AMD looks ahead to Zen 4 and RDNA 3

[midian182]

Forward-looking: Zen 3 might have only just arrived, and RDNA 2 is around the corner, but AMD is already looking toward the next generation of its GPU and CPU technologies—Zen 4 and RDNA 3—which it says will offer the same performance and efficiency gains seen in previous generational leaps.

Speaking in an interview with The Street, AMD Executive Vice President Rick Bergman said RDNA 3 would offer similar 50 percent performance-per-watt gains as its RDNA 2 predecessor.

"Let's step back and talk about the benefits of both. So why did we target, pretty aggressively, performance per watt [improvements for] our RDNA 2 [GPUs]. And then yes, we have the same commitment on RDNA 3," said Bergman.

"It just matters so much in many ways, because if your power is too high -- as we've seen from our competitors -- suddenly our potential users have to buy bigger power supplies, very advanced cooling solutions. And in a lot of ways, very importantly, it actually drives the [bill of materials] of the board up substantially. This is a desktop perspective. And invariably, that either means the retail price comes up, or your GPU cost has to come down. We focused on that for RDNA 2. It's a big focus on RDNA 3 as well."

It doesn't take Sherlock Holmes to know Bergman is taking a dig at Nvidia's RTX 30xx series and its high power demands. He also mentions the importance of AMD's Infinity Cache tech.

"On Infinity Cache, it's somewhat linked to that as well, to a certain degree. If you've been in graphics for a long time, you realize there's a pretty good correlation between memory bandwidth and performance. And so typically, the way you do it is you jack up your memory speed and widen your bus to open up performance. Unfortunately, both of those things drive up power," Bergman added.

The AMD exec said that Zen 4, which is expected to land sometime in 2022, will be another major overhaul of the company's architecture, with improvements in all areas, especially when it comes to instructions per clock (IPC). When asked if most of the gains will come from IPC improvements, rather than increased core counts or faster clock speeds, Bergman said:

"[Given] the maturity of the x86 architecture now, the answer has to be, kind of, all of the above. If you looked at our technical document on Zen 3, it was this long list of things that we did to get that 19% [IPC uplift]. Zen 4 is going to have a similar long list of things, where you look at everything from the caches, to the branch prediction, [to] the number of gates in the execution pipeline. Everything is scrutinized to squeeze more performance out."

In the area of manufacturing nodes, we already know that Zen 4 will use TSMC's 5nm process, which is also behind Apple's new M1 SoC found in the upcoming MacBook Air, MacBook Pro 13‑inch, and Mac mini. With RDNA 3, however, Bergman wasn't clear on whether it will shift from 7nm to 5nm.

"Certainly EUV is an option with that process, and there are no technical limitations holding us back. Nor are there huge technical benefits either. So that's more of a manufacturing question. There are additional nodes on 7nm that we'll take advantage of over time. But again, nothing to disclose," Bergman says.

Finally, speaking about the upcoming Radeon RX 6000-series cards, the first of which launch next week (November 18), Bergman said AMD was aiming for 1440p as the perfect resolution for a great ray tracing experience. "…that was kind of the performance level that we targeted. Now it depends on particular games and everybody's systems and so on, but I think you'll find that we have very good ray-tracing performance overall. And the game support will be strong as we go through 2021, because again, we get that great leverage. It's just built in: You support ray tracing on Microsoft or Sony [consoles], you're supporting AMD on the PC side as well."

Permalink to story.

https://www.techspot.com/news/87558-amd-looks-ahead-zen-4-rdna-3.html

 

[joaopaulopage]

This is what happens when the company focuses on design and innovation. Absolutely necessary for the x86 ecosystem, imagine if AMD had failed Ryzen, maybe ARM would be slaughtering the x86 ecosystem.

 

[Aryassen]

This is what happens when the company focuses on design and innovation. Absolutely necessary for the x86 ecosystem, imagine if AMD had failed Ryzen, maybe ARM would be slaughtering the x86 ecosystem.

Good point, and Apple actually already entered the arena with their custom chip (though with Ryzen now established, it probably will be a niche product, but it does look promising).

