ATI R600 gets leaked on the web

[kenc51]

http://level505.com/2006/12/30/the-full ... 00-test/1/

The R600 Specs are:

* 64 4-Way SIMD Unified Shaders, 128 Shader Operations/Cycle
* 32 TMUs, 16 ROPs
* 512 bit Memory Controller, full 32 bit per chip connection
* GDDR3 at 900 MHz clock speed (January)
* GDDR4 at 1.1 GHz clock speed (March, revised edition)
* Total bandwidth 115 GB/s on GDDR3
* Total bandwidth 140 GB/s on GDDR4
* Consumer memory support 1024 MB
* DX10 full compatibility with draft DX10.1 vendor-specific cap removal (unified programming)
* 32FP internal processing
* Hardware support for GPU clustering (any 2^n number, not limited to Dual or Quad-GPU)
* Hardware DVI-HDCP support (High-Bandwidth Digital Content Protection)
* Hardware Quad-DVI output support (Limited to workstation editions)
* 230W

Benchmarks here
http://level505.com/2006/12/30/the-full ... 00-test/1/

Credit goes to the Inq
http://theinquirer.net/default.aspx?article=36642

[HONkUS]

thats interesting. my only worries are that this is from the Inq and that the Oblivion benchmark is wrong. I have an 8800 GTS and I get around 43 fps with everything maxed out at 1680x1050 im going to try out my 8800 GTS again on some of those other games but these seem fishy to me

[dicecca112]

over at Xs they pointed out that almost all those benchs are off, and Dailytech pointed out the owner and creater of the site is indeed a DAAMIT employee. So take th is all with a grain of salt

If Level505's comments seem a little too pro-ATI, don't be too surprised. When asked if the site was affiliated in any way to ATI or AMD, the owner replied to DailyTech with the statement that "two staff members of ours are directly affiliated with AMD's business [development] division."

[liqnit]

It could be an off spin of ATI marketing...
But if so and the banches are wrong they will get too much bad publicity i dont think it worth it

[HONkUS]

well its DAAMIT propaganda then because those two employees would be nuts to post this and hope to keep their jobs right? Also anyone even heard of this site before? its strange that this is the ONLY review of anything they have on their entire site lol.

Also their reason for not showing us the card is BS because anyone that can put together those benchmarks can surely photoshop the mentioned "ID Tags" out of the photo.

im calling total BS!

[HONkUS]

Looks like the backfire is complete the site is now down completely!

Way to embarrass yourself DAAMIT!

[prozac26]

http://level505.com/2006/12/30/the-full ... 00-test/1/

The R600 Specs are:

* 64 4-Way SIMD Unified Shaders, 128 Shader Operations/Cycle
* 32 TMUs, 16 ROPs
* 512 bit Memory Controller, full 32 bit per chip connection
* GDDR3 at 900 MHz clock speed (January)
* GDDR4 at 1.1 GHz clock speed (March, revised edition)
* Total bandwidth 115 GB/s on GDDR3
* Total bandwidth 140 GB/s on GDDR4
* Consumer memory support 1024 MB
* DX10 full compatibility with draft DX10.1 vendor-specific cap removal (unified programming)
* 32FP internal processing
* Hardware support for GPU clustering (any 2^n number, not limited to Dual or Quad-GPU)
* Hardware DVI-HDCP support (High-Bandwidth Digital Content Protection)
* Hardware Quad-DVI output support (Limited to workstation editions)
* 230W

Benchmarks here
http://level505.com/2006/12/30/the-full ... 00-test/1/

Credit goes to the Inq
http://theinquirer.net/default.aspx?article=36642

That's fake. Level505 site just went up, so can't really see them getting a R600 to test. No pics also..

[DeusEx]

looks fake to me. they said they cant show pictures of the cards because of the tag numbers? here is an idea for you: photoshop. it takes 5 minutes and the number could be blacked out.

[werty316]

Until the R600 is out in out in the market I would take this review with a grain of salt.

[cyberneticimplant]

I think it is fake since there is no pictures.

[sushrukh]

The R600 looks considerably faster than the G80.I doubt if that test was valid though coz no pics were there during the entire test.

[Kougar]

I forgot about that "leak". Regardless of if you believe the results are faked or completely real... as far as I know ATI has not specificed the core clock speed on R600, and it's not listed in the given specs. With the latest A15 revision I've even read 1ghz was "possible", if only as a ultra pricy crème de la crème model.

So, if we take these tests at face value, we can expect them to actually show R600 with an disadvantage. The Retail R600XTX will very likely have a faster clock speed than their sample, and somewhat more refined drivers to match.

This is also ignoring that nVidia has separate shader clock speeds on their 8800GTX and 8800GT cards, so far I've not heard either way if ATI will have separate shader clock speeds on their parts or not.

[atlr]

Here's a new wrinkle in R600 rumors.
I look forward to the unveilings over the next few months.

http://www.ocworkbench.com/2007/article ... rds/g2.htm

The X2800GT looks like a good value.

[Dresden]

Saving my $$$, the 1 gig X2800XTX looks very good, hopefully the Intel price drop happens earlier then expected...

Drool


Dresden

[largon]

Those are quite strange wattage figures; 270W for the XTX?
That's funny because PCIe x16 slot can deliver 75W, 8pin: +100W and 6pin : +75W.

250W total...