Gigabytes radeon 9700 hits 400mhz core clock!!

[Prometheus]

www.anandtech.com/video/showdoc.html?i=1686
to say te least!!!Imagine what they can do with 0.13!!!! 8) Maybe 500mhz who knows.

 

[Geeforcer]

Something cought my attention: http://www.anandtech.com/video/showdoc.html?i=1686&p=9

For some reason, ATIs 9700 outperfroms Gygabytes one, sometimes by as much 10% or more.

 

[SvP]

You'll also notice that the ATI Radeon 9700 Pro that didn't overclock as high as the Gigabyte card, still managed to outperform the overclocked Gigabyte card. This goes back to the prerelease BIOS issues we mentioned earlier, and it's a trend that you'll see continue as we explore our gaming benchmarks under Unreal Tournament 2003.

 

[Prometheus]

Don't worry that gigabyte is a pre-production board.Anand states that it will have the same speed as ATI's and even more overclocking potential!!

 

[Saem]

Could it be that the card isn't overclocking and there is a false reading being taken?

 

[Prometheus]

Who knows?That gigabyte sure has a loud fan!!!

 

[SvP]

Saem - possible, but how do you explain the increase in performance then?

 

[Prometheus]

Saem - possible, but how do you explain the increase in performance then?

Good point!!!

 

[Saem]

All right, I'll revise my statement.

What if the reported clock speed is wrong; however, not necessarily meaning that it didn't overclock, simply not to the degree that is stated by the readings taken?

 

[Prometheus]

Saem, are you by any chance an nvidia fan?Why do you find it so hard to believe radeon 9700 gets 400mhz?Anand is no child to make silly mistakes!

 

[JF_Aidan_Pryde]

Is it possible that this is *hand picked* just for this purpose?

 

[ram]

What if the reported clock speed is wrong; however, not necessarily meaning that it didn't overclock, simply not to the degree that is stated by the readings taken?

You could have checked that easy. The GPU limited 3d mark 2K1 high polygon count test with 8 lights increases about the same as the core clock.

 

[SvP]

If I'm right then:
ATI's card overclocked by 16% and performance increased by about 16% in 8 light test and UT2003 (highest resolution, AA+AF)

GB's card (with "bad" BIOS) overclocked by 23% and performance increased by about 21% in 8 light test and by 13% in UT2003.

It seems to me that the cards indeed overclocked and reported MHz's weren't errors.

 

[Prometheus]

Is it possible that this is *hand picked* just for this purpose?

*Hand picked* pre-production board?Hmm why do you think that? Gigabyte boards always got higher clock speeds than ATI's.Just look for a radeon 8500 review.

 

[Dave Baumann]

Well, this is from a document ATI sent out - I've not had any notification that it can't be posted, but I may have to remove it if someone asks. Anyway:

ATI has strong relationships with the major fab companies and we are closely monitoring the progress of the new process technologies. We believe the .15u process still has some headroom to increase clock speeds beyond what was achieved with the RADEON 9700 PRO, and coupled with the availability of DDR-II memory, higher speed derivatives of this product can be expected in the coming months.

So, it would seem that R300 still has a little up its sleeve...

 

[Saem]

Wow, I got accused of being an nVidia fanboy. Hrm, neat.

Well, I've had a Mach 64, Rage II, Rage Pro all in wonder, Radeon and a Radeon 8500. And a few cards in between, I'm happy with ATI products they have served me well. Actually, they're all working and continue to serve me. Me and my ATI slaves.

In anycase, I'm merely having a hard time believing it because I find it to be a little excessively high. But I could be wrong, because ATI has had a track record for conservatively clocking their parts.

Is it possible that this is *hand picked* just for this purpose?

Maybe, but I doubt it.

Now that I think about it, it's just shy of a 30% OC, but it might be possible thanks to the help provided by extra PCB layers and nice clean signal with the onboard power connecter. Hrm, I'm still a little iffy, but at this point I'll deem it in the realm of possible to will usually need better cooling.

BTW, I do consider Anand a putz when it comes to these things, he's almost as bad as Kyle with his lets see what I can OC today. I bet the clocks in his house are made to run faster.

 

[Prometheus]

Well, this is from a document ATI sent out - I've not had any notification that it can't be posted, but I may have to remove it if someone asks. Anyway:

ATI has strong relationships with the major fab companies and we are closely monitoring the progress of the new process technologies. We believe the .15u process still has some headroom to increase clock speeds beyond what was achieved with the RADEON 9700 PRO, and coupled with the availability of DDR-II memory, higher speed derivatives of this product can be expected in the coming months.

So, it would seem that R300 still has a little up its sleeve...

Whouldn't that be getting a product launch too close to r350's Dave?

 

[ram]

Whould that be getting a product launch too close to r350 dave or is this r350 you are talking about?

What makes you think that there will be a R350 at all? What we might see is an RV350 sometime in the far future, but I guess first there will be the RV300, where the character indicates that it will be a cost reduced part.

For me it seems the R300 will be the only enthusiast part till the R400.

 

[Prometheus]

Whould that be getting a product launch too close to r350 dave or is this r350 you are talking about?

What makes you think that there will be a R350 at all? What we might see is an RV350 sometime in the far future, but I guess first there will be the RV300, where the character indicates that it will be a cost reduced part.

For me it seems the R300 will be the only enthusiast part till the R400.

It has been aounced by the ATI CEO than they will release r350 in 1H 2003 on 0.13 with ddr2 memory and DX9.1 support.

 

[ram]

It has been aounced by the ATI CEO than they will release r350 in 1H 2003 on 0.13 with ddr2 memory and DX9.1 support.

URL?