I like that AMD are not sitting on their palms and they keep driving innovation. It also helps with establishing the longevity the brand's position (if partners see that not only it is good right now, but its ecosystem keeps evolving, that definitely is a plus, and probably encourages buy-in, and/or renewal)

 

[Lounds]

Dr Su knows how to run a company.

 

[Mr Majestyk]

Intel is screwed until at least 2023 when Redwood Cove appears on 7nm at which point it'll be up against Zen 5.

 

[defaultluser]

Well, now that we have a confirmed 2022 launch date for Zen 4, that's exactly when I am planning to upgrade my 4790k to a 5600 or 5700. Hopefully we will see massive AM4 closeouts / Zen 3 price cuts.

Thanks to Microcenter, should be able to upgrade both my main gaming PC and HtPC for around $600. By then, 2x16GB ram should be under $60.

 

[poohbear]

If Nvidia ends up buying ARM and clearing all the regulatory hurdles, will that be the end of the x86 architecture? We're all holding our breath to see what Apple does with their ARM desktop version, cause their mobile laptop version is pretty damn impressive if what their saying about its performance and battery life is true!

Well, now that we have a confirmed 2022 launch date for Zen 4, that's exactly when I am planning to upgrade my 4790k to a 5600 or 5700. Hopefully we will see massive AM4 closeouts / Zen 3 price cuts.

Thanks to Microcenter, should be able to upgrade both my main gaming PC and HtPC for around $600. By then, 2x16GB ram should be under $60.

I hear you! I'm going to upgrade my 4790k to Zen 3 next year on Black Friday. I imagine by then prices will have come down dramatically as Zen 3 will be competing against Rocket Lake and Alder Lake come Black Friday 2021. Hope to get the RAM, Mobo, & their 5900X for much cheaper by then! I can sell my 4790k CPU, Mobo, & RAM for $250 despite it being 7 years old by then. We sure got our money's worth from these 4790Ks huh?

 

[Ben1978]

Do those interested in stocks think AMD is a good investment at $80? They are really socking it to Intel at present and look like they can really challenge Nvidia too, now and in the future. I'm just not sure what their other business interests are compared to Nvidia who are at over $500. Do they have the potential to get to that kind of valuation?

 

[poohbear]

Do those interested in stocks think AMD is a good investment at $80? They are really socking it to Intel at present and look like they can really challenge Nvidia too, now and in the future. I'm just not sure what their other business interests are compared to Nvidia who are at over $500. Do they have the potential to get to that kind of valuation?

The last quarter was a homerun for AMD, and if it wasn't for the news that they're buying Xilinx for an astonishing $35 billion, it'd easily be in the high 90s! It's been dropping to $74 each time the Nasdaq gets hit with news of a vaccine or political uncertainty. Wait for the next Vaccine news, which will cause people to sell their tech and rotate into some of the beaten down travel stocks. That'll be your chance to get in @ $75ish. If their deal with Xilinx goes through it'll jump. If the deal with Xilinx doesn't go through...it'll also jump!(maybe even more) In the end, AMD is firing on all cylinders and it's a growth story.

I like this quote from this article:

"And the game support will be strong as we go through 2021, because again, we get that great leverage. It's just built in: You support ray tracing on Microsoft or Sony [consoles], you're supporting AMD on the PC side as well."

Just shows you how many tailwinds AMD has under it. The only real fear is if Apple manages to release an ARM Desktop CPU that smashes AMD's X86 equivalent...if Nvidia owns ARM by then...well Nvidia and AMD's rivalry will be pushed to the limit and Nvidia usually comes out on top. But who knows...Lisa Su has been doing phenomenally.

 

[defaultluser]

If Nvidia ends up buying ARM and clearing all the regulatory hurdles, will that be the end of the x86 architecture? We're all holding our breath to see what Apple does with their ARM desktop version, cause their mobile laptop version is pretty damn impressive if what their saying about its performance and battery life is true!

It is impressive , but when you add the overhead of running emulation, it's gong to be much slower, and it will take most application makers a couple years to port their Apple stuff to ARM native.

Apple is a special-case closed ecosystem that is not going to have any long-term effect on Windows ARM notebooks becoming more popular than they already are...but, they could potentially move them in the server world.

