LWN: Comments on "Much hot air over blinking cursors"

Conservation of energy

occ — Wed, 18 Feb 2009 23:30:40 +0000

[which says that in any closed system, ]
you computer is not a closed system... at least from a thermodynamic point of view.

[Singling out a particular part of a closed system and making guesses]
there was measurement involved, not guesses...

[about the energy flowing in and out in absolute terms is bound to be silly,]
Good luck convincing your Electric Utility compagny that their meters are silly...

Much hot air over blinking cursors

occ — Wed, 18 Feb 2009 17:15:59 +0000

[actually means "[x] make my computer pointlessly waste power and drain my battery so i can't watch a complete DVD on the train ride to Berlin".]

Except that when you watch a DVD, there is no 'blinking' cursor to be seen anyway, and event if you are twisted enough to watch you dvd in a windows while to focus is in another windows, in an edit-field, it is still unlikely that the blinking cursor would cause a wake-up of the CPU, which is already busy playing your DVD.

In other words the 'actually means' above is a Red-Herring, and whether or not you battery has the charge to allow you to watch a DVD is completely unrelated with the blinking-cursor drama.

Much hot air over blinking cursors

chojrak11 — Thu, 12 Feb 2009 17:20:51 +0000

As you can see, this subject matter provokes long discussion in every place
it appears :-)

Conservation of energy

hppnq — Mon, 09 Feb 2009 14:27:23 +0000

"Saving" doesn't mean increasing or decreasing the total amount of something; it means reserving it in some particular place.

I think "saving" energy in general refers to not using it to do work that is "not needed". Even in these practical terms, it is intimately related to the law of conservation of energy, which says that in any closed system, the total amount of energy remains constant over time.

Singling out a particular part of a closed system and making guesses about the energy flowing in and out in absolute terms is bound to be silly, especially if you want to point out where other people are completely wrong and you are absolutely right in terms of what is needed and what not, and in short definitely a waste of that other difficult to grasp natural phenomenon, time.

Much hot air over blinking cursors

dimi — Mon, 09 Feb 2009 01:11:34 +0000

This claim seems rather dubious -- do you have any reference to support it?

Nevertheless, *nobody* argued against a patch to fix this brain-dead behaviour. Obviously there would be no point in blinking a cursor on a hidden console...

The entire discussion was about turning off blinking on the regular cursor used in X, something that users are used to for a long time.

Much hot air over blinking cursors

nix — Mon, 09 Feb 2009 00:46:58 +0000

You don't get it. Those six text consoles are there for almost everyone,
including non-technical users, but *even though they're not visible* and
are on different virtual consoles, they *still* wake up the video card
twice a second to blink their invisible cursors on and off.

Much hot air over blinking cursors

dimi — Sun, 08 Feb 2009 23:42:34 +0000

That's a complete non-issue. The population of users that use only real consoles (no X) is too small for words. Besides, those are the ultra-technical users that are more than capable to change that setting themselves.

BTW, I'm not even sure if the change under discussion here applies to real consoles...

Much hot air over blinking cursors

jwb — Sun, 08 Feb 2009 23:36:20 +0000

I think he's referring to the six real consoles, not xterms.

Much hot air over blinking cursors

dimi — Sun, 08 Feb 2009 22:48:33 +0000

> Oh, good. Then you know that most Linux installs these days have six
> text screens sitting in virtual space, waking up twice a second to turn
> their unseen cursor on and off.

Of course I am aware. I have addressed this problem in my first post:
A cursor blinks only in a windows that has the following properties:
- it has focus
- it's waiting for keyboard input

Moreover, only _one_ windows at a time has focus.

This is trivial to test -- just put your six terminals side-by-side.

So your entire argument is based on a false premise. Considering this,
now you can see that the 2W savings can happen only when:
- the focus is on a window that has an editor
(not that common in a typical session)
- the computer is totally idle, like the mouse is not moving,
not playing music or video, etc
- the screensaver hasn't kicked in

How often does that happen?

Continuing claiming the 2W saving is like saying we save in fact 20W because there is a spike of 20W every few seconds. It's ridiculous.

Dimi.

Much hot air over blinking cursors

PaulWay — Sun, 08 Feb 2009 22:29:44 +0000

> You are missing the point. I have not taken issue with the measurements
> that Matthew took, but with the _implied_ power savings.

