Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Desktop compositing latency is real

lofibucket.com

Post reply on HN
Print view View original

271–280 of 306 posts

Re: Desktop compositing latency is real

#271

Post by renox »

Zardoz84 wrote: post #31

> Don’t you find it a bit funny that Windows 95 is actually snappier than Windows 10? It’s really a shame that response times in modern computers are visibly worse than those in twenty years ago. My Amiga A1200 (Only have extra RAM) feels faster and more responsive that any modern computer with Windows or GNU/Linux desktop.

Except that the Amiga didn't have any memory protection, something not really possible when connected to Internet.

A more interesting comparison is BeOS: it had memory protection and it was (probably still is) way more responsive than Linux and Windows. Thad said with a SSD a computer responsiveness is good enough and if display latency bother you, buy a monitor with an high refresh rate! You'll have a fix much sooner than waiting for a software fix of the issue..

Re: Desktop compositing latency is real

#272

Post by thesmok »

mark-r wrote: post #178
falcolas wrote: post #126

Earlier quoted context omitted.

Requires a bigger chip (100+ IO pins) and more complex wiring diagrams than most inexpensive keyboard makers are willing to invest.

I don't understand why someone hasn't come out with a dedicated keyboard chip yet. If it's cheap enough you don't need to run all the keys to a single chip, you could have multiple chips that all talk over a serial bus to one that it designated the master.

Another layer of serial interfaces would make it even slower.

Re: Desktop compositing latency is real

#273

Post by rasz »

Zardoz84 wrote: post #31

> Don’t you find it a bit funny that Windows 95 is actually snappier than Windows 10? It’s really a shame that response times in modern computers are visibly worse than those in twenty years ago. My Amiga A1200 (Only have extra RAM) feels faster and more responsive that any modern computer with Windows or GNU/Linux desktop.

open a hard drive folder, wait until it draws one icon at a time and get back to me

Re: Desktop compositing latency is real

#274

Post by rasz »

laumars wrote: post #85
amiga-workbench wrote: post #54

Earlier quoted context omitted.

Try popping the start menu open on a Windows 10 machine with a mechanical hard drive.

It wasn't much better on Windows 95 on average hardware around the time of the release of Windows 95. Heck, if you dared click [Start] as soon as the desktop displayed (read: Windows hadn't yet finished booting) then the whole OS would hang for several minutes.

Windows 95 had a ton of "clever" tricks, like "OLE Chicken" aka shimming in fake OLE instead of loading real thing just to display desktop faster, executing anything triggered .dll load anyway, but official metric was blue desktop....

https://blogs.msdn.microsoft.com/oldnewthing/20040705-00/?p=...

Re: Desktop compositing latency is real

#276

Post by Gravityloss »

Zardoz84 wrote: post #31

> Don’t you find it a bit funny that Windows 95 is actually snappier than Windows 10? It’s really a shame that response times in modern computers are visibly worse than those in twenty years ago. My Amiga A1200 (Only have extra RAM) feels faster and more responsive that any modern computer with Windows or GNU/Linux desktop.

A fun experience some years back: running Wine on OSX and browsing the file system with explorer was a lot faster than with OSX:s own finder.

Re: Desktop compositing latency is real

#277

Post by sspiff »

klmr wrote: post #93
83457 wrote: post #33

Earlier quoted context omitted.

"I can send an IP packet to Europe faster than I can send a pixel to the screen. How f’d up is that?" - John Carmack

I had a hard time believing that quote at the time. John ended up answering the question I asked about it on SuperUser [1] with details about his setup. His writeup is impressive: he actually set up a high-speed camera to record the actual latency. [1] https://superuser.com/q/419070/2269

I used to work for Philips' TV branch, they used a similar setup. Instead of a camera they mounted a light sensor to the display and measured end-to-end delay for various color transitions (white-to-black, black-to-white, ...). All other manufacturers and many serious reviewers perform similar tests.

Of course, these tests are always performed with all built-in image processing disabled, those can easily add 2 frames of delay on modern televisions. MPEG deblocking and noise removal are performed by most TV brands to improve image quality at a cost of latency.

Re: Desktop compositing latency is real

#278

Post by rjsw »

raphlinus wrote: post #98

I've been experimenting with this too, in the context of the Windows front-end for xi editor. It's absolutely true that the compositor adds a frame of latency, but I have a very different take than "turn it off." First, it's possible to design an app around the compositor. Instead of sending a frame to the system, send a tree of layers. When updating the content or scrolling, just send a small delta to that tree. Fur…

> Further, Microsoft engineers have said that hardware overlays are coming, in which the graphics card pulls content from the windows as it's sending the video out the port, rather than taking an extra frame time to paint the windows onto a composited surface.

So they are copying the Xerox Alto.

Re: Desktop compositing latency is real

#279

Post by AnIdiotOnTheNet »

cat199 wrote: post #247
AnIdiotOnTheNet wrote: post #228

Earlier quoted context omitted.

I'll take a frame of latency and Wayland's not-completely-insane model over X any day.

yet strangely, wayland removes the frames-of-latency known as 'remote windowing' in it's quest for 'improvement'..

Because it vastly simplified many things, and oh look, compositors have managed to add remoting back in somehow without it!

I mean christ, windows has no support for remote windowing in it's API and still has RDP, even for single windows.

Linux people just love to complain about any change that actually makes the system better because then they can't feel quite as elitist for using it.

Re: Desktop compositing latency is real

#280

Post by VeejayRampay »

nickjj wrote: post #23

This is also why picking a good monitor is important for software development. Some monitors have tons of input lag (60-70ms) and that's the time it takes for you to see what you're typing to reach the display. This also includes the time it takes to see you move your mouse cursor too. I did a huge write up on picking a good monitor for development which can be found at: https://nickjanetakis.com/blog/how-to-pick-a-g…

We're using them at work, I'm very pleased with them. And the bang/buck ratio is great.
Post reply on HN