Jump to content
  • Advertisement
Sign in to follow this  
LockOnStrattoss

GDDR5X Vram question!

This topic is 514 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I had a discussion with some friends, i believed that GDDR5X and HBM render the 3D scene much faster in games than a normal GDDR5 because of the double prefetch that they do. I was curious if i am right.

 

Define "much".

You simply can't make a blanket statement such as "GPU A with 5X is always going to be faster than GPU B with 5" because there are so many other variables, not least of which is the type of workload you have (a workload that doesn't stress the memory subsystem is obviously not going to show a large, if any, perf increase from faster memory).

If you're looking to settle an argument with your friends, then you're both right and you're both wrong.  Sometimes higher memory speed is important, sometimes higher clock speed, sometimes more shader units, sometimes more cache, sometimes more bandwidth, sometimes two or more of these, and it all depends on how the workload is tuned.

Share this post


Link to post
Share on other sites
Advertisement
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!