net (8.6.9/8.6.9) id QAA05247 for freebsd-hardware@freebsd.org; Tue, 29 Aug 1995 16:13:34 -0700 From: Don Littlefield Message-Id: <199508292313.QAA05247@veenet.value.net> Subject: IDE patches... To: freebsd-hardware@freebsd.org Date: Tue, 29 Aug 1995 16:13:34 -0700 (PDT) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 369 Sender: hardware-owner@freebsd.org Precedence: bulk I am running 2.0.5-Release. The patches for the ATAPI IDE stuff do not seem to work with my version either. Basically when running patch on the conf.c I get 2 out of three hunks failing. Is there somewhere I can go to just pick up the newer conf.c file and use it? It seems to me that once I get that file working I will be on the road to success.. Thanks, Don From owner-freebsd-hardware Tue Aug 29 16:10:26 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id QAA28215 for hardware-outgoing; Tue, 29 Aug 1995 16:10:26 -0700 Received: from rs1.hrz.th-darmstadt.de (rs1.hrz.th-darmstadt.de [130.83.22.60]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id QAA28189 for ; Tue, 29 Aug 1995 16:10:17 -0700 Received: from [130.83.177.7] (ppp07.stud.th-darmstadt.de [130.83.177.7]) by rs1.hrz.th-darmstadt.de (8.6.10/8.6.4) with SMTP id BAA19756; Wed, 30 Aug 1995 01:09:59 +0200 X-Sender: (Unverified) Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 30 Aug 1995 01:10:08 +0200 To: -Vince- , freebsd-hardware@freebsd.org From: beckmann@powermac.stud.th-darmstadt.de (Michael Beckmann) Subject: Re: Upgrade to my machine Sender: hardware-owner@freebsd.org Precedence: bulk At 15:56 Uhr 18.8.1995, -Vince- wrote: > Hmmm okay but what drives can touch the barracuda's in terms of >speed? The IBM DFHS line of drives, absolutely. 7 ms access, 7200 rpm, MTBF 1.000.000 h. Available as both Fast - and Wide SCSI-II drives. I would always prefer these drives and not buy a Barracuda, since I have made very good experiences with IBM drives in both reliability and speed. From owner-freebsd-hardware Tue Aug 29 16:22:14 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id QAA28526 for hardware-outgoing; Tue, 29 Aug 1995 16:22:14 -0700 Received: from relay3.UU.NET (relay3.UU.NET [192.48.96.8]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id QAA28520 for ; Tue, 29 Aug 1995 16:22:11 -0700 Received: from uucp6.UU.NET by relay3.UU.NET with SMTP id QQzeyf09696; Tue, 29 Aug 1995 19:22:05 -0400 Received: from uanet.UUCP by uucp6.UU.NET with UUCP/RMAIL ; Tue, 29 Aug 1995 19:22:05 -0400 Received: by crocodil.monolit.kiev.ua (8.6.8.1/8.5) id CAA25554 for hardware@freebsd.org; Wed, 30 Aug 1995 02:19:09 +0300 Date: Wed, 30 Aug 1995 02:19:09 +0300 From: System daemons Message-Id: <199508292319.CAA25554@crocodil.monolit.kiev.ua> Apparently-To: hardware@freebsd.org Sender: hardware-owner@freebsd.org Precedence: bulk [ I've sent this already, but to wrong address... second try should be ok. ] Subject: SCSI CD Hitachi 1750S + PCI NCR cannot play sound (data ok) Date: 25 Aug 1995 03:33:32 +0300 Organization: FARM Computing Association Keywords: PCI NCR SCSI CD-ROM audio I can use the drive in data mode (installed my system from it), but when I run cdplay (or any X player, it doesn't matter), I get the following diagnostics: cd0(ncr0:2:0): ILLEGAL REQUEST asc:20,0 Invalid command operation code cd0(ncr0:2:0): phase change 2-3 10@1e9d8 resid=2. dmesg: FreeBSD 2.0.5-RELEASE #0: Mon Aug 14 13:52:02 EET DST 1995 dk@dog.farm.org:/usr/src/sys/compile/FARM CPU: i486DX (486-class CPU) real memory = 7995392 (1952 pages) avail memory = 6729728 (1643 pages) Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <8 virtual consoles, flags=0x0> ed0 at 0x300-0x31f irq 15 on isa ed0: address 00:40:33:2c:a6:13, type NE2000 (16 bit) bpf: ed0 attached sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16450 sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16450 sio2 at 0x3e8-0x3ef irq 5 on isa sio2: type 16450 sio3 not found at 0x2e8 lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface lpt1 at 0x278-0x27f on isa lpt2 not found at 0xffffffff pca0 on isa pca0: PC speaker audio driver fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 765 fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 516MB (1058400 sectors), 1050 cyls, 16 heads, 63 S/T, 512 B/S wdc1 not found at 0x170 npx0 on motherboard npx0: INT 16 interface Probing for devices on the pci0 bus: configuration mode 1 allows 32 devices. pci0:5: vendor=0x1039, device=0x496, class=bridge [not supported] vga0 rev 0 on pci0:11 ncr0 rev 2 int a irq 9 on pci0:13 reg20: virtual=0xf224b000 physical=0xf0800000 size=0x100 ncr0: restart (scsi reset). ncr0 scanning for targets 0..6 (V2 pl21 95/03/21) ncr0 waiting for scsi devices to settle (ncr0:2:0): "HITACHI CDR-1750S 0009" type 5 removable SCSI 1 cd0(ncr0:2:0): CD-ROM cd0(ncr0:2:0): UNIT ATTENTION asc:28,0 cd0(ncr0:2:0): Not ready to ready transition, medium may have changed cd present.[400000 x 2048 byte records] pci0: uses 8388864 bytes of memory from f0000000 upto f08000ff. pci0: uses 256 bytes of I/O space from 6000 upto 60ff. bpf: lo0 attached bpf: ppp0 attached bpf: sl0 attached cd0(ncr0:2:0): phase change 2-3 10@1e9d8 resid=2. cd0(ncr0:2:0): ILLEGAL REQUEST asc:20,0 Invalid command operation code cd0(ncr0:2:0): phase change 2-3 10@1e9d8 resid=2. cd0(ncr0:2:0): ILLEGAL REQUEST asc:20,0 Invalid command operation code cd0(ncr0:2:0): phase change 2-3 10@1e9d8 resid=2. [same for each request] From owner-freebsd-hardware Tue Aug 29 16:41:56 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id QAA29058 for hardware-outgoing; Tue, 29 Aug 1995 16:41:56 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id QAA29050 for ; Tue, 29 Aug 1995 16:41:44 -0700 Date: Tue, 29 Aug 1995 19:41:24 -0400 (EDT) From: -Vince- To: Michael Smith cc: hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <199508290901.SAA25428@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Tue, 29 Aug 1995, Michael Smith wrote: > -Vince- stands accused of saying: > >>> Hmmm, okay... I meant a Alpha 275Mhz compared to a P5-90... > >> > >> There are a few of those around here too, but it's kinda hard to get > >> realistic number off them, as they're rather busy most of the time. > > > > Hmmm, can a FreeBSD machine handle as big of a load as the Alpha? > > You really ought to stop and read what you just asked, and put the > question properly; perhaps : > > "What sort of Alpha will my P90 compare to?" You are right about that.. > The answer is, I dunno. Alpha boxes scale rather better, but that's > because they're designed as servers, not PC's. Hmmm, what about machines in terms like SUN's, HP's will the P90 compare to since the Alpha is a fast machine. > > I mean on Intel PCI Chipsets since even ftp.cdrom.com only has > > 128 megs of RAM and you need to use swap somehow even on servers since I > > haven't really seen anyone with a server with more the 256 megs of memory > > yet... > > That just means you haven't been around very long 8) I can still remember > what it used to feel like as a newbie talking about those sorts of numbers 8) I know some machines can do it but I mean the PC is still limited =) > Anyone who does big models of any sort uses huge amounts of memory, > as Rod already observed. That's true but who would actually need a gig of ram? Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Tue Aug 29 17:38:01 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id RAA00595 for hardware-outgoing; Tue, 29 Aug 1995 17:38:01 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id RAA00589 for ; Tue, 29 Aug 1995 17:37:56 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id RAA04867; Tue, 29 Aug 1995 17:36:58 -0700 From: "Rodney W. Grimes" Message-Id: <199508300036.RAA04867@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: vince@penzance.econ.yale.edu (-Vince-) Date: Tue, 29 Aug 1995 17:36:58 -0700 (PDT) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: from "-Vince-" at Aug 29, 95 04:47:19 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2828 Sender: hardware-owner@freebsd.org Precedence: bulk > > On Tue, 29 Aug 1995, Michael Smith wrote: > > > -Vince- stands accused of saying: > > > > Depends on what you're doing, as far as I can tell. I'll have a clearer > > > > picture when the new box (P100) comes in, and I can get some comparative > > > > numbers against the low-end (-66) alphas around here. > > > > > > Hmmm, okay... I meant a Alpha 275Mhz compared to a P5-90... > > > > There are a few of those around here too, but it's kinda hard to get > > realistic number off them, as they're rather busy most of the time. > > Hmmm, can a FreeBSD machine handle as big of a load as the Alpha? You seem to ask a lot of these types of questions, perhaps you should write 1 email message asking 20 of them, instead of creating these one and two at a time iterative question cycles :-) > > > I know what you mean but memory is still limited to 256 megs or > > > less so there is no way you can have 1 gig of physical ram I think... > > > > On what? No reason why you can't have several GB of physical memory > > if you happen to want it. There may not be any Intel PCI chipsets > > that support it (yet), but there's no hard law-of-physics limit > > that applies there; there are certainly plenty of GB+ memory > > machines kicking around. > > I mean on Intel PCI Chipsets since even ftp.cdrom.com only has > 128 megs of RAM and you need to use swap somehow even on servers since I > haven't really seen anyone with a server with more the 256 megs of memory > yet... Contact the Severs business unit of Intel Corporation, they can sell you Pentium based SMP boxes that can have up to 1G of physical memory, these are known as ``Extended Express'' series servers. Similiar products are avaliable from ALR, AST, NCR and a few others. These are very expensive back room types of Pentium servers sporting things like ECC memory, set associative write back caches per CPU chip, vendor specific CPU/Memory backplanes with PCI/EISA peripheral buses. I believe a recent Unix Review had an article on the ALR box, it is one of the more cost effective server class machines around right now. The chipset used in the Extended Express products from Intel is OEM'ed to at least 4 companies (whom, due to NDA's I can not mention) that use them to build servers, though I have never seen these products on the open market. The chipset itself can support 4G of physical address space, the 1G limit has been a memory density vs number of slots for memory cards tradeoff done by system designers. I would suspect that with next years memory density increase these boxes will sport 4G memory capacities, but then, we will also have P6 next year in volume :-). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Tue Aug 29 18:14:09 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id SAA02001 for hardware-outgoing; Tue, 29 Aug 1995 18:14:09 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id SAA01994 for ; Tue, 29 Aug 1995 18:13:52 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id KAA27237; Wed, 30 Aug 1995 10:46:34 +0930 From: Michael Smith Message-Id: <199508300116.KAA27237@genesis.atrad.adelaide.edu.au> Subject: Re: Upgrade to my machine To: vince@penzance.econ.yale.edu (-Vince-) Date: Wed, 30 Aug 1995 10:46:33 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: from "-Vince-" at Aug 29, 95 07:41:24 pm Content-Type: text Content-Length: 1649 Sender: hardware-owner@freebsd.org Precedence: bulk -Vince- stands accused of saying: > Hmmm, what about machines in terms like SUN's, HP's will the P90 > compare to since the Alpha is a fast machine. Depends lots on what you're doing with them; in a straight line, the P90 is pretty quick, but what you put around it largely determines how it will perform in an applications context. (Especially memory/cache/disk) >> Anyone who does big models of any sort uses huge amounts of memory, >> as Rod already observed. > > That's true but who would actually need a gig of ram? In one of my looking-for-work periods, I did some time as a CAD operator for the local telephone company, feeding all of their cabling details into a mixed oracle/GDS database. We were using Decstations (mostly /133's) with 64M of core and ~500M of swap. They used to thrash horribly whenever you moved around the suburb, and often ran out of swap. The database admin machine had 128M of core and several GB of swap (for working in larger areas, obviously). Ahh, fond memories... whenever you wanted a 5 min. break, a surreptitions ^\ in the GDS window kept your supervisor scurrying 8) (And none of them ever worked it out ... ) Anyway, this is rather off-topic 8) > -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Tue Aug 29 19:46:57 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id TAA04576 for hardware-outgoing; Tue, 29 Aug 1995 19:46:57 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id TAA04561 for ; Tue, 29 Aug 1995 19:46:48 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id TAA05665; Tue, 29 Aug 1995 19:44:59 -0700 From: "Rodney W. Grimes" Message-Id: <199508300244.TAA05665@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: beckmann@powermac.stud.th-darmstadt.de (Michael Beckmann) Date: Tue, 29 Aug 1995 19:44:58 -0700 (PDT) Cc: vince@penzance.econ.yale.edu, freebsd-hardware@freebsd.org In-Reply-To: from "Michael Beckmann" at Aug 30, 95 01:10:08 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1567 Sender: hardware-owner@freebsd.org Precedence: bulk > > At 15:56 Uhr 18.8.1995, -Vince- wrote: > > > Hmmm okay but what drives can touch the barracuda's in terms of > >speed? > > The IBM DFHS line of drives, absolutely. 7 ms access, 7200 rpm, MTBF > 1.000.000 h. Available as both Fast - and Wide SCSI-II drives. I have nothing against IBM drives, but I do wish to point out a little known fact about MTBF numbers, they are reported as proper physics type of signifcant digit numbers, thus, 1,000,000 means somewhere between 500,000.0 and 1,500,000.0. IE, only the first digit in that number is signficant since it has no trailing decimal point. These are based upon calculations that where written many many years ago, and given light of technological changes once they go much above about 500,000 are pretty meaningless data until you can collect some reasonable large sample of FIT rates. Basically is what I am trying to say, is treat all things with MTBF numbers >500,000 hours as if they will last there technilogical life of 3 to 5 years and it does not really matter if it is 600K or 2M. > I would always prefer these drives and not buy a Barracuda, since I have > made very good experiences with IBM drives in both reliability and speed. I will agree with that, with one caveat, IBM also makes some of the most incompatible SCSI drives on the market. Drives that claim to be SCSI-II but can't do a device inquire is one example (0664). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Tue Aug 29 20:18:29 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id UAA05704 for hardware-outgoing; Tue, 29 Aug 1995 20:18:29 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id UAA05687 for ; Tue, 29 Aug 1995 20:17:33 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id MAA27585; Wed, 30 Aug 1995 12:48:44 +0930 From: Michael Smith Message-Id: <199508300318.MAA27585@genesis.atrad.adelaide.edu.au> Subject: Re: S.O.S -2.1Stable and ASUSP54TP4 To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Wed, 30 Aug 1995 12:48:44 +0930 (CST) Cc: hardware@freebsd.org In-Reply-To: <199508300302.UAA05696@gndrsh.aac.dev.com> from "Rodney W. Grimes" at Aug 29, 95 08:02:01 pm Content-Type: text Content-Length: 3005 Sender: hardware-owner@freebsd.org Precedence: bulk Rodney W. Grimes stands accused of saying: > I would have agreed with this 3 years ago, but I suggest you go study > some of the current technology in use. I'd love to - you try getting that sort of data on this side of the planet 8) > First alpha particle disturbance in DRAM is gone, it was pretty much killed > 4 years ago with the advent of certain epoxy materials used to coat the > die with before plastic encapsolation. At todays DRAM scales a single Neat - I knew they were using plastic because they could control the isotope content more closely, I wasn't aware they'd gone to those lengths. > Static memories are suspetable to alpha particule disturbance, it just > takes a heck of a lot more to do it, and given ceramic is out of the > picture it won't occur anyway. In a cmos static memory you have to have > enough disturbance to perturb the gate voltage of one side of the latch > to cause a bit flip, about 10 micro rinkens will do it, but it usually > sends the device into latchup at the same time :-). Exactly; and to get that sort of flux, unless you're direct-beaming the device, you're going to poach anyone standing nearby 8) > The more prevelent cause today in single bit errors in both DRAM and SRAM > is cause by VCC noise and or ION contamination caused by moisture > absorbition into plastic packages before surface mount vapor phase soldering. Is that what the bother about mositure absorption is? I thought it had to do with sweating during soldering - we get lots of SMD parts coming through in moisture-proof packaging, but never got an answer out of the manufacturers as to why... > Current FIT per bit are in the 0.0002 to 0.0004 range, that is measure in > billions of power on hours. Today MTBF in a 2MB x 16 bit DRAM subsystem > is 30 to 35 years... I'd say I can live with that given that my disk > is going to go belly up in 57 years anyway :-) :-) :-) *chuckle* You're not taking vibration into account there - I'd love to see some numbers on accelerated failure due to bond wire stress from fan and disk spindle vibration Especially given the MTBBF (mean time between bearing failure 8) for modern taiwanese CPU fans is about, oh, six weeks ... > Sources of information for this and more details are in many vendors > data books, Mircon's 1995 has a few good tech notes about it, as does > Intels memory products books. Almost impossible to get over here 8( The agents for either of those two want to talk MOQ 1000+ before they'll even think about giving you data. > Rod Grimes rgrimes@gndrsh.aac.dev.com (note the new cc:) -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Tue Aug 29 20:22:19 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id UAA05835 for hardware-outgoing; Tue, 29 Aug 1995 20:22:19 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id UAA05829 for ; Tue, 29 Aug 1995 20:22:14 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id UAA05719; Tue, 29 Aug 1995 20:15:39 -0700 From: "Rodney W. Grimes" Message-Id: <199508300315.UAA05719@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: msmith@atrad.adelaide.edu.au (Michael Smith) Date: Tue, 29 Aug 1995 20:15:39 -0700 (PDT) Cc: vince@penzance.econ.yale.edu, msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: <199508300116.KAA27237@genesis.atrad.adelaide.edu.au> from "Michael Smith" at Aug 30, 95 10:46:33 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1178 Sender: hardware-owner@freebsd.org Precedence: bulk > > -Vince- stands accused of saying: > > Hmmm, what about machines in terms like SUN's, HP's will the P90 > > compare to since the Alpha is a fast machine. > > Depends lots on what you're doing with them; in a straight line, the P90 > is pretty quick, but what you put around it largely determines how it will > perform in an applications context. (Especially memory/cache/disk) > > >> Anyone who does big models of any sort uses huge amounts of memory, > >> as Rod already observed. > > > > That's true but who would actually need a gig of ram? THINK for a minute about large applications. An Intel Pentium 90/100 CPU chip as 3.3 billon transistors on it. Each cmos transitor takes at least 6 rectangles to represent the minimal transitor data and 3 contacts to hook it up, now thats 19.8G assumming I can stuff a rectange into a byte :-). We haven't even started to talk about interconnecting these 3.3 billon transistors... Can you say that a gigabyte in this world is actually a very small amount of data! -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Tue Aug 29 20:57:42 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id UAA07302 for hardware-outgoing; Tue, 29 Aug 1995 20:57:42 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id UAA07296 for ; Tue, 29 Aug 1995 20:57:37 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id UAA05821; Tue, 29 Aug 1995 20:57:00 -0700 From: "Rodney W. Grimes" Message-Id: <199508300357.UAA05821@gndrsh.aac.dev.com> Subject: Re: S.O.S -2.1Stable and ASUSP54TP4 To: msmith@atrad.adelaide.edu.au (Michael Smith) Date: Tue, 29 Aug 1995 20:57:00 -0700 (PDT) Cc: hardware@freebsd.org In-Reply-To: <199508300318.MAA27585@genesis.atrad.adelaide.edu.au> from "Michael Smith" at Aug 30, 95 12:48:44 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 4647 Sender: hardware-owner@freebsd.org Precedence: bulk > > Rodney W. Grimes stands accused of saying: > > I would have agreed with this 3 years ago, but I suggest you go study > > some of the current technology in use. > > I'd love to - you try getting that sort of data on this side of the planet 8) :-(. > > First alpha particle disturbance in DRAM is gone, it was pretty much killed > > 4 years ago with the advent of certain epoxy materials used to coat the > > die with before plastic encapsolation. At todays DRAM scales a single > > Neat - I knew they were using plastic because they could control the > isotope content more closely, I wasn't aware they'd gone to those > lengths. Once the charge on a cell dropped to about 2 alpha particles it become rather important to solve. > > Static memories are suspetable to alpha particule disturbance, it just > > takes a heck of a lot more to do it, and given ceramic is out of the > > picture it won't occur anyway. In a cmos static memory you have to have > > enough disturbance to perturb the gate voltage of one side of the latch > > to cause a bit flip, about 10 micro rinkens will do it, but it usually > > sends the device into latchup at the same time :-). > > Exactly; and to get that sort of flux, unless you're direct-beaming the > device, you're going to poach anyone standing nearby 8) Nah, just acid etch the plastic down to the die cavity and point a good UV light at it similulate similiar failure modes due to photon disturbances. > > The more prevelent cause today in single bit errors in both DRAM and SRAM > > is cause by VCC noise and or ION contamination caused by moisture > > absorbition into plastic packages before surface mount vapor phase soldering. > > Is that what the bother about mositure absorption is? I thought it had to > do with sweating during soldering - we get lots of SMD parts coming through > in moisture-proof packaging, but never got an answer out of the > manufacturers as to why... Yes, that is why you see all the mosture proofing. The problem is if the plastic obsorbos mosture prior to soldering, that mosture boils and expands quite rapidly when it hits the vapor phase, this causes micro cracks in the plastic, that then allow mosture to easily pentrate the package and cause ion contamination, which causes corrossion, which is very nasty/ugly for the chip. Answer can be obtained from any good pacakgeing engineer :-). And a few papers on the subject. > > Current FIT per bit are in the 0.0002 to 0.0004 range, that is measure in > > billions of power on hours. Today MTBF in a 2MB x 16 bit DRAM subsystem > > is 30 to 35 years... I'd say I can live with that given that my disk > > is going to go belly up in 57 years anyway :-) :-) :-) > > *chuckle* You're not taking vibration into account there - I'd love to > see some numbers on accelerated failure due to bond wire stress from > fan and disk spindle vibration Especially given the MTBBF > (mean time between bearing failure 8) for modern taiwanese CPU fans is > about, oh, six weeks ... Your back in the 80's again, bond wire stress does not exist in plastic packages, the bond wires are either a) totally gone due to TAB bonding between the die and the lead frame, or b) supported by the plastic used in encapsolation (ie, there is _no_ die cavity to speak of any more). That does not hold true of PGA or other cavity type packageing, but I haven't seen any of that stuff on disk drives lately! By American Enhance power supplies, the are made in Taiwan, but they use a ball bearing fan, and American Enhance long ago learned all about fan failures and how to fix them :-). Remeber, those low dollar clone manufactures are targeting the home PC market, where the MTPO (Mean Time Powered On) is only about 3 weeks a year :-). > > Sources of information for this and more details are in many vendors > > data books, Mircon's 1995 has a few good tech notes about it, as does > > Intels memory products books. > > Almost impossible to get over here 8( The agents for either of those two > want to talk MOQ 1000+ before they'll even think about giving you data. Hummm.. never had that problem here, simply explain that I am a contracting consultant by trade, drop a few client names, and boom.. I have all the data books I want. But then, these are all true things and I have a very lagitimate reason for needing there data books. > > > Rod Grimes rgrimes@gndrsh.aac.dev.com > > (note the new cc:) :-) Thanks... -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Tue Aug 29 21:27:42 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id VAA08723 for hardware-outgoing; Tue, 29 Aug 1995 21:27:42 -0700 Received: from oasis.txdirect.net (oasis.txdirect.net [204.57.120.34]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id VAA08712 for ; Tue, 29 Aug 1995 21:27:39 -0700 Received: (from rsnow@localhost) by oasis.txdirect.net (8.6.11/8.6.9) id XAA01968; Tue, 29 Aug 1995 23:27:26 -0500 Date: Tue, 29 Aug 1995 23:27:05 -0500 (CDT) From: Rob Snow X-Sender: rsnow@oasis To: "Rodney W. Grimes" cc: Michael Smith , vince@penzance.econ.yale.edu, hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <199508300315.UAA05719@gndrsh.aac.dev.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Tue, 29 Aug 1995, Rodney W. Grimes wrote: > > > > -Vince- stands accused of saying: > > > Hmmm, what about machines in terms like SUN's, HP's will the P90 > > > compare to since the Alpha is a fast machine. > > > > Depends lots on what you're doing with them; in a straight line, the P90 > > is pretty quick, but what you put around it largely determines how it will > > perform in an applications context. (Especially memory/cache/disk) > > > > >> Anyone who does big models of any sort uses huge amounts of memory, > > >> as Rod already observed. > > > > > > That's true but who would actually need a gig of ram? > > THINK for a minute about large applications. An Intel Pentium 90/100 CPU > chip as 3.3 billon transistors on it. Each cmos transitor takes at least ^^^^^^^^^^ > 6 rectangles to represent the minimal transitor data and 3 contacts to hook > it up, now thats 19.8G assumming I can stuff a rectange into a byte :-). > > We haven't even started to talk about interconnecting these 3.3 billon ^^^^^^^^^^ > transistors... > > Can you say that a gigabyte in this world is actually a very small amount > of data! > -- > Rod Grimes rgrimes@gndrsh.aac.dev.com > Accurate Automation Company Reliable computers for FreeBSD Isn't that [M]illion? --- Rob Snow rsnow@txdirect.net From owner-freebsd-hardware Tue Aug 29 21:55:29 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id VAA09769 for hardware-outgoing; Tue, 29 Aug 1995 21:55:29 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id VAA09762 for ; Tue, 29 Aug 1995 21:55:11 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id VAA05887; Tue, 29 Aug 1995 21:49:20 -0700 From: "Rodney W. Grimes" Message-Id: <199508300449.VAA05887@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: rsnow@txdirect.net (Rob Snow) Date: Tue, 29 Aug 1995 21:49:20 -0700 (PDT) Cc: msmith@atrad.adelaide.edu.au, vince@penzance.econ.yale.edu, hardware@freebsd.org In-Reply-To: from "Rob Snow" at Aug 29, 95 11:27:05 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2120 Sender: hardware-owner@freebsd.org Precedence: bulk > > On Tue, 29 Aug 1995, Rodney W. Grimes wrote: > > > > > > > -Vince- stands accused of saying: > > > > Hmmm, what about machines in terms like SUN's, HP's will the P90 > > > > compare to since the Alpha is a fast machine. > > > > > > Depends lots on what you're doing with them; in a straight line, the P90 > > > is pretty quick, but what you put around it largely determines how it will > > > perform in an applications context. (Especially memory/cache/disk) > > > > > > >> Anyone who does big models of any sort uses huge amounts of memory, > > > >> as Rod already observed. > > > > > > > > That's true but who would actually need a gig of ram? > > > > THINK for a minute about large applications. An Intel Pentium 90/100 CPU > > chip as 3.3 billon transistors on it. Each cmos transitor takes at least > ^^^^^^^^^^ > > 6 rectangles to represent the minimal transitor data and 3 contacts to hook > > it up, now thats 19.8G assumming I can stuff a rectange into a byte :-). > > > > We haven't even started to talk about interconnecting these 3.3 billon > ^^^^^^^^^^ > > transistors... > > > > Can you say that a gigabyte in this world is actually a very small amount > > of data! > > -- > > Rod Grimes rgrimes@gndrsh.aac.dev.com > > Accurate Automation Company Reliable computers for FreeBSD > > Isn't that [M]illion? Get a grip, it takes a million transistors just to implement the 16k of cache: 16384 bytes * 8bits/byte * 6 transistors a bit == 786432, add the decoders and other gunk and your over a million right there.... And from the 1994 i486DX2 data sheet ``Over one million transistors implement this RISC integer core'' talking about just the integer ALU and registers here.... and that was the 486 core ALU, the P54 ALU is an order or two in magnitude more complex due to being super scaler and longer pipe. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Tue Aug 29 22:11:51 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id WAA10235 for hardware-outgoing; Tue, 29 Aug 1995 22:11:51 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id WAA10211 for ; Tue, 29 Aug 1995 22:11:11 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id OAA27796; Wed, 30 Aug 1995 14:43:53 +0930 From: Michael Smith Message-Id: <199508300513.OAA27796@genesis.atrad.adelaide.edu.au> Subject: Re: S.O.S -2.1Stable and ASUSP54TP4 To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Wed, 30 Aug 1995 14:43:52 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org In-Reply-To: <199508300357.UAA05821@gndrsh.aac.dev.com> from "Rodney W. Grimes" at Aug 29, 95 08:57:00 pm Content-Type: text Content-Length: 2248 Sender: hardware-owner@freebsd.org Precedence: bulk Rodney W. Grimes stands accused of saying: > Yes, that is why you see all the mosture proofing. The problem is if the > plastic obsorbos mosture prior to soldering, that mosture boils and expands > quite rapidly when it hits the vapor phase, this causes micro cracks in the > plastic, that then allow mosture to easily pentrate the package and cause > ion contamination, which causes corrossion, which is very nasty/ugly for > the chip. Neat; means I can stop worrying about it, as we're small enough to do everything by hand 8) > Your back in the 80's again, bond wire stress does not exist in plastic > packages, the bond wires are either a) totally gone due to TAB bonding > between the die and the lead frame, or b) supported by the plastic used > in encapsolation (ie, there is _no_ die cavity to speak of any more). > That does not hold true of PGA or other cavity type packageing, but I > haven't seen any of that stuff on disk drives lately! I was actually thinking of CPUs at that point, and the extremely scungy fans that get mounted on them. (And the number of cases that seem to resonate with one or more of the spinning items inside) > By American Enhance power supplies, the are made in Taiwan, but they use > a ball bearing fan, and American Enhance long ago learned all about fan > failures and how to fix them :-). Remeber, those low dollar clone manufactures > are targeting the home PC market, where the MTPO (Mean Time Powered On) is > only about 3 weeks a year :-). Hah! Buy a Name Brand in australia? You gotta be kidding 8( Seriously, I've considered the relative possibilities in running a version of AA on my spare time here, but a) sourcing good parts is close to impossible, and b) the only 'right price' around here for PC parts is the lowest one 8( > Rod Grimes rgrimes@gndrsh.aac.dev.com -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Wed Aug 30 00:41:17 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id AAA16233 for hardware-outgoing; Wed, 30 Aug 1995 00:41:17 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id AAA16225 for ; Wed, 30 Aug 1995 00:41:01 -0700 Date: Wed, 30 Aug 1995 03:40:46 -0400 (EDT) From: -Vince- To: Michael Beckmann cc: freebsd-hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Wed, 30 Aug 1995, Michael Beckmann wrote: > At 15:56 Uhr 18.8.1995, -Vince- wrote: > > > Hmmm okay but what drives can touch the barracuda's in terms of > >speed? > > The IBM DFHS line of drives, absolutely. 7 ms access, 7200 rpm, MTBF > 1.000.000 h. Available as both Fast - and Wide SCSI-II drives. > > I would always prefer these drives and not buy a Barracuda, since I have > made very good experiences with IBM drives in both reliability and speed. Are these DeskStar or FireStar lines of drives? Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Wed Aug 30 00:45:03 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id AAA16419 for hardware-outgoing; Wed, 30 Aug 1995 00:45:03 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id AAA16382 for ; Wed, 30 Aug 1995 00:44:42 -0700 Date: Wed, 30 Aug 1995 03:44:11 -0400 (EDT) From: -Vince- To: "Rodney W. Grimes" cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <199508300036.RAA04867@gndrsh.aac.dev.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Tue, 29 Aug 1995, Rodney W. Grimes wrote: > > > > On Tue, 29 Aug 1995, Michael Smith wrote: > > > > > -Vince- stands accused of saying: > > > > > Depends on what you're doing, as far as I can tell. I'll have a clearer > > > > > picture when the new box (P100) comes in, and I can get some comparative > > > > > numbers against the low-end (-66) alphas around here. > > > > > > > > Hmmm, okay... I meant a Alpha 275Mhz compared to a P5-90... > > > > > > There are a few of those around here too, but it's kinda hard to get > > > realistic number off them, as they're rather busy most of the time. > > > > Hmmm, can a FreeBSD machine handle as big of a load as the Alpha? > > You seem to ask a lot of these types of questions, perhaps you should > write 1 email message asking 20 of them, instead of creating these one > and two at a time iterative question cycles :-) You do have a point but it's when the reply comes that I think of something else to ask =) > > > > I know what you mean but memory is still limited to 256 megs or > > > > less so there is no way you can have 1 gig of physical ram I think... > > > > > > On what? No reason why you can't have several GB of physical memory > > > if you happen to want it. There may not be any Intel PCI chipsets > > > that support it (yet), but there's no hard law-of-physics limit > > > that applies there; there are certainly plenty of GB+ memory > > > machines kicking around. > > > > I mean on Intel PCI Chipsets since even ftp.cdrom.com only has > > 128 megs of RAM and you need to use swap somehow even on servers since I > > haven't really seen anyone with a server with more the 256 megs of memory > > yet... > > Contact the Severs business unit of Intel Corporation, they can sell you > Pentium based SMP boxes that can have up to 1G of physical memory, these > are known as ``Extended Express'' series servers. > > Similiar products are avaliable from ALR, AST, NCR and a few others. These > are very expensive back room types of Pentium servers sporting things like > ECC memory, set associative write back caches per CPU chip, vendor specific > CPU/Memory backplanes with PCI/EISA peripheral buses. > > I believe a recent Unix Review had an article on the ALR box, it is one > of the more cost effective server class machines around right now. > > The chipset used in the Extended Express products from Intel is OEM'ed to > at least 4 companies (whom, due to NDA's I can not mention) that use them > to build servers, though I have never seen these products on the open > market. The chipset itself can support 4G of physical address space, > the 1G limit has been a memory density vs number of slots for memory > cards tradeoff done by system designers. I would suspect that with next > years memory density increase these boxes will sport 4G memory capacities, > but then, we will also have P6 next year in volume :-). I guess so but it seems like even though a P5 can address up to 4GB of memory, they should make motherboards support that much memory so people can make their P5's power horses =) Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Wed Aug 30 00:50:53 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id AAA16907 for hardware-outgoing; Wed, 30 Aug 1995 00:50:53 -0700 Received: from penzance.econ.yale.edu (penzance.econ.yale.edu [130.132.32.100]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id AAA16897 for ; Wed, 30 Aug 1995 00:50:36 -0700 Date: Wed, 30 Aug 1995 03:48:36 -0400 (EDT) From: -Vince- To: Michael Smith cc: msmith@atrad.adelaide.edu.au, hardware@freebsd.org Subject: Re: Upgrade to my machine In-Reply-To: <199508300116.KAA27237@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Wed, 30 Aug 1995, Michael Smith wrote: > -Vince- stands accused of saying: > > Hmmm, what about machines in terms like SUN's, HP's will the P90 > > compare to since the Alpha is a fast machine. > > Depends lots on what you're doing with them; in a straight line, the P90 > is pretty quick, but what you put around it largely determines how it will > perform in an applications context. (Especially memory/cache/disk) Hmmm, I see what you mean. Somehow it seems a P590 can handle 20 users doing the same things as a SUN SparcStation 20 alot easier than the SUN can. > >> Anyone who does big models of any sort uses huge amounts of memory, > >> as Rod already observed. > > > > That's true but who would actually need a gig of ram? > > In one of my looking-for-work periods, I did some time as a CAD operator > for the local telephone company, feeding all of their cabling details > into a mixed oracle/GDS database. > > We were using Decstations (mostly /133's) with 64M of core and ~500M of > swap. They used to thrash horribly whenever you moved around the suburb, > and often ran out of swap. The database admin machine had 128M of core > and several GB of swap (for working in larger areas, obviously). > > Ahh, fond memories... whenever you wanted a 5 min. break, a surreptitions > ^\ in the GDS window kept your supervisor scurrying 8) (And none of them > ever worked it out ... ) > > Anyway, this is rather off-topic 8) Oh, I see what you mean now since CAD does require alot =) Cheers, -Vince- vince@kbrown.oldcampus.yale.edu - GUS Mailing Lists Admin UCLA Physics/Electrical Engineering - UC Berkeley Fall '95 SysAdmin bigbang.HIP.Berkeley.EDU - Running FreeBSD, Real UN*X for Free! Chabot Observatory & Science Center From owner-freebsd-hardware Wed Aug 30 06:25:03 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id GAA00729 for hardware-outgoing; Wed, 30 Aug 1995 06:25:03 -0700 Received: from yarrina.connect.com.au (yarrina.connect.com.au [192.189.54.17]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id GAA00715 for ; Wed, 30 Aug 1995 06:24:55 -0700 Received: from frisbee.damp.apana.org.au (frisbee.damp.apana.org.au [203.2.134.193]) by yarrina.connect.com.au with ESMTP id XAA06035 (8.6.12/IDA-1.6 for ); Wed, 30 Aug 1995 23:24:42 +1000 Received: (from dpeters@localhost) by frisbee.damp.apana.org.au (8.6.12/8.6.8) id WAA03023 for hardware@freebsd.org; Wed, 30 Aug 1995 22:56:06 +0930 From: David Peters Message-Id: <199508301326.WAA03023@frisbee.damp.apana.org.au> Subject: Wangdat 3400DX and NCR SCSI woes To: hardware@freebsd.org Date: Wed, 30 Aug 1995 22:56:06 +1030 (CST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 1282 Sender: hardware-owner@freebsd.org Precedence: bulk Hi, I've got a Wangdat 3400DX hanging off a NCR 53C810-based SCSI controller. Performance sucks. This isn't a FreeBSD-specific problem; it causes problems booting (which is talking via the BIOS) as well. (Although we haven't actually tried under other operating systems) Once the system is up, throughput is around 2-3M/minute. For what it's worth, we have the jumper diagram out of the service manual; there's nothing of significance that we can find - termination options, SCSI 1/2 select, compression enable/disable. Cabling and termination isn't much of an option for problems either; the DAT has been swapped in in place of a disk (performance of the disks without the DAT is good) and performs equally badly there or in its normal external casing. We've also run the unit on an Ultrastor 34F. This boots fine, but performance is just as bad. The questions: - Is there anyone out there using one of these in a similar configuration with good results? - Does anyone know what the dipswitches 5,6,7 & 8 do? - Any ideas? Please? Mike Smith, per -- David Peters Email: dpeters@damp.apana.org.au Adelaide Hills, South Australia http://www.damp.apana.org.au/~dpeters "blublbulabah blubahahilibalba" - Kaboobi, Shazzam. From owner-freebsd-hardware Wed Aug 30 07:29:52 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id HAA03277 for hardware-outgoing; Wed, 30 Aug 1995 07:29:52 -0700 Received: from Relay1.Austria.EU.net (relay1.Austria.EU.net [192.92.138.47]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id HAA03244 for ; Wed, 30 Aug 1995 07:29:37 -0700 From: marino.ladavac@aut.alcatel.at Received: from aut.alcatel.at (dnisun.aut.alcatel.at) by Relay1.Austria.EU.net with SMTP id AA06711 (5.67b/IDA-1.5 for ); Wed, 30 Aug 1995 16:29:05 +0200 Received: from atuhc16 by aut.alcatel.at (4.1/SMI-4.1/AAA-1.29/main) id AA11372; Wed, 30 Aug 95 16:29:10 +0200 Message-Id: <9508301429.AA11372@atuhc16.aut.alcatel.at> Received: by atuhc16 (1.38.193.4/16.2) id AA05725; Wed, 30 Aug 1995 16:29:02 +0200 Subject: Re: Upgrade to my machine To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Wed, 30 Aug 95 16:29:02 METDST Cc: hardware@freebsd.org In-Reply-To: <199508300315.UAA05719@gndrsh.aac.dev.com>; from "Rodney W. Grimes" at Aug 29, 95 8:15 pm Mailer: Elm [revision: 70.85] Sender: hardware-owner@freebsd.org Precedence: bulk > > > > -Vince- stands accused of saying: > > > Hmmm, what about machines in terms like SUN's, HP's will the P90 > > > compare to since the Alpha is a fast machine. > > > > Depends lots on what you're doing with them; in a straight line, the P90 > > is pretty quick, but what you put around it largely determines how it will > > perform in an applications context. (Especially memory/cache/disk) > > > > >> Anyone who does big models of any sort uses huge amounts of memory, > > >> as Rod already observed. > > > > > > That's true but who would actually need a gig of ram? > THINK for a minute about large applications. An Intel Pentium 90/100 CPU > chip as 3.3 billon transistors on it. Each cmos transitor takes at least ^^^^^^ Minor nit. Million. Otherwise, we would have been having at least 500 megabyte RAM chips for several years now. Cheaply. I wish we did :) /Alby > 6 rectangles to represent the minimal transitor data and 3 contacts to hook > it up, now thats 19.8G assumming I can stuff a rectange into a byte :-). > We haven't even started to talk about interconnecting these 3.3 billon > transistors... > Can you say that a gigabyte in this world is actually a very small amount > of data! > -- > Rod Grimes rgrimes@gndrsh.aac.dev.com > Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Wed Aug 30 11:33:48 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id LAA19916 for hardware-outgoing; Wed, 30 Aug 1995 11:33:48 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id LAA19908 for ; Wed, 30 Aug 1995 11:33:41 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id LAA08959; Wed, 30 Aug 1995 11:33:14 -0700 From: "Rodney W. Grimes" Message-Id: <199508301833.LAA08959@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: marino.ladavac@aut.alcatel.at Date: Wed, 30 Aug 1995 11:33:14 -0700 (PDT) Cc: hardware@freebsd.org In-Reply-To: <9508301429.AA11372@atuhc16.aut.alcatel.at> from "marino.ladavac@aut.alcatel.at" at Aug 30, 95 04:29:02 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 4256 Sender: hardware-owner@freebsd.org Precedence: bulk > > > > > > > > -Vince- stands accused of saying: > > > > Hmmm, what about machines in terms like SUN's, HP's will the P90 > > > > compare to since the Alpha is a fast machine. > > > > > > Depends lots on what you're doing with them; in a straight line, the P90 > > > is pretty quick, but what you put around it largely determines how it will > > > perform in an applications context. (Especially memory/cache/disk) > > > > > > >> Anyone who does big models of any sort uses huge amounts of memory, > > > >> as Rod already observed. > > > > > > > > That's true but who would actually need a gig of ram? > > > THINK for a minute about large applications. An Intel Pentium 90/100 CPU > > chip as 3.3 billon transistors on it. Each cmos transitor takes at least > ^^^^^^ > Minor nit. > > Million. Otherwise, we would have been having at least > 500 megabyte RAM chips for several years now. Cheaply. > > I wish we did :) I will publically admit I may very well be in error here, 2 other folks have sited references that state the number is 3.3 million, and I went and checked 4 references here on this and they all say millon, but I am really having a hard time fitting the Pentium design into 3.3 million transistors from a few simple calculations. 