Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 22 Sep 2005 17:49:13 -0700
From:      =?ISO-8859-1?Q?Malachi_de_=C6lfweald?= <malachid@gmail.com>
To:        "O. Hartmann" <ohartman@mail.uni-mainz.de>
Cc:        freebsd-questions@freebsd.org, freebsd-amd64@freebsd.org
Subject:   Re: nForce4-SLI, AHCI and NCQ and FreeBSD 6.0
Message-ID:  <c090347a0509221749604cf48c@mail.gmail.com>
In-Reply-To: <43202588.9030903@mail.uni-mainz.de>
References:  <431D90E1.1030105@mail.uni-mainz.de> <c090347a050906070821e844d7@mail.gmail.com> <43202588.9030903@mail.uni-mainz.de>

next in thread | previous in thread | raw e-mail | index | archive | help
Asus's site seems to be down right now.... but at least Google caches it...

http://66.102.7.104/search?q=3Dcache:CG_qYngXXLYJ:usa.asus.com/products/mb/=
socket939/a8nsli-p/overview.htm+a8n+sli+premium+site:asus.com&hl=3Den
 It could very well be marketing hype. All I know is that one of the sellin=
g
points I chose it on was: "SATA 3Gb/s" (on that page)
site and manual both say that the Si3114 supports 3Gb/sec, but they don't
say that about the nVidia, which is why I was choosing it.
I am not sure why it would be via the PCI32 bus... checking this page (
http://www.buzzsurf.com/speed/) it says that a PCI 64-bit bus running at
66MHz would be 4Gb/sec.... maybe they reserved the pciE for the SLI?
 As far as the performance, the manual says that the performance of raid 5
is greatly affected by block size.. from what I have seen online it has to
do with whether the files being written are big enough to end up striped.
Maybe the 10Mb/s is with a very large block size and very small files? Mayb=
e
a 4k block size and larger files would do better? Just guessing here, reall=
y
not sure.
 As far as why I chose the Premium instead of the Deluxe (for my Windows
workstation and my FreeBSD server), there were a few reasons. The
workstation primarily because it allows software-controlled flipping betwee=
n
SLI and non-SLI, which I thought would be nice if I were to accidentally
play games. For the server, I was going to go with a different board, but i=
t
didn't support RAID5, so I switched to this one.
 Of course, now I am seeing that RAID5 doesn't appear very realistic with
FreeBSD :(
 Malachi

 On 9/8/05, O. Hartmann <ohartman@mail.uni-mainz.de> wrote:
>
> Hello.
> Are you really sure the onboard Sil SATA controller supports SATA II?
> When I studied the handbook of both motherboards, the A8N-SLI Deluxe and
> A8N-SLI Premium I found both motherboards are idetically equipted with
> the *not* SATA II capable SilI3114 SATA controller. By the way, as I
> know, the second SATA controller is attached via the PCI32 bus, not via
> PCIe! That means (in my opinion) there is no benefit using this
> controller. It is more a marketing GAG than a serious add-on.
>
> RAID5 performance of ICH7R and SilI is said to be very, very poor (about
> 10 MB/s read/write performance), so what benefit I do have using this
> controller with it's senseless 'capabilities' and non-PCIe attachment?
>
> This is only a thought based on my personal available informations and
> maybe wrong ...
>
> Oliver
>
> Malachi de =C6lfweald wrote:
> > I have noticed the same thing with the onboard Silicon Image controller=
.
> > I am using the A8N-SLI Premium. I am using the Silicon Image controller
> > instead of the nForce4 controller because it specifically said SATAII
> > and said it supported RAID5 (whereas the nForce did not support RAID5).
> > However, on boot it recognizes it as SATA150 (which should be SATA300).
> >
> > Malachi
> >
> > On 9/6/05, *O. Hartmann* <ohartman@mail.uni-mainz.de
> > <mailto:ohartman@mail.uni-mainz.de>> wrote:
> >
> > Hello.
> > I have a little question about the popular nForce4-SLI chipset for the
> > Socket 939 platform. The onboard SATA controller of the nForce4-SLI
> > chipset claims to be NCQ (or SATA II) capable, but nVidia implemented
> > this feature in a not-AHCI-standardised way, but as an own solution. My
> > question is: Is the FBSD 6.X driver for the nForce4-SLI chipset capable
> > of using NCQ (as I know, the driver has to enable NCQ and it's not done
> > automatically by the harddrive-controller interaction).
> >
> > Thanks in advance,
> > Oliver
> > _______________________________________________
> > freebsd-questions@freebsd.org <mailto:freebsd-questions@freebsd.org>
> > mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> > To unsubscribe, send any mail to "
> > freebsd-questions-unsubscribe@freebsd.org
> > <mailto:freebsd-questions-unsubscribe@freebsd.org>"
> >
> >
>
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?c090347a0509221749604cf48c>