Oh, good. Then you know that most Linux installs these days have six text screens sitting in virtual space, waking up twice a second to turn their unseen cursor on and off. Yes, that's not a hardware feature. Matthew's been complaining about that one since LCA 2007. As far as I know it's still the case. Whether those wake-ups are coalesced by the kernel these days is anyone's guess.

Right now I have a terminal with two tabs underneath my Firefox window. I'd be more than happy to see any evidence that X, GNOME and gnome-terminal is smart enough to not have that wake-up still occurring twice a second given that the window is completely obscured. Likewise when I stop typing in this text field the bar cursor blinks on and off. I think you're underestimating the ubiquity of cursors in our GUIs.

The real problem here is that, as I understand what Matthew said in his talk, the only way to know for sure that we didn't write anything to the screen is by having no display code entered during the period. It would require much more extensive surgery on X (that no-one wants to do) to know that no pixels were touched even though display code was entered. So from what I understand of this, putting in a patch to turn off blinking cursors really can save 2W or more.

And I'd still be happy with a 0.2W saving totalled across millions of laptops.

Have fun,

Paul

Much hot air over blinking cursors

jengelh — Sun, 08 Feb 2009 21:00:01 +0000

There is a multitude of cursors available:
* text mode cursor (80x25 and such)
* CGA/EGA/VGA video mode cursor ('mode 13h' and thelike)
* FB-generated cursor
* X11 programs' cursor

Much hot air over blinking cursors

mikov — Sun, 08 Feb 2009 20:46:03 +0000

Or, did I misunderstand you completely, and did you actually mean the text mode console? I guess mentioning VGA threw me off...

Much hot air over blinking cursors

mikov — Sun, 08 Feb 2009 20:44:05 +0000

There is no accelerated hardware support for the text cursor in any graphics card that I know, nor in the driver APIs. (It simply doesn't make sense to accelerate it - fundamentally it is very different from the mouse cursor, which is frequently supported in hardware.)

Much hot air over blinking cursors

dimi — Sun, 08 Feb 2009 14:33:57 +0000

You are missing the point. I have not taken issue with the measurements that Matthew took, but with the _implied_ power savings.

What Matthew measured is the difference in power usage for an *idle* computer with and without a blinking cursor. Great. But from here to say that we'll be saving 2W over an entire user session is balderdash. It's like measuring the power needed to do the actual blink (which may be 20W over 10ms) and claiming we'll be saving 20W.

There is this notion of a "load factor". First we applied it to the blinking itself, and reduced the 20W instantaneous power usage to about 2W average for and idle computer with a blinking cursor.

To calculate real power savings over a user session you have to ask yourself: out of a typical session (8h?) how long would the computer have a blinking cursor and be totally idle? Not a lot (please re-read my original comment). The 2W savings would easily drop to 0.2W, likely even more.

Hope this helps,
Dimi.

radioactive? You got something wrong

dirtyepic — Sun, 08 Feb 2009 09:49:43 +0000

i really must remember one of these days that sarcasm doesn't translate well through internets.

Much hot air over blinking cursors

PaulWay — Sun, 08 Feb 2009 07:43:46 +0000

> Even if you waste 2W as claimed (while the cursor is blinking on an
> otherwise idle box), that would translate to maybe 0.05-0.1W in real life.

> Maybe that's significant somewhere, but please stop claiming we're
> saving 2W. We are not.

So what you're saying is that Matthew Garrett measuring this with a meter is wrong and your unsupported and unjustified guesses are right. Matthew Garrett, who knows exactly what kind of hardware he's working with and what kind of power it consumes, is wrong and you, who offer no proof or knowledge of any kind, are right.

Yeah. Thanks for your useless contribution.

Matthew Garrett gave a talk on this at LCA, in which he basically pointed out that the GPU can be switched from updating at 60Hz to updating at 0Hz, also switching most of its functions off, whenever the graphics driver detects that we're not updating the screen regularly. It's basically about 90% of the power usage of that chip turned off in one go. The start up speed is faster than 1/60th of a second, so you can do it in the time the screen would normally refresh anyway. If that chip's using 5W - and I'd be surprised if even the Intel 915 GPUs use less than that - then that's easily going to deliver 2W of power.

Hope these facts help you,

Paul

Much hot air over blinking cursors

dimi — Sun, 08 Feb 2009 02:44:32 +0000

Yes, but we should not forget that:
* if a computer is idle for ~10min, screensaver kicks in
* the cursor blinks only on windows with editor, so if you browse the web, read email, or read a PDF document there is no such blinking
* you have to be idle at the computer for the savings to happen since you type faster than the cursor blinks

When all of these are put together, the "load factor" of savings are very small, probably just a few percentages over a typical session. Even if you waste 2W as claimed (while the cursor is blinking on an otherwise idle box), that would translate to maybe 0.05-0.1W in real life.

