• 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.

Updated AMD roadmap (rumours)

Soldato
Joined
6 Feb 2019
Posts
17,624
Highlights

* Zen3 will be followed by: Zen3+, TSMC 6nm on AM4 with ddr4 and pcie4

* Followed by: Zen 4, TSMC 5nm on AM5 with DDR5 and pcie5. Launch quarter is Q3 2022. Also Zen 4 includes an RDNA2 based iGPU, all Zen 4 CPU include an iGPU chiplet.

RDNA3: TSMC 5nm, launch quarter is Q3 2022, target performance is 50% increase per watt

Zen 3 Threadripper target launch month: August 2021

https://www.chiphell.com/thread-2314832-1-1.html

https://www.reddit.com/r/Amd/comments/mjst0p/latest_amd_cpu_roadmap/

https://mobile.twitter.com/Olrak29_/status/1378488719787786240
 
Last edited:
Soldato
Joined
21 Jan 2016
Posts
2,915
Will be interesting how Warhol stacks up against Alderlake, previous rumours had suggested relatively mild uplift for Warhol and quite a significant gain for Alderlake... with a slight process change though perhaps Warhol might be better than expected?

I do like the suggestion it’ll stay on AM4 though, can pick up a Warhol chip and stay relevant enough well into the ddr5 generation to allow the new platform to mature.
 
Soldato
Joined
30 Jan 2009
Posts
17,192
Location
Aquilonem Londinensi
All seems reasonable. Can't believe they didn't have a small iGPU on Zen already tbb, saying that APUs exist doesn't really cut it as they usual trail by many months and until recently haven't offered more than 4 cores, let alone 12/16 etc.

Zen 3+ also makes sense given the climate and Apple jumping on 5nm, leaving 7nm capacity behind. I've read elsewhere AMD were buying up all the capacity they could in advance on 7nm.

RDNA reaching 50% pert/watt improvement I can see, by going wider rather than faster, but this metric usually has lots of asterisks.
 
Associate
Joined
27 Mar 2010
Posts
1,468
Location
Denmark
Not confirmed/rumors
q5elfp5xg4r61.png
 
Last edited:
Associate
Joined
27 Dec 2008
Posts
405
Zen4 and RDNA3 coming in Q3 2022 sounds a bit late. AMD will get 5nm production capacity before the end of this year, I doubt they'll not release anything for basically a year unless they're planning to use it on APUs.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
Will be interesting how Warhol stacks up against Alderlake, previous rumours had suggested relatively mild uplift for Warhol and quite a significant gain for Alderlake... with a slight process change though perhaps Warhol might be better than expected?

I do like the suggestion it’ll stay on AM4 though, can pick up a Warhol chip and stay relevant enough well into the ddr5 generation to allow the new platform to mature.

Intel made claims of a significant uplift for Rocket Lake over Comet Lake, at best it is 10% better, at worst 10% slower, while i didn't expect it to be quite that bad i already knew it was not going to be anything like Intel's claims, Intel already have a decade + history of making wildly exaggerated claims about the performance upgrade of their next CPU.

AMD over the last few years have under promised and over delivered, i'm expecting just as much from Warhol as i am from Alder Lake.

We need to stop buying in to Intel's marketing, all the early Rocket Lake hype was embarrassing, it was never going to be that good, not from Intel, and its turned out to be a pile of crap.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
Over the years Intel's lack of generational upgrades was always blamed on AMD, never Intel's fault, always AMD's fault for their lack of competition.

AMD are pulling ahead of Intel with every subsequent release and are utterly dominant in DIY market share, Intel still can't improve on their previous generations, the stagnation remains.

Its time we saw Intel for what they really are.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
I'm not really a fan of either company, but I do agree with what some have said before, that you could've had the same performance years earlier with Intel. The Tech Deals guy explains it well


Before i start i also think the 5800X is too expensive, but that's very different from saying its over priced, for the level of performance you're getting, in a historical context, it is great for $450. Yes we are in different times now, but it was AMD who actually made that happen and AMD are competing with themselves at this point.

