AMD: Navi Speculation, Rumours and Discussion [2017-2018]

My brain reads this as Nvidia10..

 

Well, as long as it does have just an impact on the industry as GeForce 256 did

 

Navi10 is added to pre-release version of HWiNFO64


https://www.hwinfo.com/news.php

 

Well, we already have Intel CPU's with Radeon inside, so it is natural to expect AMD with GeForce inside isn't it?

 

edit: wrong thread

 

Since AMD will have no new GPUs in 2018 the slide must be wrong.

 

Vega is pretty young, I'm not surprised that 2018 is still a "Vega year". The problem is it did not deliver (or / and was too late in the first place). If Navi use the same human ressources than Vega, it's not a surprise that it's won't be ready until 2019...

 

2018 is the year, Vega-IP enters mass deployment.

 

On the other hand, the later Navi gets, the more palatable some of the speculated features might become That's assuming its design wasn't frozen at a point prior to them becoming achievable.

 

2019 is when 7nm will actually be available so it will be when Navi comes. Maybe AMD originally thought 7nm will come sooner or they planned to do 10nm Navi then cancelled?

 

Anyone seen this?

https://wccftech.com/amd-new-major-gpu-architecture-to-succeed-gcn-by-2020-2021/
https://www.tweaktown.com/news/60760/amd-working-next-gen-gpu-navi-2020-2021/index.html

"Back in October I exclusively reported that AMD would be launching Navi during SIGGRAPH 2018, in July/August this year. At the time everyone else called me crazy and that it wasn't going to launch that early, but now I've got other sources that are backing up my claims and saying that I was right. The first Navi reveal will be a professional card and not a consumer-focused Radeon RX Navi at all, that card will launch in 2019. Then a story on WCCFTech popped up which said that "Navi will be the last GCN-based architecture" and that it will be "succeeded by brand new micro-architecture in 2020/2021 timeframe, execution of GCN successor being led by new RTG leadership". This is interesting, so I reached out to some sources to clarify a few things. First, we have no idea how much the next-gen GPU architecture (post-Navi) was influenced by now ex-RTG boss Raja Koduri (love you, Raja!), who has defected and left for CPU competitor Intel. My sources said that Vega was such a "disaster" (their words, and mine)
and that the RTG team "don't know where they're going forward" and that (gulp) Navi is going to be "just as bad" as Vega."

 

Navi still based on GCN ?? Whoaw...

 

Vega isn't that bad tho, it's just not saleable upwards, which seems to be a problem with all GCN. Aside from the features that might or may not be completely bugged, the base performance of the smaller Vega parts seem to be good and power consumption seem ok too.

 

Well i dont know why this should surprising... Nothing indicated in the past years that AMD developed another GPU µarch.

 

Vega's fine, except that it's a year late compared to pascal, is a much larger ASIC (which doesn't really impact users, but does impact AMD), and assuming you rein in its wattage by keeping it in power saving profile.

I'm quite happy with my boards I must say. Current driver has good methods to keep the GPUs nice and cool, and it's even possible to game and run GPGPU apps simultaneously with acceptable or nearly no performance impact (depending on game of course; needs to be relatively shader-light game obvs...)

It's disappointing we won't see binning rasterizer or primitive shaders used after AMD made such hullabaloo about them, but I suppose that's life. *sigh* Cards do give pretty nice performance anyway tho.

 

The tweaktown article reads like speculative BS using imaginary sources.

 

Binning is being used.

 

Oh yeah? We have official confirmation of that now?

 

Yeah.
A slide, even.
It was posted somewhere here.

 

Right now, I think Vega 10's biggest problem that AMD didn't see coming (or failed to achieve with GF) is that the chip just hits a huge efficiency wall above 1.4GHz.

Vega 10 has 40% more transistors than Fiji while keeping the same number of execution units, halving the number of memory channels and bringing relatively few changes other than RPM. AMD stated that most of the additional transistors were put there to drive the clocks higher, but Vega 10 only clocks 40% higher than Fiji.
It's like the chip is gaining almost nothing from the 28nm -> 14FF transition, because 1.4GHz it's practically how much a GM200 can hit within 300W.

Then we have statements from @Nebuchadnezzar saying the 14FF process was never meant to be used with voltages anywhere near Vega's default values. And pre-launch statements from AMD officials saying Vega 64 would clock above 1700MHz (it never did, did it?).


It all points to AMD engineers planning on Vega 10 being able to clock comfortably at ~1650MHz with a ~220W TDP and 0.9-0.95V. And when the production units came up they realized the cards wouldn't clock above 1400Mhz at that vcore, so they panicked, overvolted and pushed the clocks as much as they could within a sub-300W TDP, which ended up being the 1-1.1V that we saw in the final product.
The 2-3% performance that Vega 10 loses in power saving mode makes it gain a whopping 30% in efficiency, or around 80W. And it was all just to achieve GTX1080 parity, which is a bit stupid IMO because driver tweaks usually amount to much more than that.