PowerVR SGX543
- Thread starter argor
- Start date
http://www.imgtec.com/News/Release/index.asp?NewsID=547
Let's recap: so far Renesas and Texas Instruments.
Let's recap: so far Renesas and Texas Instruments.
Plus 3 unnamed - likely to be Sony, Intel and Apple.
Apart from SONY which had already a cryptic "no name" licensing announcement in the past I don't see anything else to be granted yet. For Apple you'd need to be an Apple employee to know what's going on and no relevant announcement yet for Intel sounds weird at least. Unless of course Intel inherited from Apple some sort of uber-secrecy virus, but I can't figure out why.
I'm not saying that it can't be true, I'm merely having second thoughts just because Renesas and Texas Instruments ARE in fact lead partners for IMG.
digitimes have just posted a press release from IMG regarding SGX544MP.
The main difference is that it has:-
"support for DirectX 9 Feature Level 3 with hardware acceleration"
SGX543MP never mentioned DX9 support.
It appears that Digitimes posted ahead of the official press date, as the PR isn't even on IMG's own website.
http://www.digitimes.com/news/a20100609PR201.html
The main difference is that it has:-
"support for DirectX 9 Feature Level 3 with hardware acceleration"
SGX543MP never mentioned DX9 support.
It appears that Digitimes posted ahead of the official press date, as the PR isn't even on IMG's own website.
http://www.digitimes.com/news/a20100609PR201.html
It's on IMG's homesite now.
As for SGX543MP if you look at the XT whitepaper it states:
Gpu accelerated browsing is avalaible to which class of GPU, I mean does it need to be direct11 compliant?
Perhaps in due course we will see a 545XT or perhaps not.
That is pure 2D compositing. It does not need any shaders. AFAICS, even ff hw should be able to do it.
If you are using some AA on images, then of course things are different. But even then, you don't need shaders, let alone tessellation or DXCS.
OGL_ES2.x isn't about fixed function is it?
Yep but from one side because the design is flexible enough to reach such heights and there's at least one of IMG's partners that has asked for something alike. Anything SGX520 up to 540 and 543MP are somewhere on the SM3.0+ level, while stuff from other IHVs aren't necessarily even on that level.
I can't for the world see anything worthwhile in that announcement (and yes I'm obviously missing something) that separates the 544MP from the 543MP. I'd like to stand corrected but it sounds more like some minor additions.
They can't leave Series6 on hold forever can they?
I did some digging and could be I found the answer to that riddle:
http://en.wikipedia.org/wiki/DirectX
I couldn't find anything in reference to "DX9 Level 3".
The information is in the DirectX Graphics Documentation. The Device Feature levels accepted by Direct3D 11 have the features listed below.
Up to you to figure out which Direct3D 9 era hw maps to 9_3, 9_2 and 9_1
Also my opionion is this stuff should have been in Direct3D 10.
Code:
| 11_0 | 10_1 | 10_0 | 9_3 | 9_2 | 9_1
------------------------------+-------+----------+----------+---------+---------+------
Shader Model | 5.0 | 4.x | 4.0 | 2.0 | 2.0 | 2.0
Geometry Shader | Yes | Yes | Yes | No | No | No
Stream Out | Yes | Yes | Yes | No | No | No
Compute Shader | Yes | Optional | Optional | No | No | No
Hull and Domain Shaders | Yes | No | No | No | No | No
Texture Resource Arrays | Yes | Yes | Yes | No | No | No
Cubemap Resource Arrays | Yes | Yes | No | No | No | No
BC4/BC5 Compression | Yes | Yes | Yes | No | No | No
BC6H/BC7 Compression | Yes | No | No | No | No | No
Alpha-to-coverage | Yes | Yes | Yes | No | No | No
Extended Formats*(BGRA, etc.) | Yes | Optional | Optional | Yes | Yes | Yes
10-bit XR High Color Format | Yes | Optional | Optional | N/A | N/A | N/A
Max Texture Dimension | 16384 | 8192 | 8192 | 4096 | 2048 | 2048
Max Cubemap Dimension | 16384 | 8192 | 8192 | 4096 | 512 | 512
Max Volume Extent | 2048 | 2048 | 2048 | 256 | 256 | 256
Max Texture Repeat | 16384 | 8192 | 8192 | 8192 | 2048 | 128
Max Anisotropy | 16 | 16 | 16 | 16 | 16 | 2
Max Primitive Count | 2^32 | 2^32 | 2^32 | 1048575 | 1048575 | 65535
Simultaneous Render Targets | 8 | 8 | 8 | 4 | 1 | 1
Occlusion Queries | Yes | Yes | Yes | Yes | Yes | No
Separate Alpha Blend | Yes | Yes | Yes | Yes | Yes | No
Mirror Once | Yes | Yes | Yes | Yes | Yes | No
Overlapping Vertex Elements | Yes | Yes | Yes | Yes | Yes | No
Independent Write Masks | Yes | Yes | Yes | Yes | No | No
Instancing | Yes | Yes | Yes | Yes | No | NoUp to you to figure out which Direct3D 9 era hw maps to 9_3, 9_2 and 9_1
Also my opionion is this stuff should have been in Direct3D 10.
Didn't want to start another PowerVr/SGX related thread, so this one looked the most relevant given the recent postings.
IMG put up a whole raft of job vacancies on their website in the last couple of days, Compiler designers,EGL design engineers, microkernal design engineers etc.
however the one that caught my eye was "DX11 Graduate Design Engineer". Its interesting because IMG currently have no (at least public) DX11 compliant graphics IP.
http://www.imgtec.com/corporate/vacancydetail.asp?VacancyID=386&VacancyTitle=DX11 Graduate Design Engineer&VacancyRef=GP86
IMG put up a whole raft of job vacancies on their website in the last couple of days, Compiler designers,EGL design engineers, microkernal design engineers etc.
however the one that caught my eye was "DX11 Graduate Design Engineer". Its interesting because IMG currently have no (at least public) DX11 compliant graphics IP.
http://www.imgtec.com/corporate/vacancydetail.asp?VacancyID=386&VacancyTitle=DX11 Graduate Design Engineer&VacancyRef=GP86
Similar threads