Maybe that's significant somewhere, but please stop claiming we're saving 2W. We are not.

Conservation of energy

giraffedata — Sat, 07 Feb 2009 20:26:27 +0000

More to the point, all discussions about energy-saving that ignore the law of conservation of energy are a waste of my time.

I don't get your point. "Saving" doesn't mean increasing or decreasing the total amount of something; it means reserving it in some particular place. In the case of energy saving, these days we're usually talking about saving it in high-enthalpy carbon compounds such as coal, oil, and wood because a side effect of moving it out of there is that the carbon atoms go into low-enthalpy CO2 in the atmosphere.

On the other hand, when people talk about "conserving energy," I can see how that might sound like something that doesn't require any effort.

Much hot air over blinking cursors

mjg59 — Sat, 07 Feb 2009 16:20:35 +0000

Two wakeups per second, combined with some hysteresis on the drop back to the lower power mode to avoid rapid flicking between power states. It ends up with the card being in high power mode about 10% of the time.

Much hot air over blinking cursors

dlang — Sat, 07 Feb 2009 03:45:34 +0000

have you looked at how much power the high-end video cards eat up nowdays? ;-)

300w isn't impossible

that being said, I agree that it seems very high.

Much hot air over blinking cursors

jwb — Sat, 07 Feb 2009 02:42:53 +0000

I'm pretty sure that the 2W claim is horse hockey, as they say back home. The cursor only causes two wakeups per second to perform a trivial operation. Considering how cheap the cursor blink is, it would have to be causing a spike of something like 300W in order to cause a 2W average power increase over a long period. I know the original poster says he measured it with a wall meter, but those just aren't that good when faced with a spiky, intermittent load.

Much hot air over blinking cursors

hppnq — Fri, 06 Feb 2009 17:33:33 +0000

Personally I find this complete thread bizarre, and a complete and utter waste of energy, but your comment must be the most bizarre and pointless one of all. I don't mean this in an offending way, I am truly flabbergasted.

More to the point, all discussions about energy-saving that ignore the law of conservation of energy are a waste of my time.

Much hot air over blinking cursors

arafel — Fri, 06 Feb 2009 11:18:04 +0000

> If GNOME had "more easily accessible options" the cursor would still be
> blinking by default, power would still be consumed needlessly on devices
> around the world, and mjg59 would still be suggesting that the default
> behaviour be fixed.

Congratulations - I think you win a prize for the shortest path to misunderstanding someone's comment (at least in this thread).

There is no connection between "more easily accessible options" and "mjg59 applying a patch to change the default behaviour". Having accessible options does not in any way prevent defaults being change. Suggesting that it does is, frankly, bizarre.

What Would Steve Jobs Do?

njs — Fri, 06 Feb 2009 05:43:30 +0000

> Although, considering one is running Konsole in the GUI/WM, I suppose that a lot of other activity is keeping the CPU awake, likely making the power savings a moot point.

Well if you're curious, run powertop and find out :-).

In my experience, if I'm not touching the computer, then the CPU pretty much just goes to sleep, waking up only a few times per second.

Unless firefox is running. Sigh. (Maybe this is the fault of random people's annoying javascript, I don't know.) But *every other* app I run is fine...

What Would Steve Jobs Do?

pr1268 — Thu, 05 Feb 2009 22:36:25 +0000

and Konsole (Debian Lenny, at least).

Also on KDE 3.5.10 in Slackware 12.2. It's easy to make it blink, if you prefer (Settings → Configure Konsole). Of course, I'm using a konsolerc file that has persisted through several upgrades, so maybe I so configured this several Slackware releases ago.

I like the not-blinking-cursor.

I do, too. Although, considering one is running Konsole in the GUI/WM, I suppose that a lot of other activity is keeping the CPU awake, likely making the power savings a moot point.

Does a non-blinking cursor in a GUI-based console actually contribute to measurable power savings?

Much hot air over blinking cursors

nlucas — Thu, 05 Feb 2009 22:20:03 +0000

And what about having a power-saving option to disable the cursor blink according to the power-saving policy?

Much like the monitor brightness option, the user can configure what brightness value should be if on battery or on AC. Add it an option to blink or not the cursor according to the user wish.