So, I find his argument really strange. He makes the argument the price of the 5800X is too high (Again i agree) but the example he uses is mindbogglingly daft, he uses the example of a listener / viewer... whatever, who replaced his Ryzen 2700 with a 5800X and was pleased with the increase in FPS he gained from that with his 2080TI, he then makes the argument that if he only had a 2060 he would not have seen that up rate in performance, probably true, but he doesn't, he has a 2080TI, then he says: the 5900X is only $100 more, what an absolute moron, he cites his GPU being fast enough to warrant the CPU upgrade as if that's the only reasonable reasoning behind doing it and then gets to his point by citing the 5900X is only $100 more, as if that is some sort of big brain revelation. So 5800X only because his GPU is fast enough, but the wrong choice because the 5900X is only $100 more. The 5900X would have given this person 0 benefit over the 5800X, for $100 more.

The 5800X does have its place, its right there where you get those really high performance 8 cores that are on the precipice of the highest gaming performance possible without spending more and getting 0 benefit for it, but i maintain it is too expensive, i just wish morons wouldn't attempt the mental gymnastics to make the point.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
I'm still whatching... the 2700X and 9900K came out in the same year, yes, but they did not come out at the same time, the 2700X was months before, so depending on when he bought his 2700X there was no 9900K.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
So what's Zen 3+ offering over Zen 3? Likely to run on existing B550/570 boards?

Same DDR4, same PCIe Gen 4.

No one knows yet but its almost certain to be for existing AM4 boards, yes, like Zen to Zen +, its is Zen 3+, it will have some minor architectural tweaks, its probably a mixture of a small IPC gain and some more Mhz resulting in +10% and drop straight into any 500 series Motherboard.
 
Soldato
Joined
26 Aug 2004
Posts
5,032
Location
South Wales
I'm still whatching... the 2700X and 9900K came out in the same year, yes, but they did not come out at the same time, the 2700X was months before, so depending on when he bought his 2700X there was no 9900K.
So it was, if counting gaming only it looks like the 8700k is maybe a better comparison as it was already faster than the 2700x before it came out.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
So it was, if counting gaming only it looks like the 8700k is maybe a better comparison as it was already faster than the 2700x before it came out.

Yeah, at that time the 8700K was the best gaming CPU.

I can see why this person got the 2700.

There is a tendency these days amongst most reviewers, so far as i can tell, to compress CPU gaming reviews down to what CPU can drive the current GPU at settings which most gamers might use, so 1080P with high or highest IQ settings, i get why they do this and certainly this type of testing has its place.

The problem is doing that a 2700 can look like its got the same gaming performance as the 8700K, and the 2700 is cheaper, so you look at these charts and you see Ryzen 2700 100% £250, Core i7 8700K 106% £350, any reasonable person who doesn't understand what's really going on here is rightfully going to think the 8700K is only 6% better and £100 more expensive, but it isn't, the 2080TI can get 6% more out of the 8700K than the 2700 and that's a limitation of the 2080TI, not the 8700K, skip forward a couple of years with a faster GPU suddenly the 8700K is 25% faster than the 2700, now the 2700 is a significant bottleneck in your system.

Whose fault is that? these reviewers for not doing their job properly, IE using methods that show which CPU is actually the more powerful one, even if that means running the test game at 480P, like Anand do, the point is not that you don't play games at that resolution, the point is which CPU is actually faster so your readers can make a properly informed decision.
 
Caporegime
Joined
17 Mar 2012
Posts
47,774
Location
ARC-L1, Stanton System
I'll give you an example of this, Hardware Unboxed here, in the video where this slide came from are arguing the Ryzen 5600X is not really a much better gaming CPU than the Ryzen 3600.

As you can see on the face of it that's true, its +7%, he used a RX 6700XT or an RX 6800, can't remember which, not the fastest GPU currently available.

In the seconds slide, Steve Burke, same game, same benchmark, the difference is he used an RTX 3080.

Now the 5600X is not 7% faster, its a whopping 46% faster!!! notice HB slide all the CPU's top out at around 200 to 220 FPS, this is a limitation of the GPU he used, not the 5600X, now look at the FPS of the 3600 in GN's slide, its about 220 FPS.... Yup, this sort of crap makes my blood boil!

bKycQcc.png

YbkeDDn.png
 
Back
Top Bottom