Date: Tue, 10 Jun 1997 10:37:24 -0700 (PDT) From: Howard Lew <hlew@www2.shoppersnet.com> To: Stefan Esser <se@freebsd.org> Cc: Craig Johnston <craig@gnofn.org>, hardware@freebsd.org Subject: Re: poor memory bandwidth on ABIT IT5H rev 1.5 Message-ID: <Pine.BSF.3.91.970610101234.23498C-100000@www2.shoppersnet.com> In-Reply-To: <19970610112536.48352@mi.uni-koeln.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 10 Jun 1997, Stefan Esser wrote: > On Jun 9, Howard Lew <hlew@www2.shoppersnet.com> wrote: > > I thought the TX was better than the VX, so I compared several > > motherboards with dd if=/dev/zero of=/dev/null bs=1m count=1000 > > > > CPU is AMD K5-PR133 > > > > FIC PT2006 (Intel VX) with 256K PB Cache & FreeBSD-2.1.7.1R > > Got 65MB/s > > > > FreeTech F63T (Intel VX) with 256K PB Cache & FreeBSD-2.1.7.1R > > Got 62MB/s > > > > FreeTech F79 (Intel TX) with 512K PB Cache & FreeBSD-3.0SNAP-6/6/97 > > Got 43MB/s > > Does the code in FreeBSD-current use the FPU bcopy for the AMD K5 ??? Hmmmm... > > > > Is there something different about 3.0SNAP6/6/97? Is there any option > > parameter for the kernel config for the K5-PR133? Unfortunately, 2 > > parameters changed, so I can't tell if it is the MB or the OS. I decided > > to go with 3.0snap because of this missing TX PCI & IDE drivers. > > You won't see much of a difference between 2.1.x and -current, with > regard to chip-set support. The TX does not need any specific code, > and I doubt that the EIDE code in -current know about the TX IDE > chip ... > uh oh.. Then I think the upgrade was more of a downgrade except I have a few more memory slots and better cpu support... The funny thing is that Win95 seems to run faster. On Mode 3 EIDE, it gets about 8MB/sec on the MB according to coretest. > So you could have used 2.1.7 for the TX as well! > Gee I should have asked that earlier... 2.1.7.1R was super stable although it felt a little slower and I thought it was because of the "no driver assigned messages". I think that USB device steals an IRQ now, so now every IRQ available is being used. I think the next great chipset should support more IRQs (Is that possible)? Now I can't get NFS to work (2.1.7.1R server, 3.0 client) and sound no longer works, but I am using 4Front-Tech's driver which seems to work nicely. Maybe I'll swap the motherboards back. > > All three tests were done with 32MB EDO memory (set to best memory > > settings). > > The VX is known to perform badly with EDO, and just very slightly > better than a TX (with EDO), if the VX got SDRAM modules ... So the TX is even slower than the VX for EDO? This gives me a very bad feeling...
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.970610101234.23498C-100000>