Endorfy Arx 700 Air chassis review
Beelink SER5 Pro (Ryzen 7 5800H) mini PC review
Crucial T700 PCIe 5.0 NVMe SSD Review - 12GB/s
Sapphire Radeon RX 7600 PULSE review
Gainward GeForce RTX 4060 Ti GHOST review
Radeon RX 7600 review
ASUS GeForce RTX 4060 Ti TUF Gaming review
MSI GeForce RTX 4060 Ti Gaming X TRIO review
GeForce RTX 4060 Ti 8GB (FE) review
Corsair 2000D RGB Airflow Mini-ITX - PC chassis review
Preview: AMD VEGA GPU Architecture
It is time to talk through a few mysteries from the AMD VEGA graphics processor architecture, join us in a preview architecture breakdown of what is to be expected with AMD VEGA graphics processors.
Read the editorial right here.
« Deepcool Demonstrates Some pretty Nice Products on CES · Preview: AMD VEGA GPU Architecture
· Corsair Launches Flagship K95 RGB Platinum Mechanical Gaming Keyboard »
Preview: Desktop Kaby Lake Core i5 7600K CPU - 12/17/2016 12:31 PM
We look at the Intel Core i5 7600K processor, a respin from Intel based on the 14nm node; these processors are more energy friendly. For this review we look at the performance of this processor in a w...
Preview: AMD Radeon R9 NANO - 08/27/2015 01:56 PM
At the launch of the Radeon R9 Fury and Fury X AMD also announced something that caught the interest of many, a small form factor yet high-end graphics card aimed at powerful small game and HTPCs, yes...
mR Yellow
Senior Member
Posts: 1935
Joined: 2005-02-02
Senior Member
Posts: 1935
Joined: 2005-02-02
#5378720 Posted on: 01/06/2017 09:04 AM
What if the card demo'ed was VEGA but a mid-range version?
What if the card demo'ed was VEGA but a mid-range version?
moeppel
Senior Member
Posts: 153
Joined: 2015-06-30
Senior Member
Posts: 153
Joined: 2015-06-30
#5378721 Posted on: 01/06/2017 09:11 AM
It's "OpenGL" in the sense that Vulkan will be OpenGL's successor but beyond that they have little to nothing in common outside of being an open API that's cross-system-compatible.
Vulkan games could natively run on Linux, much like OpenGL brethren unlike DX counterparts. No emulation or black box reverse engineering necessary.
That said, Vulkan may also turn into "OpenGL 2" where it loses to DX so heavy handed that it'll sink into obscurity. It's another chance for developers and the market to remove proprietary DX and Windows enforcement, insofar Vulkan will be implementation friendly and competitive enough.
Going by the titles with Vulkan implementations (Doom & Dota 2 for instance) it certainly does show up (very) strong - all the more so on AMD cards when compared to DX11. That however also applies to most 'proper' DX12 titles.
Anyway, Mantle forced MS' hand to show up with DX12. If Mantle didn't come to live we might still be on the "DX11 forever" for eternity train. That perhaps would've suited Intel just fine with their "4 cores forever" policy I bet.
Beyond Valve and ID Software who were always quite open to open standards (i.e OpenGL) I'm not quite sure whether or not many, if any, will adapt Vulkan long term. At least there's hope.
AMD gave Mantle to Khronos Group, who then used it to create Vulkan.
It's "OpenGL" in the sense that Vulkan will be OpenGL's successor but beyond that they have little to nothing in common outside of being an open API that's cross-system-compatible.
Vulkan games could natively run on Linux, much like OpenGL brethren unlike DX counterparts. No emulation or black box reverse engineering necessary.
That said, Vulkan may also turn into "OpenGL 2" where it loses to DX so heavy handed that it'll sink into obscurity. It's another chance for developers and the market to remove proprietary DX and Windows enforcement, insofar Vulkan will be implementation friendly and competitive enough.
Going by the titles with Vulkan implementations (Doom & Dota 2 for instance) it certainly does show up (very) strong - all the more so on AMD cards when compared to DX11. That however also applies to most 'proper' DX12 titles.
Anyway, Mantle forced MS' hand to show up with DX12. If Mantle didn't come to live we might still be on the "DX11 forever" for eternity train. That perhaps would've suited Intel just fine with their "4 cores forever" policy I bet.
Beyond Valve and ID Software who were always quite open to open standards (i.e OpenGL) I'm not quite sure whether or not many, if any, will adapt Vulkan long term. At least there's hope.
EdInk
Member
Posts: 98
Joined: 2016-03-30
Member
Posts: 98
Joined: 2016-03-30
#5378739 Posted on: 01/06/2017 10:34 AM
My question is how much of all the new hardware features rely on software? Async compute was in the R9 290 and is barely used 4years later. Will buying Vega mean we get to utilise say, 70% of it true potential and the remaining is unlocked by well written code at a much later date?
I'd like AMD to demo the benefits of all these new feature. Its easy to confuse people with tech-speak.
My question is how much of all the new hardware features rely on software? Async compute was in the R9 290 and is barely used 4years later. Will buying Vega mean we get to utilise say, 70% of it true potential and the remaining is unlocked by well written code at a much later date?
I'd like AMD to demo the benefits of all these new feature. Its easy to confuse people with tech-speak.
Deleted member 164252
Unregistered
Unregistered
#5378748 Posted on: 01/06/2017 11:23 AM
^
New features like ACompute come with new APIs. There is not going to be anything new in the next years, except of already established, non-mandatory and not much supported SM6.
All of the "tech-speak" was about improving effectivity, utilization and performance of the GPU. And that is actually all that matters. Gaming GPUs are getting old because of speed, not because of missing feature sets. Just check benchmarks, check price and you are fine.
TBH even all those talks about what tech is better, whether AMD or NVIDIA are just for fun of it. It doesn't really matter... all comes down to speed vs price.
^
New features like ACompute come with new APIs. There is not going to be anything new in the next years, except of already established, non-mandatory and not much supported SM6.
All of the "tech-speak" was about improving effectivity, utilization and performance of the GPU. And that is actually all that matters. Gaming GPUs are getting old because of speed, not because of missing feature sets. Just check benchmarks, check price and you are fine.
TBH even all those talks about what tech is better, whether AMD or NVIDIA are just for fun of it. It doesn't really matter... all comes down to speed vs price.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 12613
Joined: 2011-10-22
AMD gave Mantle to Khronos Group, who then used it to create Vulkan.