AMD: Speculation, Rumors, and Discussion (Archive)

Status
Not open for further replies.
So how likely are the rumors P10 is being pushed to october. Supposedly failing validation above 850MHz?!
Source?

A few things:
1) Odd news in light of the rumors that the PS4 Neo uses Polaris and runs at 900+.
2) Not inconsistent with the unverified benchmark scores linked to an 800MHz sample.
3) Might be inconsistent with the idea that AMD was gearing for an earlier launch and was more "ready", unless this problem did come up rather unexpectedly.
4) If this was known for some time, AMD might have been hyping things for users and investors while knowing there was a problem.
5) Could this have impacted whether boards could be shown? How different could the innards of the test setups look if AMD wasn't gearing up with more final board configurations? Would AMD be going through those motions if something as basic as "getting at base clocks" came up back when they were insisting they were ready to go?
 
So how likely are the rumors P10 is being pushed to october. Supposedly failing validation above 850MHz?!
you mean based on :
http://www.overclock.net/t/1599305/vc-first-gtx-1080-benchmarks/1230#post_25149678
Polaris 10 just got pushed back to October.

http://www.overclock.net/t/1599305/vc-first-gtx-1080-benchmarks/1240#post_25149690
What I can tell you, and this is only part of the reason, is that they need the time to get their GPU up to speed. It's failing validation above about 850 MHz.

http://www.overclock.net/t/1599305/vc-first-gtx-1080-benchmarks/1290#post_25151235
I've provided plenty of references for my credibility. I stand by AMD releasing Polaris in October.

No idea, but the guy has a positive history of good leaks (first Titan specs and benchmark that were correct)
 
Well we will see shortly if its an announcement at the end of this month and no real date given for when availability will be, then we will have a clearer picture of what is going on. But they seemed to be already hitting 850 mhz at least from other rumors out there but that could be the cut down version of P10
 
Obviously P100 was a fake launch. Nothing was shown running.

original.jpg
 
Hitting as in being able to controlledly demo it and validating it may be two different things, mind you. I'm not saying this is really the case since I wouldn't know.
 
true. I don't know, but having shown demos of the cards running and waiting almost 2 quarters to launch is a large lead time.
 
He is specifically talking about about the P10 though, not the P11. The latter one may appear in laptops rather soon.

PS4 Neo might be based of the P11 as well, so it's not necessarily effected.
 
I haven't been following this, but is PS4 Neo not based on an SoC anymore but has a discrete GPU?
 
Judging by what I read there, I fail to see where there's any strong indication that the author actively hints at PS4 Neo using not an SoC anymore. In fact, sentences like „AMD has created both of its current-gen console processors so far by taking older, off-the-shelf components and disabling a couple of compute units. In effect, Xbox One got the Radeon HD 7790, while PlayStation 4 got a more capable, semi-custom Radeon HD 7870.“ do not inspire confidence in very carefully chosen wording on behalf of the author. But then, english obviously isn't my native language, so I might be dead wrong there.
 
No idea, but the guy has a positive history of good leaks (first Titan specs and benchmark that were correct)
No, he hasn't.
His account is a little over a year old, and out of the blue he's claiming that he used to own a hardware website that supposedly brought those leaks, but is pretty much gone from the Internet (no google cache, no archive.org, nothing).
Quite convenient.
He also said that several websites would start breaking the news "a couple of hours" after his post... 10 hours ago.


He could be right and that would be rather depressing for AMD (although they never promised both GPUs for Summer), but that 1 year-old forum account isn't a reputable source at all.
 
I'm thinking a chip whose performance would be between Polaris 11 and Polaris 10. If P11's performance is equivalent to a Pitcairn and P10 to Hawaii, then Vega 11 could have Tonga's performance and Polaris 11's power consumption.
Carrying only one stack of HBM2 for 4GB at 256GB/s, they could make a "premium low-power" chip that would fit into all high-margin products like macbook pros, Surface Book, slim laptop workstations, etc.

Have we had any confirmation that there isn't a HBM2 memory controller in Polaris?

If the HBM2 memory controller size hasn't grown significantly from HBM1, a 1024bit HBM2 memory controller for a single HBM2 stack would use a rather small portion of the Polaris 10 die and allow AMD to produce such a 'premium low-power' product once more HBM2 suppliers are on-line without designing a whole new chip.
I guess we'll find out if AMD publish the die photos.

(As we have seen in 'Sea Islands', AMD product family names can be more marketing than architectural. Polaris with HBM2 enabled could be re-named into the Vega family.)
 
Have we had any confirmation that there isn't a HBM2 memory controller in Polaris?

Yes, the moment AMD presented several slides claiming that Vega's main feature was HBM2.
Doesn't mean Polaris 10 couldn't use HBM1, though, but memory amount could be a problem unless they go for 8-Hi stacks.
 
Yes, the moment AMD presented several slides claiming that Vega's main feature was HBM2.
Doesn't mean Polaris 10 couldn't use HBM1, though, but memory amount could be a problem unless they go for 8-Hi stacks.
AFAIK no-one makes 8-Hi HBM1.
 
Yeah I would expect it to be a custom SOC again.
The PS4 was 18CU (technically 20 but 2 disabled) with 1152ish shader processors - I think anyway.
Cheers
 
Status
Not open for further replies.
Back
Top