AMD's Strix Point: Zen 5 Hits Mobile

https://chipsandcheese.com/2024/08/10/amds-strix-point-zen-5-hits-mobile/

by klelatti • 2 years ago

166 225 2 years ago

sm_10246• 2 years ago

IMO, the most interesting thing about this line is the battery life---within an hour of MBP3 and within 2 hours of Asus's Qualcomm. Making it comparable to ARM architectures.

Which is a little surprising because ARM is commonly believed to be much more power efficient than x86.

[1] https://youtu.be/Z8WKR0VHfJw?si=A7zbFY2lsDa8iVQN&t=277

jiggawatts• 2 years ago

because ARM is commonly believed to be much more power efficient than x86.

Because most ARM processors were designed for mobile phones and optimised to death for power efficiency.

The total power usage of the front end decoders is a single digit percentage of the total power draw. Even if ARM magically needed 0 watts for this, it couldn’t save more power than that. The rest of the processor design elements are essentially identical.

halJordan• 2 years ago

Yeah if you make a worse core and then downclock it then you will increase power efficiency. AMD thankfully only downclocks the 5c, but Intel is shipping ivy lake equivalents in their flagship products just to get power efficiency up.

arnaudsm• 2 years ago

ARM got a lot of hype since the release of the M1, but most users only compared it to the terrible Intel MBPs. Ryzen mobile has been consistently close to Apple silicon perf/watt for 5 years. But got little press coverage.

Hype can be really decorrelated from real world performance.

sm_10246• 2 years ago

I have heard that part of the reason for little coverage of ryzen mobile CPUs is their limited availability as AMD was focussing on using the fab capacity for server chips.

jsheard• 2 years ago

Any efficiency comparison involving Apples chips also has to factor in that Tim Cook keeps showing up at TSMCs door with a freight container full of cash to buy out exclusive access to their bleeding edge silicon processes. ARM may be a factor but don't underestimate the power of having more money than God.

Case in point, Strix Point is built on TSMC 4nm while Apple is already using TSMCs second generation 3nm process.

carstenhag• 2 years ago

But how is this the case? I never saw a single article mentioning that a non-Mac laptop was better.

(Random article saying M3 pro is better than a Dell laptop https://www.tomsguide.com/news/macbook-pro-m3-and-m3-max-bat... )

Filligree• 2 years ago

Ryzen mobile is consistently close, yeah. But with the sole exception of the Steam deck, I've yet to see a Ryzen mobile-bearing laptop, Windows included, which is close to the overall performance of the Macbook.

sandywaffles• 2 years ago

I think that's because all the press talks about actual battery life per laptop and the Apple Silicone laptops ship with literally double the size battery of any AMD based laptop without a discrete GPU. So while the efficiency may be close, actually perceived battery life of the Mac will he more than double when you also consider the priority Apple puts into their power control combined with a larger overall battery.

wtallis• 2 years ago

The CPU core's instruction set has no influence on how well the chip as a whole manages power when not executing instructions.

sm_10246• 2 years ago

That is fair, I was taught that decoders for x86 are less efficient and more power hungry than RISC ISAs because of their variable length instructions.

I remember being told (and it might be wrong) that ARM can decode multiple instructions in parallel because the CPU knows where the next instruction starts, but for x86, you'd have to decode the instructions in order.

IshKebab• 2 years ago

Sure, but it's not idle power consumption that's the difference between these.

dagmx7• 2 years ago

Battery tests are important, but so is how it fairs on battery (what is the performance drop off to maintain that), what’s its performance is ant its peak and how it long before it throttles when pushed.

The M series processors have succeeded in all four: battery life, performance parity between battery and plugged in, high performance and performance sustainability.

So far, very few benchmarks have been comparing the latter three as part of the full package assessment.

GrumpyYoungMan• 2 years ago

The display, RAM, and other peripherals are consuming power too. Short of running continuous high CPU loads, which most people don't do on laptops, changes in CPU efficiency have less apparent effect on battery life because it's only a fraction of overall power draw.

Panzer042• 2 years ago

5hr Battery life in laptops is mostly a function of how well idle is managed, i think. The less work you can do while running the users core program, the better. I'm not sure how much impact CPU efficiency really has in that case.

If you are running a remotely demanding program (say, a game) , your battery life will be bad no matter what (ie. <4hrs) unless you choose a very low TDP that performs badly always.

