Nuvia: don’t hold your breath

[Artemis]

 

[Jimmyjames]

I recall the days when people required proof.

 

[Jimmyjames]

Don’t trust any power measurements unless you know they used Kill-a-watt.

 

[Artemis]

I recall the days when people required proof.

Purely amazing you posted that right as I posted the documents — you’re gonna love what I just posted from leaked Dell Internal documents in a 300+ page PDF. Like I said, I actually do look into these things — more than you and more than most of this forum or the AMD etc ones. It’s a rare quality these days, I know.

 

[Artemis]

Don’t trust any power measurements unless you know they used Kill-a-watt.

I’m pretty sure Dell’s internal engineering do exactly that. So do QC’s etc. It’s necessarily bound by the battery size anyways.

 

[dada_dave]

x.com

x.com

I checked by downloading the leaked Dell papers, and what he says is correct. This is categorically in Apple territory if you had a cluster map of idle or low load power basically.

That was a considerably better take than the last one I saw of his ... which was pretty bad.

And even when comparing to MTL (someone else here, Jimmy, whom I am not trying to pick on but it’s a relevant point, made fun of them for comparing to ADL — Qualcomm didn’t do that, that was Dell, and MTL is not as much of an improvement as QC brings so this is still relevant) it’s still an advantage.

That was me too, I mean I don't know if Jimmy did it elsewhere, but I was poking fun at Dell for comparing themselves to older Apple laptops like the base M2 and M1 Pro. And I maintain it was still funny, though not as funny as their marketing description of their target customer is. Still vaguely wonder how Apple describes their target demographics, you know beyond dollar signs that go "Mooooo".

I’d be very careful with any claims from Twisted Andy. This is the person who claimed that Speedometer was not a valid benchmark but octane and kraken are. A person who openly stated that they look at data to see which fits their idea of reasonable before using it. A bullshitter in other words.

Andy is definitely terrible broadly and hates Apple, his comparisons are completely full of it in that vein and he cherry picks - I’m fairly unbiased and am aware of that, it’s very irritating.

But you can download the documents from dell yourself from Scribd with a subscription, the idle power is indeed that low. This is all Dell’s own internal measurement, not QC marketing, and I hate to say it but Andy is right.

I'd never heard of him before.

That’s probably right. But that’s a valid way to do things.

Never implied otherwise. Can be more expensive of course, but certainly valid and is often better.

I think people here etc are going to be surprised, Qualcomm, MediaTek/Nvidia are NOT like AMD and Intel when it comes to energy and power and you’re going to see some gaps diminish in the next few years.

Like, even if Apple has an ST lead still, some of these gaps are in absolute and don’t change much from node to node, like idling.

Sure I mean they're all operating on similar paradigms vs Intel/AMD on CPU core design.

 

[Jimmyjames]

Purely amazing you posted that right as I posted the documents — you’re gonna love what I just posted from leaked Dell Internal documents in a 300+ page PDF. Like I said, I actually do look into these things — more than you and more than most of this forum or the AMD etc ones. It’s a rare quality these days, I know.

I think it’s great that you have graced us with your presence. Amazed you tolerate the peasants.

 

[Artemis]

I think it’s great that you have graced us with your presence. Amazed you tolerate the peasants.

Look I apologize for being abrasive. Andy I agree with you is a capital B bullshitter btw. I used to get mad at him for BSing about Apple Silicon. I don’t trust him, but I do trust him when I can verify the numbers — which is exactly what I did because he has an Apple hate boner.

 

[Artemis]

I’m pretty fair here so like yeah of course Jimmy is right about Andy. Don’t listen to that guy generally — his M4 comparisons are patently dishonest — he takes high averages from M3 etc.

But like, if he says the sky is blue, well ok shrug.

 

[Jimmyjames]

I’m pretty fair here so like yeah of course Jimmy is right about Andy. Don’t listen to that guy generally — his M4 comparisons are patently dishonest — he takes high averages from M3 etc.

But like, if he says the sky is blue, well ok shrug.

Yes, we should definitely take an internal Dell document as absolutely accurate without waiting for independent verification.

 

[exoticspice1]

Look Andy pops up like a freaking Pokemon in tall grass whenever Apple launches a new SoC. He does pop out for other hardware but he doesn't write his long windwinded essays about how everything is better than Apple SoC.

