DOOM updated with Vulkan support

[IEC]

I'll try in necropolis as well.

Edit: getting regression there.

NoAA: 94fps
FXAA: 100fps
TSAA: 91fps

What's the rest of your system? You shouldn't be getting a FPS increase by enabling FXAA.

 

[kondziowy]

I tested Advanced Research Complex, no camera movement, I wait until train stops and make measurement.

NoAA: 77fps
FXAA: 71fps
TSSAA: 74fps

Something wrong on your end

 

[Yakk]

With all this extra "free" performance, how is VR shaping up for DOOM?

This may well tip me over to get a Vive.

 

[3DVagabond]

With low level api's it's up to the game developer to improve performance, not the driver.

FUD nVidia is responsible for the performance of it's own hardware.

Nvidia better start making games then.

 

[Irenicus]

With Nvidia it's getting pretty simple, upgrade every two years and you are golden. It's good that my GTX 670 died earlier this year, so I don't have to witness this humiliation first-hand anymore, lmao. 1080 is doing very well in the above test, however. Great card indeed, just a tad overpriced, imo.

Slight tangent but, your name and magic carpet avatar sort of ties in with DOOM, magic carpet was the game that lost out to the standard corridor/ground shooter. It was apparently slower to run on 4x86 intel hardware, but when I got my very first computer in 1996, It had a cyrix 5x86 and ran magic carpet very well. And it was INFINITELY better than doom at the time. You could fly around in free space, and had varied spells that could not only blast your enemies to bits on the ground or in the sky, you could create castles and volcanoes and fire meteors and CARVE tendrils into the earth better than any so called GOD game ever made.

rip magic carpet
rip bullfrog
rip the loss of a franchise archetype that never was, because the easier to run doom took over, that or people have terrible taste.

 

[poofyhairguy]

Kepler is going dooooown boys. And that is tested with async compute disbled(SMAA) so further gains possible for AMD.

Why is the 280X beating the 380X?

I got so excited about us finally getting a full Tonga but jeez it's disappointing.

 

[Yakk]

Why is the 280X beating the 380X?

I got so excited about us finally getting a full Tonga but jeez it's disappointing.

Bus width probably make the tiny difference.

However both can also be called "Titan Slayers"!

 

[StinkyPinky]

I fired it up yesterday on my 290x and nearly fell out my chair at the gains. Massive massive fps gains, holy crap. Ultra at 1440p smooth as butter. On a card that's three years old. Well played.

 

[Silverforce11]

Now id Software has come out and said it. Multi-Engine aspect of Async Compute is the key here.

http://www.eurogamer.net/articles/d...n-patch-shows-game-changing-performance-gains

Senior engine programmer Jean Geoffrey goes into depth on the profound advantages that async compute brings to the table.

"When looking at GPU performance, something that becomes quite obvious right away is that some rendering passes barely use compute units. Shadow map rendering, as an example, is typically bottlenecked by fixed pipeline processing (eg rasterisation) and memory bandwidth rather than raw compute performance. This means that when rendering your shadow maps, if nothing is running in parallel, you're effectively wasting a lot of GPU processing power.

"Even geometry passes with more intensive shading computations will potentially not be able to consistently max out the compute units for numerous reasons related to the internal graphics pipeline. Whenever this occurs, async compute shaders can leverage those unused compute units for other tasks. This is the approach we took with Doom. Our post-processing and tone-mapping, for instance, run in parallel with a significant part of the graphics work. This is a good example of a situation where just scheduling your work differently across the graphics and compute queues can result in multi-ms gains.

"This is just one example, but generally speaking, async compute is a great tool to get the most out of the GPU. Whenever it is possible to overlap some memory-intensive work with some compute-intensive tasks, there's opportunity for performance gains. We use async compute just the same way on both consoles. There are some hardware differences when it comes to the number of available queues, but with the way we're scheduling our compute tasks, this actually wasn't all that important."

Combined with posts on twitter, looks like id Software access the DMAs for their megatexture streaming and rasterizer for shadow maps, all the while shaders are doing graphics work. Post processing is the filler for shader utilization.

 

[brandonmatic]

Now id Software has come out and said it. Multi-Engine aspect of Async Compute is the key here.

http://www.eurogamer.net/articles/d...n-patch-shows-game-changing-performance-gains



Combined with posts on twitter, looks like id Software access the DMAs for their megatexture streaming and rasterizer for shadow maps, all the while shaders are doing graphics work. Post processing is the filler for shader utilization.