A laptop at idle should be able to manage ~5w power consumption sumtpion regardless of AMD/intel/Apple processor, but it's largely on the OS to achieve that.

9999000009992• 2 years ago

I have a 365 AMD laptop.

The battery is great if your doing very light stuff, Call of Duty takes it's battery down to 3 hours.

Macs don't really support higher end games, so I can't directly compare to my M1 Air.

sedatk• 2 years ago

How does “great” translate to hours?

cubefox5• 2 years ago

Unlike AMD and Qualcomm, Apple uses an expensive TSMC 3nm process, so you would expect better battery life from the "MBP3". I assume they used the process improvements to increase performance instead.

hajile24• 2 years ago

Perf per watt is higher for M1 on N5 vs Zen5 on N4P, so the problems go deeper than just process.

X Elite also beats AMD/Intel in perf/watt while being on the same N4P node as HX370.

https://www.notebookcheck.net/AMD-Zen-5-Strix-Point-CPU-anal...

double0jimb0• 2 years ago

I didn’t watch this link, but my Zenbook S 16 only gets remotely close to my M2 MBA battery life if the zenbook is in whatever is Windows 11 ‘efficiency’ mode, and then it benchmarks at 50% of the M2.

I don’t think the two are remotely comparable in perf/watt.

acchow• 2 years ago

within an hour of MBP3

Not a good way to measure. The Zenbook S16 has a larger 78Wh battery vs the MacBook Pro’s 69.6Wh.

So that’s 11% less battery life despite 12% more battery capacity.

nullc• 2 years ago

Speaking of Zen 5, are there any rumors on when 128 core turin-x will ship?

apatheticonion• 2 years ago

Are there any mini PCs with Zen 5?

wmf3• 2 years ago

You could probably put a 9700X in a MS-A1. Besides that it will take a few months.

hajile24• 2 years ago

I think there's a lot of hope for a Strix Halo mini-PC.

We currently have 7840HS+6650M with two massive heatsinks just barely coming in at what you would consider a large mini-PC rather than a SFFPC.

Just one chip cuts the heatsink demand in half. The cores should be faster and it moves from 28CU up to 40CU and from RDNA2 to RDNA3.5. As long as the cross-core-complex latency isn't as bad as the HX370, I think it could be a real winner for a long time as it's basically an upgraded PS5 that runs Linux/Windows.

moffkalast• 2 years ago

Hell I'd settle for a Z1 SBC or even mini PC. They can't seem to keep up with demand for these new chips to put them into any sort of products that don't have complete mass appeal. It's impossible to find one that's not in a handheld gaming console. I doubt they'll even make enough of the Halo to cover laptops.

sliken• 2 years ago

"AMD Strix Point expected to debut in October, claims AOOSTAR" was a new headline I've seen. Seems about right that laptops (higher margin parts) land a few months before the SFFs.

I'm excited about the strix halo that has more cores, double the GPU, and double the memory badwidth (to keep the GPU fed).

JonChesterfield• 2 years ago

There seem to be some on hawk point but not many. I'd like to replace a PN50 (4800u system) from years ago but am attached to it being small enough to fit in the cable tidy under the desk - the 4" form factor seems to have grown a little over time.

CyberDildonics• 2 years ago

What does "strix point" mean?

icegreentea2• 2 years ago

It's AMD's code/product name for their mobile CPUs with the Zen5 microarchitecture.

The last AMD laptop code name was "Hawk Point" - strix is a mythological bird. Who knows if AMD will keep with this naming scheme.

jml7c5• 2 years ago

I sit firm in my belief that the best thing Microsoft could do for their laptop ecosystem is to add support for a "max fan speed" slider somewhere prominent in the Windows UI.

People want the option to make their laptop silent or nearly silent. And when users do need the power, they generally prefer a slightly slower laptop at a reasonable volume rather than the roar of a jet engine.

Laptop manufacturers want their devices to score high on benchmarks. The best way to do that is to add a fan that can become very loud.

The incentives are not aligned.

All laptops should be designed to operate passively 100% of the time, if the owner so chooses. I doubt manufacturers will go that route unless Microsoft nudges them towards it. It would have downstream effects on how review sites benchmark laptops (i.e., at various power draws/noise levels producing a curve rather than a single number), which would have downstream effects on what CPU designers optimize for. It'd be great for consumers.

latchkey• 2 years ago

I don't know anything about Windows, but at least on Mac, I've been using TGPro for years [0]. I'd assume there is something similar in the Windows world.

