Anand Part 4 up: Desktop Perf.

Toggle sidebar Toggle sidebar
Wow! Opteron owns the P4C for gaming when using an MX440 PCI graphics card! I'm impressed!

I think I'll wait until somebody can bench with an AGP card before I get a chubby though.

Still...I didn't think the Opteron at 1.8 gig would walk all over the P4 3.0C like that.

Didn't ASUS just announce an Opteron board with AGP?



<font color=green>The Netherlands is where you go when you're too good for heaven.</font color=green> :tongue:


<P ID="edit"><FONT SIZE=-1><EM>Edited by Twitch on 04/25/03 00:27 AM.</EM></FONT></P>
 
Very surprising indeed especially after seeing the A64 scores over there at xbitlabs. One wonders why it would perform so well in those games. Perhaps the memory latency helps a lot.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
 
Well, I still wanna see some AGP benchies. I read something to the effect that the IMC might actually be a bottleneck when streaming AGP...I may have misunderstood the context so I'm not sure about that, but I'm not going to get too excited over benchmarks run on two systems with ultra-high-end processors and PCI graphics...

I will say it does make one take pause though, doesn't it.


<font color=green>The Netherlands is where you go when you're too good for heaven.</font color=green> :tongue:
 
The benches I read about were "over 15000 with a Radeon 9700 PRO," which says nothing. Athlon 2600+ beats 15,000 with a 9700 PRO. That's my point. Since I am pretty much nothing but a hard-core gamer, I am on the verge of getting excited, but I am going to force myself to maintain my composure until more data is available. :smile:


<font color=green>The Netherlands is where you go when you're too good for heaven.</font color=green> :tongue:
 
I will say this: *<i>If</i> these benchmarks are reflective of how A-64 will perform across-the-board in games using AGP graphics, and <i>if</i> AMD can release a 2.0 GIG or even a 2.2 GIG version of A-64, then Prescott better be pretty damned good, or AMD will own gamers...

However, the article did point out that the Opteron has twice the memory bandwidth that A-64 will have, and I worry that might hamper A-64's performance.


<font color=green>The Netherlands is where you go when you're too good for heaven.</font color=green> :tongue:


<P ID="edit"><FONT SIZE=-1><EM>Edited by Twitch on 04/25/03 01:12 AM.</EM></FONT></P>
 
It was pointed out at the Anandtech forums that indeed this may be a PCI issue. From previous benchmarks from Anandtech, a P4 2.4 GHz with the same video card, but using the AGP version (Geforce 4 MX 460 I believe) scored twice what the P4 3.0C system did. The nForce3's PCI implementation is much better than Canterwood's as they seem to think over at the forums.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
 
I not sure about the PCI by itself but the Ht can decrease the lantency and not stack the bandwith betwwen the Northbridge and the south bridge

[-peep-] french
 
If that´s true, we´ll have to wait for an AGP-supporting mobo to show up around review sites to get a real impression. And if indeed the 3.0C scores twice as much on AGP, then these benchmarks don´t really mean anything!
 
a 2.4 P4 with an AGP card scored twice as better than the 3.0C system with the PCI version of the vid card. So it's safe to say that for now, these results are questionable. Without further testing, there's no way to say for sure that it's a PCI thing.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
 
in any case that just become a lantency benchmark as bandwith cannot be use on canterwood.Agp implentation cannot be use and run on a very old card.

[-peep-] french
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom