Discussion AMD SoC Halo series GPU discussion

Page 20 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Joe NYC

Platinum Member
Jun 26, 2021
2,931
4,301
106
Yeah but only do much they can do with memory on board.
They always have the nuclear option tho.

AMD may have to aim higher, at apple.

In 2026, it is possible that the principal competitors of AMD in client space will be Apple and NVidia.
 

Hitman928

Diamond Member
Apr 15, 2012
6,604
12,103
136
Well, when the 14" macbook with the same SoC and a simliar battery size beat the Strix Halo I would say yes.

Also why do you ignore the resolution difference? Thats the biggest

This would mean that the 16 inch screen is taking, at minimum, an additional 22 W over the 14 inch screen (it would actually be more than this but this is an easy baseline number to calculate). Does that sound reasonable? This large difference in power consumption doesn't show up in any lighter load tests so I don't know how it could be legitimate.

The difference in resolution between the 14 inch and 16 inch Macbook Pro is not a large difference.
 

poke01

Diamond Member
Mar 8, 2022
3,382
4,625
106
Does that sound reasonable? This large difference in power consumption doesn't show up in any lighter load tests so I don't know how it could be legitimate.
I found that weird too, but I did also check for lighter loads.
Here:
 

Hitman928

Diamond Member
Apr 15, 2012
6,604
12,103
136
I found that weird too, but I did also check for lighter loads.
Here:

The Max CPUs take more power, especially with additional cores. Just compare the M4 Max 14 core and M4 Pro 14 core.

Edit: I don't know what the difference is, but the fully loaded result I shared makes way more sense than Just Josh's numbers and it's not that hard to do a sanity check on it as at full load, the CPU power consumption will be the dominant factor by far. My guess, as I said earlier, is that it most likely comes down to a difference in power profiles used for the different tests.
 

The Hardcard

Senior member
Oct 19, 2021
314
397
106
Gamers aren't the ones who decided the first product with Strix Halo would be the Republic of Gamers Flow Z13. Asus did.

Gamers also didn't create this slide with gaming performance to announce Strix Halo:

View attachment 117559

That would be AMD, the company who developed the chip.

Trying to get their money and designed it for them are two separate things. Viagra was not developed for erectile dysfunction, but they are taking all that money. Companies will take anybody’s money if they are giving it away.

Yeah it can, 60-70W is all the thing needs.
I don’t believe that either the CPU or GPU alone is maxed out at 70W. I expect desktop Strix Halos to cross far past 100W on workloads for each one. I also expect performance that the Z13 is not capable of. A 200 billion parameter will require the GPU and the memory bus to be fully energized for many minutes at a time. i expect the Z13 to fall far behind a desktop.
 
Reactions: Tlh97 and marees

poke01

Diamond Member
Mar 8, 2022
3,382
4,625
106
don't know what the difference is, but the fully loaded result I shared makes way more sense than Just Josh's numbers and it's not that hard to do a sanity check on it as at full load, the CPU power consumption will be the dominant factor by far. My guess, as I said earlier, is that it most likely comes down to a difference in power profiles used for the different tests.
What was wrong with Just Josh numbers? I think thats the confusion here.

EDIT: Is this referring to the Strix Halo numbers and/or the numbers I shared for the video playback battery test?
 

eek2121

Diamond Member
Aug 2, 2005
3,277
4,823
136
Strix Halo looks pretty amazing, hopefully someone does a proper thin/light laptop...the thick tablet thing doesn't really do it for me. OH AND 1.2 POUNDS FOR A POWER BRICK? Don't try to tell me it is needed, that phablet uses significantly less power than my laptop.

The Max CPUs take more power, especially with additional cores. Just compare the M4 Max 14 core and M4 Pro 14 core.

Edit: I don't know what the difference is, but the fully loaded result I shared makes way more sense than Just Josh's numbers and it's not that hard to do a sanity check on it as at full load, the CPU power consumption will be the dominant factor by far. My guess, as I said earlier, is that it most likely comes down to a difference in power profiles used for the different tests.

I've been disappointed in quite a few of the Apple products. I've used M1 and M2 based Macbook Pros for work, and I've found they don't quite live up to the hype. Both of the laptops I had used significantly more power for combination loads involving heavy usage of disk, IO, and CPU, and none of it was GPU related. When I was working, they'd last 4 hours with a mix heavy/light workload and enough screen brightness to get things done without straining eyes. Don't get me wrong, they are efficient and I was happy to use them, but not quite the champs that some claim them to be.
 
Reactions: igor_kavinski

adroc_thurston

Diamond Member
Jul 2, 2023
5,365
7,547
96
I don’t believe that either the CPU or GPU alone is maxed out at 70W
You don't have to "max it out".
Unlike mobile dGFX, where bandwidth is sized for proppa desktop SKUs, here you'll be limited by our friend L5X@8000 anyway.
I expect desktop Strix Halos to cross far past 100W on workloads for each one
no.
A 200 billion parameter will require the GPU and the memory bus to be fully energized for many minutes at a time
It's a membw benchmark.
Don't get me wrong, they are efficient and I was happy to use them, but not quite the champs that some claim them to be.
most BL claims are about them being amazing facebook machines (which they are!).
 

poke01

Diamond Member
Mar 8, 2022
3,382
4,625
106
I've been disappointed in quite a few of the Apple products. I've used M1 and M2 based Macbook Pros for work, and I've found they don't quite live up to the hype. Both of the laptops I had used significantly more power for combination loads involving heavy usage of disk, IO, and CPU, and none of it was GPU related. When I was working, they'd last 4 hours with a mix heavy/light workload and enough screen brightness to get things done without straining eyes. Don't get me wrong, they are efficient and I was happy to use them, but not quite the champs that some claim them to be.
That’s the thing it’s impressive when you find out they don’t throttle on battery. You get the full CPU performance.
The default for x86 laptops when unplugged is they throttle CPU performance in exchange for better battery life.

