Michellstar
Regular
Well, what was the reason of AMD pulling out "sea island"´s pdf??
Xbox One: DirectX 11.1+ AMD GPU , PS4: DirectX 11.2+ AMD GPU what's the difference?
- Thread starter onQ
- Start date
Sorry don't understand?
Sea Island is gcn2.0 though, we're talking about gcn1.1.
I would go inferring anything like that; I'm merely pointing out the capabilities of AMD's discrete GPU's and how that plays into the IP sets that are available. I'll leave others to consider how they implemented elsewhere. I think people reading the thread should also play closer consideration as to how IP's are developed and implemented as well.
The terms "Southern Islands", "Sea Islands", etc. do not refer to an IP level, they refer to a group of ASIC's on a roadmap that only really correlates to silicon implementation resources. While there has been close correlation to particular IP and a number of ASIC's under a roadmap family name (eg: Evergreen, Southern Islands) that's not a rule and mixed IP sets will fall under roadmap names (eg Northern Islands, Sea Islands).
So seeing that there isn't any DirectX 11.2 hardware out yet & PS4 is using a DirectX 11.2 GPU does that mean the GPU is a future GPU that's not out yet while Xbox One is a customized GPU that is already out?
Before I thought it was going to be the other way around with the PS4 using a older GPU & customizing it while the Xbox One was using a newer GPU that haven't been tested on the market yet.
Before I thought it was going to be the other way around with the PS4 using a older GPU & customizing it while the Xbox One was using a newer GPU that haven't been tested on the market yet.
It could be that one is GCN 1.0 and the other is GCN 1.1, or that they're both 1.1, or that they're both 1.0. In any case they can say it's a direct 11.2 gpu, from what I understand.
It seems likely that one or both is GCN 1.1. In any case, they seem to be customized to some extent.
In terms of being a directx 11.2 gpu, I'm not sure which of these features require hardware level support, but both consoles do tiled resources and gpu overlay, from what we know.
http://msdn.microsoft.com/en-us/library/windows/apps/bg182880.aspx
No such thing as DX 11.2 hardware. Theres only GCN1.0 and 1.1. Ostensibly 1.0 maps DX 11.2 API to tier 1 Tiled resource capability and 1.1 maps 11.2 to Tier 2.
Its all still the same ip family. There will be future derivatives of both. Not sure how to reconcile that with the inquirer article.
SenjutsuSage
Newcomer
I basically look at this stuff regarding tier 1 and tier 2 as pretty much being similar to what AMD did with their Barts series cards and their improved tessellator compared to Cypress, or the refinements done to the texture filtering engine. Barts also had AMD's 7th generation tessellator, and then that was enhanced further in Cayman and classified their 8th generation tessellator.
So pretty much, with regards to GCN, we're likely dealing with similar such refinements to how it handles, at a hardware level, specific DX 11.2 features since it's an architecture that's in ongoing development, even if they don't necessarily call it something else, or tell people what those lower level changes are.
So pretty much, with regards to GCN, we're likely dealing with similar such refinements to how it handles, at a hardware level, specific DX 11.2 features since it's an architecture that's in ongoing development, even if they don't necessarily call it something else, or tell people what those lower level changes are.
What would be the thought process behind (anyone) believing Sony would choose a current or older AMD architecture for PS4? Microsoft had an established relationship with AMD/ATI "console wise" since XB360 development... so one would think XB1 GPU was in the process of being developed first - "WAY BEFORE" Sony and AMD came to any agreement on working together. I would think Sony engineers would be smart enough to know/use something within AMD's roadmap that would suffice their needs beyond 2013. Sony is known for using exotic wares... except this time around, they wanted something developer friendly, but, robust enough to serve their purposes beyond 2013.
Last edited by a moderator:
I know the hardware isn't DX 11.2 hardware but it has DX 11.2 feature set.
The fact that it's a APU\SoC I was thinking this year APU using last year GPU as part of the APU.
Are you sure you thought this post through?
liquidboy
Regular
MS's relationship with AMD is probably more with ATI especially in the early days ... lots of shared work going on between MS and ATI that resulted in many famous patents of which many believe are the final XB1 architecture.
We believe its the final architecture because lots of the architects that are presenting XboxOne are the names on these patents .. and some are even on ATI patents as early as 2007
... yes MS has special arcitects called "partner architects" , they're role is to work with partners at an architecture level .. eg. Mark S. Grossman - partner hardware architect at MS (partners with ATI) http://www.linkedin.com/pub/mark-grossman/0/856/695
MS's relationship with AMD is probably more with ATI especially in the early days ... lots of shared work going on between MS and ATI that resulted in many famous patents of which many believe are the final XB1 architecture.
We believe its the final architecture because lots of the architects that are presenting XboxOne are the names on these patents .. and some are even on ATI patents as early as 2007
eg. Mark S. Grossman - partner hardware architect at MS (partners with ATI) http://www.linkedin.com/pub/mark-grossman/0/856/695
Partner could mean that his bonus is tied in with the revenue sharing, like as a law firm's partner.
So I've read comments that somehow suggested that Sony made custom design to the GPU by bumping the ACEs from 2 to 8 (?), hence the compute command queue length was increased to 64. The ACE is still AMD's IP, not Sony's. On the other hand, the X1's GPU actually have 2 compute cmd processors + 2 gfx cmd processor, and I couldn't find similar design in other Radeon products. This would suggest that this could be MSFT/AMD joint IP.
SenjutsuSage
Newcomer
I looked at the GCN architecture pdf, and I can't find anything that suggests there are two graphics command processors, which confuses the heck out of me regarding the Xbox One. I see two compute command processors, but I assume that's just another name for the two ACEs that come with regular GCN. Microsoft renamed quite a few things on their own GPU compared to what AMD calls specific components, but 2 graphics command processors?
The command processors are on the upper left in purple (graphics) and blue (compute)
The ACEs are on the lower left in green (2 swizzle copies, 1 swizzle copy with encode, 1 swizzle copy with decode).
Betanumerical
Veteran
The command processors are on the upper left in purple (graphics) and blue (compute)
The ACEs are on the lower left in green (2 swizzle copies, 1 swizzle copy with encode, 1 swizzle copy with decode).
The ACE's(Purple/Blue) are not the Move Engines (Green), and he is indeed correct. After all ACE stands for Asynchronous Command Processor and a second note, both the PS4 and the XBONE come with 2 Graphics Command Processors, the only difference in this regard is that the PS4 has 8 Compute Command Processors whilst the XBONE has 2.
Ah yes you are correct. my bad.
