Yes. You do. If I'm casually browsing non-intensive web pages, e-reading, or watching a Netflix movie, or even encoding a video, the background is doing IO rate limited system updates and basic data logging, etc, the vast majority of the time the CPU demands are very low, but frequent enough that putting a CPU to sleep is completely out of the question.

An A53 has a much lower ceiling, but a much better middle tier power usage level, than the A57. Yes, if you want to run a benchmark the A53 is not a good bet (and is generally worse in a workload power usage), but it is a very good bet for most real world usage.

Encoding a video should not be CPU, so let's get that out of there. Most of the rest is more GPU dependent than CPU dependent. The amount of CPU time you should be spending on these tasks is really low.

For example, web browsing and reading, the CPU should be asleep most of the time.

Now, don't get me wrong, there are plenty of very silly little tasks for A53's to do, but what you listed are not those tasks.

It's more things like "syncing" or something that is a poll loop and event bound, not something that is in any way CPU bound. Period. CPU bound stuff is not something for the A53's in this to tackle. It makes battery life worse. That is what the actual, in-the-field data says.

" a much better middle tier power usage level, than the A57"

Truthfully, for most A53 cores, this is true only in the dreams of the chip designers.

"but it is a very good bet for most real world usage."

Then what, pray tell, do you expect the A57's to be doing in this world?

And why, in practice, has big.little and other things not shown any better battery life at all if it's really a better way of doing things.

I have no doubt it may be a better way of doing things in the future, but it ain't right now ;)

This is 100% false. I do bringup on these platforms. I know a lot about them.

"I'm going to favor the industry's interpretation of this a bit more than your anecdotal, occasional compiler-writer knowledge. "

Dude, i literally do the toolchain bringup on these platforms you claim i have no knowledge of. This is not "anecdotal occasional compiler-writer knowledge". This is "I get paid to make the stuff you keep talking about work fast and get good battery life out of it".

If what you say was true, then one would expect that when they were brought up, they would have worse performance and better battery life. Instead, the exact opposite is true in most cases. They have slightly better performance, and much worse battery life. Try it sometime.

Battery life is gotten back mainly through .. wait for it ... compiler optimization ... to speed up the software so things can sleep faster.

Not to "move things onto the little cores to save power", as you seem to think.

Since you want to attack my experience, remind me again, what background do you have in this again? (FWIW - I would stay away from this argument line as it is unlikely to serve you well)

As swetland (who was android's main kernel guy for a long time) pointed out to you, you literally have no idea what you are talking about when it comes to CPU sleeping and what the main power policies are.

BTW, much earlier i asked if you had anything other than self-serving industry press releases to back up your claims. I'm guessing the answer is "no", given what you've written.

Anyway, i'm going to stop responding now, because i'm just some anecdotal compiler writer, and clearly i can't compete with your vast knowledge store and experience on this one.

The fact that your response to someone like swetland is "Not to mention that their comment is simply dated and wrong now regardless." is hilariously arrogant and amazing.

I'll just continue to find it funny that you want to cite chip manufacturers (who uh, might be slightly self-serving in selling their architectures, and produce zero actual phones except for samsung. Except, of course, that the division of samsung that produces the phones is separate and does not really talk to the chip making division. I'm sure you know this), when i am talking about the actual qualification and testing of real world phones, and not "whatever random crap they made up to put in their marketing material".

You can try to gussy that up as "actual field observations of SOC vendors" or whatever, but that just shows further ignorance in how SOC vendors actually operate.

Again, I urge you: Buy an evaluation board or two. They are cheap. Put an OS on them. Try out what you claim is true. Measure the power draw. I think you will be surprised.

Have a good one!

1) I can easily envision a scenario where he really is an expert but is not necessarily allowed to come out and say "I am a senior embedded systems engineer working for XYZ Mobile Devices" and has years of experience and is sitting there laughing at you. I should know because there are a couple of completely unrelated subjects that get widely discussed on the internet where I am an expert but can't just flat out state all my qualifications. I try to contribute sometimes in discussions about these subjects but often people say I have no idea what I'm talking about.

2) You seem to believe that he cannot possibly be an expert. He disagrees, and offers no absolute proof (and really, does he have to do that? Why can't we just be nice to each other).

3) I am embarrassed for you because you have seemed to be very mean and dismissive. Even if you're right, using phrases like "You are a Hacker News "Expert" and a bore" and "Save the world from your ignorance" just makes me feel that you're being rude.

That said there's plenty of lightweight threads and processes that are mostly IO or event bound, do very little compute, and will run in roughly the same time on a small core as a large core, but the small cores have a lower static power footprint (base cost to have them powered up at all), so big.little works out well there.

A sleeping core is obviously the most efficient core, however you have said nothing to discount what I said. Putting a core to sleep is a very costly activity, which is exactly why it happens at the millisecond scale.

Before that happens, the core will likely have been frequency scaled to more appropriately fit the window's loading. I mean, we know this is the case right now, and that "run fast and go to sleep" is not the primary policy. It's "run as fast as appropriate for the workload to fill a quantum, and sleep when there is no workload". There are many if not most workloads that are externally bound, or event triggered enough that sleep is completely out of the question and running faster does nothing.

Your words have been used (out of context and inappropriately) to bolster DannyBee when they represent, literally, all that is wrong with Hacker News.

Modern CPUs have many different levels of "going to sleep" and many of them can be switched in and out of very quickly.