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

221–230 of 306 posts

Re: Desktop compositing latency is real

#221

Post by mzzter »

ahartmetz wrote: post #156
mzzter wrote: post #133

Smartphones suffer from input latency too, though I’m unsure of the underlying cause (curious how iOS handles window/view drawing). It only seems to be getting worse, though I haven’t done tests on this. While each new model undoubtedly has better tech specs, the interface responsiveness doesn’t seem to improve.

Ghetto latency test: finger-scroll alternatingly up and down very quickly, see at which frequency your finger and scroll position are 180° out of phase, i.e. you finger is up while the contents are down or vice versa. Smartphones seem to be fine according to that test. Android is very good and iOS is even better.

Hm I get about 4 up and downs per second before the scroll position is 180deg out of phase in Safari on iPhone 7+. That translates to about 125 (1000/8) ms latency?

Re: Desktop compositing latency is real

#222

Post by krylon »

mschuster91 wrote: post #139
krylon wrote: post #136

My work laptop (the only Windows computer I use) runs Windows 7, and I intend to keep it that way as long as Windows 7 still gets updates. This article just confirms my bias, and I freely admit I am biased. I do not like Windows very much to begin with, but as far as Windows goes, I think Windows 7 ____ing nailed it (for people without touchscreens, anyway). On a related note, I have noticed that Outlook 2013 exhibit…

> On a related note, I have noticed that Outlook 2013 exhibits a notable lag between a keystroke and a character appearing in the message window. Try switching to text-only mails, no zoom... and if you must write HTML mails, do not have an image that is larger size than the window. As soon as there is an image that doesn't fit into the window at 100% zoom Outlook begins to crawl.

My work computer is stuck with Office 2007, because I have Office 2007 Professional, and I need Access about once per year for an arcane reason. Office Pro is fairly expensive, so for the time being, I am stuck with 2007. I am still not entirely sure if I should be happy or sad about it. ;-) But I have used Outlook 2013 on coworkers' computers every now and then, and it was pretty laggy.

These days, I do a lot more programming than sysadmin'ning and help desk, but when I was the IT support guy at our company, my overall impression of Office 2013 was not very good. I have seen it just stop working on a handful of computers (out of about 75-80, so to that is a lot), in such a way that I could only "fix" it by uninstalling and reinstalling Office from scratch. On one of our CAD workstations, Outlook and Autodesk Inventor started a feud where an update to MS Office caused Inventor to crash, and the subsequent reinstallation of Inventor caused Outlook to crash when we tried to write an email. (Then we reinstalled Office, and then suddenly things worked magically, so I remain clueless as to what happened.) The latter may be Autodesk's fault as much as Microsoft's (I get the vague impression that they care even less about their software crashing than Microsoft, as long as the license is paid for). But the impression I get is that MS Office has suffered quite a bit over the years. Therefore I am not entirely unhappy about being stuck on Office 2007. I do miss OneNote, the one program from their Office suite I really like, but I have org-mode, so I can manage. ;-)

EDIT: Sorry for venting, that one has been building up for a long time.

Re: Desktop compositing latency is real

#223

Post by Groxx »

cpburns2009 wrote: post #106
jdietrich wrote: post #91

Earlier quoted context omitted.

That problem is largely solved if you use 8P8C connectors with a rubber boot protecting the latching tab. https://info.pcboard.ca/wp-content/uploads/2017/04/RJ45-Conn...

Even with the protective guard the little plastic tabs inevitably break on me. I can't even say those have fared any better than the unprotected versions.

My main complaint is that the protective rubber often gets stiff as it ages, which can make it pretty dang hard to remove old cables (until you resort to pliers).

Re: Desktop compositing latency is real

#224

Post by tormeh »

wruza wrote: post #56
fredley wrote: post #37

Earlier quoted context omitted.

Layers of abstraction take you further away from the metal. The more layers of abstraction your keypress must traverse before rendering is complete and the photons have reached your retina, the longer it will be until that happens. Layers of abstraction make complex tasks more reachable by a larger number of programmers by reducing the amount of specialist knowledge about those lower layers required to do the job. Th…

Numerous games, including those having complex graphics and behavior, can render 120+ frames per second and realtime interactions (physics, optics, reactions) on pretty average hardware. I don’t think that game scripters who make final things like scenery or ui face complexity much harder than those in gtk/qt/wpf/htmljs widget programming. Details would be interesting though, since I’m no game developer. If true, it…

I think you severely underestimate the difficulty of making a game engine.

Re: Desktop compositing latency is real

#225

Post by colejohnson66 »

satysin wrote: post #143
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…

Indeed. I would rather take an excellent monitor, keyboard and mouse over a faster machine without them. Sadly this means I am pretty much limited to desktops as finding a laptop with a colour accurate 120Hz+ display with excellent keyboard seems to be impossible unless I look at some gaming laptop monstrosity that weights 5KG and has a 1 hour battery life. If anyone reading this know of a good laptop for development…

Razer Blades? Not sure about refresh rate, but they have amazing keyboards and displays from what I hear

Re: Desktop compositing latency is real

#226

Post by gjm11 »

satysin wrote: post #143
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…

Indeed. I would rather take an excellent monitor, keyboard and mouse over a faster machine without them. Sadly this means I am pretty much limited to desktops as finding a laptop with a colour accurate 120Hz+ display with excellent keyboard seems to be impossible unless I look at some gaming laptop monstrosity that weights 5KG and has a 1 hour battery life. If anyone reading this know of a good laptop for development…

I can't offer you 120Hz, but the ThinkPad P51 with 4K screen has excellent colour gamut and accuracy and the best laptop keyboard in the business (with the possible exception of the old-style ThinkPad keyboard currently available only in the limited-edition "ThinkPad 25"). Battery life is pretty good.

Downsides: relatively bulky, pricey, no touchscreen or stylus input. (Upside of that: matt rather than glossy screen, which I think is difficult to combine with a touchscreen.)

Re: Desktop compositing latency is real

#227

Post by fra0 »

badminton1 wrote: post #194

If you want to have the most minimal Windows setup: - Don't use an antivirus - Stop unused services running in the background (e.g: services.msc) - Turn off all vision effects including compositing and animation Then you might want to set up a firewall to block all the nonsense like SMB, NetBIOS, etc. You can also set up a cheap old machine to act as your firewall, reverse proxy cache, antivirus/antispam, etc. You ca…

Pick up an LTSB release and remove the desktop (custom shell).

It looks a lot like Arch / Debian with first class hardware support.

Re: Desktop compositing latency is real

#228

Post by AnIdiotOnTheNet »

badsectoracula wrote: post #83
wruza wrote: post #56

Earlier quoted context omitted.

Numerous games, including those having complex graphics and behavior, can render 120+ frames per second and realtime interactions (physics, optics, reactions) on pretty average hardware. I don’t think that game scripters who make final things like scenery or ui face complexity much harder than those in gtk/qt/wpf/htmljs widget programming. Details would be interesting though, since I’m no game developer. If true, it…

There isn't much of a difference between a GUI toolkit you'd find in a desktop application and the GUI framework you'd see in a game - the most likely difference will be that the game GUI will be redrawn every frame whereas the desktop GUI wont (and there are game GUI frameworks that cache their output to avoid redrawing the entire widget tree every frame). The difference when it comes to why games can be snappier is…

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

Re: Desktop compositing latency is real

#229

Post by digi_owl »

jririr828283 wrote: post #88
542458 wrote: post #74

Earlier quoted context omitted.

The obvious ones: Ethernet cables are big (imagine one on a phone), and aren't designed to be plugged and unplugged the massive number of times that USB cables are (spring contacts wear out, little clips break).

USB cables have little to do with the question of sending pixels via IP Outside HDMI and some DisplayPort cables, all display connectors click or screw in And we’re talking about desktop OS and displays

I'm not even sure about sending it as IP, rather as raw Ethernet frames (or simply just using the cables).

Re: Desktop compositing latency is real

#230

Post by stinos »

emn13 wrote: post #130

I will gladly believe this is a real problem, but this page does not demonstrate that (at least not convincingly); the metric used is simply too poor. To quote: > I used my own hacky measurement program written in C++ that sends a keypress to the application to be benchmarked. Then it waits until the character appears on screen. Virtual keypresses were sent with WinAPIs SendInput and pixels copied off screen with Bit…

Ideally, you'd measure from real keypress upto visible light;

This, I think. When measuring latencies for e.g. reaction time experiments for neurophysiology I'd always use a photodiode or similar to figure out when things actually get on the screen as that's the only thing which matters. IIRC this was always with V-Sync on, but still with custom DirectX applications it was impossible to get lower than 2 frames of latency on Windows 7. So it has always been a bit mistery to me how these articles talk about latency of keyboard to screen of right now'?

Post reply on HN