I proved him wrong many times. He actually thinks AMD laptop CPUs the high end ones are more efficient than the M3 Max. He writes falsehoods that are truth in his reality. He also lies too much, I called him out for the iPhone 15 Pro perf slowdown which he said Apple did but Apple never did.

I mute him but I had to unmute him during the M4 launch because Vadim and others believed his speil.

 

[Artemis]

That was a considerably better take than the last one I saw of his ... which was pretty bad.

Andy is terrible. @jimmy is totally right about that broadly. But again if he says “wind blows north” and a lot of signs point that way, and I can see it with my own eyes, then ok broken clock thing.

But broadly, ignore him. He is an Intel/AMD Zealot of the Nth degree and a liar. And I do mean basically a liar.

That was me too, I mean I don't know if Jimmy did it elsewhere, but I was poking fun at Dell for comparing themselves to older Apple laptops like the base M2 and M1 Pro. And I maintain it was still funny, though not as funny as their marketing description of their target customer is. Still vaguely wonder how Apple describes their target demographics, you know beyond dollar signs that go "Mooooo".

Oh yeah I would just ignore the customer descriptions.

The truth re comparing to the M1/2 is that
A) the Snapdragon is late
B) apple is so far ahead on power efficient SoCs that the M1/2 are still relevant. I would take an M2 over like, Strix Point or Lunar Lake from what I know just purely based off the hardware.

I'd never heard of him before.


Never implied otherwise. Can be more expensive of course, but certainly valid and is often better.

Oh yeah for sure. Wasn’t accusing you really just pointing out that’s fair!

Sure I mean they're all operating on similar paradigms vs Intel/AMD on CPU core design.

Yeah.

Yes, we should definitely take an internal Dell document as absolutely accurate without waiting for independent verification.

It’s simply about the accuracy within the frame of his claim: did the Dell documents say that? Yes, they did, and they are internal leaks which is fairly nonstupid. You can wishcast around that part but I’m willing to bet Snapdragon reviews are going to find something eerily similar RE: battery life vs Intel and AMD. It’s really silly to die on this hill when I just handed you something anyway — I agree the guy sucks.

 

[Jimmyjames]

It’s simply about the accuracy within the frame of his claim: did the Dell documents say that? Yes, they did, and they are internal leaks which is fairly nonstupid. You can wishcast around that part but I’m willing to bet Snapdragon reviews are going to find something eerily similar RE: battery life vs Intel and AMD. It’s really silly to die on this hill when I just handed you something anyway — I agree the guy sucks.

Just so we’re clear. My contention is not that the figures are wrong, or a lie. It’s that we should wait and see if they are accurate. If that is what it means to “die on this hill” then bury me. Your own screenshot says “Estimatation”. Let’s wait and see. This shouldn’t be an issue to those not looking to show off their galaxy brain.

 

[Artemis]

Just so we’re clear. My contention is not that the figures are wrong, or a lie. It’s that we should wait and see if they are accurate. If that is what it means to “die on this hill” then bury me. Your own screenshot says “Estimatation”. Let’s wait and see. This shouldn’t be an issue to those not looking to show off their galaxy brain.

I’m not looking to show off any galaxy brain btw I’m just tired of poor standards and misunderstandings people have. Even on Twitter you’ll note repliers agreed with me, and then you blocked them. “Just so we’re clear.”

Should we wait and see? Yes, of course!

But it’s an entirely different claim from “Andy is misrepresenting the data” which is what he usually does. Qualcomm having better battery life than MTL and ADL laptops would check out too unless you think MS is false advertising today.

So directionally it also makes sense.

But yes , reviews are the final say.

 

[Artemis]

Moving on here, the Surface 13.8 has a 52wh nominal battery and a 2.2K resolution touchscreen, and the X Plus lasted for 20H video, or 13H web browsing. Video was still connected to WiFi unlike many tests.

Worse than Dell’s claim but different display used.

But still pretty dang good I think

 

[Artemis]

Look Andy pops up like a freaking Pokemon in tall grass whenever Apple launches a new SoC. He does pop out for other hardware but he doesn't write his long windwinded essays about how everything is better than Apple SoC.

I proved him wrong many times. He actually thinks AMD laptop CPUs the high end ones are more efficient than the M3 Max. He writes falsehoods that are truth in his reality. He also lies too much, I called him out for the iPhone 15 Pro perf slowdown which he said Apple did but Apple never did.

