• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Question Zen 6 Speculation Thread

Page 390 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Saylick said:
C'mon, that's a bad bet. You don't think if the engineers told Lisa that they could only must 6.95 GHz that Lisa wouldn't immediately ask, "What? No, get to a clean 7 GHz for marketing purposes"?
Click to expand...
That's why they'll go tell her only 6.9 Ghz is possible:
"We can do 6.9 Ghz"
"What? No, get to a clean 7 GHz"
"We would if it were 6.95, but we're already clock stretching at 6.9"

Meanwhile in a parallel universe:
Engineer: "we can do 6.95"
Lisa: "go for clean 7.0, impress"
Anandtech: "they're holding back at 7.0, a real limit would more random like 7.15"

Meanwhile I'm looking forward for something like 6.5 Ghz and everyone going out of their mind because Zen 6 has "obviously missed it's clock target". Zen 6.9 would be nice though, rolls of the tongue.
 
gdansk said:
6250mhz max
Click to expand...
I'm not falling for the Zen 5 hype again, when we have a slide that described zen 5 perfectly accurately (and zen 6 almost certainly as well).

So I'm setting my baseline at 6.25 GHz tops and ~9% actual IPC uplift. Everything above that is a very welcome surprise. (Those are very conservative and I'm quite sure AMD is capable of delivering a lot more, but my Zen 5 bets were conservative as well, and they missed those by a lot)

I don't know how many remember, but we had almost exactly the same discussion here with early Zen 5 leaks similarly reporting 2 Ghz-įsh clocks in GeekBench (but actually running close to stock speeds just reporting wrong, eventually ending up only like 10-15% slower than the final product, despite people here believing there would be 40 or even 60% IPC gains)

And we have a certain guy hyping 7Gh as absolute certainty, that in 2024 repeated Ad Nauseam that Zen 5 was such an incredible jump AMD deliberately downplayed themselves to not Osborne Zen 4 products and claimed it will surely cost no less than 999$:

Abwx said:
If we extrapolate from the Computerbase test, and using the 70W lower bound to not overestimate the calculation, then this Strix Point ES perform 26% better than a 7940HS@70W and about 52% better than a 7940HS@45W, so assuming a run at 45W that would make 1.52x the perf with 1.5x the core count, and this would also imply 20% better perf/clock.
Click to expand...

adroc_thurston said:
You're close!
Very close.
Click to expand...

Kaffeekenan said:
Isn't 20% more IPC kinda low? You said something around 30-40 if I remember correctly? So were you wrong or what happened?
Click to expand...

Gideon said:
Whichever way you slice it, 20% better perf per core is a "run of the mill" incremental update from AMD. It's nowhere near "Osborning everything before it" or if it were, Zen4 with its 30% ST perf uplift would have been a bigger deal
Click to expand...

adroc_thurston said:
Man this is the worst attempt to own me in eons.
Click to expand...


I'd be happy to be wrong, and would love to see AMD get back on track with their 20% IPC per gen (which are way too long apart anyway with 22+ month cycles) and hopefully get decent clock speed bumps from 2 nodes, but those are more likely to manifest in MT workloads than ST peak clocks.

All in all I'd much rather see wider and slower cores than stupidly high clock speeds. 7 GHz ain't gonna translate to laptops anyway. There we'd still be at 5.5 - 6 Ghz in ST loads, guzzing 5x more energy than the ARM competition to perform worse (this specifically in ST, while being competitive in MT wrokloads perf/watt). And If AMD continues the trend, the battery life would still be the same that Zen 2 Lucienne had in early 2021. All of that not because AMD coudn't do any better, but because they don't care about consumer market at all.
 
Not exactly a 7GHz believer, but personally, anything less than about 6.5GHz on desktops would be disappointing as hell with only 10-15% IPC, Zen5's mediocrity and AMD's cadence being taken into consideration. If they can't muster that much, they may as well throw in the towel. Apple, Qualcomm and stock ARM cores will just keep lapping them.
 
techjunkie123 said:
Well we're talking stable clocks. Not raptor lake zoom zoom into boom boom.

If they hit 6.5+ on Zen 6 it'll be great.
Click to expand...
Those were the stable clocks.
1.42v is still below what Z4 runs at.
Gideon said:
All in all I'd much rather see wider and slower cores than stupidly high clock speeds
Click to expand...
Everyone is pumping a mix of meth and paint thinner.
Gideon said:
7 GHz ain't gonna translate to laptops anyway. There we'd still be at 5.5 - 6 Ghz in ST loads
Click to expand...
Nope.
7G 1.35v DT means mobile with the same CCD sits at like 6.7.
 
Gideon said:
guzzing 5x more energy than the ARM competition to perform worse (this specifically in ST, while being competitive in MT wrokloads perf/watt).
Click to expand...
- If you're comparing to Apple, then AMD of course stands no chance against a competitor who has owns all of their stack.
- If you're comparing to Crapdragon, pfft, no way.
 
Tachyonism said:
- If you're comparing to Apple, then AMD of course stands no chance against a competitor who has owns all of their stack.
- If you're comparing to Crapdragon, pfft, no way.
Click to expand...
Single-threaded the Oryon v4 will beat Zen6. At much lower power draw. Multi-thread that’s tbd, still likely quite a bit less power draw.
 
511 said:
Good luck cooling that thing on mobile
Click to expand...
They already ship 1.3v parts in mobile. Works pretty dang well you see.
jdubs03 said:
Single-threaded the Oryon v4 will beat Zen6. At much lower power draw. Multi-thread that’s tbd, still likely quite a bit less power draw.
Click to expand...
atta boy. Fun's about to start.
Tachyonism said:
- If you're comparing to Apple, then AMD of course stands no chance against a competitor who has owns all of their stack.
Click to expand...
"stack" is a meme. Apple just has fatter cores.
 
511 said:
i meant xtor density which is quite a bit dense than N4P hence you need better cooling in mobile
Click to expand...
Good thing that process shrinks also provide incremental Cac reduction.
Again, the heat density there is nothing special.
It'll only be a problem if you're trying to push >250W thru AM5 (that IHS sucks).
 
adroc_thurston said:
"stack" is a meme. Apple just has fatter cores.
Click to expand...

For the most part this is true, but Apple never has to worry whether their OS will be aware of some aspect of the CPU design that might restrict performance.

Meanwhile there have been numerous cases where Zen loses performance on Windows because the scheduler doesn't understand the way the CPU is designed and makes bad choices.

Most of the performance gap is down to Apple having a wider design. The larger caches almost certainly play a big part in this as well for certain workloads.
 
Mopetar said:
but Apple never has to worry whether their OS will be aware of some aspect of the CPU design that might restrict performance.
Click to expand...
Yes they do.
Apple s/w is utter garbage vomit.
Mopetar said:
Meanwhile there have been numerous cases where Zen loses performance on Windows because the scheduler doesn't understand the way the CPU is designed and makes bad choices.
Click to expand...
nope, CPPC2 is ancient and works very well.
Mopetar said:
The larger caches almost certainly play a big part in this as well for certain workloads.
Click to expand...
L1's are larger.
Overall x86 land has more in terms of total caching packages (but it has to compensate for skinnier cores).
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top