• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Navi 23 ‘NVIDIA Killer’ GPU Rumored to Support Hardware Ray Tracing, Coming Next Year

Status
Not open for further replies.
Soldato
Joined
19 Dec 2010
Posts
12,019
You should know that Charlie Demerjian is one of if not the most reliable source of credible information in the industry.
If you don't take what he writes in his articles, then I have nothing more to discuss on the matter of "factual" things.

I presume this is Sarcasm?
 
Permabanned
Joined
2 Sep 2017
Posts
10,490
I presume this is Sarcasm?

I presume it is not. Who told us the truth about hot and unmanufacturable Fermi, who told us the truth about intel not being able to produce things on N10 node?

Nvidia’s Fermi GTX480 is broken and unfixable
https://www.semiaccurate.com/2010/02/17/nvidias-fermigtx480-broken-and-unfixable/

Intel kills off the 10nm process
https://semiaccurate.com/2018/10/22/intel-kills-off-the-10nm-process/

Why SemiAccurate called 10nm wrong
https://semiaccurate.com/2019/01/25/why-semiaccurate-called-10nm-wrong/
 
Man of Honour
Joined
13 Oct 2006
Posts
90,813
Only Fermi wasn't unfixable and broken at least not to the degree he was implying - sure there were some issues with power and thermals but nothing like as bad as he was claiming.

The chip is unworkable, unmanufacturable, and unfixable. If Nvidia management has any semblance of engineering sense, it will declare its usual victory with Fermi and focus resources on Fermi II, something it is still promising for 2010. The changes needed to fix the first Fermi are basically not doable until the chip is shrunk to 28nm, the size is just too big.

This puts any hope for Nvidia out until 2011 for anything but PR stunts. The derivative parts for Fermi exist only on paper, they haven’t taped out yet. If Nvidia does tape them out, they will have the same size, yield and power problems as Fermi GF100. ATI will price them into the red, so there is no way Nvidia can make money on the derivatives, and there is no way it can fix the problems in time to matter either. Nvidia will not have any economically viable DX11 parts until the last days of 2010 if everything goes well from here.

Yet nVidia managed to get the GTX480 out the door, commercially viable albeit only just, in early 2010...

If you've been following this as long as I have you'd know Charlie has only been right about half the time and often wrong because he allows a dislike of nVidia to blind him.
 
Last edited:
Man of Honour
Joined
13 Oct 2006
Posts
90,813
You keep writing the same thing, without countering my argument with anything meaningful but rumours without any factual numbers.

£900 graphic card is TOO EXPENSIVE regardless if it is AMD or Nvidia. FULL STOP.
Anything over £600 is too much also, and that from me who bought a £700 295X2 and £730 GTX1080Ti.

I would be surprised if AMD pulls a 500mm2 GPU at 7nm at sub £1000 price. As I doubt Nvidia going to keep all these RT & Tensor cores if goes to 7nm.
More likely Ampere is 12mn until the MCM design is on. Otherwise expect £2000+ RTX3080 and over £2500 for the Ti models.

And all these are related to yield numbers in relation to the size of the chips. If we move to MCM design, I wouldn't be surprised to see something like an RTX2080Ti at sub £400.

nVidia is not dumping, or paring back, RT cores any time soon.
 
Permabanned
Joined
2 Sep 2017
Posts
10,490
Only Fermi wasn't unfixable and broken at least not to the degree he was implying - sure there were some issues with power and thermals but nothing like as bad as he was claiming.



Yet nVidia managed to get the GTX480 out the door, commercially viable albeit only just, in early 2010...

If you've been following this as long as I have you'd know Charlie has only been right about half the time and often wrong because he allows a dislike of nVidia to blind him.

Technically, Charlie was right about GF100. It was launched as late as third revision of the chip A3 and never got a fix.
Later, nvidia had to release Fermi 2.0, GF110 which had less transistors and smaller die area.

But we all know how it all went, with all the memes of the grill Fermi.
 
Permabanned
Joined
2 Sep 2017
Posts
10,490
Navi-21.png

https://www.techpowerup.com/gpu-specs/amd-navi-21.g923
 
Associate
Joined
7 Apr 2017
Posts
1,762
Art that die size and on 7nm vs 12nm if this doesn't totally obliterate the rtx 2080ti then it's a huge fail.

Still think AMD will manage to s*** the bed on this one and snatch defeat from the jaws of victory.
 
Permabanned
Joined
2 Sep 2017
Posts
10,490
Art that die size and on 7nm vs 12nm if this doesn't totally obliterate the rtx 2080ti then it's a huge fail.

Still think AMD will manage to s*** the bed on this one and snatch defeat from the jaws of victory.

There is still room to make it bigger - Fiji was 596 mm^2.

21 B transistors is only 12.9% more transistors than the 18.6 B transistors in RTX 2080 Ti.

So, this would be almost like an optical shrink of TU102.

21 B transistors is really disturbing, unless AMD found a way to increase the performance out of these transistors.
 
Status
Not open for further replies.
Back
Top Bottom