-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
***The Official Vulkan API Thread***
- Thread starter shankly1985
- Start date
Think someone had posted that on reddit Greg, I thought I was having a da ja vue moment
I posted it on reddit
Thanks Greg.
FPS are broadly similar, its only when the FPS are higher that Vulkan gets more.
I don't think anything there would be GPU bound, is it?
GPU sits at 99% all the time (I have a G19 keyboard that displays AB), so not CPU bottlenecked anyways.
Greg perhaps a better understanding of where the low level API is working would be to compare them both at 1080P with high tho not highest settings....
We are looking for CPU bound performance.
Running it at low settings often turns off features dependant on CPU calc so that would be a good measure either.
We are looking for CPU bound performance.
Running it at low settings often turns off features dependant on CPU calc so that would be a good measure either.
Overall easier to work with than OpenGL yes.
Greg perhaps a better understanding of where the low level API is working would be to compare them both at 1080P with high tho not highest settings....
We are looking for CPU bound performance.
Running it at low settings often turns off features dependant on CPU calc so that would be a good measure either.
It is still pretty much single threaded in truth and just a wrapper in The Talos principle according to the devs, so hopefully they will keep at it and get it multi threaded and then it would be a better test and I will redo at a very low res.
Look I don’t disagree Vulkan is based on Mantle. What I disagree with is the majority of work was done by AMD and that the majority of the credit belongs to AMD. There was a large amount of intensive work done by other vendors to make Vulkan what it is.
Vulkan isn’t just Mantle with a few small changes and upgrades. Look at the article you linked. “Khronos has worked to make it very clear that multiple vendors are responsible for contributing IP that ultimately went into Vulkan.”
AMD gave Mantle as a baseline to start work on. The other vendors then contributed major IP to fix all the problems which Mantle had. Yes AMD deserve credit but why should AMD get all the credit? AMD created the baseline but they did not do the majority of the work. Its kind of like AMD creating the foundations of a building and getting the credit for doing the entire building when all they did was the bottom 10%. The other 90% deserve credit as well.
Last edited:
Caporegime
- Joined
- 9 Nov 2009
- Posts
- 25,303
- Location
- Planet Earth
And the article says Vulkan is BASED on the Mantle framework. The other vendors are contributing to modifications to the Mantle framework which means it can run cross platform.
Its almost like saying that since Seat,Skoda and VW use VW platforms and engines that VW did not do much of the important work in defining the common platform and the engines it contains. Its like saying the Mitsubishi F2 which is DERIVED from the F16 with significant modifcations to the wings and massive changes to the avionics,did not owe a lot of what it is down to what the F16 did before.
AMD spent a lot of R and D time in getting Mantle out by themselves,and if it were not for the fact they gave it away for FREE,where do you think Vulkan would be now??
It most likely would be still in the planning stages NOW,and probably would have cost more LONG TERM.
The whole direction of where Vulkan was heading was altered when AMD gave Mantle away - if using your logic of AMD not contributing that much to Vulkan were true,then why did the Khronos Group jump at the opportunity to use Mantle as the basis of Vulkan??
All that R and D prior was just ditched for the new direction Mantle pushed it in.
It was obvious the original version of OpenGL Next was going nowhere!
This is what you don't get - AMD could have just made Vulkan die a death,especially since Vulkan benefits Linux users the most,an area AMD does not give a damn about going by their poor Linux support,and kept Mantle only for their VR stuff where it would be more useful,now that DX12 will be the defacto standard for most Windows games,and will do most of what Mantle promised for Windows and probably will be more amenable to the AMD GCN uarch than DX11 was.
Mantle was realistically more a bandaid for AMD DX11 performance and for devs to understand to get the most out of AMD GPUs,which will probably help when DX12 comes along.
Last edited:
Caporegime
- Joined
- 9 Nov 2009
- Posts
- 25,303
- Location
- Planet Earth
I believe a while back I read on a website a comment by Nvidia on how important Mantle was as a contribution to Vulkan,so I am still not sure what Pottsey is argueing about when the head of the Khronos Group is actually a senior Nvidia bloke called Neil Trevett. The fact that the Khronos Group jumped at the chance of using Mantle as the basis of Vulkan instead of just developing something else like Apple did with Metal(who is a member too BTW),indicates how important a contribution it was for them to base Vulkan on it. Why would an important member like Nvidia want to be using competitors tech?? They obviously thought it was important enough to agree to use it.
I am sure Khronos Group members like Nvidia,Intel and Google between them have more resources to make a clean sheet API than taking something AMD has worked on with their limited funds and only functioned on their newer cards,and was a bit hit and miss due to its experimental nature,and adapting it for their own means.
Last edited: