• 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.

Vulkan and DX12 on new GPU'S

It's exactly this, GCN architecture has a lot of hardware onboard that simply goes unused in DX11 as it's not setup to utilise the command system of DX11 efficiently, AMD played the long game a d hedged on people utilising their hardware sooner, it never really happened for DX11. The hardware still requires power for the gpu to work and that power is effectively wasted.

Nvidia has not bothered with a lot of these hardware features and setup their hardware for the here and now, this has let them be a lot more streamlined and efficient for DX11 and require Less power as they dont have the hardware onboard wasting it.

However this does mean their dx12 performance and Vulcan is not as great as they have to tackle it with software only where as AMD hardware gains an upper hand, as their hardware finally gets to stretch its legs and the results speak for themselves mostly.

Nvidia ultimately will have to design these features into their hardware or lose ground on AMD cards, which means they will become less efficient as they add extra hardware that requires power.

I will say look at Polaris vs Pascal in DX12 and Vulcan, if we compare the 480 to the 1060 you will see the 1060 wins most DX11 and uses less power, the 480 reverses this trend mostly in DX12 and especially in Doom Vulcan, although the power remains higher on AMD. You are probably correct you say ultimately it will even the playing field once Nvidia build for DX12 their efficiency will take a hit but performance will improve.

As far as I can tell Polaris is the first true DX12 architecture, Pascal is still designed for DX11, Volta will probably be Nvidias first DX12 focused Hardware, but I expect Vega to be out before then and that will brute force close the DX11 gap and widen the DX12 gap

I wouldn't go as far as calling this short :D but it is succinct and informative. This to me provides the missing link. Now the power gap makes more sense and I suppose that makes the additional watts easier to swallow AND provides the rationale for the longevity versus Nvidia.

Thank you for this missing information. I reckon that this pushes me firmly towards the Red Camp based on my requirements.

Only decision now is between 4GB for the here and now or more likely bite the bullet for "future proofing" and get the 8GB.

Thank you again for the enlightenment.
 
I wouldn't go as far as calling this short :D but it is succinct and informative. This to me provides the missing link. Now the power gap makes more sense and I suppose that makes the additional watts easier to swallow AND provides the rationale for the longevity versus Nvidia.

Thank you for this missing information. I reckon that this pushes me firmly towards the Red Camp based on my requirements.

Only decision now is between 4GB for the here and now or more likely bite the bullet for "future proofing" and get the 8GB.

Thank you again for the enlightenment.

Buy 8gb if your intending to keep it for a while. The extra vram will layoff when you start getting games that are hitting your monitors fps limit as you will be able to turn everything texture wise to max
 
As far as I can tell Polaris is the first true DX12 architecture, Pascal is still designed for DX11

Depends what you mean "true" dx12.

AMD HD7790 and 8770 have 12_0 support level since "forever". And is exactly the same as with AMD 260/290 (2013!!), 360/390, Xbone, 285/380, Fury and Polaris cards.

11_0 level of support which is needed for DX12 compatible card, can be even found in GTX600 and the AMD 7700 all way back to 2012!!!!
 
I've seen this being claimed by certain people on this forum before, I specifically mentioned async compute so if it's wrong as you're saying can I have a source please I'd like to read up on it.

D.P is wrong, Maxwell does no Async at all, it would through drivers but has never been activated and used. In fact the driver specifically tells Maxwell cards to ignore any async requests as it causes a performance hit.
Futuremark has their benchmark designed to always try and use Async on even Maxwell, Maxwell simply cannot do it.

Sources: Futuremark dev for TimeSpy, and PcPer talking directly with NVIDIA

http://www.pcper.com/reviews/Graphi...Looking-DX12-Asynchronous-Compute-Performance

https://steamcommunity.com/app/223850/discussions/0/366298942110944664/

Now, let’s talk about the bad news: Maxwell. In my discussion with NVIDIA about this topic, I was told that async compute support isn’t enabled at the driver level for Maxwell hardware, and that it would require both the driver and the game engine to be coded for that capability specifically.

