MSI KT3 Ultra 2 Vs Gigabyte 7VRXP Motherboard Shootout - PAGE 1

- Tuesday, September 17th, 2002 Like Share



Article Index

1.Introduction, Specs & MSI Box Contents
2.Gigabyte Box Contents, OC & Sandra
3.Sysmark, PCMark & 3DMarks
4.Final Thoughts



Get updates when we publish new articles

Comments

Sort by date: ascending descending
0 thumbs!
^
Mitch20021 Nov 29, 02
im looking at a new motherboard, that MSI board sounds very nice, the main reason im looking is because the board i have MSI K7266 Pro 2 refuses to run cool, there is a pic of the temps on the review are those temps after using it for ages or just turning it on? And is there a GeForce card on board? Im looking for a AMD motherboard that runs cool on the northbridge.

My email is supersaiyan2goku@hotmail.com if you know anything about the temps in the email please send them to me
0 thumbs!
^
spill Sep 22, 02
note:

Publicly Known Issue:

I just returned a 7xrp v1.2. It resets bus speeds whenever you power off the PC. DO ANY REVIEWS SAY THAT?

And I personally had to boot it up about 3 times to achieve a bus speed of over 166. It took a boot at 166 to get the 1/5 divider to kick in. DO YOU WANT THAT FOR THE NEXT YEAR? ARG.

that sucks but it did run at 180x10.5
0 thumbs!
^
Redemption Sep 19, 02
While Neumann's article does have some mistakes, I think some of the comments the people made on that thread are sort of out of proportion and isn't even all that valid. Almost a mob mentality.

For instance, two people make fun of Neumann's comment:

quote Speedy-J

ROFL overall,
quote Neumann
"the MSI Ultra2 led the way in this benchmark. The most probable cause for this is the fact that the FSB is overclocked +1Mhz."
Well if you look at the performance numbers, the MSI only lead the benchmarks by around 1-2% (1% in nearly all results except for the Sisoft FPU results, where it was ~2%). The 1Mhz increase in FSB, or even if it is just a 0.66Mhz increase in FSB leading to a 20Mhz difference in total clock speed, is around 1% difference. 1% increase in performance, 1% difference in clock speeds... I don't think the conclusion that the clock speed has a high chance of influencing the results is so far fetched, even if clock speeds and performance differences aren't necessarily matched proportionally.

Secondly, to dispel any further untruths: there is indeed a 2100XP Thoroughbred chip and I think several websites have them: from AMD themselves. Our 2100XP Thoroughbred is being used because 2200 Thoroughbred's were in short supply when we made our request for a chip.

quote amdmb.com
The first thing I did upon hearing this news was get a couple of new Thoroughbred processors. Unfortunately, the 2200+ CPU is in very high demand with most of the parts being eaten up by large OEMs, and thus I wasn't able to attain any more than the one I had. Instead, I got two more 2100+ processors based on the Thoroughbred core. Though these had a 66 MHz stock speed deficit, the new core is what I really wanted to test, and so the 2100+ processors fit my needs nicely.

excerpt from: http://www.amdmb.com/article-display.php?ArticleID=188
If you don't like Amdmb.com, try this review of a 2100XP Thoroughbred at PCStats:

http://www.pcstats.com/articleview.cfm?articleID=1162

Digit-Life has a chart showing 2000 and 2100XP Thoroughbred core chips here:
http://www.digit-life.com/articles/athlonxp2200thoroughbred/

Zdnet UK has a chart for power consumption comparing Palomino vs Thoroughbred core chips, and they show the 2100 Thoroughbred:

http://reviews.zdnet.co.uk/review/21/1/1397.html

And finally, here's Geek.com's AMD Thoroughbred chart showing the 2100XP Thoroughbred availabability in Jun 2002:

http://www.geek.com/procspec/amd/thoroughbred.htm

Neumann didn't say he was testing on a 2100 Thoroughbred because he was naive, he said because AMD sent us a 2100 Thoroughbred to use in our benchmarking.

I think maybe some of the forum members need to educate themselves too. Not all "ripping" is valid either, just like not all parts of reviews can be trusted at all times.


Thanks for your comments guys, we do keep these in mind when trying to improve our articles. For instance, I think the oversight on the clock divider is the most jarring one. There's definitely a huge limitation in overclocking potential without at least the 1/5 divider.
0 thumbs!
^
webkido13 Sep 19, 02
I just read those comments over at ocworkbench. They are really ripping this "review" (more like a benchmarking contest with motherboard pictures) apart.
0 thumbs!
^
pheonix2 Sep 19, 02
There were quite a few wrong comments in the review. I came across the review here which points out some of them.
http://www.ocworkbench.com/ocwbcgi/ultimatebb.cgi?ubb=get_topic&f=21&t=000613
0 thumbs!
^
Keystone Sep 18, 02
Oops, of course I meant "synchronous" and not "asynchronous."

Other than those oversights, it's was a fairly decent comparison by Neumann.
0 thumbs!
^
webkido13 Sep 18, 02
I totally agree with Keystone. I've had a Gigabyte 7VRXP (V1.1) since it was first released. Unfortunately, I just recently upgraded my CPU and just then realized that I could not stably obtain a 166 mhz fsb without the lacking 1/5 divider. Stay far away from this board if you are thinking about overclocking. Also, multiplier adjustements have to be done through jumpers and not through the BIOS which should not be the case in this day and age.

I think all of this information should have been a part of the review so people don't put out good money for something that isn't really what it's claimed to be. Also, according to Gigabyte tech support neither Version 1.1 nor the new Version 2 of this board support the 1/5 divider. There you have it. Go buy the MSI board. That's what I'll do.
0 thumbs!
^
Keystone Sep 17, 02
Neumann wrote: "With 1Mhz increments up to 355Mhz you get more options than the MSI but I really doubt if you can achieve 355Mhz FSB. "

Unfortanately he did not explore the strong limitations with Gigabyte's set-up and dip-switch that only provides the limited choices of a PCI divider of 3 or 4. Stable asynchronous speeds (RAM/FSB) of 166MHz are not possible with the Gigabyte, while it is with the MSI motherboard.

Since Gigabyte still has not updated the BIOS to support PCI divider of 5 or even 6 and that the board's design may be limited to only 2 choices via the dipswitch, this board is not only unsuitable to many overclockers, but is unlikely to support any upcoming Athlon XPs that are on a 333MHz DDR FSB.

My first choice was a 7VRXP but after the frustration of not attaining a stable 166MHz FSB and the lack of stable Geforce 4 support (latest revision solved the problem and RMAs are available for older versions), I gave up and replaced the board with an MSI KT3 Ultra2.
^
Sponsored
Sort by date: ascending descending
Add your comment:
Name *:  Members, please LOGIN
Email:  We use this to display your Gravatar.

Sign in with
Comment *:
(0.1708/d/web3)