In normal conditions my M1 mac can control its fans just fine, but when I travel to hot places like Vietnam... I just keep the fans on more often and my machine doesn't get nearly as hot. I end up having to open it up after a few months and clean out the fans, but that's fine.

[0] https://www.tunabellysoftware.com/tgpro/

kevin_thibedeau• 2 years ago

Stop demanding paper thin laptops. My work Dell rarely turns on its fan unless an AV scan is in progress and even then it's rather tolerable. It isn't a fashionable thickness so has plenty of internal volume for heat distribution.

rafaelmn6• 2 years ago

MacBook Air is thin and fanless, so it can be done.

tuna74• 2 years ago

I want a thin and fanless laptop. You might want something else.

TacticalCoder• 2 years ago

Stop demanding paper thin laptops

Then use a desktop. Like most people I want my laptop paper thin: at least Apple understands that correctly. My daily laptop driver is a "LG Gram" which is especially slick, thin and light (lighter than any Mac laptop) and it's no slouch: 24 GB of RAM for example. And it's basically quiet: I don't even know if it has any fan (I own it since years and never heard a fan).

I'll take a slightly slower laptop if it means it's much quieter. But there's no way I'm going back to the bricks we used to have in 90s/2000s.

If you need a 4090 GPU, buy a desktop and call it a day. For everything else, you can get plenty of power, fast NVMe M.2 SSD, lots of RAM in a paper thin laptop.

shmerl• 2 years ago

You can use Linux and hwmon.

mqus• 2 years ago

add support for a "max fan speed" slider somewhere prominent in the Windows UI.

Isn't that what the "power settings" do? It's a slider at the bottom right, hidden in a tray icon. Sure, it only has three positions and also influences battery consumption but it pretty much does what you want. (Not sure if windows 11 kept this though)

nightski• 2 years ago

I don't mind fans at all, in fact I find fan noises a little soothing (a childhood thing, we didn't have AC). Everyone has different priorities, personally I'd prefer to not have throttled performance.

sandreas• 2 years ago

Did you know NBFC (Notebook Fan Control)? It's old, but still works on some devices and you can create custom profiles via XML.

https://github.com/hirschmann/nbfc

jwells892• 2 years ago

It might also be good to mandate 10+ hours of battery life when the laptop is in power saving mode. A number of laptops that'd otherwise have decent battery life are hampered by things like half-baked power management of discrete GPUs that doesn't completely cut power supply to those components. Manufacturers should be more heavily testing under this mode.

devbent• 2 years ago

A few misbehaving CSS filters can make my discreet GPU turn on and at that point my battery life is a goner. Not sure who to blame in that scenario.

There was an old bug in FF around 2018 where a tab using a GPU would prevent a Windows laptop from ever sleeping. That ended up destroying that laptop's battery after it got thrown in my backpack and overheated a couple times.

automatic6131• 2 years ago

I sit firm in my belief that the best thing Microsoft could do for their laptop ecosystem is to add support for a "max fan speed" slider somewhere prominent in the Windows UI

Until then, there's https://github.com/Rem0o/FanControl.Releases

mkhalil• 2 years ago

A closed source application for controlling one's fan...umm no thank you.

I never will understand the reasoning behind why people are so afraid of releasing their source code. Looks like a weekend project; does he expect to make a living out of a weekend project?

cpv• 2 years ago

Some recent asus ROG/TUF 2022+ models have fine tuning available via "armoury crate" or "g-helper" (non-proprietary, fan/community supported, code on github).

Disabling a dGpu and reducing power can yeld some impressive results for battery life. Allows also defining fan speed curves depending on temperature.

akvadrako• 2 years ago

This exists.

My friend's new ASUS ROG G16 let's you control maximum power draw for the CPU / GPU and draw the fan curve (temp → RPM).

The only thing that seems missing is a (temperature → max power curve) to handle spikes.

aurareturn18• 2 years ago

One of these has to be true (or both true):

1. ARM is inherently more efficient than x86 CPUs in most tasks

2. Nuvia and Apple are better CPU designers than AMD and Intel

Here are results from Notebookcheck:

Cinebench R24 ST perf/watt

* M3: 12.7 points/watt

* X Elite: 9.3 points/watt

* AMD HX 370: 3.74 points/watt

* AMD 8845HS: 3.1 points/watt

* Intel 155H: 3.1 points/watt

In ST, Apple is 3.4x more efficient than Zen5. X Elite is 2.4x more efficient than Zen5.