I mute him but I had to unmute him during the M4 launch because Vadim and others believed his speil.

Yeah it’s classic. Absolutely classic.

Those AMD SoCs could only be similarly efficient under

- perfectly threaded
- bad power measurements (just taking like CPU power modeled in SW alone will hurt Apple more than it will help btw)
- in a benchmark like CB23 with SMT

i’ve been meaning to do a writeup about this, but the comparisons under heavy load understate how efficient Apple Silicon is for everyday users and power users both. Even if it were just as efficient under load, it wouldn’t matter. That’s not why Apple users really like the M3 Pro/Max — in real use idle power matters and so does the ability to siphon dumb tasks to E cores, or get responsive ST at lower watts.

This is also why the 7840u to M2/3 comparisons at multithreaded benchmarks and full load are dumb. It misses the point.

 

[Jimmyjames]

I’m not looking to show off any galaxy brain btw I’m just tired of poor standards and misunderstandings people have. Even on Twitter you’ll note repliers agreed with me, and then you blocked them. “Just so we’re clear.”

I’m under no obligation to tolerate bullshit, no matter how confident the one person you happen to know was blocked is.

You might want to look at your own citations if you’re talking about “standards”. Mentioning Andrei, Geekerwan and iirc Notebookcheck as proof of your correctness when all three state they use powermetrics or the api that powers it, isn’t a great look.

You may decide to ignore the report mentioned earlier if software measurements bother you. Guess what the MS funded research uses? lol.

 

[Artemis]

I’m under no obligation to tolerate bullshit, no matter how confident the one person you happen to know was blocked is.

You might want to look at your own citations if you’re talking about “standards”. Mentioning Andrei, Geekerwan and iirc Notebookcheck as proof of your correctness when all three state they use powermetrics or the api that powers it, isn’t a great look.

You might want to ignore the report mentioned earlier if software measurements bother you. Guess what the MS funded research uses? lol.

LOL

Andrei used powermetrics — the old one that was better than we have today that still included DRAM — to show that it sucked. So weird you keep coming back to “he used it”. Dude he used it and thinks it sucks and that was why he published about it.



And he did that before they removed DRAM measurements too, so even if it were accurate it’d still have that issue. Mind you, the accuracy issue both ways.


Geekerwan only recently started using Apple’s internal modeling tool, which may also include other things PowerMetrics modifies.

And regardless, Andrei also called out those same 3.6W/7W A17/M4 measurements you posted here or elsewhere as “nonsense” (or bullshit, I can’t remember) in the Chips n Cheese discord and he would actually know. I happen to agree, the A17 especially is very unlikely.

The 11W M4? He says that’s reasonable.

Btw, notebookcheck does a skimmed powermetrics but their main measurements are via the wall, with an external monitor these days.

And second of all, that’s fine! RE: MS funded research, but I’d rather have it from the VRMs there too and compare to Apple. You can get directional ideas such as that an NPU might be more efficient than a CPU, I’m just not a huge fan of using first party APIs and firmware tools to compare two SoCs on power when we can do other ways and when the horse’s races are close.

Battery rundown quite frankly is probably one of the best ways to do things, and use multiple suites to do so.

 

[Cmaier]

Right.

My memory is that the low latency trope only lasted a little while and once the actual tests made the rounds people dropped it. But I'm not in every community so I can't comment on it too widely. Also, to be fair to people, though I never actually reiterated it myself, if someone had asked me prior to Anandtech's articles on the topic, "do you think on-package memory has lower latency?", then I would have said "yeah sure, I guess that makes sense".

i still say it does. whether you choose to take advantage of it is another matter. Depending on your package strategy, the RAM is located either right above the CPU, or adjacent to it. Call it 10 mm away. That means the time of flight is 60ps. Put the memory in DIMMS nearby on the motherboard instead. Call that 10cm away, That’s 600ps. So let’s call it 500ps difference. Assume your CPU runs at 4GHz. That means your CPU cycle time is 250ps. So it takes 2 extra cycles to address the RAM, and 2 extra cycles to read the RAM. Boo.

That’s assuming you optimize your design to take advantage of all that.

 

[Artemis]

Going to reiterate this last part tbh:

Battery rundown quite frankly is probably one of the best ways to do things, and use multiple suites to do so. Still imperfect, but it’s a good one. I think between that, software suites, and wall testing, if we can’t get professionals these in combination are good choice.