On Sat, 2002-03-23 at 16:49, Felix Miata wrote:

> I currently have four machines with AMD chips and one (486) with Intel.
> If I had to buy a motherboard today, it would probably be one that uses
> an Intel CPU. Most AMD motherboards use performance crippled VIA
> chipsets. How good the CPU is matters little when the primary bottleneck
> is the I/O bus.

I don't have a problem with your bottleneck statement; but I do have a
problem with it being applied to VIA chipsets.  My research does not
substantiate what you are saying.  I have a VIA based Abit KT7A-RAID
mobo and have been running it for over a year with an Athlon 800 T-bird
overclocked to 964 mhz.  If you've got more URL's backing up what you
say, by all means post them.  Otherwise I'd suggest checking out the
latest VIA KT266 mobo roundup at tomshardware.  There's also the
upcoming KT333A Via chipset for which there's a preview here:

http://www6.tomshardware.com/mainboard/02q1/020220/kt333-11.html

As for your choice in brand, I think that you are sadly mistaken if you
are planning on going with intel.  The only thing that intel has going
for it happens to be better fab plants.  The pentium 4 was castrated and
put in a new fab process just so they could clock the chip higher. 
Which they do...but it happens to be a marketing move and not a
technically superior one; the higher clock comes at a huge price. 
Literally.  In addition to that I think it's poor judgement to ignore
what G�rard Perreault has just said concerning Intel.  Their alliance
with M$ alone is enough to refrain from trusting them with your money;
not even taking into account the technical mea culpa you will incur.

>> IMHO, via has been making great chipset for ages.

>I don't think so. I found the article that lead to my comment:
>http://www.tecchannel.de/hardware/817/index.html

Since you seem to be currently doing some research, I suggest that you
go here and look at what the experts are saying about the pentium 4
architecture.

http://www.emulators.com/pentium4.htm#Pentium 4

The people that emulate the processors are the ones that have to deal
with the TRUE NATURE of the processors. Darek Mihocka is the author of
the above article and founder of Emulators, Inc.


Oh, and by the way, I checked out your link to the url "that led to your
comment."  I'm kind of surprised, because it does not look like you've
read it yourself.  The article states that by using the latest VIA
patches, the PCI bus performance is raised to equal that provided by
Intel's chipset.

Don't you read your own research?

In case you missed it the first time, I've copied and pasted it here for
your convenience.  Complete content, at this URL:

http://www.tecchannel.de/hardware/817/11.html

Text only excerpt for whomever it may concern here:


______________________________________________________________________
Update: VIAs official patch, Version 1.04

According to VIA, the new version of their patch will increase
performance with RAID cards from ACARD, Adaptec and Highpoint and not
only with Promise-cards anymore. We did a few test with all these cards.
In addition to that, we also checked the new patch with the Advance
29133, another Ultra-ATA/133-PCI-Controller, which is not on VIAs list
yet.

This time, the results are more satisfying. The increase in burst
transfer rates is between 20 to 30 MBytes/s with all cards. This brings
VIAs PCI-Bus to the level of performance found with Intels chipset. Even
with the Advance card these results can be obtained.

Burst transfer rates

PCI card No Patch [MBytes/s] Patch 1.04 [MByte/s]

ACARD AEC-6280 89,6 106,5

Adaptec 29160 78,2 104,5

Advance 29133 82,9 105,8

HighPoint RocketRAID 133 81,2 99,9

Promise Ultra133 TX2 90,1 111,3

All scores taken with Maxtors DiamondMax D740X and VIAs P4XB Motherboard
(P4X266A chipset)

In addition to this, VIAs patch 1.04 now works with older chipsets not
featuring the V-Link-Bus as well. We checked this with a KT133A and a
MVP3. Burst transfers with the KT133A went up to 108,3 MBytes/s from
78,2 MBytes/s. The MVP3 even scored 102,2 MBytes/s with the patch, and
64,1 without it. We took these scores with the Promise Ultra133 TX2.

However, it remains to be seen whether the new patch also remedies some
of the stability issues many readers have reported. 
_____________________________________________________________________

> -- 
> "And we know that all things work together for good to them that
> love God, to them who are the called according to His purpose."
>                                                 Romans 8:28 KJV
> 
>  Team OS/2 ** Reg. Linux User #211409
> 
> Felix Miata  ***  http://mrmazda.members.atlantic.net/

Until Dr Tom says something different than what he's posted so far, I'm
sticking with AMD/VIA.  IMO it's a winning combination.  As for me and
my house, we buy AMD.


Shooing away the fog,

LX



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


Want to buy your Pack or Services from MandrakeSoft? 
Go to http://www.mandrakestore.com

Reply via email to