Flaws on Oxide

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

krumme

Diamond Member
Oct 9, 2009
5,956
1,595
136
This looks the same for borderlands 1/2, batman and any other TWIMTBP title. Some call it concrete tessellation performance advantage

Lol. I had to look if yoj could actually say that in english but you can.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
So much money and years of development for so little effect on typically very stong 6c cpu. Perfectly illustrates the need for a new low level api. The oxide engine will show that and to a shocking degree.

Direct3D needs to be improved (and it will be), but we're not being CPU limited with native DX11 titles. Out of all the games I have, the only game I'm really CPU limited in is Borderlands 2, and that's DX9.

DX11 has given us games with stunning amounts of detail and great performance, games that wouldn't even be possible on DX9 without some major compromises.....like the Witcher 3.

It's amazing how much flak Direct3D is getting, when it literally singlehandedly saved PC gaming from complete fragmentation.
 

el etro

Golden Member
Jul 21, 2013
1,581
14
81
Last edited:
Feb 19, 2009
10,457
10
76
Anyone who has played Eve will realize independent turrets on space ships that track targets with its own AI is ridiculous on CPU bottlenecking, in large fleet fights, players had to TURN OFF turrets (and many other effects) from visualizing because it would bog down the strongest of CPUs. As soon as you get into a large fleet fight, it ended up with dots shooting invisible stuff at other dots.. if you want a responsive game.. or you could be the dude that leave everything on, and be the first to die since your game is a slideshow.

Draw calls are the bane of complex RTS.
 

Imouto

Golden Member
Jul 6, 2011
1,241
2
81
I think the fact that AMD doesn't support it has more to do with it than anything. Plus, as others have mentioned, it took a long time before NVidia was able to actually implement it in their drivers, as it's apparently a very difficult thing to do properly.

As a matter of fact, they are still refining it.

You're just making the point of Mantle stronger. As someone else said already DX is so messed up that not only GPU makers want to make a new API, even game developers are willing to use it.

But unlike Mantle, which is vendor specific, DCL is part of the DX11 specification. AMD has been unable to properly implement it since DX11 came out years ago, whether due to lack of resources or sheer incompetence is anyone's guess..

I'd bet for giving up after seeing the minimal improvements in CiV5, AC3 and the likes. Why throw money into that sink?

Anyway, it will come back to haunt them when Project CARS, the biggest next gen multiplatform racing game, is building their DX11 renderer around deferred context rendering which will result in much better performance on NVidia hardware.

There are already plenty of games working way better with Nvidia hardware and that doesn't mean jack. In fact if that would mean anything BF4 is a way bigger title and already works better with AMD hardware. FYI Project CARS is a game for diehard racing simulation fans. I don't think it's going to be popular at all.
 

tonyfreak215

Senior member
Nov 21, 2008
274
0
76
FYI Project CARS is a game for diehard racing simulation fans. I don't think it's going to be popular at all.

And a bad one at that. :/ As a backer, I'm very disappointed with how it handles. Looks gorgeous, but their physics just feel off.
 

Red Hawk

Diamond Member
Jan 1, 2011
3,266
169
106
I dont know if anyone have Rust here, but that will be a good game to try, as Rust is done on Unity, Unity has support for MT rendering on OpenGL, and im not 100% sure but i think it does as well on DX11, Rust is a game that you can run on DX9, DX11 and OpenGL, you just need to launch with with proper arguments.

http://docs.unity3d.com/Documentation/Manual/CommandLineArguments.html

From what I've read, Unity lacks a lot of features that PC and console engines like Unreal or Cryengine do. It's an engine that was created for mobile games and is used on PC primarily because it's free and easy to program for. I doubt it will be very performance-heavy to begin with.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
I'm not talking about DCL.

See this:



Dated from December 4. Not favors any architecture.



That's why i don't like PClab tests.

Just one more reason why I don't trust TPU, and I never take their reviews seriously. Their drivers are always out of date.

Anyway, why do their benchmarks differ so much from PClabs.pl's benchmarks?

It likely has to do with the area they benchmarked more than anything else. Some areas are more GPU bound, and others more CPU bound. I think the latter would give NVidia the upper hand, where as the former would be agnostic.
 

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
OK here we go.

