Jump to content

They do most of the computational work. You can't compare CUDA core for CUDA core nor stream processor for stream processor because it all comes down the architecture that's being used. CUDA cores and stream processors cannot be compared directly.

"It pays to keep an open mind, but not so open your brain falls out." - Carl Sagan.

"I can explain it to you, but I can't understand it for you" - Edward I. Koch

"I didn't die! I performed a tactical reset!" - Apollolol

Link to post
Share on other sites

CUDA is Nvidia's form of cores, this is also one of the more popular languages/ API used in compute tasks.

Stream are AMD's form of cores, better at OpenGL iirc.

Just remember: Random people on the internet ALWAYS know more than professionals, when someone's lying, AND can predict the future.

i7 9700K (5.2Ghz @1.2V); MSI Z390 Gaming Edge AC; Corsair Vengeance RGB Pro 16GB 3200 CAS 16; H100i RGB Platinum; Samsung 970 Evo 1TB; Samsung 850 Evo 500GB; WD Black 3 TB; Phanteks 350x; Corsair RM19750w.

 

Laptop: Dell XPS 15 4K 9750H GTX 1650 16GB Ram 256GB SSD

Spoiler

sex hahaha

Link to post
Share on other sites

So, if a person is just gaming, will they see any performance difference on one vs the other? What about rendering videos?

You can't compare them directly as I already stated. It comes down to the architecture - certain applications may favor NVIDIA cards or favor AMD cards. AMD cards tend to excel at OpenGL driven applications.

"It pays to keep an open mind, but not so open your brain falls out." - Carl Sagan.

"I can explain it to you, but I can't understand it for you" - Edward I. Koch

"I didn't die! I performed a tactical reset!" - Apollolol

Link to post
Share on other sites

CUDA is Nvidia's form of cores, this is also one of the more popular languages/ API used in compute tasks.

Stream are AMD's form of cores, better at OpenGL iirc.

 

To clarify though, all cores from both vendors are considered stream processors, they just both use different methods of computing.

 

It's probably more complicated than that, but that's how I understand it in my little mind.

Link to post
Share on other sites

Ok, so, what are ROPs?

ROP stands for Raster Operation Pipeline and its function is below

ROP aka render output unit is what produces the final pretty pixel for the frame buffer. All the shading and texture goop is blended, transformed and what not to the final pixel value displayed on your screen by such a thing.

http://forums.anandtech.com/showthread.php?t=1829459

"It pays to keep an open mind, but not so open your brain falls out." - Carl Sagan.

"I can explain it to you, but I can't understand it for you" - Edward I. Koch

"I didn't die! I performed a tactical reset!" - Apollolol

Link to post
Share on other sites

Thanks! So, more ROPs are better, yes?

Per the same architecture, yes.

"It pays to keep an open mind, but not so open your brain falls out." - Carl Sagan.

"I can explain it to you, but I can't understand it for you" - Edward I. Koch

"I didn't die! I performed a tactical reset!" - Apollolol

Link to post
Share on other sites

So, if a person is just gaming, will they see any performance difference on one vs the other? What about rendering videos?

I wouldn't say it's something you should look at for the performance. For rendering videos it depends on the editor. Vegas likes OpenCL a little more then cuda while Premiere pro like Cuda a little more then OpenCL

Computing enthusiast. 
I use to be able to input a cheat code now I've got to input a credit card - Total Biscuit
 

Link to post
Share on other sites

So, are ROPs basically what makes the FPS difference between something like the GTX960 and the GTX980Ti?

ROPs definitely play a part, but CUDA cores are also important. In the case of something like a 660 Ti vs 760, you will notice the 760 has less CUDA cores but more ROPs. Because of this, it generally beats the 660 Ti in most titles. However, in compute intensive applications, the 660 Ti may pull ahead due to the higher amount of CUDA cores.

 

Edit: Small corrections made.

Edited by Godlygamer23

"It pays to keep an open mind, but not so open your brain falls out." - Carl Sagan.

"I can explain it to you, but I can't understand it for you" - Edward I. Koch

"I didn't die! I performed a tactical reset!" - Apollolol

Link to post
Share on other sites

ROPs definitely play a part, but CUDA cores are also important. In the case of something like a 660 Ti vs 760, you will notice the 760 has less CUDA cores but more ROPs. Because of this, it generally beats the 660 Ti in most titles. However, in compute intensive games, the 660 Ti will pull ahead due to the higher amount of CUDA cores.

Example of a compute intensive game?

Link to post
Share on other sites

Example of a compute intensive game?

Slight correction. It's compute intensive applications, not games. Even so, it appears to only be in CLBenchmark 1.1 in fluid simulation.

http://www.anandtech.com/show/7103/nvidia-geforce-gtx-760-review/15

"It pays to keep an open mind, but not so open your brain falls out." - Carl Sagan.

"I can explain it to you, but I can't understand it for you" - Edward I. Koch

"I didn't die! I performed a tactical reset!" - Apollolol

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×