Overclock.net banner
1 - 6 of 6 Posts

·
Registered
Joined
·
140 Posts
Discussion Starter · #1 ·
I was just looking at specs for the GTX 260 and 280s, and the 4850s and 4870s. On New Egg their Stream Processors category seems to be quantified differently. nVidia's seems to have a lower number always, but listed as Processing Cores, while ATI's have a higher number of Processing Units.
Sorry if this topic has been covered, I searched for an answer but was able to find nothing.
I guess all I'm asking is what is the difference between the two.
Thanks,
Victor
 

·
Registered
Joined
·
1,413 Posts
The extra stream processors give you looser anti alising (sp) That way you dont rob your games of speed for the AA.
 

·
Banned
Joined
·
5,411 Posts
You can't compare them. Their two different technologies used to accomplish the same task, but in different ways.

I'll wait for someone who knows what their talking about to explain further.
 

·
Iconoclast
Joined
·
31,768 Posts
NVIDIAs stream processors do more work per clock, and are clocked quite a bit higher.

ATI's equivalent does less work perclock, are clocked slower, but are apparently small enough to fit ass loads on the die.

They both do very similar things, and as benchmarks show, both strategies are rather comparable in final performance.

Also, stream processors usually have nothing to do with AA (except in cases of differed shading). AA performance is limited by the number and clock of the ROPs, and to a lesser extent, in most cards, memory bandwidth.
 

·
Registered
Joined
·
1,655 Posts
ATI has Stream processors arranged in units of five to make up a 1 5-isssue shader (800 stream processors /5=160 Shaders). IIRC each one of those stream processors can do 1 operation-per-clock. Nvidia has a more straight forward approach with 1 shader= 1 shader. However you can't compare ATIs and Nvidias numbers.

Quote:

Originally Posted by Blameless View Post
NVIDIAs stream processors do more work per clock, and are clocked quite a bit higher.

ATI's equivalent does less work perclock, are clocked slower, but are apparently small enough to fit ass loads on the die.

They both do very similar things, and as benchmarks show, both strategies are rather comparable in final performance.

Also, stream processors usually have nothing to do with AA (except in cases of differed shading). AA performance is limited by the number and clock of the ROPs, and to a lesser extent, in most cards, memory bandwidth.
ATI moved AA operations from ROPs to the Stream Processors with R600. On nvidia cards AA is still handeled by ROPs. However on ATI cards the ROPs still have a impact on AA performance.(Improved pixel fill-rate increased AA performance with Rv770).
 

·
Registered
Joined
·
140 Posts
Discussion Starter · #6 ·
Thanks for the quick replies, makes more sense now
 
1 - 6 of 6 Posts
Top