AMD: There is no such thing as full support for DX12 today

by Hilbert Hagedoorn on: 09/01/2015 09:03 AM | source: | 72 comment(s)

After yesterdays turmoil on a lacking DX12 feature for Nvidia AMD’s Robert Hallock shares that Fury X is also missing a number of DX12 features.

The good man replied in a Reddit thread on the DX12 Async shader/compute feature that is missing from NVIDIA’s graphics cards, and then claimed that there is no such thing as “full support” for DX12 on the market today. Which obviously was already known as AMD never claimed full DX12 support on all feature levels with their GCN architecture

“I think gamers are learning an important lesson: there’s no such thing as “full support” for DX12 on the market today.” said Robert and continued:

“There have been many attempts to distract people from this truth through campaigns that deliberately conflate feature levels, individual untiered features and the definition of “support.” This has been confusing, and caused so much unnecessary heartache and rumor-mongering.

Here is the unvarnished truth: Every graphics architecture has unique features, and no one architecture has them all. Some of those unique features are more powerful than others.

Yes, we’re extremely pleased that people are finally beginning to see the game of chess we’ve been playing with the interrelationship of GCN, Mantle, DX12, Vulkan and LiquidVR.”

When somebody asked what are the aspects of DX12 that the FuryX is missing, Hallock replied and listed them.

“Raster Ordered Views and Conservative Raster. Thankfully, the techniques that these enable (like global illumination) can already be done in other ways at high framerates (see: DiRT Showdown).”

So it is simple, currently no graphics card with full 100% DirectX 12 support, that means that some games are to favor AMD and others Nvidia.

I want to add thing here, yesterday Hallock was all over this downplaying Nvidia and evangelizing how good their GPUs are, and now he is taking a step back with these answers on Reddit.

It's all marketing mud-fighting and attacking each other these days in-between Nvidia and AMD. Fun fact: on the AMD GPU Tech day for Fury, I myself literally confronted and asked about the DX12 supported feature levels to Hallock, and in this case Hallock himself absolutely refused to give a valid answer at the time as he very well knew that AMD would not fully support DX12 either.

5 pages 1 2 3 4 5

airfathaaaaa
Unregistered

#5151591 Posted on: 09/01/2015 12:23 PM

What did I say in the other thread?

Just ignore the junk from both sides. Sad they have to resort to petty crap. Or it's just trolls making **** up..and posting benchmarks and tables.

We seen it before. Hardly original.

yeap im a troll now
prove me wrong then on the specs of mx cards 2.0 as far as everyone knows ***** is real pretty much every card below titan x and 980ti(suspected to have 2 dma's) is gimped for sure
if that is a troll post then by any means report it

there is a reason that everyone is going mad on nvidia forums and no its not because of trolls

Lane
Senior Member

Posts: 6361
Joined: 2005-02-25

#5151595 Posted on: 09/01/2015 12:28 PM

Something to look, theres a bit of problem on the MS front.

Many features who was optional for 12_1 have suddenly moved as essential, and even some features as 12_0 who was optionnal have then suddenly moved back to non optionnal or vice versa... ( let alone the problem with Tier1-2-3 ) who move up and down from needed to optionnal.

Maybe MS should have look at the 12_0-12_1 a bit differently and take all features supported by all, and then put the other on the optional side.

Even today, there's not one driver from AMD, Intel and Nvidia who expose all features available on their architecture
..
Each driver we find new cap-bit who was not exposed or only partially.

example: Cat 15.8beta: - Geometry Shader bypass performance cap-bit finally exposed.

pharma
Senior Member

Posts: 995
Joined: 2003-09-10

#5151612 Posted on: 09/01/2015 12:48 PM

While this is inconvenient that some active fella changes it over again, he only adds stuff to it, and he is not altering originally present information.

He added Async shaders, cross-node sharing, UAVs at every stage, maximum sample count for UAV-only rendering, Logical Blend Operations.

Well, and he removed MS's reference. But checking what he added, they are mostly same for AMD and nV, so it does not look like he is taking sides.
(Tho, I would state for async shader on nV as partial, since it works. Just not as intended => may mean poor implementation.)

Currently there is only one source of the table and that is at Wiki. A slew of experts modify the Wiki table from MS, Intel, Nvidia and AMD on a daily basis. Do you really think his "cut & paste" interpretation of what the table should look like is in some way superior to theirs?

Considering his resulting table it's understandable why he did not include a link to the source.

omnimodis78
Member

Posts: 90
Joined: 2010-07-30

#5151656 Posted on: 09/01/2015 01:41 PM

It is precisely these sort of things that make me appreciate the stupidity of the typical consumer. You buy something and expect it to support 100% a feature which goes live a year later, and even then it's insignificant because the games won't even exist for some time after that. But the "green team" and "red team" go on the defensive/offensive muddying the issue with feelings and uneducated opinions. Yeah not much quantitative facts in my post either, but seriously, can your GPU play current gen games perfectly? Yes. Does AMD show advantage is one single AMD-preferred demo? Absolutely. Does it matter at all?

