• 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.

Star Swarm - Oxide bench thread

So one core should have very high usage and all the others low to none? Not the case, i have profiled it and im seeing nearly equal usage on all 6 cores and no single core higher than 50%

With deferred contexts on nvidia drivers ARE multithreaded and things are even more equal, but still low utilisation and low fps and low gpu utilisation
 
So one core should have very high usage and all the others low to none? Not the case, i have profiled it and im seeing nearly equal usage on all 6 cores and no single core higher than 50%

With deferred contexts on nvidia drivers ARE multithreaded and things are even more equal, but still low utilisation and low fps and low gpu utilisation

Because the DX API is limited, not the CPU.

Mantle is simply a much more efficient better connected API, it make more use of the CPU's performance.
 
Because the DX API is limited, not the CPU.

Mantle is simply a much more efficient better connected API, it make more use of the CPU's performance.

Right, the reason DX cant handle lots of drawcalls is because it is poor at multi threading them, you know this because its in the AMD presentation about mantle, so poor threading = maxing out at least 1 cpu core and then bottleneck

so why is it NONE of my cpu cores hits 100% with star swarm, but that is still a cpu bottleneck?

Already covered by Andy's quoted post humbug. Circles.
 
"Quote:
Originally Posted by andybird123
Right, the reason DX cant handle lots of drawcalls is because it is poor at multi threading them, you know this because its in the AMD presentation about mantle, so poor threading = maxing out at least 1 cpu core and then bottleneck

so why is it NONE of my cpu cores hits 100% with star swarm, but that is still a cpu bottleneck?"
 
"Quote:
Originally Posted by andybird123
Right, the reason DX cant handle lots of drawcalls is because it is poor at multi threading them, you know this because its in the AMD presentation about mantle, so poor threading = maxing out at least 1 cpu core and then bottleneck

so why is it NONE of my cpu cores hits 100% with star swarm, but that is still a cpu bottleneck?"

(The CPU is Bottlenecked) By the DX API because it cannot process for the Drawcalls the CPU can calculate, so while The GPU usage is dropping through lack of CPU performance, its not the CPU 'sitting on the other side of the DX API' that is the problem, its the API getting in its way.

The GPU asked DX for 8K parallel Drawcalls, The API said; i can only give you 5K, thats all i can process. the CPU on the other end calculating those Drawcalls is saying to DX, 5K? is that all you want? *yawn* the GPU back to the DX API. ok, i will have to slow things down at this end so you can keep up.

DirectX, or rather DirectX 11, is Multi-threaded. "DX Deferred Path" Google it its used in BF3 and 4, the problem with DirectX is DirectX, DX Deferred or not.
 
Last edited:
(The CPU is Bottlenecked) By the DX API because it cannot process for the Drawcalls the CPU can calculate, so while The GPU usage is dropping through lack of CPU performance, its not the CPU 'sitting on the other side of the DX API' that is the problem, its the API getting in its way.

Timothy Lottes-

99dcbb6803f13a55e36158995a552d67.jpg


As a PC guy who knows hardware to the metal, I spend most of my days in frustration knowing damn well what I could do with the hardware, but what I cannot do because Microsoft and IHVs wont provide low-level GPU access in PC APIs. One simple example, drawcalls on PC have easily 10x to 100x the overhead of a console

AMD aka IHV, have now supplied Mantle-the low-level GPU access API for the PC platform that Lottes and many more have screamed out for.

Lottes is Nvidia through and through, considering he works with Epic who are heavily involved with Nvidia, is he talking animal waste too???

The Starswarm STRESS test, is NOT a BM, it's main design was to showcase that Mantle removes DX limitations and what it brings for AMD gpu's, NOT to gimp DX on both vendors-nothing more, nothing less.

Core/GPU usage on DX is low because the DX API CANNOT push the data through remotely fast enough, where as Mantle can.

The sheer amount of fud in here is incredible, because you can't see past the colour of your gpu and/or because your gpu isn't supported.

The amount of anger from Nvidia users when PhysX(wrongly) gets slated, is justified imho, pity the same users who displayed such anger are giving it back in spades, poor, very, very poor.:(
 
I'm just wondering tommy why Deferred Context is off by default if it increases fps and makes the mantle/dx improvement less or non existent ?
If Deferred Context doesnt work on amd? (dont have a 290 here atm gone away in my mates rig) then its understandable or is it something that works on both vendors?
be nice if AMD user could try see what it does for them if it works i'm too lazy to try it on my 5870 rig atm
 
Last edited:
I'm just wondering tommy why Deferred Context is off by default if it increases fps and makes the mantle/dx improvement less or non existent ?
If Deferred Context doesnt work on amd? (dont have a 290 here atm gone away in my mates rig) then its understandable or is it something that works on both vendors?

Apart from the 4 year old P-II x6 in Mantle easily keeping up with the latest £750 CPU from Intel.

edit>

be nice if AMD user could try see what it does for them if it works i'm too lazy to try it on my 5870 rig atm

it wont work on your 5870 anyway.

I will have a look at the DC thing tomorrow.
 
http://www.behardware.com/articles/770-2/report-amd-radeon-hd-5870-and-5850.html

DC should work fine ?
seems it was tested on a 4890 even
We tested this on a Radeon HD 4890 with a demo supplied by Microsoft in its SDK and which works on all Direct3D 10 cards. This represents a scene limited by the CPU,

- Rendering in immediate context mode: 55 fps
- Rendering in deferred context mode, not multithreaded: 48 fps
- Rendering in deferred context mode, multithreaded: 78 fps

might be since way back when this was tested DC when it wasnt multithreaded (assuming a single core cpu) was slower would be a valid reason for the Demo now it to be disabled by default
 
Last edited:
http://www.behardware.com/articles/770-2/report-amd-radeon-hd-5870-and-5850.html

DC should work fine ?
seems it was tested on a 4890 even
We tested this on a Radeon HD 4890 with a demo supplied by Microsoft in its SDK and which works on all Direct3D 10 cards. This represents a scene limited by the CPU,

- Rendering in immediate context mode: 55 fps
- Rendering in deferred context mode, not multithreaded: 48 fps
- Rendering in deferred context mode, multithreaded: 78 fps

Its nice to know DC does work on AMD cards but its Mantle that i was talking about, i thought thats what you meant :)
 
Back
Top Bottom