So you might get similar battery life under heavy load with an x86 laptop but at the cost of performance whereas with an M* laptop you get similar battery life to an x86 laptop under heavy load but you don’t lose performance.


use them as netflix machines and the lead will be even bigger.
 

itsmydamnation

Diamond Member
Feb 6, 2011
3,020
3,779
136
That’s the thing it’s impressive when you find out they don’t throttle on battery. You get the full CPU performance.
The default for x86 laptops when unplugged is they throttle CPU performance in exchange for better battery life.

So you might get similar battery life under heavy load with an x86 laptop but at the cost of performance whereas with an M* laptop you get similar battery life to an x86 laptop under heavy load but you don’t lose performance.


use them as netflix machines and the lead will be even bigger.
Yeah in the real world this isn't really true ( atleast on every Lenovo AMD laptop I've owned, 3 ), you loose a touch of ST sure , but the total number of "P" cores means you just chew modern workloads at low power anyway. like my Zen3 will hold mid 3ghz on 16threads at 100% cpu in 15watts where that is 8 threads for sql and 8 threads for processing.

So your 16core halo you could set like a 25watt STAMP and probably hold mid 4ghz at 100% cpu outside SIMD meme's. At that point the only thing close is going to be M4Max ..... and at what price point?
 

coercitiv

Diamond Member
Jan 24, 2014
7,082
16,340
136
That’s the thing it’s impressive when you find out they don’t throttle on battery. You get the full CPU performance.
The default for x86 laptops when unplugged is they throttle CPU performance in exchange for better battery life.
Windows has different performance profiles for battery vs. plugged. If you really want more performance on battery, it can be done with a button switch and it will stay that way indefinitely. You could argue the Apple way is better user experience by using a different default option, I can certainly argue Microsoft has made a mess by keeping both "power plans" and "power modes" while allowing OEMs to introduce their own power management on top.

All that being said though, full CPU performance on battery is not necessarily the superior alternative you keep pushing for it to be. People have different usage scenarios for their machines, and even different user scenarios when plugged vs. unplugged. If I got a Mac today, I would certainly look into maximizing performance while plugged in and optimizing performance while on battery. I'm also aware other people may prefer it differently.

 

uzzi38

Platinum Member
Oct 16, 2019
2,745
6,627
146
It's cool, but only long term will prove whether it'll be a lasting product or not.

The appeal for iGPUs are:
-Battery life
-Form factor
-Cost, of which can greatly justify the above two

You go high end, and you lose the battery life advantage, even against dGPU systems, and the since the manufacturer of the SoC isn't doing it for charity but an opportunity for more revenue, it's no longer a free one like iGPUs are but whether you are actually saving money over a dGPU.

From what we've seen STX-H seems very similar to what STX+dGPU does for idle power. Potentially a little worse, sure, but if you've ever used an optimus laptop, you know how unreliable it is. Sometimes the dGPU will just wake up for a task and eat into your battery and you have little control over it.

Strix Halo does allow for a more seamless experience overall.

You are losing the flexibility of a dGPU, and it isn't all about savings either as you need to start putting dedicated VRAM, most likely with exotic packaging techniques.

What do you mean "you need to start putting dedicated VRAM", the whole point of an APU is that you have a single shared pool of memory.

And on Strix Halo, just regular old LPDDR5X-8000 memory modules are used.

The only real spot where you lose flexibility is if you need that APU to compete with multiple different dGPU dies, say for example you build a big IOD to compete with AD104, but you also need to cut it down to compete with AD106 based laptops. That... Doesn't seem to be the case here, from what it looks like. Strix Halo looks very focused on competing with AD106.

If it's say $500 cheaper on that Asus system, then you must consider how much of that $500 is artificial tax by both Nvidia and the system manufacturer versus actual cost of production. Remember Nvidia has absolutely monopoly on laptops, even more lopsided than on desktops. Will it even be $100 cheaper on a <$1500 system?

That's speculation we don't really have a good basis for, frankly. What we do have are the prices of the old Z13 and the prices of the new ones. The top end of the older Z13 outstrips even the 128GB 395 model in terms of pricing. No matter how you look at it, the main issue with Strix Halo pricing is that evidently the 390 is priced much too close to the 395. That's it.

By the way, there are hidden costs that people aren't realising with a dGPU setup. GDDR modules and cooling them (it will eat into the chassis' cooling budget), if you want the most seamless experience you can get with a dGPU you need a mux and all the advanced Optimus stuff (which adds to board cost significantly). And evidently with the entire Strix Halo package competing with just AD106 on it's own on a perf/W basis, evidently Strix Halo is more power efficient when you factor in CPU+GPU at the same time. These are all going to add up.
 

ToTTenTranz

Senior member
Feb 4, 2021
386
739
136
Trying to get their money and designed it for them are two separate things. Viagra was not developed for erectile dysfunction, but they are taking all that money. Companies will take anybody’s money if they are giving it away.

Most of the IOD is occupied by a gigantic GPU with Infinity Cache, AMD themselves presented gaming benchmarks in the announcement slide deck and the very first product on shelves to bring Strix Halo was the Asus Republic Of Gamers Flow Z13 that even says the word GAMERS on the top.




But it wasn't designed for games. AMD just accidentally put there their biggest PC iGPU ever. The whole thing just slipped in there by accident during design.
They really didn't mean to compete with discrete GPUs like at all.

It's like Viagra.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |