Discussion RDNA4 + CDNA3 Architectures Thread

[DisEnchantment]

With the GFX940 patches in full swing since first week of March, it is looking like MI300 is not far in the distant future!
Usually AMD takes around 3Qs to get the support in LLVM and amdgpu. Lately, since RDNA2 the window they push to add support for new devices is much reduced to prevent leaks.
But looking at the flurry of code in LLVM, it is a lot of commits. Maybe because US Govt is starting to prepare the SW environment for El Capitan (Maybe to avoid slow bring up situation like Frontier for example)

See here for the GFX940 specific commits

History for llvm/lib/Target/AMDGPU - llvm/llvm-project

The LLVM Project is a collection of modular and reusable compiler and toolchain technologies. - History for llvm/lib/Target/AMDGPU - llvm/llvm-project

github.com

Or Phoronix

More AMD "GFX940" Enablement Work Landing In LLVM - Phoronix

www.phoronix.com

There is a lot more if you know whom to follow in LLVM review chains (before getting merged to github), but I am not going to link AMD employees.

I am starting to think MI300 will launch around the same time like Hopper probably only a couple of months later!
Although I believe Hopper had problems not having a host CPU capable of doing PCIe 5 in the very near future therefore it might have gotten pushed back a bit until SPR and Genoa arrives later in 2022.
If PVC slips again I believe MI300 could launch before it

This is nuts, MI100/200/300 cadence is impressive.



Previous thread on CDNA2 and RDNA3 here

Question - Speculation: RDNA3 + CDNA2 Architectures Thread

Man I have been dying to make this one for a while now. First rumours for RDNA3 are here so new thread time! Just going to start off with this one for now: kopite7kimi on Twitter: "@VideoCardz Ah, I mean a simple mcm design with 10240 cores is not enough. Because the lift from RDNA2 to RDNA3...

forums.anandtech.com

 

[Mopetar]

There was literally no hype around RDNA2 or 3 or anything; the plebs always expect another Vega from AMD and get mad whenever it's not and AMD isn't forced to drive pricing down.

I'll direct your attention to the RDNA2/3 rumor threads on this very forum as clear counter-proof to your claim. Anyone who was around and posting in those knew how wild things got.

RDNA2 was actually an instance where AMD even managed to deliver on the hype to some regard.

 

[adroc_thurston]

-TBF the hype around RDNA3 was kinda nuts

There wasn't any.
Everyone expected RDNA2-ish perf/w bump with more watts to get more goodies on the top end.

I'll direct your attention to the RDNA2/3 rumor threads on this very forum as clear counter-proof to your claim

Again, random forums are whatever, mainstream perception matters and your mainstream plebbitor expects anudda Vega.
Always.

 

[GodisanAtheist]

Again, random forums are whatever, mainstream perception matters and your mainstream plebbitor expects anudda Vega.

-There is no mainstream perception. The mainstream does not theory-craft future products. The mainstream buys whatever is in front of them within their budget so long as it is branded Nvidia at any given time and are happy with it.

There is industry/insider and enthusiast perception, and that's what matters in this case.

 

[maddie]

There wasn't any.
Everyone expected RDNA2-ish perf/w bump with more watts to get more goodies on the top end.

Again, random forums are whatever, mainstream perception matters and your mainstream plebbitor expects anudda Vega.
Always.

You pushed much higher performance expectations, just not here.

 

[Panino Manino]

AMD fans always create such a massive amount of hype around their future products that even if the card turns out to be 10% on top for 75% of the price they'd still be disappointed because it doesn't do their taxes and make them a ham sandwich afterwards.

I think what people expect from AMD is to launch a card that it's faster than Nvidia's top competing card.
What we expect that is that the new card will be faster than the older card, with each competing brands alternating at the top. But for how many years AMD don't have a card that is faster than Nvidia's top? Since the days of the 980ti?

We see AMD "doing the impossible" against Intel with CPUs, we were expecting that with time and engineers the GPU site would improve, but besides "surviving" things feels pretty stagnant. They're not even trying to best Nvidia and the software side that with each generation becomes more important is still lagging.

Some times AMD gives some hope that things can change, but than a new arch like RDNA 3 falls bellow their own expectations, so of course some will be disappointed.

There wasn't any.
Everyone expected RDNA2-ish perf/w bump with more watts to get more goodies on the top end.

There was some hype as always, but I think this time the hype deflated before launch as soon as people know that AMD was going backwards and changing to dual issue.

 

[adroc_thurston]

You pushed much higher performance expectations

Yea, which is where they missed.
This was supposed to clock a whole lot faster!

But for how many years AMD don't have a card that is faster than Nvidia's top?

This is merely a question of building a big enough config.

but besides "surviving" things feels pretty stagnant

Ughhh have you seen MI300?
That's very GPU.

 

[Panino Manino]

Ughhh have you seen MI300?
That's very GPU.

Things may be better on the markets CDNA is aimed at, but I'm talking about the consumer market and RDNA3.

This is merely a question of building a big enough config.

If they can't or don't want doesn't matter.

 

[adroc_thurston]

but I'm talking about the consumer market and RDNA3.

Client dGP is filled with fickle children thus a big, overkill part is a harder sell there.

If they can't or don't want doesn't matter.

It really-really does.

 

[Timorous]

This was supposed to clock a whole lot faster!

Technically it does, just uses an unreasonable amount of power. Presumably it was supposed to clock well in excess of 3 GHz at 300W+ rather than 600W+

 

[adroc_thurston]

Presumably it was supposed to clock well in excess of 3 GHz at 300W+ rather than 600W+

Yes!
Navi31 at 350W was a 3.5GHz design once.

 

[Joe NYC]

Yea, which is where they missed.
This was supposed to clock a whole lot faster!

This is merely a question of building a big enough config.

Ughhh have you seen MI300?
That's very GPU.

We will see if AMD brings any of the technologies from MI300 to the mass market client.

Zen 5 and RDNA4 are already too far along, but we will see what the next gen brings.

I think 2 ideas will make their way into the following gen products. One, mentioned by Lisa, that MCM is the next motherboard.

2nd, a takeaway from Mi300 is that N6 base die will be the next interposer, in the client oriented MCMs. With chips stacked on top of it, perhaps even memory chips in a non-standard HBM.

 

[Gideon]

OUCH. GG if true.

https://twitter.com/x/status/1687388515670056960

Navi4 lineup will not have any high-end GPUs

Think of it like RDNA1 or Polaris generation.

Judging by the predating @uzzi38 reaction this looks to be true:

https://twitter.com/x/status/1687379360276185088

It's almost public knowledge now (and probably will be eventually) but congratulations guys, you did it!

The way things are going soon there really will be only one desktop GPU manufacturer left. Just what you all wanted!

I can understand the RDNA1 generation as AMD was totally broke for most of the time it was under development, but ...

Frankly, I don't get why they did it. Are the economics really just not there? Or rather is it a sign that chiplet-based GPUs aren't really working out yet?

 

[uzzi38]

OUCH. GG if true.

https://twitter.com/x/status/1687388515670056960

Judging by the predating @uzzi38 reaction this looks to be true:

https://twitter.com/x/status/1687379360276185088

I can understand the RDNA1 generation as AMD was totally broke for most of the time it was under development, but ...

Frankly, I don't get why they did it. Are the economics really just not there? Or rather is it a sign that chiplet-based GPUs aren't really working out yet?

They had big plans and cancelled them because they felt they didn't have the time to chase them.

There are potentially two reasons why:

1. Validation for what they were trying to do would bring the halo parts too close to RDNA5 in terms of timescale.

2. They're giving up (or if you want to phrase it nicely, deprioritising) desktop graphics.

If you want to ask me which one I think it is, then to put it kindly, I'm leaning towards the second personally.

I could VERY easily be wrong, but this is the way I'm leaning for now.

 

[Joe NYC]

https://twitter.com/x/status/1687379360276185088

If there is Karma, NVidia will raise prices another 25-30%. $1,999 for 5090.

 

[BeverlyHillsMoon]

It's over bros

 

[Heartbreaker]

OUCH. GG if true.

https://twitter.com/x/status/1687388515670056960

That doesn't really make sense, why should we believe it?

Polaris generation would, only be nothing about RTX 5060 competitor, which makes no sense.

 

[SteinFG]

Don't think AMD will abandon 24GB cards. But the amount of CUs can decrease. Maybe back to 60 or 80.

 

[jpiniero]

That doesn't really make sense, why should we believe it?

Polaris generation would, only be nothing about RTX 5060 competitor, which makes no sense.

People only buy AMD GPUs when their margins suck. I'm still of the opinion that AMD badly needs a True Chiplet architecture... and if that still needs more time (or they've just decided it's not worth it) then yeah this is the inevitable result. I'll admit I didn't think this would happen with RDNA4 though.

It's probably the Strix Halo part but with GDDR7.

 

[Tup3x]

Well, if they could make APUs that match or beat consoles, many would be happy with just those.

 

[Heartbreaker]

People only buy AMD GPUs when their margins suck.

If NVidia margins are as egregiously fat as people here claim, there should be plenty of room for AMD to still have healthy margins, at it's slightly better than NVidia pricing.

 

[DisEnchantment]

Usual gaslighting started, which can be found in plenty in the original RDNA3 thread. Going back to the beginning of the thread and the end of the thread is pretty obvious what transpired.
Not sure why they even want to make something bigger. Just fix RDNA3 if they are totally incompetent.

What is going to happen to @TESKATLIPOKA what tables and projections will he make then?

 

[Leeea]

We all saw this coming.


The next gen could be very good cards. Boring and uninspired, but with all the errata worked out making for a very smooth user experience. Nobody will buy them, but that is normal.


I would be very surprised if they drop the price at all. AMD will just let Nvidia climb to the stratosphere.

 

[coercitiv]

They had big plans and cancelled them because they felt they didn't have the time to chase them.

AMD no longer has "the time" to make big chips, meanwhile some AI startups are buying cases of consumer GPUs. Comedy gold.

 

[tamz_msc]

If this rumor is true, then it is the logical conclusion of the fact that gaming GPUs are barely turning a couple of dozen millions in profit for AMD.

 

[Heartbreaker]

If this rumor is true, then it is the logical conclusion of the fact that gaming GPUs are barely turning a couple of dozen millions in profit for AMD.

IMO, this is another nonsense rumor. Only going to about Polaris level ( about RTX 5060 competition) next generation makes no sense at all.

I could see them abandoning the very top level of competition, but not the midrange.

Their best selling RX 6000 part is the 6700 XT.

Having only low end parts means they could may also even get buried by Intel Battlemage GPUs.

Nothing about his makes sense, thus it's nonsense.