'

Nvidia's working on its own version of AMD's Smart Access Memory tech

Nvidia's alternative should be coming to Ampere via future software updates

Nvidia's working on its own version of AMD's Smart Access Memory tech

Nvidia's working on its own version of AMD's Smart Access Memory tech

When AMD revealed its "Smart Memory Access" (SAM) technology, there was one question that came to most analysts. That question was "how will Nvidia respond". 

During discussions with Nvidia, Gamers Nexus has gotten an answer to this question. That answer is simple, "NVIDIA hardware supports this functionality and will enable it on Ampere GPUs through future software updates." Nvidia's representative also stated that the "resizable BAR" is part of the PCIe specification, making the technology something which Nvidia and Intel could enable with ease. AMD may have been first with Smart Access Memory, but they won't be the last to implement such a feature. 

Below is Gamers Nexus' Twitter thread on their conversations with Nvidia.  

 

Nvidia's working on its own version of AMD's Smart Access Memory tech  
What is Smart Access Memory? 

When discussing the matter, AMD claimed that their Smart Access Memory technology was a unique innovation which would be exclusive to systems with Radeon RX 6000 series graphics cards, AMD 500 series AM4 motherboards (with up to date firmware) and AMD's Ryzen 5000 series processors. The technology grants users of All-AMD systems increased access to GPU memory, delivering  CPU/GPU performance boost. 

On most Windows PCs, CPUs only have access to around 256MB of GPU memory, a tiny portion of their frame buffer. Smart Access Memory allows the CPUs of All-AMD systems to access the full frame buffer of modern Radeon graphics cards. 

AMD's internal data has shown that their compatible systems receive performance boosts of around 6% on average when Smart Memory Access is enabled, giving AMD-powered systems a performance edge over their rivals. With Smart Access Memory, AMD plans to convince system integrators to ship systems with Ryzen CPUs and Radeon graphics, and PC builders to invest in All-AMD hardware setups. 

Nvidia's solution will work with AMD and Intel processors?  

If Gamers Nexus' statement is correct, Nvidia's answer to AMD's SAM technology should work on all systems, regardless of your chosen CPU manufacturer. This makes the fact that AMD has tied their SAM technology to Ryzen 5000 series processors disappointing, though AMD's implementation may offer functionality which extends beyond BAR resizing. 

Right now, it remains to be seen whether or not Nvidia's answer to AMD Smart Access Memory will be on par with what AMD will offer. While Nvidia's statement has said that they "seeing similar performance results", that doesn't mean that they are receiving the same performance gains. 


Nvidia's working on its own version of AMD's Smart Access Memory tech  

Let's wait and see

Does AMD's Smart Access Memory offer something that Nvidia can't? Will Nvidia's solution be on par with what AMD's offering? Answers to these questions will come in time. 

Nvidia's response to AMD's SAM technology feels reactionary. AMD will be shipping products which support Smart Access Memory next week with their Radeon 6000 series, and Nvidia only has the vague promise of "future software updates". That said, AMD now needs to be clear about why Smart Access Memory is exclusive to Ryzen 5000 series and Radeon 6000 series products. If Nvidia can support a similar feature on all CPU platforms, why can't AMD? 

You can join the discussion on AMD's Smart Access Memory and Nvidia's response on the OC3D Forums

Nvidia's working on its own version of AMD's Smart Access Memory tech  

«Prev 1 Next»

Most Recent Comments

12-11-2020, 20:34:50

NeverBackDown
So is it known if this is feature of the the PCI-SIG spec is tied to Gen 4? Could be a reason we won't see it on older cards. If it is not, you should take into account the greater bandwidth of Gen 4 would allow more efficient memory access and lessen any overall latency if more than one fetch was needed over a smaller bandwidth.Quote

12-11-2020, 21:08:31

Kleptobot
Quote:
Originally Posted by NeverBackDown View Post
So is it known if this is feature of the the PCI-SIG spec is tied to Gen 4? Could be a reason we won't see it on older cards. If it is not, you should take into account the greater bandwidth of Gen 4 would allow more efficient memory access and lessen any overall latency if more than one fetch was needed over a smaller bandwidth.
From what i can tell the resizable BAR spec came in with PCIe 2.0. Expandable Resizable BAR came in with PCIe 3.0. The change was primarily concerned with how many bits were available to address memory space. Allowing DMA of larger address spaces across PCIe. There was no change to this spec in PCIe 4 at least non that has been published

I'm not sure why this wasn't implemented sooner really. Alternatively they could have chunked the graphics memory before this was a thing and had multiple DMA chunks across the whole or a larger section for the available graphics memory.

I suppose it comes down to, what the low hanging fruit was in terms of time to implement vs performance gain

Edit:
Did some more reading, its to do with addressable memory space in the OS being shared between system memory and GPU memory. 32Bit OSes couldn't support it unless the addressed system memory took a hit, so its only now with the widespread adoption of 64bit that this is feasible.Quote

13-11-2020, 04:08:27

g0ggles1994
Like I said before, this needs to be CPU agnostic. Nvidia will make it so because they don't have any CPUs to support; granted don't go out their way to support older CPU generations but make it available to newer ones, both AMD and IntelQuote

13-11-2020, 04:40:12

tgrech
It did sound like AMDs version was more than just the expandable BAR, making use of their large last level caches on their latest parts for caching the CPU-GPU transfers to improve effective bandwidth and reduce average latency one would assume, but if a good chunk can be gained from making use of the spec alone I assume AMD would need to follow up with that option too now.Quote

13-11-2020, 04:58:34

looz
Quote:
Originally Posted by g0ggles1994 View Post
Like I said before, this needs to be CPU agnostic. Nvidia will make it so because they don't have any CPUs to support; granted don't go out their way to support older CPU generations but make it available to newer ones, both AMD and Intel
They just bought ARM, be prepared for a full Nvidia system.... In 2030 Quote
Reply
x

Register for the OC3D Newsletter

Subscribing to the OC3D newsletter will keep you up-to-date on the latest technology reviews, competitions and goings-on at Overclock3D. We won't share your email address with ANYONE, and we will only email you with updates on site news, reviews, and competitions and you can unsubscribe easily at any time.

Simply enter your name and email address into the box below and be sure to click on the links in the confirmation emails that will arrive in your e-mail shortly after to complete the registration.

If you run into any problems, just drop us a message on the forums.