I think there is some ability to run work asynchronously in Maxwell but it will likely never see the light of day. If NVIDIA were going to enable it, they would have done so for the first wave of DX12 titles that used it (Ashes of the Singularity, Hitman) or at the very least for 3DMark Time Spy – an application that the company knows will be adopted by nearly every reviewer immediately and will be used for years.

The reason Maxwell doesn't take a hit is because NVIDIA has explictly disabled async compute in Maxwell drivers. So no matter how much we pile things to the queues, they cannot be set to run asynchronously because the driver says "no, I can't do that". Basically NV driver tells Time Spy to go "async off" for the run on that card.

If NVIDIA enables Async Compute in the drivers on Maxwell, Time Spy will start using it. Performance gain or loss depends on the hardware & drivers.
 
Last edited:
Where iis the link which states that? Who said that Kepler and Maxwell are getting legacy driver?

You know better then Nvidia and their employees?
ManuelG
NVIDIA Rep



http://forums.guru3d.com/showthread.php?t=409017

Most of the optimizations we add to our drivers are not architecture specific so they should help most GPUs

Means no more special optimization/s for that maxwell, only general. Means, maxwell gets the same as the fermi or kepler .
 
Last edited:
D.P is wrong, Maxwell does no Async at all, it would through drivers but has never been activated and used. In fact the driver specifically tells Maxwell cards to ignore any async requests as it causes a performance hit.
Futuremark has their benchmark designed to always try and use Async on even Maxwell, Maxwell simply cannot do it.

Sources: Futuremark dev for TimeSpy, and PcPer talking directly with NVIDIA

http://www.pcper.com/reviews/Graphi...Looking-DX12-Asynchronous-Compute-Performance

https://steamcommunity.com/app/223850/discussions/0/366298942110944664/

Maxwell async is real on with half software and half hw level. Never ever going to work with all the dx 12 programs the same way. No point to wait for it. Nv may enable something which can do in certain program some boost , but that's it. If ever nv going to do it.
Basically no question, amd from now on. I am saying as an nvidia owner.


I would not be a happy owner of maxwell card. I am half happy with fermi. DX 12 driver still missing. Vulkan is kinda left in the dark but it is possible to make the support to fermi.
 
Last edited:
This is why AMD should be taken seriously when it comes to DX12 and Vulkan.


Yes it should taken seriously on a broken, early access,game filled with negative reviews, performance issues, and a animation level of Xbox 360. It should be taken seriously based on that.
 
Last edited:
D.P is wrong, Maxwell does no Async at all, it would through drivers but has never been activated and used. In fact the driver specifically tells Maxwell cards to ignore any async requests as it causes a performance hit.

Thanks, This is basically what I thought was the case and I was using it as a positive for buying AMD,
I don't understand why people have to jump on other peoples statements like that, As I've stated in other threads if you change cards often go Nvidia but if you want to make that card last you in my opinion go AMD.


If people want to give reasons to buy the 1060 over the 480 based on the prospective buyers needs they should do that instead of jumping on other members posts with generalised negative responses.
 
Last edited:
Nvidia has sucessfully marketed their cards,unlike AMD who keep making mistakes,so they have at least 70% of the sales marketshare,so don't care! People will buy Nvidia based on the strength of the brand alone. This is why the GTX660,GTX660TI,GTX760 and GTX960 all have fared worse than the equivalent AMD cards of the time over the longer run.

The signs are already there for the GTX1060 and people will cling to the hope like with all these magical updates for Maxwell cards like my one,which won't appear even when the successor is released and people still live in hope.

AMD keeps making mistakes and keeps giving Nvidia chances too. Look at the launches like the HD4000 and HD5000 series?? ATI didn't make mistakes.

Almost every single AMD card launch in the last 4 years has had some sort of issue which Nvidia can exploit.

The RX480 launch was successful but AMD still did not learn from the R9 290 series launch which was apparently successful too and they lost momentum.
 
Last edited:
Almost every single AMD card launch in the last 4 years has had some sort of issue which Nvidia can exploit.

The RX480 launch was successful but AMD still did not learn from the R9 290 series launch which was apparently successful too and they lost momentum.

I agree, although I also think resources plays a massive roll. They have gotten better but they're still not good enough. Their rather on the ball with Drivers lately, and fix for the 480 along with performance increases is already there.

The main issue is; those problems should never have been there. It's bad publicity AMD really does not need.

Remember the Fury X as well? It's an amazing card, and still doing well, even more so with newer APIs; but it's drivers and support for the first 6 months were appalling; and add in the whole crappy first gen pump noise issues as well.

Now that AMD has RTG, they are improving; but just not fast enough for many. I hope with the new 490/Fury they get it right from the start.
 
Yes it should taken seriously on a broken, early access,game filled with negative reviews, performance issues, and a animation level of Xbox 360. It should be taken seriously based on that.

Listen to yourself?
One it's not an early access game
two most of itd negative reviews are because it's an episodic title
Three it's doesn't matter if it's, animation system based on your opinion is that of a 360 game.

It is still a side by side benchmark on a DirectX 12 game.

How about Quantum Break? Check out the 480 performance in combat, it doesn't drop it keeps a solid performance.
 
Last edited:
Listen to yourself?
One it's not an early access game
two most of itd negative reviews are because it's an episodic title
Three it's doesn't matter if it's, animation system based on your opinion is that of a 360 game.

It is still a side by side benchmark on a DirectX 12 game.

How about Quantum Break? Check out the 480 performance in combat, it doesn't drop it keeps a solid performance.

Please use common sense. If a Fury X cannot maintain 60fps @ 1080p then you consider this a good port? It is ported from a HD 7790 ,which is in Xbox one. A Fury X (8.6 Tflops) cannot maintain proper 60 fps ,which is ported from HD 7790 (1.3 Tflops).

You can win arguments by providing example like Doom and ATOS but not by posting this rubbish.
 
Please use common sense. If a Fury X cannot maintain 60fps @ 1080p then you consider this a good port? It is ported from a HD 7790 ,which is in Xbox one. A Fury X (8.6 Tflops) cannot maintain proper 60 fps ,which is ported from HD 7790 (1.3 Tflops).

You can win arguments by providing example like Doom and ATOS but not by posting this rubbish.

The RX480 is pretty much holding 60fps so i reckon the Fury X should easily do it. At least in this bench sequence.
 
You can win arguments by providing example like Doom and ATOS but not by posting this rubbish.

Any time someone has mentioned and showed DOOM or ATOS you've dismissed them. :rolleyes:

Also all 3 vids you linked are well over a month old, with older drivers, and older version of the game. Got an updated one showing of all the latest updates like the GTX 1060/480 one?

Even in all of them the Fury X is beating the 980Ti and Titan X, just like the 480 is beating the 970 and 1060.
So what's your point? That AMD is doing better in all the tests linked so far; with the only exception in DX12 being ROTTR?
 
Last edited:
Imo don't wait, it won't be worth it. I'd say the decision should actually boil down to vendors, rather than chipmakers. What I mean by that is, if you can get a PowerColor Devil RX480, go for that, it's the best card in this price range imo, anything below that is just a step down. Only problem is, it's still in pre-order stage & afaik the August 12th date isn't a sure thing. The second option would be the MSI or Asus ones, but again, low stock, overpriced - not worth it.
So if you can't get the Devil 480, look for 1060s. Here I would only really recommend ones that are both really good cards & have excellent warranty/cs. So it really leaves only EVGA & Zotac. Unfortunately, for EVGA, same problem as Devil, not available yet.

If you can wait 2 weeks or so - Devil 480, no question about it. If you can't, go for Zotac AMP 1060.

But look, if you care about longevity going Nvidia is a poor decision, anyone that's looked at how 290x vs 780 ti is now, or 380 vs 960, or 280 vs 760, etc. It's not even close, and it's clear that AMD is much better than Nvidia in both DX 12 & Vulkan for cards in the same performance range (obviously AMD has no real high-end cards to compete with atm - so 1080/Titan XP are still better in terms of absolute performance). The way they are approaching their architecture & that they're doing console hardware means you will see much better support going forward and see actual performance gains a year from now as well - with the 1060 what you see is what you get, right now, it will never get better, nor worse (except relatively by comparison, but that's different).
 
Back
Top Bottom