16k of static sram takes >768K transistors assuming a 6 transistor cell, now perhaps the cache is done in quasi-static 4 cell cmos sram, but then how do they do stop clock ??? Perhaps the cache is invalidated when coming out of stop clock mode, or maybe that is what happens during the cycles it takes to shut down. Also the claims that the RISC core of the 486 is in excess of a million transistors in the 1994 data book sets makes me wonder how one could do 2 of these for the dual issue super scaler pipe in anything less than 2 million transistors. Now we need to add in the MMU, TLB, and BIU to my already almost 2.7 Million transistors, okay, 600K transistors, 200K gates, yea, I suppose I might squeeze it into that... oh, and the microcode, that must be atleast 80 bits by 4k deep, but thats not much given ROM is 1 transistor per bit, plus decoders (and 4k decoderes aren't much when you use TGMX's :-)) Perhaps the 3.3 million transistor count is the RISC super scaler core excluding the cache, MMU, TLB and BIU _and_ microcode. I don't directly consult for the processor group at Intel so I don't have ready access to the right data. Published data like this is unclear as to exactly what it is. Anyway, the real point I was trying to make is there are things that get _HUGE_ in the CAD applications with respect to memory needs. I have seen single chip logic synthises runs go well into the 1.2G of data segment size on a day in day out basis, and so many other similiar sized things in my 15 years as a design automation specialist that it would make the average PC unix user's head spin when you look at a 1G main memory and 8G of local disk plus 1TB of back room disk, and a room for of engineers screaming that `we need more!!!''. I've watched an MIS group run and hide 2 weeks after we installed HSPICE on a 3084 and the engineers fired up some rather large test jobs to see if MIS's claim that the 3090 could do what we where doing on DN10000's just as well. Needless to say the 3090 buckled under once the job went to allocate the 400MB needed for state storage of the simulation since it was an MIS configured machine sporting all of 128MB of main memory. Needless to say, the purchase request for 14 additional 768MB DN10000's went past MIS after this without blinking an eye :-). > /Alby > > > 6 rectangles to represent the minimal transitor data and 3 contacts to hook > > it up, now thats 19.8G assumming I can stuff a rectange into a byte :-). > > > We haven't even started to talk about interconnecting these 3.3 billon > > transistors... > > > Can you say that a gigabyte in this world is actually a very small amount > > of data! I will hold by that statement, been there too many times. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Wed Aug 30 12:26:07 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id MAA26115 for hardware-outgoing; Wed, 30 Aug 1995 12:26:07 -0700 Received: from Relay1.Austria.EU.net (relay1.Austria.EU.net [192.92.138.47]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id MAA26101 for ; Wed, 30 Aug 1995 12:25:59 -0700 From: marino.ladavac@aut.alcatel.at Received: from aut.alcatel.at (dnisun.aut.alcatel.at) by Relay1.Austria.EU.net with SMTP id AA16910 (5.67b/IDA-1.5 for ); Wed, 30 Aug 1995 21:25:43 +0200 Received: from atuhc16 by aut.alcatel.at (4.1/SMI-4.1/AAA-1.29/main) id AA14125; Wed, 30 Aug 95 21:25:48 +0200 Message-Id: <9508301925.AA14125@atuhc16.aut.alcatel.at> Received: by atuhc16 (1.38.193.4/16.2) id AA05827; Wed, 30 Aug 1995 21:25:41 +0200 Subject: Re: Upgrade to my machine To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Wed, 30 Aug 95 21:25:41 METDST Cc: hardware@freebsd.org In-Reply-To: <199508301833.LAA08959@gndrsh.aac.dev.com>; from "Rodney W. Grimes" at Aug 30, 95 11:33 am Mailer: Elm [revision: 70.85] Sender: hardware-owner@freebsd.org Precedence: bulk Rod Grimes wrote: > 16k of static sram takes >768K transistors assuming a 6 transistor cell, > now perhaps the cache is done in quasi-static 4 cell cmos sram, but then > how do they do stop clock ??? Perhaps the cache is invalidated when > coming out of stop clock mode, or maybe that is what happens during the > cycles it takes to shut down. But is it really static? I seem to recollect that original '486 could not change clock rate on the fly, unlike '386 (allegedly, for some on chip dynamic storage problems.) Furthermore, since the L1 cache is on the same chip with the rest of the CPU, refresh can be done completely transparently in the cycles when cache is not read/written to. However, I do not know if there are such spare cycles in Pentium case; there should be some at '486. > Also the claims that the RISC core of the 486 is in excess of a million > transistors in the 1994 data book sets makes me wonder how one could do > 2 of these for the dual issue super scaler pipe in anything less than > 2 million transistors. I cannot argue about that as I'm not familiar with the document, but some manufacturers (e.g. Motorola, TI) use the word 'core' when they refer to the sillicon part of an IC. This, and knowing that Intel's DX4 has 16K cache on chip, it doesn't sound too impossible. > Now we need to add in the MMU, TLB, and BIU to my already almost 2.7 Million > transistors, okay, 600K transistors, 200K gates, yea, I suppose I might > squeeze it into that... oh, and the microcode, that must be atleast > 80 bits by 4k deep, but thats not much given ROM is 1 transistor per bit, > plus decoders (and 4k decoderes aren't much when you use TGMX's :-)) > Perhaps the 3.3 million transistor count is the RISC super scaler core > excluding the cache, MMU, TLB and BIU _and_ microcode. > I don't directly consult for the processor group at Intel so I don't have > ready access to the right data. Published data like this is unclear as > to exactly what it is. Well, a rough guess can be made from the die size and the manufacturing process. This way one could get the high limit (connections ignored in favor of transistors.) The die is, what, 11 mm by 10 mm? Process is .35 micron? Let's say that a 3x3 grid can house a transistor (can it? I have no idea) then you can put cca. 1 million of them onto 1 square mm. There is about 110 square mm of area available. This, of course, proves nothing, neither did I try to do so. It is an honest attempt to estimate how many did they really need. But, then again, who cares :) > Anyway, the real point I was trying to make is there are things that > get _HUGE_ in the CAD applications with respect to memory needs. I No kiddin' :) > > > Can you say that a gigabyte in this world is actually a very small amount > I will hold by that statement, been there too many times. Definitely agree. Too bad it's so damnably expensive for the lowly common folks (like me :( ) Tschuess, /Alby > -- > Rod Grimes rgrimes@gndrsh.aac.dev.com > Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Wed Aug 30 13:01:31 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id NAA27331 for hardware-outgoing; Wed, 30 Aug 1995 13:01:31 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id NAA27324 for ; Wed, 30 Aug 1995 13:01:28 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id NAA09303; Wed, 30 Aug 1995 13:01:03 -0700 From: "Rodney W. Grimes" Message-Id: <199508302001.NAA09303@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: marino.ladavac@aut.alcatel.at Date: Wed, 30 Aug 1995 13:01:03 -0700 (PDT) Cc: hardware@freebsd.org In-Reply-To: <9508301925.AA14125@atuhc16.aut.alcatel.at> from "marino.ladavac@aut.alcatel.at" at Aug 30, 95 09:25:41 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3800 Sender: hardware-owner@freebsd.org Precedence: bulk > > Rod Grimes wrote: > > > 16k of static sram takes >768K transistors assuming a 6 transistor cell, > > now perhaps the cache is done in quasi-static 4 cell cmos sram, but then > > how do they do stop clock ??? Perhaps the cache is invalidated when > > coming out of stop clock mode, or maybe that is what happens during the > > cycles it takes to shut down. > > But is it really static? I seem to recollect that original '486 could > not change clock rate on the fly, unlike '386 (allegedly, for some on > chip dynamic storage problems.) Not possitive on the true staticness of it. Though the 486 can not change clock rates, the SL enhance 486 can, as well as all Pentiums. This is the clock stop SL/enhancement features. To change clock frequencies you do a stop clock, shut the clock clear down, the start the clock up at the new frequency. > > Furthermore, since the L1 cache is on the same chip with the rest of the > CPU, refresh can be done completely transparently in the cycles when > cache is not read/written to. However, I do not know if there are such > spare cycles in Pentium case; there should be some at '486. I am pretty sure they are not fully dynamic, thus they do not require refresh, as fully dynamic would mean having to do a write back cycle after the destructive read that dynamic memory cells use. (You dump the stored charge of the cell into the bit line when you read a DRAM cell). > > Also the claims that the RISC core of the 486 is in excess of a million > > transistors in the 1994 data book sets makes me wonder how one could do > > 2 of these for the dual issue super scaler pipe in anything less than > > 2 million transistors. > > I cannot argue about that as I'm not familiar with the document, but some > manufacturers (e.g. Motorola, TI) use the word 'core' when they refer to > the sillicon part of an IC. This, and knowing that Intel's DX4 has 16K > cache on chip, it doesn't sound too impossible. My sighted data was for the 486DX2, not the 486DX4, sorry for leaving that detail out (I thought I had it in there, but above it appears missing :-(). > > Now we need to add in the MMU, TLB, and BIU to my already almost 2.7 Million > > transistors, okay, 600K transistors, 200K gates, yea, I suppose I might > > squeeze it into that... oh, and the microcode, that must be atleast > > 80 bits by 4k deep, but thats not much given ROM is 1 transistor per bit, > > plus decoders (and 4k decoderes aren't much when you use TGMX's :-)) > > > Perhaps the 3.3 million transistor count is the RISC super scaler core > > excluding the cache, MMU, TLB and BIU _and_ microcode. > > > I don't directly consult for the processor group at Intel so I don't have > > ready access to the right data. Published data like this is unclear as > > to exactly what it is. > > Well, a rough guess can be made from the die size and the manufacturing > process. This way one could get the high limit (connections ignored in > favor of transistors.) > > The die is, what, 11 mm by 10 mm? Not noted in the data books, not avaliable in die form :-(. > Process is .35 micron? A80502-60 and 66 are 5V 0.8 micron technology sporting ``3.1M transistors'' A80502-75/90/100 are 3.3V 0.6 micron technology sporting ``3.3M transistors'' > Let's say > that a 3x3 grid can house a transistor (can it? I have no idea) then > you can put cca. 1 million of them onto 1 square mm. There is about > 110 square mm of area available. Smallest transistor I could build with 0.6 micron technology is 1.8 micron by 1.2 micron, and that is assuming multiple transistors in the same well, quite common in cmos logic design. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Wed Aug 30 16:52:01 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id QAA08970 for hardware-outgoing; Wed, 30 Aug 1995 16:52:01 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id QAA08964 for ; Wed, 30 Aug 1995 16:51:59 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id JAA29910; Thu, 31 Aug 1995 09:25:14 +0930 From: Michael Smith Message-Id: <199508302355.JAA29910@genesis.atrad.adelaide.edu.au> Subject: Re: Upgrade to my machine To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Thu, 31 Aug 1995 09:25:14 +0930 (CST) Cc: marino.ladavac@aut.alcatel.at, hardware@freebsd.org In-Reply-To: <199508301833.LAA08959@gndrsh.aac.dev.com> from "Rodney W. Grimes" at Aug 30, 95 11:33:14 am Content-Type: text Content-Length: 3013 Sender: hardware-owner@freebsd.org Precedence: bulk Rodney W. Grimes stands accused of saying: > > > THINK for a minute about large applications. An Intel Pentium 90/100 CPU > > > chip as 3.3 billon transistors on it. Each cmos transitor takes at least > > ^^^^^^ > > Minor nit. > > > > Million. Otherwise, we would have been having at least > > 500 megabyte RAM chips for several years now. Cheaply. > > > > I wish we did :) > > I will publically admit I may very well be in error here, 2 other folks > have sited references that state the number is 3.3 million, and I went > and checked 4 references here on this and they all say millon, but I am > really having a hard time fitting the Pentium design into 3.3 million > transistors from a few simple calculations. > > 16k of static sram takes >768K transistors assuming a 6 transistor cell, > now perhaps the cache is done in quasi-static 4 cell cmos sram, but then > how do they do stop clock ??? Perhaps the cache is invalidated when > coming out of stop clock mode, or maybe that is what happens during the > cycles it takes to shut down. Have a look at a die photo of one of those puppies, Rod. The cache occupies a _very_ substantial amount of real estate. (This is why on-chip caches are so small) > Also the claims that the RISC core of the 486 is in excess of a million > transistors in the 1994 data book sets makes me wonder how one could do > 2 of these for the dual issue super scaler pipe in anything less than > 2 million transistors. Because lots of the support logic is common. (clock generation, propagation etc.) > Perhaps the 3.3 million transistor count is the RISC super scaler core > excluding the cache, MMU, TLB and BIU _and_ microcode. No, that's the whole shebang. There's not much microcode, and you ought to know that shrapnel logic actually uses very few gates. > I don't directly consult for the processor group at Intel so I don't have > ready access to the right data. Published data like this is unclear as > to exactly what it is. >From some vendors at least 8) DEC are pretty clear that there are ~9M transistors on the biggest of the Alpha processors. Check out the die photos on the PowerPC homepage (off motorola's web site). The pictures for the 604 and 620 in particular offer good examples of how big the various bits are. > Anyway, the real point I was trying to make is there are things that > get _HUGE_ in the CAD applications with respect to memory needs. I You're spot on here. > Rod Grimes rgrimes@gndrsh.aac.dev.com -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Wed Aug 30 17:49:01 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id RAA10512 for hardware-outgoing; Wed, 30 Aug 1995 17:49:01 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id RAA10505 for ; Wed, 30 Aug 1995 17:48:47 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id RAA10127; Wed, 30 Aug 1995 17:47:16 -0700 From: "Rodney W. Grimes" Message-Id: <199508310047.RAA10127@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: msmith@atrad.adelaide.edu.au (Michael Smith) Date: Wed, 30 Aug 1995 17:47:16 -0700 (PDT) Cc: marino.ladavac@aut.alcatel.at, hardware@freebsd.org In-Reply-To: <199508302355.JAA29910@genesis.atrad.adelaide.edu.au> from "Michael Smith" at Aug 31, 95 09:25:14 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 541 Sender: hardware-owner@freebsd.org Precedence: bulk ... > Have a look at a die photo of one of those puppies, Rod. The cache > occupies a _very_ substantial amount of real estate. (This is why > on-chip caches are so small) Sorry, your right. I think I've been staring at CPU chips with no cache on them for too long again... time to pull out the wall size 486 poster and tape it to my forehead while I put on the conical hat <:-). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Thu Aug 31 05:25:19 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id FAA04172 for hardware-outgoing; Thu, 31 Aug 1995 05:25:19 -0700 Received: from Relay1.Austria.EU.net (relay1.Austria.EU.net [192.92.138.47]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id FAA04164 for ; Thu, 31 Aug 1995 05:25:16 -0700 From: marino.ladavac@aut.alcatel.at Received: from aut.alcatel.at (dnisun.aut.alcatel.at) by Relay1.Austria.EU.net with SMTP id AA11101 (5.67b/IDA-1.5 for ); Thu, 31 Aug 1995 14:24:54 +0200 Received: from atuhc16 by aut.alcatel.at (4.1/SMI-4.1/AAA-1.29/main) id AA23659; Thu, 31 Aug 95 14:24:56 +0200 Message-Id: <9508311224.AA23659@atuhc16.aut.alcatel.at> Received: by atuhc16 (1.38.193.4/16.2) id AA05934; Thu, 31 Aug 1995 14:24:49 +0200 Subject: Re: Upgrade to my machine To: rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Date: Thu, 31 Aug 95 14:24:49 METDST Cc: hardware@freebsd.org In-Reply-To: <199508302001.NAA09303@gndrsh.aac.dev.com>; from "Rodney W. Grimes" at Aug 30, 95 1:01 pm Mailer: Elm [revision: 70.85] Sender: hardware-owner@freebsd.org Precedence: bulk Rod Grimes wrote: > > > > Rod Grimes wrote: > > [ is the L1 cache static? ] > Not possitive on the true staticness of it. Though the 486 can not > change clock rates, the SL enhance 486 can, as well as all Pentiums. > This is the clock stop SL/enhancement features. To change clock > frequencies you do a stop clock, shut the clock clear down, the start > the clock up at the new frequency. > > > > Furthermore, since the L1 cache is on the same chip with the rest of the > > CPU, refresh can be done completely transparently in the cycles when > > cache is not read/written to. However, I do not know if there are such > > spare cycles in Pentium case; there should be some at '486. > I am pretty sure they are not fully dynamic, thus they do not require > refresh, as fully dynamic would mean having to do a write back cycle > after the destructive read that dynamic memory cells use. (You dump > the stored charge of the cell into the bit line when you read a DRAM > cell). Since I do not know much about chip innards, and manufacturing technologies, this question is really a shot in the dark, but: would it be possible to implement the cache in a following manner: cache itself is dynamic. The line that is presently read is buffered in some fully static memory, and refreshed after the read completes. Basically, this implies a L0 cache, fully static but very small. The likelihood that the same L1 cache line will be read in the next cycle is small, and if it occurs, the data is still available in L0 cache. It should be sufficient to have only a few lines of L0. If the L1 cache is being refreshed and the requested lines are not available in L0, read is stalled (cannot be noticed by the user as cycle counting has been progressively impossible with introduction of caches and pipelines.) > My sighted data was for the 486DX2, not the 486DX4, sorry for leaving > that detail out (I thought I had it in there, but above it appears > missing :-(). My mistake. You did, in the previous mail. > > Well, a rough guess can be made from the die size and the manufacturing > > process. This way one could get the high limit (connections ignored in > > favor of transistors.) > > > > The die is, what, 11 mm by 10 mm? > Not noted in the data books, not avaliable in die form :-(. > > Process is .35 micron? > A80502-60 and 66 are 5V 0.8 micron technology sporting ``3.1M transistors'' > A80502-75/90/100 are 3.3V 0.6 micron technology sporting ``3.3M transistors'' > > Let's say > > that a 3x3 grid can house a transistor (can it? I have no idea) then > > you can put cca. 1 million of them onto 1 square mm. There is about > > 110 square mm of area available. > Smallest transistor I could build with 0.6 micron technology is 1.8 micron > by 1.2 micron, and that is assuming multiple transistors in the same well, > quite common in cmos logic design. So, it would seem that one needs at least a 4x3 grid to house a transistor (we need some room between transistors, right? Or did you include that gap in the numbers above?) This gives about 230,000 transistors per sq. mm. for a total of cca. 25 million on a 11x10 mm die. This would be the upper theoretical limit. I would guess that the cache is really implemented (at least partially) dynamically. /Alby > -- > Rod Grimes rgrimes@gndrsh.aac.dev.com > Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Thu Aug 31 05:58:40 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id FAA04626 for hardware-outgoing; Thu, 31 Aug 1995 05:58:40 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id FAA04620 for ; Thu, 31 Aug 1995 05:58:38 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id WAA01554; Thu, 31 Aug 1995 22:30:49 +0930 From: Michael Smith Message-Id: <199508311300.WAA01554@genesis.atrad.adelaide.edu.au> Subject: On-chip caches (was re: Upgrade to my machine) To: marino.ladavac@aut.alcatel.at Date: Thu, 31 Aug 1995 22:30:48 +0930 (CST) Cc: rgrimes@gndrsh.aac.dev.com, hardware@freebsd.org In-Reply-To: <9508311224.AA23659@atuhc16.aut.alcatel.at> from "marino.ladavac@aut.alcatel.at" at Aug 31, 95 02:24:49 pm Content-Type: text Content-Length: 1851 Sender: hardware-owner@freebsd.org Precedence: bulk marino.ladavac@aut.alcatel.at stands accused of saying: > > > Rod Grimes wrote: > > > > [ is the L1 cache static? ] > > Since I do not know much about chip innards, and