AMD: Speculation, Rumors, and Discussion (Archive)

Status
Not open for further replies.
I see mostly in 60fps ( 55-56 in the low, with 63+in high ) range who put it between the 980 and FuryX if i compare with this review http://www.guru3d.com/articles_page...merock_premium_edition_g_panel_review,24.html

Ofc, need to see the setting used.
Thats at max settings.The youtube test is custom, nearer to high:
350x700px-LL-670f9c8a_1080.jpeg
 
Well, videocardz' reasoning is lacking, comparing it to Hawaii (no compression, low GDDR5-clocks) and Fiji (totally different memory subsystem, and almost twice as many CUs contending for bandwidth) for conclusion.
 
"Long way to go" he says, so probably not finished silicon I suppose. Unless verification process and finalizing drivers and reference board design takes a huge amount of time... *grr!*
It can take time from working silicon to getting product out too, both Polaris 10 & 11 were demonstrated live in december as working chips
 
Is it just me or NDA breaking when it comes to AMD launches is far more rampant than Nvidia ones? Probably just me.
 
Is it just me or NDA breaking when it comes to AMD launches is far more rampant than Nvidia ones? Probably just me.

I recall some leaks prior to a few launches for both, but I haven't cataloged NDA breaking or really kept track. It's hard to know how much is NDA breaking versus fabrication since we don't have reference to the actual information currently under NDA.

It would take a fair amount of dredging up rumors over the years to get a good picture of it.
I thought the speculation was for the most recent example is that the card was procured outside of an NDA.
The odds of getting a larger volume product are higher, particularly if the launch is expected to be broad enough that samples and shipments cannot get as much monitoring for leakage. Lower price can also mean it may not have as much of an economic case to secure it enough.

Given how AMD has hyped this, its interactions with developers like Oxide, and how Koduri is tweeting up possible tapeouts, I'm not sure all of this is entirely without purpose. Polaris or later hardware in the PC and console space is being disclosed left and right for various reasons, and the combination of broader attack surface and various PR measures make it more likely.


I don't understand the appeal of video reviews/leaks/whatever. Who wants to watch a 10min video to get the information can be read in a few seconds?

I CAN'T READ!
 
Maybe Vega chips are in the same development stage both Polaris 10/11 was at that time. Early silicon, but functional.
 
Well, videocardz' reasoning is lacking, comparing it to Hawaii (no compression, low GDDR5-clocks) and Fiji (totally different memory subsystem, and almost twice as many CUs contending for bandwidth) for conclusion.
Yeah, on the plus side he meant to had clarified the ROPs/texture units with a 'source'.
Wrong context by the article in comparing it to Hawaii, should had been Tonga.
That is if the info is right anyway; but no-one still reports if Polaris 10 is an evolution of Tonga or Hawaii as I have asked in the past, or those that do know are under NDA.
Cheers
 
I don't think we should use some of the higher-level details of an implementation to say Polaris 10 is an evolution of Hawaii.
Hawaii and Tonga belong to two presumably separate IP ranges, with Tonga being from the more recent pool.
The number of CUs or width of the memory bus don't really figure into the technological "DNA".

I'm assuming Polaris would not revert back to Hawaii's older GCN revision and use that as a starting point rather than continue to make progress from where Tonga and Fiji had gone.
 
In another user's video the uploader comments the gpu gets quite hot.
Stands to reason, if it's just a tiny extruded solid hunk of alu rather than a dense thin-fin sink soldered to a vapor chamber filling most of the shroud as has been the typical choice these past few years.
 
Status
Not open for further replies.
Back
Top