Here is an alleged photo of AMD Radeon Big Navi Radeon RX 6000
Yesterday Lisa Su from AMD announced that big Navi is going to be announced next month (and there have been far too many announcements about announcements). But meanwhile, in Asia, there's always a guy that has something to show. And as it turns out, that would be a photo of big Navi.
The photo shows the backside' of the allegedly a Big Navi prototype PCB. You can see that Big Navi indeed is big as well as eight memory SMT traces with paper labels "Typical XT ASIC" references for a "16 Gb Samsung GDDR6 memory."
Also, excellent cooler :) (again prototype sample)
That's a 256-bit memory interface then. The chip thus should be paired with 16 GB of GDDR6 memory; we don't think AMD would be silly enough to go with 8GB. It's an engineering sample of some kinds, hence all the paper labels and such. Well, we'll see it at the October 28 launch.
Update Your NAS Server - There is a Vunerability in Samba - 05/26/2017 08:14 AM
There is a new a new vunerability discovered with Samba it affects versions 3.5 (released March 1, 2010) and onwards, the defacto standard for providing Windows-based file and print services on Unix ...
AMD: There is no such thing as full support for DX12 today - 09/01/2015 09:03 AM
After yesterdays turmoil on a lacking DX12 feature for Nvidia AMD’s Robert Hallock shares that FuryX is also missing a number of DX12 features. ...
Senior Member
Posts: 18517
Joined: 2008-08-28
If its real and its second best navi its strange that it has 16gb. Not that i complain though. It spells doom for 3070.
Senior Member
Posts: 1309
Joined: 2003-09-14
I'm going with the 'Bits and Bytes' theory, that this is 16Gbit chips of Samsung GDDR6 being shiowcased, along with the A0 rev of the card.
It looks like those stickers were put there to show off the components to people, maybe during a backroom demo a long while ago ?
If Nvidia can keep their cards under wraps until launch day, I suspect AMD can too.

(I'm also secretly hoping that the differentiation between PRO and Consumer GPU's at AMD will enable the removal of Double Precision hardware from the consumer side, and increase the number of 'simpler' shader units. Remember what happened between the Radeon 2xxx series and the 3xxx series ? a large jump in shader count.
Could this be why Nvidia upped the shader count so much ? because of a possible 80CU / 128 shaders per CU card at AMD ?)
Senior Member
Posts: 140
Joined: 2015-12-21

At the same time a 3080 has the same number of ROP, TMU, Int ALU and similar clock speed than a 2080ti. In the other hand, yes, the Ampere card has 2x the FP ALU and more memory bandwidth, but unless you look at cases where those were serious bottleneck for a 2080ti, the improvement won't be massive.
Nvidia is paying the price of Samsung 8nm and zero improvement to clock speed....
Senior Member
Posts: 1434
Joined: 2004-02-14
Well if it actually has 16GB it's going to be 256-bit memory controller, no other way that will actually work using an 8 x 2GB memory configuration. 8 x 32-bit memory controllers = 256-bit. The # of memory controllers has to match the # of chips on the PCB. If they do GDDR6X same as the RTX 3080, it would have 608GB/s using 19Gbps GDDR6X memory.
Short of them doing 16 x 1GB configuration which I also doubt as that would make for a very crowded PCB layout. Also kinda rules out them putting them on the back of the PCB too as I don't think the backplate would provide enough cooling for GDDR6 or GDDR6X. Not to mention a 16 x 1GB memory config would end up being a 512-bit memory bus.
Senior Member
Posts: 1434
Joined: 2017-02-14
If its real(I don't believe it is) its probably Navi22 which is the bigger but not biggest die. There is no way the largest Navi die will have a 256bit memory controler.