manufacturing technologies, > this question is really a shot in the dark, but: > > would it be possible to implement the cache in a following manner: cache > itself is dynamic. The line that is presently read is buffered in some No. Lookup time, time to refresh etc. pretty much preclude this. > fully static memory, and refreshed after the read completes. Basically, > this implies a L0 cache, fully static but very small. The likelihood that > the same L1 cache line will be read in the next cycle is small, and if it > occurs, the data is still available in L0 cache. It should be sufficient > to have only a few lines of L0. If the L1 cache is being refreshed and > the requested lines are not available in L0, read is stalled (cannot be > noticed by the user as cycle counting has been progressively impossible > with introduction of caches and pipelines.) You lose here on propagation delay; you'd have to deepen the pipeline, and with speculative execution the problem just gets worse. > This would be the upper theoretical limit. I would guess that the cache is > really implemented (at least partially) dynamically. I'd be _really_ suprised to see anything other than low-gain static cells in use in an L1 cache; speed is too important to trade off for anything else. > /Alby -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[ From owner-freebsd-hardware Thu Aug 31 06:33:15 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id GAA05331 for hardware-outgoing; Thu, 31 Aug 1995 06:33:15 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id GAA05325 for ; Thu, 31 Aug 1995 06:33:12 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id GAA11309; Thu, 31 Aug 1995 06:32:28 -0700 From: "Rodney W. Grimes" Message-Id: <199508311332.GAA11309@gndrsh.aac.dev.com> Subject: Re: Upgrade to my machine To: marino.ladavac@aut.alcatel.at Date: Thu, 31 Aug 1995 06:32:27 -0700 (PDT) Cc: hardware@freebsd.org In-Reply-To: <9508311224.AA23659@atuhc16.aut.alcatel.at> from "marino.ladavac@aut.alcatel.at" at Aug 31, 95 02:24:49 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 5640 Sender: hardware-owner@freebsd.org Precedence: bulk > > Rod Grimes wrote: > > > > > > Rod Grimes wrote: > > > > [ is the L1 cache static? ] > > > Not possitive on the true staticness of it. Though the 486 can not > > change clock rates, the SL enhance 486 can, as well as all Pentiums. > > This is the clock stop SL/enhancement features. To change clock > > frequencies you do a stop clock, shut the clock clear down, the start > > the clock up at the new frequency. > > > > > > Furthermore, since the L1 cache is on the same chip with the rest of the > > > CPU, refresh can be done completely transparently in the cycles when > > > cache is not read/written to. However, I do not know if there are such > > > spare cycles in Pentium case; there should be some at '486. > > > I am pretty sure they are not fully dynamic, thus they do not require > > refresh, as fully dynamic would mean having to do a write back cycle > > after the destructive read that dynamic memory cells use. (You dump > > the stored charge of the cell into the bit line when you read a DRAM > > cell). > > Since I do not know much about chip innards, and manufacturing technologies, > this question is really a shot in the dark, but: > > would it be possible to implement the cache in a following manner: cache > itself is dynamic. The line that is presently read is buffered in some > fully static memory, and refreshed after the read completes. Basically, > this implies a L0 cache, fully static but very small. The likelihood that > the same L1 cache line will be read in the next cycle is small, and if it > occurs, the data is still available in L0 cache. It should be sufficient > to have only a few lines of L0. If the L1 cache is being refreshed and > the requested lines are not available in L0, read is stalled (cannot be > noticed by the user as cycle counting has been progressively impossible > with introduction of caches and pipelines.) That is the standard implementation of static column dynamic memory. You have what are known as the the static column data latches used to refresh the DRAM cells during the precharge period. Thing is you can not run any other read or write cycle even if you had a second set of L0 latches as the bit lines are shared. You would have to have 2 bit lines and 2 transfer transistors per storage cell, and that would double the size of your array. The cache array in the pentium is already extremly compex on the data side as it can dual issue data (one to each pipe) and due a store all in 1 clock cycle. This is known as tri ported memory, and given a closer look at the Pentium die photograph suspect the cache is using close to 1.5M transitors, 1M for the data cache and .5M for the instruction cache. > > My sighted data was for the 486DX2, not the 486DX4, sorry for leaving > > that detail out (I thought I had it in there, but above it appears > > missing :-(). > > My mistake. You did, in the previous mail. Okay, I am not looseing my mind then :-), though I have made an order of magnitude error in million vs billion. > > > Well, a rough guess can be made from the die size and the manufacturing > > > process. This way one could get the high limit (connections ignored in > > > favor of transistors.) > > > > > > The die is, what, 11 mm by 10 mm? > > Not noted in the data books, not avaliable in die form :-(. > > > > Process is .35 micron? > > A80502-60 and 66 are 5V 0.8 micron technology sporting ``3.1M transistors'' > > A80502-75/90/100 are 3.3V 0.6 micron technology sporting ``3.3M transistors' > > > > Let's say > > > that a 3x3 grid can house a transistor (can it? I have no idea) then > > > you can put cca. 1 million of them onto 1 square mm. There is about > > > 110 square mm of area available. > > > Smallest transistor I could build with 0.6 micron technology is 1.8 micron > > by 1.2 micron, and that is assuming multiple transistors in the same well, > > quite common in cmos logic design. > > So, it would seem that one needs at least a 4x3 grid to house a transistor > (we need some room between transistors, right? Actaully no, when building something like a 3 input CMOS nand gate the source of one transistor is the drain of the next so I need a 3 gate P-channel and a 3 gate N-channel. That takes 3 * 0.8 for the gate widths, 4*0.8 for the source, the drain, and the shared S/D's in the middle. So width is 7*0.8, length is fixed at 2 * 0.8 (I don't have an Intel process spec, but am assuming from typical high density BiCMOS process work I have done in the past that cell lenghts are typically 2 to 4 times the feature size.) So my 6 transistor 3 input nand gate is 2 * ((7 * 0.8) + (2 * 0.8)) or 14.4 microns square, or 2.4 microns square per transitor. We should probably assume this to be a reasonble areal usage rate. > Or did you include that gap in the numbers above?) See above, it is not that simple :-) > This gives about 230,000 transistors per sq. mm. > for a total of cca. 25 million on a 11x10 mm die. > > This would be the upper theoretical limit. I would guess that the cache is > really implemented (at least partially) dynamically. Nope, I dont think so, I can do a tri ported qusi static 6 trans/cell 8Kx8 (really it is a 1kx64) static cache in about 1 million transistors using TGMX's style read columns and decoders. Wish I had my DN5500 and mentor tools up and running, I would go lay one out and run a gate/transitor count report on it just as a final sanity check to my back of the envelope calculations :-). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-hardware Thu Aug 31 11:08:12 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id LAA15580 for hardware-outgoing; Thu, 31 Aug 1995 11:08:12 -0700 Received: from cliff.corp.cubic.COM (cliff.corp.cubic.com [149.63.66.25]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id LAA15572 for ; Thu, 31 Aug 1995 11:08:10 -0700 From: price@earth.corp.cubic.COM Received: from earth.corp.cubic.COM (earth.corp.cubic.com [149.63.65.10]) by cliff.corp.cubic.COM (8.6.9/8.6.9) with ESMTP id LAA16483 for ; Thu, 31 Aug 1995 11:01:24 -0700 Received: (price@localhost) by earth.corp.cubic.COM (8.6.9/8.6.4) id LAA05063 for hardware@freebsd.org; Thu, 31 Aug 1995 11:05:00 -0700 Date: Thu, 31 Aug 1995 11:05:00 -0700 Message-Id: <199508311805.LAA05063@earth.corp.cubic.COM> To: hardware@freebsd.org Subject: 7850 motherboards Content-Length: 535 Sender: hardware-owner@freebsd.org Precedence: bulk A slew of Pxxx motherboards are hitting the streets that contain built-in pci adaptec aic7850, I repeat 7850, controllers. These are quality boards with built-in adaptec bios. They save the PCI consumer a couple of hundred dollars by not having to buy a scsi controller. I have used these boards successfully with: OS2, WinNT, Win95, Netware 3.x-4.x, Dos. Linux does not work. FreeBSD 2.0.5 does not work. Consumers would like to know i[it,if] FreeBSD will support the adaptec aic7850 controller that these boards are using. dp From owner-freebsd-hardware Thu Aug 31 11:51:13 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id LAA16746 for hardware-outgoing; Thu, 31 Aug 1995 11:51:13 -0700 Received: from persprog.com (persprog.com [204.215.255.203]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id LAA16737 for ; Thu, 31 Aug 1995 11:51:05 -0700 Received: from novell.persprog.com by persprog.com (8.6.9/4.10) id OAA26592; Thu, 31 Aug 1995 14:39:02 -0400 Received: from NOVELL/SpoolDir by novell.persprog.com (Mercury 1.12); Thu, 31 Aug 95 14:38:57 +0500 Received: from SpoolDir by NOVELL (Mercury 1.12); Thu, 31 Aug 95 14:38:54 +0500 From: "David Alderman" Organization: Personalized Programming, Inc. To: price@earth.corp.cubic.COM, hardware@freebsd.org Date: Thu, 31 Aug 1995 14:38:49 EST Subject: Re: 7850 motherboards X-Confirm-Reading-To: "David Alderman" X-pmrqc: 1 Priority: normal X-mailer: Pegasus Mail/Windows (v1.21) Message-ID: <1D7C874812@novell.persprog.com> Sender: hardware-owner@freebsd.org Precedence: bulk > From: price@earth.corp.cubic.COM > Date: Thu, 31 Aug 1995 11:05:00 -0700 > To: hardware@freebsd.org > Subject: 7850 motherboards > A slew of Pxxx motherboards are hitting the streets that contain built-in pci > adaptec aic7850, I repeat 7850, controllers. These are quality boards with built-in adaptec bios. They save the PCI consumer a couple of hundred dollars by not > having to buy a scsi controller. I have used these boards successfully with: > OS2, WinNT, Win95, Netware 3.x-4.x, Dos. > Linux does not work. FreeBSD 2.0.5 does not work. Consumers would like to know i[it,if] FreeBSD will support the adaptec aic7850 controller that these boards are > using. dp > Just out of curiosity, what Adaptec controller is aic7850 equivalent to? Is it a 2940 or a 1522 or something in between? ====================================== When philosophy conflicts with reality, choose reality. Dave Alderman -- dave@persprog.com ====================================== From owner-freebsd-hardware Thu Aug 31 12:02:36 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id MAA17288 for hardware-outgoing; Thu, 31 Aug 1995 12:02:36 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id MAA17282 ; Thu, 31 Aug 1995 12:02:35 -0700 Message-Id: <199508311902.MAA17282@freefall.FreeBSD.org> X-Authentication-Warning: freefall.FreeBSD.org: Host localhost didn't use HELO protocol To: price@earth.corp.cubic.com cc: hardware@freebsd.org Subject: Re: 7850 motherboards In-reply-to: Your message of "Thu, 31 Aug 95 11:05:00 PDT." <199508311805.LAA05063@earth.corp.cubic.COM> Date: Thu, 31 Aug 1995 12:02:35 -0700 From: "Justin T. Gibbs" Sender: hardware-owner@freebsd.org Precedence: bulk >A slew of Pxxx motherboards are hitting the streets that contain built-in pci >adaptec aic7850, I repeat 7850, controllers. These are quality boards with bu >ilt-in adaptec bios. They save the PCI consumer a couple of hundred dollars b >y not >having to buy a scsi controller. I have used these boards successfully with: >OS2, WinNT, Win95, Netware 3.x-4.x, Dos. >Linux does not work. FreeBSD 2.0.5 does not work. Consumers would like to kn >ow i[it,if] FreeBSD will support the adaptec aic7850 controller that these boa >rds are >using. dp If you send me a MB, I will make it work. It took less than two weeks for me to get our sequencer code working on the aic7870, I would expect less for the aic7850. You can contact me at Walnut Creek CDROM (1-800-786-9907) if you're interested. I would, of course, send the MB back to you as soon as I was done with the work. The MB would have to have a CPU in it of some sort though, since I don't have any spare. -- Justin T. Gibbs =========================================== Software Developer - Walnut Creek CDROM FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-hardware Thu Aug 31 12:12:57 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id MAA17719 for hardware-outgoing; Thu, 31 Aug 1995 12:12:57 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.FreeBSD.org (8.6.11/8.6.6) with SMTP id MAA17712 ; Thu, 31 Aug 1995 12:12:56 -0700 Message-Id: <199508311912.MAA17712@freefall.FreeBSD.org> X-Authentication-Warning: freefall.FreeBSD.org: Host localhost didn't use HELO protocol To: "David Alderman" cc: price@earth.corp.cubic.com, hardware@freebsd.org Subject: Re: 7850 motherboards In-reply-to: Your message of "Thu, 31 Aug 95 14:38:49 EST." <1D7C874812@novell.persprog.com> Date: Thu, 31 Aug 1995 12:12:56 -0700 From: "Justin T. Gibbs" Sender: hardware-owner@freebsd.org Precedence: bulk >> From: price@earth.corp.cubic.COM >> Date: Thu, 31 Aug 1995 11:05:00 -0700 >> To: hardware@freebsd.org >> Subject: 7850 motherboards > >> A slew of Pxxx motherboards are hitting the streets that contain built-in pc >i >> adaptec aic7850, I repeat 7850, controllers. These are quality boards with >built-in adaptec bios. They save the PCI consumer a couple of hundred dollars > by not >> having to buy a scsi controller. I have used these boards successfully with >: >> OS2, WinNT, Win95, Netware 3.x-4.x, Dos. >> Linux does not work. FreeBSD 2.0.5 does not work. Consumers would like to >know i[it,if] FreeBSD will support the adaptec aic7850 controller that these b >oards are >> using. dp >> > >Just out of curiosity, what Adaptec controller is aic7850 equivalent >to? Is it a 2940 or a 1522 or something in between? Its an aic7870 without the external memory port and space for only three concurrent commands (the aic7870 handles 16 by default and up to 255 if you hook SRAM up to the memory port - the 3940 controllers do this). > >====================================== >When philosophy conflicts with reality, choose reality. >Dave Alderman -- dave@persprog.com >====================================== -- Justin T. Gibbs =========================================== Software Developer - Walnut Creek CDROM FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-hardware Fri Sep 1 05:01:54 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id FAA29452 for hardware-outgoing; Fri, 1 Sep 1995 05:01:54 -0700 Received: from itp.ac.ru (itp.ac.ru [193.233.32.4]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id FAA29435 ; Fri, 1 Sep 1995 05:01:30 -0700 Received: from alt.itp.ac.ru (alt.itp.ac.ru [193.233.32.9]) by itp.ac.ru (8.6.11/8.6.5) with ESMTP id QAA06193; Fri, 1 Sep 1995 16:01:38 +0400 Received: (from ks@localhost) by alt.itp.ac.ru (8.6.11/8.6.11) id QAA00264; Fri, 1 Sep 1995 16:11:44 +0400 Date: Fri, 1 Sep 1995 16:11:44 +0400 From: Sergey Kosyakov Message-Id: <199509011211.QAA00264@alt.itp.ac.ru> To: hardware@freebsd.org Subject: RISCom/N1 Cc: questions@freebsd.org Sender: hardware-owner@freebsd.org Precedence: bulk Dear all, can we use our RISCom/N1 sync port card (SDL Communications) with FreeBSD 2.0.5 ? As I know BSDI supports this card. Sincerely, Sergey. From owner-freebsd-hardware Fri Sep 1 18:29:59 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id SAA05284 for hardware-outgoing; Fri, 1 Sep 1995 18:29:59 -0700 Received: from tippy.vnet.net (tippy.vnet.net [166.82.197.240]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id SAA05272 for ; Fri, 1 Sep 1995 18:29:56 -0700 Received: (from root@localhost) by tippy.vnet.net (8.6.12/8.6.9) id BAA00262; Sat, 2 Sep 1995 01:30:39 GMT Date: Sat, 2 Sep 1995 01:30:38 +0000 () From: Chris Madison To: hardware@freebsd.org Subject: Fujitsu M1606S-512 SCSI Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk Also known as the M1606 SAU - atleast by the marking on the hd itself. During boot time it is detected as a M1606-512 with the following geo info: C:3457 H:6 "and an average of 102 sectors/track" The problem I have is that the Geometry is wrong:-( I purchased it from a company far away and didn't receive a manual or any documentation. While using sysinstall I get an error about the geometry being wrong. I am just curious if anyone out there may have a the same drive and who would be nice to send me the geo info. PLEASE!!!!! Didn't think about this until now, but: Would it be the correct and obvious thing to do to use the boot info:-) thanx chris From owner-freebsd-hardware Sat Sep 2 00:00:15 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id AAA28653 for hardware-outgoing; Sat, 2 Sep 1995 00:00:15 -0700 Received: from fieber-john.campusview.indiana.edu (Fieber-John.campusview.indiana.edu [149.159.1.34]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id AAA28647 for ; Sat, 2 Sep 1995 00:00:14 -0700 Received: (from jfieber@localhost) by fieber-john.campusview.indiana.edu (8.6.11/8.6.9) id BAA24421; Sat, 2 Sep 1995 01:59:29 -0500 Date: Sat, 2 Sep 1995 01:59:28 -0500 (EST) From: John Fieber To: hardware@freebsd.org Subject: Quantum fireball? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk I'm in the market for a 1G SCSI drive. Any opnions on the Quantum Fireball? Any other suggestions? I'd prefer to get one with a 5 year warranty (the fireball is 3). How about the Empire 1080? They are about as common as hens teeth in the latest Computer Shopper and quite a bit more expensive. -john == jfieber@indiana.edu =========================================== == http://fieber-john.campusview.indiana.edu/~jfieber ============ From owner-freebsd-hardware Sat Sep 2 10:42:27 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id KAA00668 for hardware-outgoing; Sat, 2 Sep 1995 10:42:27 -0700 Received: from fieber-john.campusview.indiana.edu (Fieber-John.campusview.indiana.edu [149.159.1.34]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id KAA00662 for ; Sat, 2 Sep 1995 10:42:25 -0700 Received: (from jfieber@localhost) by fieber-john.campusview.indiana.edu (8.6.11/8.6.9) id MAA27116; Sat, 2 Sep 1995 12:41:36 -0500 Date: Sat, 2 Sep 1995 12:41:36 -0500 (EST) From: John Fieber To: hardware@freebsd.org Subject: More 1gig hard drive questions Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk Well, from my last query I recommendations for the Quantum Capella. However, after spending all morning going through Computer Shopper and calling every place that mentioned carrying Quantum drives I've come to the conclusion that the 1 gig version simply cannot be bought. SO, any other suggestions for 1gig drives with a 5 year warranty that can actually be purchased? I recall Jordan mention some time back that he had good luck with Segate Hawk drives. Anyone want to confirm or refute? -john == jfieber@indiana.edu =========================================== == http://fieber-john.campusview.indiana.edu/~jfieber ============ From owner-freebsd-hardware Sat Sep 2 18:07:29 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id SAA16806 for hardware-outgoing; Sat, 2 Sep 1995 18:07:29 -0700 Received: from kryten.atinc.com (kryten.Atinc.COM [198.138.38.7]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id SAA16794 for ; Sat, 2 Sep 1995 18:07:26 -0700 Received: (jmb@localhost) by kryten.atinc.com (8.6.9/8.3) id UAA26694; Sat, 2 Sep 1995 20:59:40 -0400 Date: Sat, 2 Sep 1995 20:59:39 -0400 (EDT) From: "Jonathan M. Bresler" Subject: Re: Fujitsu M1606S-512 SCSI To: Chris Madison cc: hardware@freebsd.org In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk i got one and am very happy with it. here is my disklabel and fdisk output # /dev/rsd1c: type: SCSI disk: sd1s1 label: flags: bytes/sector: 512 sectors/track: 32 tracks/cylinder: 64 sectors/cylinder: 2048 cylinders: 1041 rpm: 5400 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: # size offset fstype [fsize bsize cpg] a: 81920 62 4.2BSD 0 0 0 # (Cyl. 0*- 40) b: 81920 81982 swap # (Cyl. 40*- 80) c: 2131126 62 unused 0 0 # (Cyl. 0*- 1040*) e: 614400 163902 4.2BSD 0 0 0 # (Cyl. 80*- 380) f: 1352886 778302 4.2BSD 0 0 0 # (Cyl. 380*- 1040*) ******* Working on device /dev/rsd1d ******* parameters extracted from in-core disklabel are: cylinders=1041 heads=64 sectors/track=32 (2048 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=1041 heads=64 sectors/track=32 (2048 blks/cyl) Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 0 is: sysid 165,(386BSD) start 62, size 2131126 (1040 Meg), flag 80 beg: cyl 0/ sector 1/ head 1; end: cyl 1010/ sector 62/ head 33 The data for partition 1 is: The data for partition 2 is: The data for partition 3 is: Jonathan M. Bresler jmb@kryten.atinc.com | Analysis & Technology, Inc. FreeBSD Postmaster jmb@FreeBSD.Org | 2341 Jeff Davis Hwy play go. | Arlington, VA 22202 ride bike. hack FreeBSD.--ah the good life | 703-418-2800 x346 From owner-freebsd-hardware Sat Sep 2 19:36:12 1995 Return-Path: hardware-owner Received: (from majordom@localhost) by freefall.FreeBSD.org (8.6.11/8.6.6) id TAA23395 for hardware-outgoing; Sat, 2 Sep 1995 19:36:12 -0700 Received: from tippy.vnet.net (tippy.vnet.net [166.82.197.240]) by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id TAA23380 for ; Sat, 2 Sep 1995 19:36:05 -0700 Received: (from root@localhost) by tippy.vnet.net (8.6.12/8.6.9) id WAA27229; Sat, 2 Sep 1995 22:35:20 -0400 Date: Sat, 2 Sep 1995 22:35:19 -0400 (EDT) From: Chris Madison To: "Jonathan M. Bresler" cc: hardware@freebsd.org Subject: Re: Fujitsu M1606S-512 SCSI In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hardware-owner@freebsd.org Precedence: bulk On Sat, 2 Sep 1995, Jonathan M. Bresler wrote: >//... Wonderful! Thanx! chris ========================================================================== root@tippy.vnet.net | Bill Gates is the Anti-Christ cmmadiso@uncc.edu | Windows95 is the False Prophet ==========================================================================