I don't see how people would not like it if the default was to disable blinking when one wants to save every mW, but stay on by default when the power-saving is not important.

For the others where it is really a question of taste, just disable/enable it, knowing they will reduce the battery life by their selection (much like the decision on the monitor brightness).

Much hot air over blinking cursors

jlokier — Thu, 05 Feb 2009 20:07:32 +0000

I think their point is the GPU wakeups cost more power than CPU-only wakeups.

Which begs the question - Do you really need the GPU to do something as trivial as draw a tiny line or rectangle? The answer is no.

Text editors and terminals since time immemorial have had "special hacks" for efficient single-character updates and cursor drawing, bypassing the slower, fully general methods of screen refresh.

Much hot air over blinking cursors

ncm — Thu, 05 Feb 2009 19:20:06 +0000

We just need to have the blinking cursor implemented in hardware, like in the old days, so it still blinks when all the intelligence has gone to sleep.

I'll start worrying about cursor-blink wakeups when the wakeup rate on my laptop goes below 50/sec.

Much hot air over blinking cursors

Yorick — Thu, 05 Feb 2009 18:42:41 +0000

The advanced ABC 800 did this in 1981, probably more for reasons of ergonomics rather than power consumption (blinking stuff makes it harder to concentrate on other things on the screen).

Much hot air over blinking cursors

nix — Thu, 05 Feb 2009 16:59:39 +0000

It's true: xdaliclock *is* the spawn of Satan. Or at least makes your
machine run hotter. :)

What Would Steve Jobs Do?

mb — Thu, 05 Feb 2009 15:50:09 +0000

and Konsole (Debian Lenny, at least).

I like the not-blinking-cursor.

Much hot air over blinking cursors

NAR — Thu, 05 Feb 2009 15:48:17 +0000

But that's pretty useless to detect Linux hangs (i.e. is the whole computer locked or just the application?). Also there are CPU load applets, network usage applets, etc. that update every second. Anyway, if the GPU uses too much power, isn't it caused by the fact that nowadays GPUs are huge 3D-computing machines that are used to draw totally 2D office documents? Shouldn't they have some kind of low-power mode to be used for 2D graphics?

radioactive? You got something wrong

incase — Thu, 05 Feb 2009 14:03:46 +0000

Those hockey pucks were not radioactive, they where just blinking in the infrared range.

Much hot air over blinking cursors

jamesh — Thu, 05 Feb 2009 13:52:17 +0000

So it is a good thing that the clock defaults to not showing the seconds then.

Much hot air over blinking cursors

NAR — Thu, 05 Feb 2009 11:16:14 +0000

Every transition requires that the system wake up to make the cursor change, and wakeups increase power usage.

Like updating of the seconds value in the clock applet?

Much hot air over blinking cursors

russell — Thu, 05 Feb 2009 09:49:48 +0000

So your total behind "power to the people" unless they have a different preference to you. At which point they become "stupid", and that makes them easier to ignore.

And that's my whinge. GNOME never listens.

Much hot air over blinking cursors

epa — Thu, 05 Feb 2009 09:31:00 +0000

When I first read the story I assumed it was referring to the console... what else is 'the cursor' after all?

Much hot air over blinking cursors

jengelh — Thu, 05 Feb 2009 08:03:04 +0000

I wonder - does this also affect the (non-framebuffer) VGA console, or is it that the blinking there is done entirely by the video hardware without causing wakeups?

What Would Steve Jobs Do?

tzafrir — Thu, 05 Feb 2009 07:49:34 +0000

Likewise are:

xterm/uxterm
mlterm
mrxvt

here (Debian Lenny)

Much hot air over blinking cursors

jdub — Thu, 05 Feb 2009 06:22:10 +0000

People will whinge about everything. They will whinge about the most amazing things. If the people who really want the blinking are so angry about the lack of user-clickable option, so be it.

Everyone else (who isn't whinging about their favourite bugbear) knows that it's just the kind of thing you find at the top of a very slippery slope, and further investigation of the slope generally leads to the kind of vomitous mess we laugh about in other software. You know, like Windows... and all the hilarious crap you see on DailyWTF.

It's what we call an "un-break-me option".

An option which says "[x] blink cursor" actually means "[x] make my computer pointlessly waste power and drain my battery so i can't watch a complete DVD on the train ride to Berlin".

I'm totally behind all that "power to the people" stuff, but it needs to be balanced by a very healthy disgust for "stupid to the people".