I mean if they are using the best nodes and the fact it uses 2.5d fanouts, that means higher BOM across the stack, which must be offset by higher ASPs.Z6 is only a newer version. I don't THINK that AMD intends on creating a more "Premium" product with Z6, it is simply the newest itteration of their product lineup. I expect it will START at the top as it always does, then new variants down the chain get released and the last variant is always the integrated graphics super cheap system chip or in this case, Ryzen 10XXX-G.
Definitely less than TR and we have to see benchmark for NVL 52 Cores but i would expect 2-2.2X the Multi Performance vs ARL-S in cinememe and if it can beat a 64C Zen4 Threadripper at ~300W I would call it very good for core heavy workload for Zen 6 I am expecting it would beat 64C Zen3 threadripper easily.What do you think the total system cost will be for Nova Lake-S 52C vs corresponding AMD TR? Which will be the cheapest solution for comparable MT perf?
At least currently, the AMD TR platform and total system cost is quite expensive compared to regular DT.
Calling ARL-S bulldozer a bit too much it sucks in games and few workloads and that's it it's not hot and power hungry like bulldozer and didn't suck at ST .I mean if they are using the best nodes and the fact it uses 2.5d fanouts, that means higher BOM across the stack, which must be offset by higher ASPs.
The performance bump should be high enough this generation to justify it with Z5 being slowly phased out with price cuts as needed, but far slower than before.
Remember how much Z2 cost vs Z1, and especially Z3 vs Z2 because Intel had no answer at launch.
Z6 will be priced > any prior Zen generation at launch unless Intel has some sort of answer.
Intel is the Bulldozer now.
The Lion Cove part of ARL-S is almost Bulldozer. In the Statuscore utility, Skymont does 17000 MIPS while the Lion Cove cores despite 500 MHz higher clockspeed, only manage 11000 MIPS. That's a shameful show of force by the "big" cores.Calling ARL-S bulldozer a bit too much it sucks in games and few workloads and that's it it's not hot and power hungry like bulldozer and didn't suck at ST .
After recent events, I too am a Merz missile, long live Bismarck 2.Lmao Merz be upon ye.
Well the perf difference between launch and now is like Bulldozer -> Piledriver.Calling ARL-S bulldozer a bit too much it sucks in games and few workloads and that's it it's not hot and power hungry like bulldozer and didn't suck at ST .
In theory yesSonoma Valley is a replacement for Mendocino>>Dali?
It was reversed with Alder Lake and Raptor Lake but than raptor lake had to blow itself.Well the perf difference between launch and now is like Bulldozer -> Piledriver.
And that bad impression effectively doomed Arrow Lake, Intel has not changed the status quo that was established with Z3, and so Ryzen remains the premium brand.
Skymont is an insanely good PPA Core but it has flaws as well including it is slow in heavy vector workloads.The Lion Cove part of ARL-S is almost Bulldozer. In the Statuscore utility, Skymont does 17000 MIPS while the Lion Cove cores despite 500 MHz higher clockspeed, only manage 11000 MIPS. That's a shameful show of force by the "big" cores.
Strix Halo's die-to-die interconnect is logically almost unaltered to Granite Ridge's and Turin's. I expect them to improve on that in Medusa, i.e. make it wider or/and clock it higher. (Might nevertheless be backwards compatible with Strix Halo's pinout, width, speeds, and power states though, but such backwards compatibility has its cost.)
Zen 4 Ryzen/EYPC/Instinct was the last time AMD stuck with the LEGO® principle. They abandoned it in Zen 5: They have different 8c CCDs for Granite Ridge, Turin, and Strix Halo (Zen 5 thread reference). Of course they are not from-the-ground-up different by any means, but they are not anymore mere different bins at the end of the same conveyor belt.
What is X3D for 400, Alex.It was reversed with Alder Lake and Raptor Lake but than raptor lake had to blow itself.
If they can get the same MT perf cheaper from 52C Nova Lake-S than from TR, why would they go with TR? If the performance is the same, they won’t get any increased productivity for going with TR. They will just waste money.Because the only applications where that many cores would matter are used by professionals. Professionals won't bat an eye at the higher price of TR as it pays for itself many times over in the increased productivity.
LoL when did I said the framecoper thing . It's the truth Alder Lake reversed the Zen3 dominance than 13900K was good vs 5700X3D than 7800X3D launched which only took gaming crown and than with Zen 5 X3D they took every crown.What is X3D for 400, Alex.
Don't give me the FrameCoper magical "100% stable" timings nonsense or the other lies from that circle of degenerates.
Zen 4 TR would have PCI-E lane advantage and memory size advantage the bandwidth advantage should be less when we compare Quad Channel DDR5-5200 vs Dual channel 8000 JEDEC and if Intel's IMC is to go by 10K shouldn't be difficult on NVL.If they can get the same MT perf cheaper from 52C Nova Lake-S than from TR, why would they go with TR? If the performance is the same, they won’t get any increased productivity for going with TR. They will just waste money.
Sure, 64C TR will be faster, but 16/32C TR will likely not be a sensible option unless AMD drops the price significantly on those. But then the price jump for going to 64C will be insane if the price remains unchanged on only that variant.
For PCI-E lanes and memory size advantage it depends on the use case whether it’s of any importance. For a lot of use cases it’s not, and what you get with regular DT is good enough.Zen 4 TR would have PCI-E lane advantage and memory size advantage the bandwidth advantage should be less when we compare Quad Channel DDR5-5200 vs Dual channel 8000 JEDEC and if Intel's IMC is to go by 10K shouldn't be difficult on NVL.
Yeah, it's meant for people that are serious about multi thread performance. A tiny volume of customers will want a mainstream platform crippled by memory bandwidth. The number of people that buy 16 cores zen parts right now is tiny.What do you think the total system cost will be for Nova Lake-S 52C vs corresponding AMD TR? Which will be the cheapest solution for comparable MT perf?
At least currently, the AMD TR platform and total system cost is quite expensive compared to regular DT.
Not crippled by memory bandwidth. See previous post by 511.Yeah, it's meant for people that are serious about multi thread performance. A tiny volume of customers will want a mainstream platform crippled by memory bandwidth. The number of people that buy 16 cores zen parts right now is tiny.
Assuming LBT doesn't kill the 52 core part or move it to workstation platform. Somehow I have a hard time believing this will materialize. Gonna end up costing $1000 or more if AMD doesn't have a good answer for it.TR will only make sense if you need 64+ cores.
Yeah, but that’s still far cheaper than corresponding TR, especially when taking total system cost into account.Assuming LBT doesn't kill the 52 core part or move it to workstation platform. Somehow I have a hard time believing this will materialize. Gonna end up costing $1000 or more if AMD doesn't have a good answer for it.
10K MT/s wouldn't be possible without Memory on Package it requires strong signalling also Nova Lake Halo exits with a beefy GPU 2026 is going to be a showdown with both companies having products across all the stacks competing.If Medusa Halo has quad channel 256GB 10,000 MT/s RAM in addition to being attractively priced with PCIe 5.0 x16 slot, it will solve multiple problems at once:
1) Superb iGPU
2) Enough bandwidth for LLMs and AVX-512 workloads
3) No buying DIMMs and hoping they will hit the needed speed.
4) Less heat generation and lower noise and compact chassis
I will pay $2000 for it today if AMD wants a beta tester
I said it just in case.LoL when did I said the framecoper thing.
Wrong, it got Intel back to parity in MT and kinda gaming.It's the truth Alder Lake reversed the Zen3 dominance
$800Zen 6 24 Core would be $900 ish.
Well 1.5X cores and expensive node and packing they would easily charge it$800
AMD isn't that greedy though the CPU may get scalped hard.