Are there any other idTech games coming out soon? Or is this just limited to Doom?

 

[Silverforce11]

Are there any other idTech games coming out soon? Or is this just limited to Doom?

Not sure what other games Bethesda will make on id6. But Multi-Engine usage applies to DX12/Vulkan, whether devs want to use it or not, is a different matter.

But AMD's pushing: http://radeon.com/radeon-civ6-dx12/

 

[Yakk]

Now id Software has come out and said it. Multi-Engine aspect of Async Compute is the key here.

http://www.eurogamer.net/articles/d...n-patch-shows-game-changing-performance-gains



Combined with posts on twitter, looks like id Software access the DMAs for their megatexture streaming and rasterizer for shadow maps, all the while shaders are doing graphics work. Post processing is the filler for shader utilization.

That's interesting.

Traditionally although very nice, each operation individually would bottleneck the GPU in a huge way, but by throwing all these individually GPU resource wasteful operations into the Async Scheduler, ID is able to saturate all parts of the GPU very, very effectively to make an awesome looking AND PERFORMING game.

Well done ID.

 

[IllogicalGlory]

Kepler is going dooooown boys. And that is tested with async compute disbled(SMAA) so further gains possible for AMD.

R7 370 competing with Titan.

Incidentally, now we see why Doom performs so well on consoles, why Digital Foundry's 750 Ti rig was getting humiliated, and why you needed an R9 380 even to keep up. Now we can probably match console performance (at the same settings) with an R7 370 or an R9 270(X).

 

[Mahigan]

Why is the 280X beating the 380X?

I got so excited about us finally getting a full Tonga but jeez it's disappointing.

Yeah... once you enable TSSAA you end up with this..

Because the AA is being handled as a compute task in parallel with the Graphics etc tasks.

So you gain quite a bit.

Oh... and it is clearly not just about AMDs bad OpenGL drivers. How many other games can you name where a FuryX slaughters a GTX 1070 and comes within stricking distance of a GTX 1080 (once AA is used)?

Clearly... Vulkan favors AMD and that is not surprising considering that Vulkan is based on Mantle (and to some extent so is DX12).

Imagine an aftermarket overclocked RX 480... it would come within stricking distance of a GTX 1070. AMD have yet to release their highend cards too.

 

[.vodka]

Now id Software has come out and said it. Multi-Engine aspect of Async Compute is the key here.

http://www.eurogamer.net/articles/d...n-patch-shows-game-changing-performance-gains



Combined with posts on twitter, looks like id Software access the DMAs for their megatexture streaming and rasterizer for shadow maps, all the while shaders are doing graphics work. Post processing is the filler for shader utilization.

That's interesting.

Traditionally although very nice, each operation individually would bottleneck the GPU in a huge way, but by throwing all these individually GPU resource wasteful operations into the Async Scheduler, ID is able to saturate all parts of the GPU very, very effectively to make an awesome looking AND PERFORMING game.

Well done ID.

Amazingly some thick skulls out there still try to downplay, put this feature down and won't even try to understand what is it about in addition to what it enables in the hands of competent developers. It's revolutionary, on top of what you can do now thanks to DX12/Vulkan that you couldn't before. Yes, more work required... Is it worth it? So far, yes.

Console style of hardware utilization in the PC that further enhances the platform for those who want to go the extra mile, what's not to like? Only people with agendas will try to paint it in a negative light.


The landscape has changed, hope nV stops scamming their users with hardware that falls apart in a year or two and starts properly supporting the feature at some point. It's needed.

 

[majord]

Are there any other idTech games coming out soon? Or is this just limited to Doom?

Quake Champions?

 

[tweakboy]

Is DX12 ,, windows 10 only ?

 

[Det0x]

Are there any other idTech games coming out soon? Or is this just limited to Doom?

Atleast Quake Champions (pc exclusive)

Quake. Is. Back.

The fast, skill-based arena-style competition that turned the original Quake games into multiplayer legends is making a triumphant return with Quake Champions. Running at an impressive 120hz with unlocked framerates, id Software’s new multiplayer shooter is getting ready to take PC gaming by storm once more.

Quake Champions features a roster of unique characters, each with their own distinctive abilities, allowing you to fight the way you want. . Get your first look at Quake Champions and some of its heroes in the cinematic reveal trailer, debuted during the #BE3 2016 Showcase.
To stay up to date with all the latest information on Quake Champions, head to Quake.com.

