Speculation: Spring refresh for Ryzen

[Vattila]

As I've posted elsewhere, I think AMD fell short of their frequency targets for the Ryzen 3000 series. A later launch than expected, statements relating to frequency challenges, Max Boost redefinition, boost issues, 3900X shortages, and now the 3950X delay from September to November, do all corroborate this. I guess they aimed for 5 GHz, but fell ~8% short.

However, TSMC is firing on all cylinders. Amongst a number of process roadmap announcements, they now offer N7P as a refinement of their N7 process on which the Zen 2 chiplet is built. N7P is compatible with N7 design rules, so it should provide a fairly simple and cost-effective opportunity to optimise existing designs.

With that in mind, is there a chance that AMD will do a spring refresh based on a faster stepping of the CPU chiplet?

 

[DrMrLordX]

Hard to believe it's almost 7/7 and we still haven't seen any leaks or reviews of the XT chips.

 

[amrnuke]

Hard to believe it's almost 7/7 and we still haven't seen any leaks or reviews of the XT chips.

AMD must have really really clamped down on stuff. I wonder if they are tracing their sample parts somehow (since Zen2 / RDNA1) and some people got into trouble, and now AMD are using that as a warning.

 

[Hitman928]

I'm interested in seeing availability. The latest Intel release has been rough, at least in terms of the high end models, not sure about the lower end stuff. It'll be interesting to see if AMD is able to provide enough chips for wide availability or not.

 

[DrMrLordX]

I'm interested in seeing availability. The latest Intel release has been rough, at least in terms of the high end models, not sure about the lower end stuff. It'll be interesting to see if AMD is able to provide enough chips for wide availability or not.

AMD hasn't got a wafer supply problem. Intel's 14nm problems have been ongoing for awhile now. Die sizes keep rising etc.

AMD must have really really clamped down on stuff. I wonder if they are tracing their sample parts somehow (since Zen2 / RDNA1) and some people got into trouble, and now AMD are using that as a warning.

That was my thinking. They really should have showed hardware on June 16th though.

 

[lightmanek]

Well, it's 7/7 here already (UK) and no leaks ...
I'm waiting on benchmarks just to see if I should buy 3900XT or just save a bit and go with 3900X instead. Enjoyed my little R5 3600 and had fun overclocking it the last few weeks, but coming from TR1920X, I'm sad every time I open Task Manager in CPU view

Bring on the leaks guys!

 

[Hitman928]

AMD hasn't got a wafer supply problem. Intel's 14nm problems have been ongoing for awhile now. Die sizes keep rising etc.



That was my thinking. They really should have showed hardware on June 16th though.

I'm not worried about the wafer supply, more about the packaging and distribution network. The 3300x and 3100 have been very hard to find since they launched from what I can tell. I know AMD prioritized OEMs for these chips, but you would think they would have at least a decent amount in retail by now.

 

[VirtualLarry]

I know that AMD designates a "Golden core" for each chiplet, and uses those for ST boosts that go above a certain level, but it would be nice if, perhaps, hypothetically, that a 3600XT could boost multi-core to 4.2Ghz, while AT THE SAME TIME boosting a particular single-threaded load to 4.5-4.7Ghz on the "Golden core".

IMHO that would actually be perfect for gaming, and allow AMD the clear lead in gaming, because most game engines have a single CPU-heavy "master thread", and while they are multi-threaded, and can put loads on all available cores, there is benefit for allowing this CPU-heavy master thread to clock higher than the other cores/threads.

AMD could call this "AMD Advanced Game BOOST". I think that it would work.

 

[amrnuke]

AMD could call this "AMD Advanced Game BOOST". I think that it would work.

AMD Zen 2 Refresh
Now with AMD Game Cache* and AMD Advanced Game BOOST**
* for this iteration, may not have any clear effects on gaming, as we altered so much in Zen 2 compared to Zen/Zen+ that even we aren't sure where the improvements came from
** your kilometerage very verily may vary

 

[Panino Manino]

Oh, Anandtech's complete results mighty arrive a bit late.

https://twitter.com/x/status/1280129949672910849

 

[moinmoin]

The 3300x and 3100 have been very hard to find since they launched from what I can tell.

In Germany 3100 is readily available according to Geizhals (28 shops have it), 3300X on the other hand appears to be rare indeed (just 3 shops) and price jumped above MSRP at the start of this month. Mindfactory doesn't even list the 3300X to begin with (whereas 3100 is available and sold ~400 pieces so far), so availability appears to depend on markets for both separately.

Oh, Anandtech's complete results mighty arrive a bit late.

And such a typical Windows problem to have as well.

 

[jpiniero]

Have to say I am a little dissapointed in the all core frequencies the reviewers got. It's not much more than the X model.

 

[DisEnchantment]

Remove cooler, reinstate launch prices and offer new buyers a 100 MHz extra boost clock which they already are getting on late non XT models anyway.
Hmmm... 🤔

Lol, This whole exercise is probably to remove the Boxed Cooler . Another 20-25 bucks margin on top.

However, in the long run this being a SEP only, AMD could price these CPUs cheaper than they already are compared to non XT models and they could keep them along side the Zen3 DT 4000 series CPUs

 

[tamz_msc]

So these run hotter than the -X models, have margin of error improvements in performance, at most give a +200MHz higher all core manual OC, don't come with a cooler and cost up to 40% more than existing SKUs.

A big meh.

 

[Hitman928]

At Newegg the 3900XT and 3600XT are in stock as of 6:30 a.m. PT. The 3800XT is out of stock.

 

[Antey]

super super boring release

 

[Hitman928]

So these run hotter than the -X models, have margin of error improvements in performance, at most give a +200MHz higher all core manual OC, don't come with a cooler and cost up to 40% more than existing SKUs.

A big meh.

Yeah, I was hoping for slightly more on the manual overclock given the reddit posts, but other than that it's about as I expected. Not much to be excited about here unless you are really concerned with running single threaded only apps on a 6+ core CPU for some reason.

 

[jpiniero]

Lol, This whole exercise is probably to remove the Boxed Cooler . Another 20-25 bucks margin on top.

However, in the long run this being a SEP only, AMD could price these CPUs cheaper than they already are compared to non XT models and they could keep them along side the Zen3 DT 4000 series CPUs

I still think that Renoir desktop will replace the X and perhaps non-X models, and Renoir will arrive once stock runs low.

 

[beginner99]

What's the purpose of this release? Sounds like a money grab to me or is Zen 3 much later than expected? Can't see them releasing Zen3 on desktop in 2020 with this refresh.

 

[jpiniero]

Can't see them releasing Zen3 on desktop in 2020 with this refresh.

Works with the rumor that it'll be only 12 and 16 core. You'll notice that there is no 16 core XT model.

 

[DisEnchantment]

What's the purpose of this release? Sounds like a money grab to me or is Zen 3 much later than expected? Can't see them releasing Zen3 on desktop in 2020 with this refresh.

You'd be surprised to learn that Lisa talked about the XT launch and Zen3 "later this year" in the same video (at 84 seconds mark).

 

[Kenmitch]

What's the purpose of this release? Sounds like a money grab to me or is Zen 3 much later than expected? Can't see them releasing Zen3 on desktop in 2020 with this refresh.

Reading between the lines often leads to wrong conclusions. Probably best to just wait and see what happens in the end. It's not like they upped the price compared to the original launch sku's.

Purpose....Reinstate price point? Fresh reviews? Exposure?

If Zen 3 was going to be delayed AMD would have already Covid-19'd out most likely.

Works with the rumor that it'll be only 12 and 16 core. You'll notice that there is no 16 core XT model.

If so then why do the 3900XT?

 

[ondma]

Reading between the lines often leads to wrong conclusions. Probably best to just wait and see what happens in the end. It's not like they upped the price compared to the original launch sku's.

Purpose....Reinstate price point? Fresh reviews? Exposure?

If Zen 3 was going to be delayed AMD would have already Covid-19'd out most likely.



If so then why do the 3900XT?

Well, for the higher end models, they *did* effectively increase the price by removing the cooler.

 

[tamz_msc]

You'd be surprised to learn that Lisa talked about the XT launch and Zen3 "later this year" in the same video (at 84 seconds mark).

That's a reiteration of what she said earlier. It's as vague as it can be.

 

[exquisitechar]

That's a reiteration of what she said earlier. It's as vague as it can be.

Nothing vague about it, we’re getting Zen 3 later this year. They already confirmed that when they debunked the 5nm Vermeer in 2021 rumor. I don’t know why some people insist on denying this. Now, it is likely that only a part of the full lineup will launch.

 

[tamz_msc]

Nothing vague about it, we’re getting Zen 3 later this year. They already confirmed that when they debunked the 5nm Vermeer in 2021 rumor. I don’t know why some people insist on denying this. Now, it is likely that only a part of the full lineup will launch.

"Zen 3 launches later this year" - could mean EPYC or Ryzen. She doesn't clarify.