On 10/16/2021 1:48 PM, J. P. Gilliver (John) wrote:
> Chrome. (The browser; reasonably up-to-date.)
>
> Often, when I look at task manager to see what's slowing things, or Process Monitor to see what's doing all the disc access, I find there are a lot of chrome.exe processes.
>
> Are they one per tab I have open? If so, is there any way to figure out what PID corresponds with which tab (other than killing them)?
>
> and:
>
> when I go back to a tab I haven't been to for a while (but that is open): hovering over the tab, I usually get a popup thumbnail of what that tab looked like last time I looked at it. But if I actually click the tab, it often reloads - even if it was something I wouldn't have expected to change, like a Google map view. Is there any way to tell Chrome not to reload a tab when I switch to it, until/unless I actually click reload?
>
> Windows 7 HP 32-bit, but I thought might not be OS-specific, hence the crosspost.

There is a "continuous display paradigm".

Firefox is proud of its "FPS" or frames per second display.
They compute the view, 60 times per second. And they must have
copied this idea from Chrome (because Mozilla never invents
anything for themselves, the inspiration always comes from Chrome).

This is not a more conventional "exposure event, redraw screen"
world we live in. Instead, the windows compute the contents
continuously. It's possible some of this is done with
compositing (so the multimedia process draws the video window(s),
the content process asks the video card to render text, the video
card overlays make this look seamless).

I'm not claiming to know the details of this. This is
merely an illustration of how upside-down things are
in the year 2021.

You can tell it is compositing, just based on how quickly
some menus "flash" and disappear as you slide a mouse over
a menu area on some web pages. It's happening with "video card
speed", not "CPU speed". When a CPU is asked to compute a graphic
on the fly, it's a touch slower at it.

It's all totally unnecessary.

When a browser window is iconified, obviously the Javascript
is still running. But not all parts of the display system are
kept consistent (which is why the tab view is "out of date").
But the Javascript is hella busy, as just yesterday, I found
my browser had done 32GB of I/O read operations. And I'd like
to know what exactly it was scraping off my disk drive... :-/
Was it scanning for every occurrence of "Top Secret" ?

Paul

On Sun, 17 Oct 2021 at 01:22:28, Paul <nospam@needed.invalid> wrote (my
responses usually follow points raised):
>On 10/16/2021 1:48 PM, J. P. Gilliver (John) wrote:
>> Chrome. (The browser; reasonably up-to-date.)
>> Often, when I look at task manager to see what's slowing things, or
>>Process Monitor to see what's doing all the disc access, I find there
>>are a lot of chrome.exe processes.
>> Are they one per tab I have open? If so, is there any way to figure
>>out what PID corresponds with which tab (other than killing them)?

(JJ has answered that bit - I've responded separately.)

>> and:
>> when I go back to a tab I haven't been to for a while (but that is
>>open): hovering over the tab, I usually get a popup thumbnail of what
>>that tab looked like last time I looked at it. But if I actually click
>>the tab, it often reloads - even if it was something I wouldn't have
>>expected to change, like a Google map view. Is there any way to tell
>>Chrome not to reload a tab when I switch to it, until/unless I
>>actually click reload?
>> Windows 7 HP 32-bit, but I thought might not be OS-specific, hence
>>the crosspost.
>
>There is a "continuous display paradigm".

Yes. Fair enough for videos, or at a pinch things like rolling news (or
I suppose Twitter - I'm not yet into Facebook or any of the others); but
not, for example, a map!
>
>Firefox is proud of its "FPS" or frames per second display.
>They compute the view, 60 times per second. And they must have
>copied this idea from Chrome (because Mozilla never invents
>anything for themselves, the inspiration always comes from Chrome).

The phrases "for its own sake" or "for the sake of it" come to mind )-:.

(For Firefox, I use such an old version that _some_ of these things
don't apply. It's still my default browser as far as actual settings are
concerned, but for a lot of things I know to go to Chrome as I know the
old F won't render them.)
>
>This is not a more conventional "exposure event, redraw screen"
>world we live in. Instead, the windows compute the contents

Don't I know it )-:.

>continuously. It's possible some of this is done with
>compositing (so the multimedia process draws the video window(s),
>the content process asks the video card to render text, the video
>card overlays make this look seamless).
>
>I'm not claiming to know the details of this. This is
>merely an illustration of how upside-down things are
>in the year 2021.

So you don't know any way to tell Chrome not to do it.
>
>You can tell it is compositing, just based on how quickly
>some menus "flash" and disappear as you slide a mouse over
>a menu area on some web pages. It's happening with "video card
>speed", not "CPU speed". When a CPU is asked to compute a graphic
>on the fly, it's a touch slower at it.
>
>It's all totally unnecessary.

Oh, but we've got to keep using up all the hardware resources, haven't
we - otherwise the poor hardware manufacturers ... (-:
>
>When a browser window is iconified, obviously the Javascript
>is still running. But not all parts of the display system are
>kept consistent (which is why the tab view is "out of date").

Pity there's no way (unless anyone knows otherwise?) to tell chrome to
_pause_ the javascript (idle thought: should it have a capital J or
not?) on a tab-by-tab basis, or even just for all but the
currently-focused tab.

>But the Javascript is hella busy, as just yesterday, I found
>my browser had done 32GB of I/O read operations. And I'd like
>to know what exactly it was scraping off my disk drive... :-/

Ditto.

>Was it scanning for every occurrence of "Top Secret" ?

Or 16-digit numbers, the word password, flesh tones, ...
>
> Paul
>
John
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

I'm not a great fan of new technology. I don't change my phone every time the
bell rings - Sir David Attenborough, RT 2016/1/23-29