Teaser @ https://www.youtube.com/watch?v=sa-6fQyNkZo

 

[Red Hawk]

Now id Software has come out and said it. Multi-Engine aspect of Async Compute is the key here.

http://www.eurogamer.net/articles/d...n-patch-shows-game-changing-performance-gains

Combined with posts on twitter, looks like id Software access the DMAs for their megatexture streaming and rasterizer for shadow maps, all the while shaders are doing graphics work. Post processing is the filler for shader utilization.

If people weren't sold on the benefits of asynchronous compute, DooM should be the final proof. Massive gains. And it's also another piece of evidence that Nvidia has missed the boat with async compute with Maxwell and even Pascal.

""Yes, async compute will be extensively used on the PC Vulkan version running on AMD hardware," lead programmer Billy Khan tells us"

Sounds like they didn't even bother implementing it on Nvidia hardware, indicating that Nvidia can't even make use of it.

Why is the 280X beating the 380X?

I got so excited about us finally getting a full Tonga but jeez it's disappointing.

Yeah that's a little puzzling. Even with regards to asynchronous compute, isn't Tonga supposed to have 8 compute engines as opposed to 2 on Tahiti? So you'd think that Tonga could leverage an advantage with compute.

 

[Bacon1]

Yeah that's a little puzzling. Even with regards to asynchronous compute, isn't Tonga supposed to have 8 compute engines as opposed to 2 on Tahiti? So you'd think that Tonga could leverage an advantage with compute.

That test was done without async compute as they used MSAA not TSSAA which disables async compute.

 

[badb0y]

Kepler is going dooooown boys. And that is tested with async compute disbled(SMAA) so further gains possible for AMD.

I am still salty about Kepler.

 

[Mikeduffy]

Silverforce11
A few weeks ago you responded to a post of mine about the benefits of Asynchronous Compute - anyways, you specifically mentioned the exact same situation as ID Software has brought up here:
"When looking at GPU performance, something that becomes quite obvious right away is that some rendering passes barely use compute units. Shadow map rendering, as an example, is typically bottlenecked by fixed pipeline processing (eg rasterisation) and memory bandwidth rather than raw compute performance. This means that when rendering your shadow maps, if nothing is running in parallel, you're effectively wasting a lot of GPU processing power.

Just wanted to say that you were correct and these asynchronous compute deniers should apologize for all their misinformation.

 

[sweetusernames]

A lot of people say AMD is more forward looking / creates GPUs that do well for the future, but it seems like AMD is a company that is creating their own goddamn future. They invented Async compute, pushed Mantle/DX12/Vulkan down everyone's throat with the consoles (which wouldn't have even been possible without AMD's efforts in APUs, which began all the way back when they bought ATi), invented HBM, and now who knows what they're planning to do with multi-GPU in the future.

I am somewhat concerned for the new APIs though, what if AMD decides to do a huge arch change and suddenly GCN is left in the dust like Kepler is today given the low level nature of Vulkan and DX12?

 

[coercitiv]

I am somewhat concerned for the new APIs though, what if AMD decides to do a huge arch change and suddenly GCN is left in the dust like Kepler is today given the low level nature of Vulkan and DX12?

The next wave of consoles already have the answer you are looking for, and the answer is NO.

 

[krumme]

Silverforce11
A few weeks ago you responded to a post of mine about the benefits of Asynchronous Compute - anyways, you specifically mentioned the exact same situation as ID Software has brought up here:
"When looking at GPU performance, something that becomes quite obvious right away is that some rendering passes barely use compute units. Shadow map rendering, as an example, is typically bottlenecked by fixed pipeline processing (eg rasterisation) and memory bandwidth rather than raw compute performance. This means that when rendering your shadow maps, if nothing is running in parallel, you're effectively wasting a lot of GPU processing power.

Just wanted to say that you were correct and these asynchronous compute deniers should apologize for all their misinformation.

We started discussing Mantle 2.5 years ago. For reference this is the closing of the thread.
http://forums.anandtech.com/showthread.php?t=2344633&page=281

Before looking forward its always a good thing to look back and evaluate. Two things i would like to point out:

We can be happy it turned out the gains was even better than what we hoped for fps games and not only eg RTS where the drawcall gains back then showed us the gigantic potential of a new type of thin API. Looking forward to CIV 6 btw and see what it can do here.

We can discuss these things with good mods without thread being closed, even if the discussion is far more heated than then. Takes a lot of non biased modding this time. Good work guys. It needs to be said openly. Sorry,- but imo we didnt quite have that non biased modding back then at the gpu section.