Rust DX9


Rust OpenGL


Rust DX11


I no need to go intro details and going intro more draw call limited areas, i can see with my own eyes that DX11 is firing up 3 cores on there(Core 0 is probably loaded with trash and OS), while OpenGL and DX9 only 1... And this is UNITY ENGINE guys, UNITY!
 
Last edited:

Imouto

Golden Member
Jul 6, 2011
1,241
2
81
I no need to go intro details and going intro more draw call limited areas, i can see with my own eyes that DX11 is firing up 3 cores on there(Core 0 is probably loaded with trash and OS), while OpenGL and DX9 only 1... And this is UNITY ENGINE guys, UNITY!

You sure ain't doing something wrong? That's a massive CPU usage increase over DX9 for exactly zero perf increase.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
You're just making the point of Mantle stronger. As someone else said already DX is so messed up that not only GPU makers want to make a new API, even game developers are willing to use it.

DX is not messed up. How could anyone with any sense make a comment like that, when PC games have never looked or performed better!?

We have the Witcher 3, a game that will feature no loading screens at all, and have no chapters. It's completely open World from beginning to end.

And it's DX11 native. A game like that would not even be possible on DX9 without serious sacrifices. And your precious Mantle uses HLSL, a DX11 feature to work btw..

Even BF4 performs excellently using the Direct3D path, with triple digit frame rates at ultra settings.

I'd bet for giving up after seeing the minimal improvements in CiV5, AC3 and the likes. Why throw money into that sink?
Right, almost 50% is "minimal." But when Mantle claims 45% improvement it's all of a sudden spectacular!

BTW, here's a taste of what's to come with Project CARS. A 7850 vs a GTX 660 test off, and the GTX 660 has well over 50% faster frame rates than the 7850, courtesy of DX11 multithreading.

There are already plenty of games working way better with Nvidia hardware and that doesn't mean jack. In fact if that would mean anything BF4 is a way bigger title and already works better with AMD hardware. FYI.
You're way behind the times there dude. Mantle aside, after multiple patch and driver updates, BF4 does not work better on AMD hardware. It actually runs faster on NVidia, especially in multiplayer where the CPU plays a bigger role in performance.

Read the latest reviews at GameGPU, HardOCP, PClabs.pl or any website that reviews using MP, and you'll see that NVidia now has the lead.

Project CARS is a game for diehard racing simulation fans. I don't think it's going to be popular at all
You can think what you like, but Project CARS has tons of youtube channels devoted to it. Also, it's the biggest multiplatform racing game for next gen, that alone will make it a success.
 

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
You sure ain't doing something wrong? That's a massive CPU usage increase over DX9 for exactly zero perf increase.

I just checked with gpu-z and gpu load is at 100% on DX9 and DX11, no way fps are going to get any better than that.
 
Last edited:

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
Despite multiple people on this thread saying that it only uses 1 or two cores.

The actual average CPU usage this benchmark is current showing under D3D is only showing a couple of cores of usage, as low as 1 core in the worst areas. Based on what Oxide has said in the past this is due to DX not being able to support the very high (30k+) draw calls and the result is that the engine itself has nothing to do while the single thread for D3D is left processing the draw calls and slows the whole thing down. I do see up to 4 threads at times but the benchmark spends most of its time around 1 to 2 cores of usage overall.

Presumably when people test this with Mantle it will show higher utilisation as the engine and game wont be held up by the low fps caused by DX draw call limits and the result will be a much faster and higher utilised machine, both CPU and GPU along with higher FPS.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
I dont know if anyone have Rust here, but that will be a good game to try, as Rust is done on Unity, Unity has support for MT rendering on OpenGL, and im not 100% sure but i think it does as well on DX11, Rust is a game that you can run on DX9, DX11 and OpenGL, you just need to launch with with proper arguments.

http://docs.unity3d.com/Documentation/Manual/CommandLineArguments.html

OpenGL doesn't have true multithreading like DX11 if I remember correctly.

With OpenGL, each core can upload data to the GPU, but not at the same time like with DX11, which makes it much more limited.

What OpenGL has though are bindless extensions, which are vendor specific.. I don't have the technical knowledge to explain what bindless extensions are, but I know they can increase draw call output tremendously....more than Mantle in fact.
 

Paul98

Diamond Member
Jan 31, 2010
3,732
199
106
As dots? Shouldn't be hard, considering they all look identical. And motion blur at that distance is pointless.



Despite multiple people on this thread saying that it only uses 1 or two cores.

The program doesn't know they will be dots, it has to do work on the objects to know what will be needed, what will be shown what will not. It isn't magic. Unless you want to cull objects that are far away you will end up doing matrix math on the vertices to get the objects to the screen.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
And a bad one at that. :/ As a backer, I'm very disappointed with how it handles. Looks gorgeous, but their physics just feel off.

Well the game doesn't release till later on this year, so they have plenty of time to tweak the physics..
 

jj109

Senior member
Dec 17, 2013
391
59
91


So what's preventing Oxide from doing a proper DX11 implementation? Apparently DX11 CL isn't broken enough to prevent BF4 from using it and improving CPU performance massively.

The actual average CPU usage this benchmark is current showing under D3D is only showing a couple of cores of usage, as low as 1 core in the worst areas. Based on what Oxide has said in the past this is due to DX not being able to support the very high (30k+) draw calls and the result is that the engine itself has nothing to do while the single thread for D3D is left processing the draw calls and slows the whole thing down. I do see up to 4 threads at times but the benchmark spends most of its time around 1 to 2 cores of usage overall.

Presumably when people test this with Mantle it will show higher utilisation as the engine and game wont be held up by the low fps caused by DX draw call limits and the result will be a much faster and higher utilised machine, both CPU and GPU along with higher FPS.

D3D itself is only using one thread in the demo. When the D3D thread is bottlenecking the game the simulation threads stall also, which leads to people calling the game single threaded b/c they look at the task manager and see 1 core being used. Not quite accurate, but in the same ballpark.


I just checked with gpu-z and gpu load is at 100% on DX9 and DX11, no way fps are going to get any better than that.

What GPU are you using that's choking on that simple scene?
 
Last edited:

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
I think the point is that the engine is not optimised for DX, and from what im seeing, not even using all DX11 features.

Thats also not "console style optimizations" as AMD own slides told us either, who render useless things in a console?

That benchmark is out just for "hey i can run 500% faster in mantle", who cares of the final results or if its done properly on DX11...

What GPU are you using that's choking on that simple scene?

HD5850 i gona try to make DX9 cpu limited just for the hell of it.
 
Last edited:

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
The program doesn't know they will be dots, it has to do work on the objects to know what will be needed, what will be shown what will not. It isn't magic. Unless you want to cull objects that are far away you will end up doing matrix math on the vertices to get the objects to the screen.

Which is why it should have an LOD system.

Object so far away, its simply at dot = 1 draw call (assumption need someone with more experience to confirm),

Multiple identical objects= batched draw call,

Objects so far away = forget about motion blur.

It looks like oxide is running the same draws on each object despite them being identical objects that are very far away (like independently revolving turrets on a dot). Also objects seem to be rendered, even if they are occluded by an explosion.
 

Spjut

Senior member
Apr 9, 2011
928
149
106
So what's preventing Oxide from doing a proper DX11 implementation? Apparently DX11 CL isn't broken enough to prevent BF4 from using it and improving CPU performance massively.

Except BF4 isn't using it. Just like the vast majority of other DX11 games.
 

Paul98

Diamond Member
Jan 31, 2010
3,732
199
106
What I find funny is so many here seem to think they know more about what can be done with DX11 than the developers themselves.
 

jj109

Senior member
Dec 17, 2013
391
59
91
That benchmark is out just for "hey i can run 500% faster in mantle", who cares of the final results or if its done properly on DX11...

Well, the claim was "hey I run 300% faster in Mantle". They left out the part where D3D was running only 1 thread on a 2C/4T CPU, so the 300% claim is pointless.

Except BF4 isn't using it. Just like the vast majority of other DX11 games.

BF4 is. Repi's tweet says they are supporting it. Repi's slides say the only major roadblock is "performant drivers". Nvidia had the "performant" drivers problem solved since driver release 270, as shown by the sudden and confusing Civ V FPS improvement.

You are welcome to go back to the Mantle thread and see my BF4 run where my Nvidia GPU was running 35% faster in CPU-bound situations vs hardcoreware's Kaveri release benchmarks.

What I find funny is so many here seem to think they know more about what can be done with DX11 than the developers themselves.

I find it sad that so many here think that developers trying to sell a platform will tell them the whole truth.
 
Last edited:
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/    |