NVIDIA and AMD GPUs can finally combine memory

DirectX logoBack in 2009, NVIDIA and AMD GPUs were allowed to combine their resources together for the first time ever, bypassing the need for AMD CrossFire and NVIDIA SLI technologies. Or at least that was the plan.

Unfortunately, the controller never managed to make history, as there was little performance gain, but instead you have plenty of graphical artifacts, unstable gameplay, poor game support and overall poor drivers.

Add to that a pretty high price tag for integrating the technology into a motherboard and it all turned into a flop.

With time, the technology might have been refined, but the failure to turn the world on its head from the start heralded the early death of that idea. Even though AMD CrossFire and NVIDIA SLI are hardly perfect performance stacking solutions.

NVIDIA and AMD GPUs can finally combine memory

Essentially, you take two video boards powered by a similar GPU architecture and install them side by side on a motherboard.

However, the memory of the video boards does not actually stack. So if you use two video boards with 4 GB VRAM each, the configuration doesn't actually give the PC 8 GB memory to work with. Not really. Because of this, the performance gains of SLI or CrossFire are usually of around 50%, if that.

Now, though, this may change, and it's not so much because of anything AMD and NVIDIA did, but because of low level application programming interfaces.

Mantle is the application programming interface that AMD developed for games, giving them higher-end graphics for better performance.

DirectX 12 is the latest version of Microsoft's DirectX and the API that NVIDIA-optimized games will use from now on.

These are low level APIs, which means that they define individual components, provide detail rather than overview, and deal with rudimentary functions instead of complex overall ones. Thus, they are concerned more with individual system components and how they operate.

High-level APIs, in comparison, are abstracted and define overall goals and systemic features, dealing with the system as a whole or with large parts of it at once.

As low-level APIs, Mantle and DX12 should allow game developers to enable buffer stacking and do things like split-frame rendering (each GPU and respective renders would handle half the screen). According to AMD representative Robert Hallock at least.

Alas, while this doesn't mean that games will suddenly be able to use a stacking memory approach (and thus, reach better performance), developers will have to configure the games to allow for it.

Source: Softpedia

Tags: AMD, DirectX, NVIDIA, videocards

Add comment

Your name:
Sign in with:
Your comment:

Enter code:

E-mail (not required)
E-mail will not be disclosed to the third party

Last news

Galaxy Note10 really is built around a 6.7-inch display
You may still be able to download your content
Facebook, Messenger and Instagram are all going away
Minimize apps to a floating, always-on-top bubble
Japan Display has been providing LCDs for the iPhone XR, the only LCD model in Apple’s 2018 line-up
The 2001 operating system has reached its lowest share level
The entire TSMC 5nm design infrastructure is available now from TSMC
The smartphone uses a Snapdragon 660 processor running Android 9 Pie
The Samsung Galaxy A5 (2017) Review
The evolution of the successful smartphone, now with a waterproof body and USB Type-C
February 7, 2017 / 2
Samsung Galaxy TabPro S - a tablet with the Windows-keyboard
The first Windows-tablet with the 12-inch display Super AMOLED
June 7, 2016 /
Keyboards for iOS
Ten iOS keyboards review
July 18, 2015 /
Samsung E1200 Mobile Phone Review
A cheap phone with a good screen
March 8, 2015 / 4
Creative Sound Blaster Z sound card review
Good sound for those who are not satisfied with the onboard solution
September 25, 2014 / 2
Samsung Galaxy Gear: Smartwatch at High Price
The first smartwatch from Samsung - almost a smartphone with a small body
December 19, 2013 /

News Archive



Do you use microSD card with your phone?
or leave your own version in comments (15)