The ARM efficiency advantage does exists, but because it takes more instructions to get the same amount of work done, you need an absolute fetch monster to get around this. It's going to be a toughb road for both Apple and Nvidia trying to keep-up with their current yearly 20% IPC increases

 

[JamesBlond]

Well, now that we have a confirmed 2022 launch date for Zen 4, that's exactly when I am planning to upgrade my 4790k to a 5600 or 5700. Hopefully we will see massive AM4 closeouts / Zen 3 price cuts.

Thanks to Microcenter, should be able to upgrade both my main gaming PC and HtPC for around $600. By then, 2x16GB ram should be under $60.

I bought my RTX2070, i7-8700k 16GB, SSD just over £900, since I have had it I have made £700 and decided to run some cheap Lighting Network Nodes, paid off in 8 months, taking all my profits moving to 5900x and 6800XT, and I only have to add around £300 which I will get when I sell my current rig... dont waste your computer power, make money with it, get its life out of it and make it worth your while

 

[Endymio]

Do those interested in stocks think AMD is a good investment at $80? ... I'm just not sure what their other business interests are compared to Nvidia who are at over $500. Do they have the potential to get to that kind of valuation?

You're looking at the wrong metric. AMD's P/E at present is a lofty 113 to 1. NVidia's is somewhat lower, at 98. Companies for which growth is expected to be flat or light generally have P/E's in the 10-30 range, so investors are already pricing AMD to have monstrous profit growth over the next few years. The shares are a buy only if you think that growth is going to be even more astonishing than what they're already betting on.

 

[mbrowne5061]

ARM isn't killing x86, and x86 isn't going anywhere. People keep thinking these are two competing technologies, when they're almost complimentary (when they interact at all). x86 and ARM use completely different instruction sets, and excel at completely different tasks. ARM is meant to optimize performance in terms of power usage, and x86 is just meant to maximize computations regardless of power. You might see ARM gain traction in the low-power notebook segment, maybe even the low-mid range stuff too. But you'll likely never see it enter the mainstream on the desktop - especially not on a workstation.

All this hype about a 'fight' between ARM and x86 just reminds me of way back in the day when WiFi and Bluetooth were first showing up on the scene. Everyone was talking about 'which one would win', not realizing that they were two different technologies that were designed for different tasks that only seemed similar on the surface.

 

[Markoni35]

...so investors are already pricing AMD to have monstrous profit growth over the next few years. The shares are a buy only if you think that growth is going to be even more astonishing than what they're already betting on.

And that's why I see a warning sign in this story. Every time stocks (or cryptocurrency, or derivatives, or whatever) are predicted to have a steep growth, the opposite happens. Markets and predictors are often manipulated by the top players to create an illusion for the naive investors. It's hard to deduct when the growth will turn into free fall. Remember what happened in 2007/2008?

 

[Ben1978]

You're looking at the wrong metric. AMD's P/E at present is a lofty 113 to 1. NVidia's is somewhat lower, at 98. Companies for which growth is expected to be flat or light generally have P/E's in the 10-30 range, so investors are already pricing AMD to have monstrous profit growth over the next few years. The shares are a buy only if you think that growth is going to be even more astonishing than what they're already betting on.

OK, thanks. So you think much of the success that I'm expecting is already priced in. I'll wait for a pullback in that case. I already have some at $75. Only just started investing in March like a lot of other people (and stupidly bought gold instead of buying the dip). I'll need to actually put my tech knowledge to use in the future to spot success stories like AMD before general investors do.

 

[Ben1978]

You're looking at the wrong metric. AMD's P/E at present is a lofty 113 to 1. NVidia's is somewhat lower, at 98. Companies for which growth is expected to be flat or light generally have P/E's in the 10-30 range, so investors are already pricing AMD to have monstrous profit growth over the next few years. The shares are a buy only if you think that growth is going to be even more astonishing than what they're already betting on.

Thanks. Very useful reply.

 

[Endymio]

OK, thanks. So you think much of the success that I'm expecting is already priced in. I'll wait for a pullback in that case.

It's priced in by rational investors. But irrational ones are harder to predict. A wave of stories in the popular media telling the public what us here already know might loft AMD's share prices even higher. But if so, I believe it would give those gains back quickly. Rationally, I don't see AMD justifying a $130+ valuation, unless it also does to NVidia what it did to Intel. That GPU market is where future growth lies, not CPUs.