Gigabytes radeon 9700 hits 400mhz core clock!!

Prometheus said:
Correction:It's not ATI CEO but ATI executives from the japaneze announcement of radeon 9700.There is a typo where it says DX8.1 it's DX9.1. Go to www.reactorcritical.com 7 August:"Ati talks about it's future plans"

This still raises the question if a) Reactorcritical understands japanese at all and b) PC Watch made the correct interpretation based on the answers by ATI.

=)
 
This is good news and think about the possible consequences.

Think about a simplified R300 with:
- .13 micron process
- 500Mhz
- 4 pixel pipelines (2Gigatexels/s)
- 2 vs pipelines (360Millions polygons/s)
- half the R300 die size
- 128bits DDR-II
- probably 80% of the R300 performance in most situations
- for the maisntream market during 2003 :)
 
Geeforcer said:
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.


If you haven't noticed that's and early beta of the card . Sure the final version of ATi's solution gets higher scores . It's also a matter of BIOS optimisation which will only come in the final product because until then only compatibility issues are fixed and not performance tweaks .
 
JF_Aidan_Pryde said:
Is it possible that this is *hand picked* just for this purpose?

No hand picked card is able to hit a 25% improvement in clock speed if the base design is not capable of that .
 
With a little help with a advanced cooler like this one, possibly over 400 mhz on the core could be possible :p

http://www.pcpowercooling.com/products/cooling/card/video_cool/video_cool.htm


This one works very well and low profile too :D
video_cool.jpg
 
I can't for one draw a conclusion from 3dmark scores or an unreleased game. Up to now overclocking the core by a high persentage and a much lower persentage the memory didn't yield as good results as when it's the other way around.

Albeit I don't understand the urge to get the 9700 even faster (it's screamingly fast as it is), faster speced ram would be a far better idea I think. Or let Gainward build Radeons; I'm sure that they'd do a brilliant job with those babies ;)

PS: How old is Anand now anyway?
 
I was thinking about this last night. Sure ATI was planning .13 and ddr2 for the fall. Its the next natural progression *if* they've topped out as fas as possible on .15 . But then yeilds start coming back... and they realize that maybe they wont need .13 to compete with NV30. They can just use the EXACT same R300 design, but used faster chips (400Mhz+) and faster Ram, and still be even, if not ahead.

And it REALLY would reduce the costs of going .13 to be able to put it off a product cycle. Things just seem to be working out perfectly for ATI.

--Braham
 
If ATi is planning a .13um refresh of the R300 It's probally been in the works for some time now. They can't simply put the .15um design on a .13um process if NV30 turns out to be a step above them.


I wonder how NVIDIA feels being ATi's .13um lab rat, or I guess you could say NVIDIA is being visionary and innovative.
 
GetStuff said:
If ATi is planning a .13um refresh of the R300 It's probally been in the works for some time now.

still doesnt mean they have to manufacture it, even if its ready to go that far... imagine.. if the r300 .15 can do 400+Mhz, and the r350 is "essentially" a shrunken r300, it should be quite a bit higher than 400MHz now, not just the 300MHz as before. everything falls right into place. And theres nothing to stop them from waiting a cycle (releasing a 400Mhz .15 R300 with faster ram in this cycle), then introducing the .13 with even faster ddr2 and core (because of process refinement in that extra cycle) than they originally expected.

get my point? dont know if i got it across or not. (if R300 yeilds clock better than expected, and the R350 is "based significantly" on the R300, then it stands to reason that the R350 would have higher clock yeilds than expected as well, not to mention being able to wait a cycle for 0.13u to become more reliable) by "originally expected" i mean when they made that statement in japan. from a manucaturing and supply point of view, a lot has happened since then.

--Braham
 
As ATi declared they will do both :

1) Launch a faster setup on the curent R300 design as a quick answer to the NV30 launch ... probably @ 400/750 ;

2) Launch the R350 in spring->May ;

And then , in autumn , the R400 will probably arrive .
 
David G. said:
As ATi declared they will do both :

1) Launch a faster setup on the curent R300 design as a quick answer to the NV30 launch ... probably @ 400/750 ;

2) Launch the R350 in spring->May ;

And then , in autumn , the R400 will probably arrive .

It really seems like the momentum is shifting to ATi...Nvidia better hope the NV30 is good, or they stand to lose a good deal of marketshare over the next couple years.
 
David G. said:
As ATi declared they will do both :

1) Launch a faster setup on the curent R300 design as a quick answer to the NV30 launch ... probably @ 400/750 ;

2) Launch the R350 in spring->May ;

When did ATi say anything about R350? Did I miss something?
 
Who cares really, don't you think fear of what another company has to offer will Nvidia going? To me it doesn't matter, as long as the price is low and the performance is high. These company's have to know that we don't all have 400 bucks to spend on a new fandangled gadget. Well... unless we love our credit card company's that much.
 
Above said:
Ailuros said:
PS: How old is Anand now anyway?
Old enough to have a grown-up bank account and not want to tell you what money goes in there.

I didn't or wouldn't have asked anyway about his financial situation. Your point?
 
I remember reading about Microsoft developing DX9.1 somewhere but I don't remember where.I think it's due out late spring,but some of the better informed people here can help you more on this.
 
Back
Top