Dazz
Senior Member

Posts: 771
Joined: 2001-08-12

#5151657 Posted on: 09/01/2015 01:41 PM

After reading this : https://www.reddit.com/r/nvidia/comments/3j5e9b/analysis-async-compute-is-it-true-nvidia-cant-do/ and running the DX12 async compute test program. I would say Maxwell can do async compute and do it well as long as the command list is below 31 (GCN can do 128 so that's the difference - but until 31 - 980 Ti is way faster)

Really i thought i read that Maxwell is meant to be able to do 32, while GCN 1.0 is 2, GCN 1.1 is 4 and GCN 1.2 is 8.

Denial
Senior Member

Posts: 12123
Joined: 2004-05-16

#5151666 Posted on: 09/01/2015 01:45 PM

Lol that reddit thread, did you guys even read the comments? Seems like no one even knows how to interpret the results including the guy who wrote the program.

Barry J
Senior Member

Posts: 2759
Joined: 2006-02-23

#5151670 Posted on: 09/01/2015 01:49 PM

Interesting read DX12 is going to be interesting

Ryu5uzaku
Senior Member

Posts: 6732
Joined: 2006-09-24

#5151671 Posted on: 09/01/2015 01:53 PM

Really i thought i read that Maxwell is meant to be able to do 32, while GCN 1.0 is 2, GCN 1.1 is 4 and GCN 1.2 is 8.

It is 1 + 2 mixed and 1 + 8 and 1 + 8 and compute only 2, 8 and 8 (Tho 260 series gcn 1.1 seems to be 1 + 2 while 290 are 1 +

. But yes maxwell should have 1 + 31 mixed and 32 for compute only.

pharma
Senior Member

Posts: 995
Joined: 2003-09-10

#5151674 Posted on: 09/01/2015 01:55 PM

Lol that reddit thread, did you guys even read the comments? Seems like no one even knows how to interpret the results including the guy who wrote the program.

The program was an attempt to produce some data, is not definitive and believe was done by a graphics developer (a lot of folk there are game developers where the program originated) -- it's basically a work in progress and may continue to be so well after the first DX12 game (Fable Legends) is released in Oct. 2015.

Denial
Senior Member

Posts: 12123
Joined: 2004-05-16

#5151683 Posted on: 09/01/2015 02:00 PM

I'm familiar with Beyond3D. I'm just pointing out that there is a huge argument going on in that thread about how to interpret the results of that program. So posting the reddit thread here and saying anything definitive about the results is misleading.

Barry J
Senior Member

Posts: 2759
Joined: 2006-02-23

#5151687 Posted on: 09/01/2015 02:04 PM

I agree but it is interesting

BedantP
Senior Member

Posts: 220
Joined: 2013-08-31

#5151689 Posted on: 09/01/2015 02:04 PM

I bought a 960 because of DX 12. Damn, could've gotten a Refurb 290 or New 280X.
No real difference now, is it?
And now the AMD cards will get even more powerful because of better async support and low latency, consoles will have a boost too. #PCMR will end soon.

Chillin
Senior Member

Posts: 6815
Joined: 2006-01-18

#5151693 Posted on: 09/01/2015 02:09 PM

Does any of this even matter?

Seriously, you guys are at each others neck over HYPOTHETICAL DETAILS! FFS, even if Nvidia did support every godamn thing in DX12(_1) and AMD only supported the most basic things, it wouldn't really matter too much today. These architectures are literally a year old at this point (even older if you take into account they are revisions), and will be two years old by the time some real DX12 games come around; not to mention for how much longer they were in development before the final DX12 spec. If you bought a year old architecture that were both known to have varying levels of support right now to play hypothetical games a year from now with future features, then you are just throwing away your money.

Pascal is what you should be screaming at if it doesn't support nearly everything, same goes with whatever AMD's next architecture is.

Turanis
Senior Member

Posts: 1404
Joined: 2014-08-15

#5151720 Posted on: 09/01/2015 03:06 PM

Pascal is like a ghost,nobody knows nothing.And HBM2 is still on Amd&Hynix hands.
Until will come on the market we still have old "gen" DX11.2/DX12_0 cards which are still good.

Fox2232
Senior Member

Posts: 9094
Joined: 2012-07-20

#5151723 Posted on: 09/01/2015 03:09 PM

That is because they have a lot of tests for compute running alone from 1 to 128 depth.
Then there is just ONE value for graphical rendering time, so one should not over look it.
And finally 1~128 test for both Graphics + compute.

Proper display for those data are 3 graphs:
1st: showing 1~128 for compute only ( lowest values & best case scenario if async shaders were magical)
2nd: 1~128 theoretical result with taking Graphical rendering time and adding to it values from 1st graph (results in highest values and worst case scenario where execution is not done in parallel at all)
3rd: showing 1~128 values for compute+graphical running at once (real world result comparable with best/worst case scenario)

And I would add 4th graph showing percent difference for each 1~128 value where it would show how much it went from worst case to best case scenario.

5 pages 1 2 3 4 5

Post New Comment

Click here to post a comment for this news story on the message forum.