- Mar 3, 2017
- 1,622
- 5,892
- 136
Peak core clock speeds are another factor in the design of cache hierarchies, from what I understood.but you are misunderstanding Apple's CPU Design philosophy. They have no need for an L3, by virtue of the fact that their L1/L2 is so huge.
You do realize we are talking about TimeSpy, right?More and/or higher performing CPU cores = mode demanding client to the memory controller that shares its resources with the GPU, in an Unified Memory Architecture (UMA).
In UMA systems, memory requests from the CPU actually reduces the GPU's effective bandwidth disproportionately. Probably because serving memory requests from one causes a latency to serving memory requests from the other. In simpler terms, if the memory controller is too busy with the CPU, the GPU will stall waiting for it.
Sony even had a slide about this in their internal guidelines for PS4 developers:
View attachment 97007
This is also one of the most probable reasons why the PS5 Pro isn't doing any substantial upgrade to the CPU over the PS5. Sony probably wants all the extra bandwidth from the ~28.5% faster memory to go to the new GPU alone.
Of course of that massive bandwidth starving you claim.Graphs or numbers of what? 18CUs being 33.(3)% more CUs than 12 CUs? 😐
You do realize we are talking about TimeSpy, right?
So where are those numbers?Because the full GPU in Phoenix is bottlenecked by memory bandwidth.
🤤A 16 core CPU + 40CU GPU will perform very well for a notebook in most of workloads, render included.
What kind of cooling solution can passively dissipate 125W?so the cooling solution can disipate silently 125W
So where are those numbers?
No one said anything about being pointless. Nowadays the GPU's compute capabilities aren't used exclusively for playing games.Or give us an explanation why would AMD pointlessly use up precious die space for more CU, If It can't feed It.
Remember when a bunch of AMD executives took hundred of thousands pages of document before fleeing to Nvidia?
More info on this?
I never said passively, just silent. Actually You can cool pasively around 65-80W with current passive cases as mine, but that won't be enough to cool a little monster like this. So I'd expect something like Beelink and Minisforum do now, but bigger and beefier. Their current solutions cool a 65W in less than a liter, silently-ish. Double the volume and the cooling solution and it might fly. These mini systems are virtually silent when idle and mid load, just when pushed to the limit you hear themWhat kind of cooling solution can passively dissipate 125W?
Well, two "sources" just told me that Zen 5 is actually 10% faster than Zen 6!
Btw if Gurman is correct, M4 Max is coming at the end of this year, so Strix Halo will have to compete with that.
I wonder what battery life can be expected at 125 W though.A 16 core CPU + 40CU GPU will perform very well for a notebook in most of workloads, render included.
AMD APUs and Gaming Laptops with AMD dGPU have (usually. Can't generally say for all oems) excellent battery life. Why would STX Halo be different? Chiplets might eat into active and idle power, but it's also using novel packaging, so losses might be quite low.I wonder what battery life can be expected at 125 W though.
Or is it in reality supposed to be an always plugged in movable desktop PC posing in a laptop form factor?
I'd like to see it available on AM5 socket too, but I guess that won't happen.
So roughly what battery life can be expected when running at full load (or close to that)? And what will the weight of such a laptop be to accomodate the large battery?AMD APUs and Gaming Laptops with AMD dGPU have (usually. Can't generally say for all oems) excellent battery life. Why would STX Halo be different? Chiplets might eat into active and idle power, but it's also using novel packaging, so losses might be quite low.
The battery life is function of so many factors. It depends on how much the OEMs would limit the power profile of Stryx Halo in battery mode. Also, if the table linked a couple of pages ago is legit, SH will have also a 55W iteration, how good the performance could be at that power is another question, though. Not that if I had such a powerful notebook I would use it in battery mode when needing max performance, though.I wonder what battery life can be expected at 125 W though.
Or is it in reality supposed to be an always plugged in movable desktop PC posing in a laptop form factor?
I'd like to see it available on AM5 socket too, but I guess that won't happen.
Are you asking seriously or are you trying to waste people's time with the full knowledge there are laptops that run significantly higher GPU power limits than the entire Strix Halo package, much less including CPU power consumption.So roughly what battery life can be expected when running at full load (or close to that)? And what will the weight of such a laptop be to accomodate the large battery?
Not going to spill the beans but it's safe to assume it won't use the basic organic substrates that AMD uses for their desktop packages. A package like that would be awful for a mobile platform, and unlike Dragon/Fire Range Strix Halo is designed for laptops first, not a desktop part repurposed for laptop usage.What packaging will Strix Halo use?
Yes.Are you asking seriously
Yes.
So roughly what battery life can be expected when running at full load, and at what weight?
G14 2024 (1.5KG) is lighter than the Macbook Pro 14 with M3 Max (1.62KG). If the same device were to run STX-Halo instead, I'd expect some minor weight savings due to only one combined heatsink for the APU package would be needed rather than one seperate for APU and dGPU like right now. Also a heatsink for the GPU's GDDR6 would be unnecessary, as there would be none. But these are just speculation.Yes.
So what battery life can be expected when running at full load, and at what weight?
That’s crazy lol. Going to be such a good budget part. What’s the bus width looking like, 64 or 128?Phoenix2 and Mendo aren't the same price bracket, Mendocino is a cheaper part. Sonoma Valley is a direct replacement for the same socket as Mendocino, so it should be considerably cheaper than PHX2.
64b, that's the limit Mendocino had so I'd be downright shocked if Sonoma Valley didn't have it too. They share the same socket after all, 64b LPDDR5 is almost certainly a platform limitation.That’s crazy lol. Going to be such a good budget part. What’s the bus width looking like, 64 or 128?
Yeah I expect Kraken to be a bit smaller than PHX, but not by much. Just basically offer similar levels of MT and iGPU performance (better at lower power, worse at higher power if I were to guess) with a better NPU and ST performance. So similar price for an overall mostly improved product.Kraken is the replacement for Phoenix and Phoenix2 both in a certain sense, yeah?
Offering Zen 5 + RDNA 3.5 on mainstream devices is almost a gift from the heavens😃.Yeah I expect Kraken to be a bit smaller than PHX, but not by much. Just basically offer similar levels of MT and iGPU performance (better at lower power, worse at higher power if I were to guess) with a better NPU and ST performance. So similar price for an overall mostly improved product.