Cinebench R24 MT perf/watt

* M3: 28.3 points/watt

* X Elite: 22.6 points/watt

* AMD HX 370: 19.7 points/watt

* AMD 8845HS: 14.8 points/watt

* Intel 155H: 14.5 points/watt

In MT, Apple is 1.9x more efficient than Zen4 and 1.4x more efficient than HX 370. I expect M3 Pro/Max to increase the gap because generally, more cores means more efficiency for Cinebench MT. X Elite is also more efficient but the gap is closer. However, we should note that in a laptop, ST matters more for efficiency because of the burst behavior of usage. It's easier to gain in MT efficiency as long as you have many cores and run them at lower wattage. In this case, AMD's Zen5 12 core setup and 24 threads works well in Cinebench. Cinebench loves more threads.

One thing that is intriguing is that X Elite does not have little cores which hurts its MT efficiency. It's likely a remnant of Nuvia designing a server CPU, which does not need big.Little but Qualcomm used it in a laptop SoC first.

Sources: https://www.youtube.com/watch?v=ZN2tC8DfJnc

https://www.notebookcheck.net/AMD-Zen-5-Strix-Point-CPU-anal...

trynumber9• 2 years ago

The Snapdragon X Elite is on the same node and when actually doing a lot of work (i.e. cores loaded) it is close enough to HX 370 while delivering similar throughput.

Why wouldn't the inherent inefficiency of x64 be as noticeable in MT when all the inefficient cores are working? Because it is running at lower clocks? Then what allows it to match the SDXE in throughput? Does that need to lower its clock even more? I'm not seeing what makes it inherent.

rafaelmn6• 2 years ago

From you link - Intel is topping the performance charts (alongside AMD in SC) - they probably tune power usage agressively to achieve these results.

I would guess it's more to do with coming from desktop CPU design to mobile vs. phones to laptops.

aurareturn18• 2 years ago

From you link - Intel is topping the performance charts (alongside AMD in SC) - they probably tune power usage agressively to achieve these results.

Cinebench 2024 ST:

* M3: 142 points

* X Elite: 123 points

* AMD HX 370: 116 points

* AMD 8845HS: 102 points

* Intel 155H: 108 points

Amongst each company's best laptop ST SoCs, no, Intel and AMD are far behind in both ST scores and perf/watt.

If you're referring to desktop speeds, then yes, Intel's 14900k does top the charts in ST Cinebench but it likely uses well over 100w.

I mostly care about laptop SoCs. In the case of the M3, it doesn’t even have a fan.

AnthonyMouse17• 2 years ago

1. ARM is inherently more efficient than x86 CPUs in most tasks
2. Nuvia and Apple are better CPU designers than AMD and Intel

The third possibility is that they just pick a different point on the efficiency curve. You can double power consumption in exchange for a few percent higher performance, double it again for an even smaller increase.

The max turbo on the i9 14900KS is 253 W. The power efficiency is bad. But it generally outperforms the M3, despite being on a significantly worse process node, because that's the trade off.

AMD is only on a slightly worse process node and doesn't have to do anything so aggressive, but they'll also sell you whatever you want. The 8845HS and 8840U are basically the same chip, but former has around double the TDP. In exchange for that you get ~2% more single thread performance and ~15% more multi-thread performance. Whereas the performance per watt for the 8840U is nearly that of the M3, and the remaining difference is basically the process node.

aurareturn18• 2 years ago

The third possibility is that they just pick a different point on the efficiency curve. You can double power consumption in exchange for a few percent higher performance, double it again for an even smaller increase.

This only makes sense if the Zen5 is actually faster in ST than the M3. In this case, the M3 is 1.24x faster and 3.4x more efficient in ST than Zen5.

AMD's Zen5 chip is just straight up slower in any curve.

The max turbo on the i9 14900KS is 253 W. The power efficiency is bad. But it generally outperforms the M3, despite being on a significantly worse process node, because that's the trade off.

It's not a trade off that Intel wants. The 14900KS runs at 253w (sometimes 400w+) because that's the only way Intel is able to stay remotely competitive at the very high end. An M3 Max will often match a 14900KS in performance using 5-10% of the power.

jeswin2• 2 years ago

In the same article, if you picked their R23 benchmarks the advantage vanishes in MT benchmarks; the 3nm M3 Max is actually behind 4nm Strix Point in efficiency.

aurareturn18• 2 years ago

Cinebench R23 is hand optimized for X86. It uses Intel Embree engine underneath.

