The AMD Execution Thread [2007 - 2017]

Status
Not open for further replies.
Those couldn't function reliably at the speeds they were clocked at, regardless of power consumption.
I'd say this is closer to the Prescott extreme editions, just less justifiable from a power and performance perspective.
 
Apple: "Everyone should be using OpenCL"

Apple Unveils New Mac Pro
http://www.techpowerup.com/185451/apple-unveils-new-mac-pro.html
Two AMD workstation GPUs will be accompanying the Ivy Bridge Xeon E5 processor.

http://vr-zone.com/articles/who-are-the-winners-and-losers-of-apples-wwdc-keynote/36544.html
The Mac Pro ships with a pair of AMD Firepro GPUs. These will have mighty horsepower: 384-bit memory buses, 528 GB/s of total bandwidth, and support for multiple 4K displays. They also aren’t on Team CUDA. During the keynote a presenter solidified Apple’s support for OpenCL with a comment along the lines of “everyone should be using OpenCL.”
Click to expand...
 
It's a dual-Tahiti with all CUs enabled, but a custom design with a custom PCB. Nothing else would fit into Apple's weird cylindrical case.
 
http://arstechnica.com/apple/2013/06/a-critical-look-at-the-new-mac-pro/

Apple may be trying to not-so-delicately nudge everyone to move their code from CUDA to OpenCL, but I’ve seen a first-hand failure of AMD’s OpenCL support with V-Ray RT for Maya. Chaos Group built V-Ray RT on OpenCL, but after extensive work trying to get the GPU variant of its RT render engine running on AMD hardware and an effort by yours truly to light a fire under Apple and AMD, Chaos Group gave up and ported it to CUDA instead. So V-Ray RT’s GPU mode only works with OpenCL and CUDA—on Nvidia hardware.
Click to expand...

If you don't want to get work done, use Apple & AMD OpenCL.
 
UniversalTruth said:
Sure, OEMs are more "important" than end-users :rolleyes: Have no idea what the logic behind it is...
Click to expand...

OEM's are a lot less likely to blow cheap mobos and massively over-rated PSU's apart, mostly because returns are bad for business.

Sure but I don't know whether there are some opitmisations or other stuff going on...
Click to expand...
I'd take a guess at a better turbo optimisation. 5 GHz is the "max turbo" and AMD is being coy on the fully loaded turbo.

I do believe the 220W rumours are bullshit though.
 
jimbo75 said:
OEM's are a lot less likely to blow cheap mobos and massively over-rated PSU's apart, mostly because returns are bad for business.

I'd take a guess at a better turbo optimisation. 5 GHz is the "max turbo" and AMD is being coy on the fully loaded turbo.

I do believe the 220W rumours are bullshit though.
Click to expand...

Unfortunately, that's not a rumour:

IMG0041750.png


I don't really understand who's supposed to want this. Someone who really doesn't care about power and wants maximum performance, perhaps, but they might as well get a Core i7 4770K (which is cheaper and perhaps slower in a few cases, but mostly comparable) or a 3930K (6-core) which is about the same price, or perhaps cheaper, and probably faster in 99% of cases.
 
jimbo75 said:
OEM's are a lot less likely to blow cheap mobos and massively over-rated PSU's apart, mostly because returns are bad for business.

I'd take a guess at a better turbo optimisation. 5 GHz is the "max turbo" and AMD is being coy on the fully loaded turbo.

I do believe the 220W rumours are bullshit though.
Click to expand...

Hmm, sound logical and interesting.

However, do you know when AMD intends to change the manufacturing process so this performance gets to more sane limits?
 
AMD doesn't have direct control over the manufacturing process, although the SOI node they are the customer for may afford them some leeway in requesting something of Globalfoundries. AMD's rapidly moving away from being in the position of being catered to, which has the side effect of them being on the hook for special charges as a result.
 
220W are confirmed by PCGH and Planet3DNow. They got the info from AMD directly. Apparently it is mentioned on some internal slides that they are not allowed to share at this time.
 
boxleitnerb said:
220W are confirmed by PCGH and Planet3DNow. They got the info from AMD directly. Apparently it is mentioned on some internal slides that they are not allowed to share at this time.
Click to expand...

IMG0041750.png


UniversalTruth said:
Hmm, sound logical and interesting.

However, do you know when AMD intends to change the manufacturing process so this performance gets to more sane limits?
Click to expand...

This new chip should be about ~15% faster than the current fastest Vishera. Steamroller is supposedly at least that much better than Piledriver, so my guess is that we'll see this level of performance in a 125W chip some time next year.
 
Status
Not open for further replies.

Similar threads

F
Dedicated gaming GPUs bottom of the priority list?
2
Replies
34
Views
3K
Seanspeed
S
F
RDNA5 Speculation
Replies
2
Views
1K
Erinyes
E
T
Re-litigating Radeon market performance *spawn
2 3
Replies
40
Views
3K
Dictator
D
D
Intel, AMD, Microsoft & Others Team Up For Develop The “UALink”, Direct Competitor To NVIDIA NVLink
Replies
3
Views
868
pcchen
P
xpea
Datacenter GPU market supply and demand (2024)
2
Replies
28
Views
3K
Erinyes
E
Back
Top