DOTA 2 launches Vulkan support

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Bacon1

Diamond Member
Feb 14, 2016
3,430
1,018
91
Vulkan only helps cpu limited situations?

No it should help everywhere, its less overhead for the GPU so it can do more actual computing and less "talking" about what its going to do.

It will take a while for these types of gains to be shown for most game engines though, as they have to be designed around that not just tacked on. So you might not see any gains with DOTA 2.
 

2blzd

Senior member
May 16, 2016
318
41
91
Yea I'm going to pass for now. Looks like you have to have the crimson driver package installed for AMD cards.

No thanks. That driver f'd up all my After Effects and Premiere Pro plugins. So I rolled back
 

MajinCry

Platinum Member
Jul 28, 2015
2,495
571
136
No it should help everywhere, its less overhead for the GPU so it can do more actual computing and less "talking" about what its going to do.

It will take a while for these types of gains to be shown for most game engines though, as they have to be designed around that not just tacked on. So you might not see any gains with DOTA 2.

The gains are to be found in games like Dark Souls, Skyrim, Mount & Blade, Total War, Freespace 2, most MMOs (raids, guild vs guild, world vs world, etc), that sorta thing.

Good example in Skyrim; each NPC makes around 40-50 draw calls, due to all the textures used on them (body texture, scars, brows, tattoos, blood, hair, etc.). If you overlook Whiterun, you're making around 3.5k draw calls, which brings me Phenom II to a glorious 27 fps. Get rid of all the NPCs, and oh my, I'm down to ~2.8k, and I'm at a good ol' 35-40fps.

With these modern APIs, we're not talking a couple thousand draw calls being the performance barrier. We're now talking several tens of thousands of draw calls (per core!) being the limit.

If OSG gets Vulkan functionality, I'd love to see what modders do to the ol' OpenMW; an open world engine with all the objects you can throw at it. Mmm.


I'd also like to see how this affects 3d rendering. Might make the 8350 the utter king, if 3DS Max (or whatever) adds support for Vulkan/D3D 12. parallel draw calls AND parallel rendering, rather than just the latter on it's own.
 

TheELF

Diamond Member
Dec 22, 2012
3,993
744
126
I'd also like to see how this affects 3d rendering. Might make the 8350 the utter king, if 3DS Max (or whatever) adds support for Vulkan/D3D 12. parallel draw calls AND parallel rendering, rather than just the latter on it's own.

This doesn't make any sense,render software is already able to use your GPU+CPU for rendering and for quite some time,look at luxrender/luxmark.
In those programs you just send the scene(minimal amount of data and only once) to all available render units and wait for them to finish.
In a game it's constant back and forth and the faster you turn your mouse,for example, the harder the whole system has to work.
 

MajinCry

Platinum Member
Jul 28, 2015
2,495
571
136
This doesn't make any sense,render software is already able to use your GPU+CPU for rendering and for quite some time,look at luxrender/luxmark.
In those programs you just send the scene(minimal amount of data and only once) to all available render units and wait for them to finish.
In a game it's constant back and forth and the faster you turn your mouse,for example, the harder the whole system has to work.

Unless you've got a remotely complex scene, then ya might be having a bit of an issue.

Ain't most renders making, what, 100k draw calls per frame? 1m? It's gotta be working hard on ya compy doing that.
 

Raising

Member
Mar 12, 2016
120
0
16
Does G-sync work with vulcan ?

Even at 144fps there's stuttering while edge panning the camera, which never happens using dx11, v-sync off ingame option stutters even more..

Think I've been spoiled by gsync, now every time it's off I notice it in a heartbeat.
 
Last edited:

Bacon1

Diamond Member
Feb 14, 2016
3,430
1,018
91

Thanks

Something looks completely broken on the 980 Ti. Is it not rendering the screen properly? Its 2.25x faster than the 980 @ 4k in Vulkan and 1.84x in OpenGL

For showing just how early Vulkan is, the GTX 950 and GTX 960 were both crashing when trying to run Dota 2 with the Vulkan renderer. Like the OpenGL results, the cards aside from the GTX 980 Ti were performing odd to say the least.

Lots of crashing and either the 980 TI was running too fast (improper rendering?) or every single other card was running way too slow.

Not looking good :\
 

SPBHM

Diamond Member
Sep 12, 2012
5,058
410
126
something else that is interesting is the 285 performance, it beats the 290 on OGL by a small margin and on Vulkan by a significant margin, it even beats the gtx 980...

the whole game/drivers/APIs don't seem to be working as expected...
 

cytg111

Lifer
Mar 17, 2008
23,564
13,123
136
+1 for Vulkan
+1 for the perfomance boost

Lets hope adoptation continues
 
Last edited:

Bacon1

Diamond Member
Feb 14, 2016
3,430
1,018
91
The Vulkan dev from Valve kinda debunked those results in the thread on Phoronix.

He is using a Titan X.

He did make another post later with a link to two other demos to use as the run, but he hardly debunked it. He gave some guesses why the results might be wrong, but no proof.
 

TheELF

Diamond Member
Dec 22, 2012
3,993
744
126
Vulkan only helps cpu limited situations?

No driver thread=the other threads get to share the freed up resources=moar FPS.
Dota2 endgame(last 10 mins) in dx11 and Vulkan playing on top of each other.
https://www.youtube.com/watch?v=-cDAWW8nkGU
Different sessions obviously but you get the jist of it.

If your GPU is already running at 100% then you might get some improvement with Vulkan if it is more efficient in talking to the GPU then Dx11 but I doubt it,if someone can test and show us it would be very appreciated I believe.
 
Last edited:

Sunaiac

Member
Dec 17, 2014
83
22
81
He is using a Titan X.

He did make another post later with a link to two other demos to use as the run, but he hardly debunked it. He gave some guesses why the results might be wrong, but no proof.

So maybe the new results posted are proof enough.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |