A market in which customers are running Cinebench 24/7/365 _does_ _not_ _exist_.
This fits AMD quite well. They've been lagging behind Intel's cheap core spam. So if this config makes it cheaper than a 2*12c they will pounce on that option.Zen6 10950X
8+24+4+4CU, 100+ TOPS
N2
2026Q3
$799
This doesn't fit AMD at all. It looks more like a wishlist from a random forum user.This fits AMD quite well. They've been lagging behind Intel's cheap core spam. So if this config makes it cheaper than a 2*12c they will pounce on that option.
Hollywood?A market in which customers are running Cinebench 24/7/365 _does_ _not_ _exist_.
No it’s from presentation slides, I just told you a few posts back.This doesn't fit AMD at all. It looks more like a wishlist from a random forum user.
Well, specs are fine. But the exact price figure is not. Mind you, we are 5 Qs from the launch.No it’s from presentation slides, I just told you a few posts back.
It is LITERALLY just a scaled up Strix Point, with one CCX having P cores and the other having dense cores. The only practical difference is that they are also on different CCDs.This doesn't fit AMD at all. It looks more like a wishlist from a random forum user.
2MB of L3 makes more sense if you move all the L3 off the CCD and into a stacked die. It helps hide the latency hit.I figured as much but some early "leaks" or "rumors" whatever you want to call them seem to suggest Zen 7 is moving to 2MB L2. I don't see it happening but it's too early to really speculate IMHO.
What about L1, especially L1I?Zen4-6 all share the same 1M L2 config.
Fair enough; however, believing that N2 would be used for ALL Zen 6 markets flies in the face of history and common sense. N2 will ALSO be limited in supply, expensive, and lower yield than N3.Why would there be a link?
That would require AMD disclosing these things, which they rarely do, and such decisions aren't exactly put on PPT slides, either.
Intel might have booked so much N3b capacity so far in advance that the booked wafers were more than enough to satisfy the weak demand for their products even at a "meh" yield rate.
Yield can also mean clock speed variation, plus compared to mature processes, even a 65% yield rate can be considered poor.
Pat himself also hinted at Lunar Lake being a lossy product for them, so there's that.
Agree. Unfortunately there is no forum rule stating you can't be intentionally obnoxious.Otherwise he is free to speculate just like anyone else. I (and others) may not like they way he posts a lot with short curt responses or never being able to admit when he was wrong but he's free to do so.
Best speculation I have seen in this thread for some time .Zen6 10950X
8+24+4+4CU, 100+ TOPS
N2
2026Q3
$799
I think that AMD has been much more financially responsible (lately) than Intel. This is also why I am speculating that desktop Zen 6 will (like Zen 5) not use the bleeding edge process node while the DC will.So fiscally unavailable. If it yields so low and has costs too high, there is no money to be made, so it is effectively unavailable, even if it's possible to use it to create a chip. I'm sure that Intel can make chips using 18a-p, but only in the test lab. It might be available, but your accountants would have you drug out and executed.
Agree. I think it a possibility that the IOD will contain 4c of Zen 6c though.I would absolutely want 12+24 over 12+12. All day, any day. I think we only get 12 core CCDs on desktop/mobile, so I think the discussion is moot.
Agree. Some semblance of an educated debate would be greatly appreciated .Leak or don't leak. If someone is walking a tightrope between revealing partial information without revealing sources, there are more succinct (and less combative) ways to do it than what we're seeing here.
Which presentation slides?The info is from presentation slides. Will not spoonfeed anyone with source, but just trust me on it...
Agree, and good point.A market in which customers are running Cinebench 24/7/365 _does_ _not_ _exist_.
Who even said that.N2 will ALSO be limited in supply, expensive, and lower yield than N3.
It's no different at all, which is why it's the latest available node unless things go awry.How is Zen 6 any different from Zen 5 with respect to the production processes it will use?
Who said N3B had issues and forced AMD to use N4P on Zen 5?Who even said that.
Again I ask, if AMD can compete (and exceed) Intel from a node behind (as they are clearly doing today), why would AMD shell out the money, and take the risk on having everything for Zen 6 on N2?It's no different at all, which is why it's the latest available node unless things go awry.
And they didn't so you get to enjoy N2p (outside of valuetown, anyway).
A lot depends on TSMC priceWho said N3B had issues and forced AMD to use N4P on Zen 5?
Back to reality here for a minute.
When has a new process EVER exceeded yields on a more mature process? When has anyone ever even expected this to be true?
Do you also deny that N2 is more expensive than N3P?
Again I ask, if AMD can compete (and exceed) Intel from a node behind (as they are clearly doing today), why would AMD shell out the money, and take the risk on having everything for Zen 6 on N2?
I do agree that much depends on if AMD believes they can compete with Nova Lake with Zen 6 on N3P or if they believe they need the extra boost from N2.
All this being said, I would love for AMD to "shoot the moon" on Zen 6 and go all out. I just don't see it in the cards as they have no reason to.
AMD.Who said N3B had issues and forced AMD to use N4P on Zen 5?
N5?When has a new process EVER exceeded yields on a more mature process?
A tiny bit.Do you also deny that N2 is more expensive than N3P?
There are competition besides Intel you know.Again I ask, if AMD can compete (and exceed) Intel from a node behind (as they are clearly doing today), why would AMD shell out the money, and take the risk on having everything for Zen 6 on N2?
No one cares about NVL.I do agree that much depends on if AMD believes they can compete with Nova Lake with Zen 6 on N3P or if they believe they need the extra boost from N2.
You can blame Intel. Oh and it won’t be vanilla N2/N2P. Contrary to what other posts have stated, AMD and TSMC have worked closely on optimized libraries for developing high frequency, high performance CPUs since/before Zen 3 (i.e. a custom process). I don’t know what next-gen will ultimately be called, but I am sure more will come out soon. Until then? 🍻Fair enough; however, believing that N2 would be used for ALL Zen 6 markets flies in the face of history and common sense. N2 will ALSO be limited in supply, expensive, and lower yield than N3.