That's why Cinebench R23 heavily favors x86.

kllrnohj4• 2 years ago

Performance per watt is a bad metric. You want instead performance for a given power budget (eg, how much performance can I get at 15w? 30w? etc...)

Otherwise you can trivially win 100% of performance/watt comparisons by just setting clocks to the limit of the lowest usable voltage level.

For example compare the 7950X to the 7950X@65w using the officially supported eco mode option: https://www.anandtech.com/show/17585/amd-zen-4-ryzen-9-7950x...

Cinebench R23 MT:

7950X stock: 225 points/watt

7950X @ 65w eco mode: 482 points/watt

Over 2x perf/watt improvement on the exact same chip, and a power efficiency that tops the charts, beating every laptop chip in that notebookcheck test by a large amount as well. And yet how many 7950x owners are using the 65w eco mode option? Probably none. Because perf/watt isn't actually meaningful. Rather it's how much performance can I get for a given power budget.

aurareturn18• 2 years ago

Performance per watt is a bad metric.

It isn't. It needs context.

Sure, you can get an Intel Celeron to have more perf/watt than an M1 if you give the Celeron low enough wattage.

The key here is the absolute performance.

In this case, both the M3 and the X Elite are not only significantly more efficient than both Zen4 and Zen5 in ST, they are also straight up faster while being more efficient.

kllrnohj4• 2 years ago

1. ARM is inherently more efficient than x86 CPUs in most tasks

I'm not sure how you're reaching the conclusion of "most tasks" when Cinebench R24 is the only test you used because R23, which doesn't agree, was rejected for hand-wavey nebulous reasons, and nothing else was tested.

R24 is hardly a representative workload of "most tasks" nor is it claiming/trying to be.

hajile24• 2 years ago

Anandtech shows[0] that M3 is massively ahead in integer performance, but slightly behind in float performance on Spec 2017.

Integer workloads are by far the most common, but they tend to not scale to multiple cores very well. Most workloads that scale well across cores also benefit from big FP/SIMD units too.

Put another way, the real issue with R24 is that it makes HX370 look better than it would look in more normal consumer workloads.

[0] https://www.anandtech.com/show/21485/the-amd-ryzen-ai-hx-370...

aurareturn18• 2 years ago

I'm not sure how you're reaching the conclusion of "most tasks" when Cinebench R24 is the only test you used because R23, which doesn't agree, was rejected for hand-wavey nebulous reasons, and nothing else was tested.

There are no hand-wavey nebulous reasons.

Cinebench R23 uses Intel Embree engine, which is hand optimized for x86 CPUs. That's why x86 CPUs look far better than ARM CPUs in it.

If there is an application that is purely hand optimized for ARM, and then compiled for x86, do you think it's fair to use it to compare the two architectures?

SPEC & GB6 mostly agrees with Cinebench 2024.

sudosysgen3• 2 years ago

Why are you comparing a chip optimized for performance with a chip optimized for efficiency? Take an Ultrabook Zen5, not the HX370.

imtringued2• 2 years ago

Read bandwidth from a single cluster caps out at just under 62 GB/s. The memory controller has a bit more bandwidth on tap, but you’ll need to load cores from both clusters to get it.

Except for DRR5-7500 it isn't just "a bit more" it is actually double at 120GB/s. This might pose a challenge for LLM inference, which absolutely needs the full 120GB/s.

mshockwave• 2 years ago

Switching from individual schedulers to unified one for integer execution makes sense to me, but I still don’t quite understand why FP execution units do the opposite, could somebody explain why?

flostk• 2 years ago

In this interview Mike Clark from AMD explains that a little, search for "third scheduler":

https://chipsandcheese.com/2024/07/15/a-video-interview-with...

The way I understand it, it's a combination of a unified scheduler for floating point being difficult to implement because many FP instructions need multiple cycles, and FP code being more regular in practice, so you don't need the scheduler to be as powerful.

EVa5I7bHFq9mnYK• 2 years ago

Do all new processors include NPUs now? What if I don't need AI, should I still pay for the unneeded transistors? If only things could be made modular/reconfigurable.

Alifatisk• 2 years ago

I wonder if the NPUs can help other parts of the computer offload some tasks anyway?

fulafel2• 2 years ago

You still pay for unused hw when you do need AI since it's incredibly unlikely your software can use it.

AMD's Strix Point: Zen 5 Hits Mobile

Related Stories