AMD: R7xx Speculation

Status
Not open for further replies.
(Deleted my post since CJ was faster.) CJ, you forgot that X800 vanilla/Pro/GT/GTO were all partially deactivated chips and might I add the Radeon X550 XT and X1650 XT.
 
I misworded my sentence.
By scalar values I meant scalar threads.
I was running out of words for what to call things once I nested elements within a batch and nested items within the groups of 5.

By your description of the operand restrictions, it would be more likely that it would be 4 threads inside each batch item in a 64 element batch, hence emulating a 256-length vector machine.
Yeah, I think 4 scalars per element normally works out quite well, e.g. single precision optimised matrix multiply computes 32 scalars per invocation and the main compute loop has an average ALU utilisation of 4.4 per clock.

Multiples of 4 also works out nicely in terms of issuing TEX instructions, since each TEX clause is a maximum of 8 instructions.

Oh and this shader allocates 39 vec4s (plus 1 temporary) :oops:

Jawed
 
=>Sunday: Just a guess, but if GDDR5 is in fact QDR instead of DDR, which it may very well be, and GPU-Z detects it like that, then 900 × 4 = 3600 MHz
 
first time hear?! where can you buy these cards?


speaking about the mystery… what’s this card with 750 MHz GPU clock and 900MHz memory clock?

http://www.tweaktown.com/news/9691/index.html

They were limited to Asia.

Basically a Radeon HD 3800 with 128-bit memory bandwidth.

=>Sunday: Just a guess, but if GDDR5 is in fact QDR instead of DDR, which it may very well be, and GPU-Z detects it like that, then 900 × 4 = 3600 MHz

Makes sense. Wasn't the rumor that the Radeon HD 4870 was using the slowest GDDR5 at 1.800Mhz?
 
Yeah, but with this approach reviewers media are spreading skewed picture about Crysis playability on particular card.
So far I only noticed that TR has adopted “in-game” measurement of performance, and these scores from GTX 280 review are extremely interesting if you compare GTX260 vs. HD3870 X2!
crysis-high-1920.gif

With all the info about RV770 that we know so far, HD4850 could very well be faster than HD3870 X2! This means that 200 USD HD4850 could be better choice for Crysis then 400 USD GTX260! Wouldn’t that be a TWI-NOT-MTBP :D

But then we have this:

http://www.extremetech.com/image_popup/0,,iid=209888&aID=228640&sID=27867,00.asp

Even the 260 is beating the GX2 here. Perhaps its down to DX10 performance. The benchmark you posted above is in DX9 while the extremetech one is DX10.
 
Do Crysis benchmarks in High Quality or Very High Quality mode have AF enabled?

I'm surprised at the scores of the 3870 because I distinctly remember it getting creamed in 4xAA 16xAF settings a while ago.
 
Do Crysis benchmarks in High Quality or Very High Quality mode have AF enabled?

I'm surprised at the scores of the 3870 because I distinctly remember it getting creamed in 4xAA 16xAF settings a while ago.
If you're referring to the Techreport tests, that remains true... in the flyby.

Apparently, for in-game tests, ATI does a lot better.
 
Do Crysis benchmarks in High Quality or Very High Quality mode have AF enabled?

I'm surprised at the scores of the 3870 because I distinctly remember it getting creamed in 4xAA 16xAF settings a while ago.

Well, just remember that ATI cards are like a wine. They age very well with the drivers. :)
 
HD4850 vs HD4870 :

Xtreme preset :

HD4850 vs 4870

GT1 : 8.62 - 12.57
GT2 : 7.13 - 12.72

FT1 : 651.67 - 779.86
FT2 : 3.41 - 5.49
FT3 : 18.11 - 20.58
FT4 : 14.23 - 17.02
FT5 : 27.59 - 33.58
FT6 : 48.92 - 53.73

GPU Score : 2692 - 4316

Source : TweakTown

http://www.tweaktown.com/news/9691/index.html

From where i got the results for comparison with HD 4850 :

http://img101.imageshack.us/img101/3754/vantagextremestock2808tn4.jpg
http://img101.imageshack.us/img101/8016/vantageud4.jpg

Look at the pixel fill rate test 5.49 GP/s . More than 16 ROPs ?
 
New article on Tweaktown:

The thing is though, I’ve tested the HD 4850, and what’s interesting is that the card is going to be a mid-range solution. But in Crossfire? - Wow! - With the amount of people who have CF boards as well, thanks to the Intel CPU and chipset being the products of choice, this is going to be a great solution.

...

The GTX 280 is slower than a pair of CrossFired 4850s for the most part; it’s also $250 AUD dearer. The price on the HD 4850 in AUD is around $250, so $500 for a pair while the GTX 280 is starting at $750.

....


As frustrated at NVIDIA as I am, who has really annoyed me is ATI. The fact that the NDA has been pushed back is just killing me. Knowing the results on the HD 4850 has completely changed my outlook on the GTX 280. At the moment, a lot of review websites have compared the GTX 280 to the 9800 GX2 and they see the performance being similar with a single core, then they see the new technology that’s implemented into the card and find it all to be pretty amazing. Performance talks though, and if these people had seen the HD 4850 run in CrossFire, I don’t doubt that their opinion on the cards would have likely changed.


http://www.tweaktown.com/articles/1467/how_nvidia_stuffed_the_gtx_280/index.html
 
Status
Not open for further replies.
Back
Top