Slashdot Mirror
Does Your LCD Play Catch-Up To Your Mouse?
Dishes of Ryan writes "I fell in love with the idea of an LCD monitor, so I ended up buying a nice, shiny Dell 2001FP. However, nowhere, and I mean *nowhere* did I read about LCDs having an input lag on them. For instance, if I scoot the mouse across the screen, there is a noticeable delay between when I move the mouse and when the cursor moves. To prove it to people, made a video showing exactly what I mean. You can almost forget being king of the hill on twitch FPS games like Unreal Tournament. Are there any other Slashdotters out there that are as annoyed as I am? What did you do?"
I wonder if it's the display that's lagging, or the video drivers? The last time I recall seeing an LCD display "lag" was back before the days of TFT screens, where your mouse would "submarine". (disappear while it was on the move)
I work for the Department of Redundancy Department.
I'll second this. I'm running a Sumsung 193v flat panel bought at Sam's club, on an old dual PIII-800 with an NVIDIA GeForce FX 5200 card.
No lag here, at all. And this stuff ain't exactly cutting-edge.
Did it occur to you that maybe you have a hardware problem with *your* system?
I just thought of something you might want to try. LCDs are a bit different than CRTs in that they are completely digital. Since the monitor is digital, it sometimes requires calibration when used with an analog connector. Check your manufacturers specs for the EXACT resolution AND refresh rate that they recommend. The monitor will run in other modes, but it supposedly won't do them as well.
:-)
Once you've set your resolution and refresh rate, be sure to use the auto-adjust button if your monitor has it. When I first got mine, I thought the picture looked like crap. Then I found the auto-adjust. With a push of a button, I suddenly saw the crispest text I'd ever seen in my life. Quite an improvement over CRT displays.
Javascript + Nintendo DSi = DSiCade
That monitor is actually a BenQ monitor with Dell's name on it. It's a great monitor, and has a 16ms response time, so it shouldn't lag at all in normal use. You should try video drivers or maybe even the mouse itself. There simply is no reason a good monitor such as that one (congratulations on your purchase. That's the best cost to performance monitor out right now.) should show lag in a normal situation. I have used a 25ms LCD, and it doesn't lag in normal use. Call Dell after if driver's don't work.
It's the buffering in the driver.
Flat Panels *will* ghost and blur, however they do not lag.
What causes this is buffering of execution commands in the drivers, which makes some games at certain resolutions lag really really bad on input.
Change drivers, and it will usually go away.
I've used lots of LCDs, including plenty of DELL LCDs. The LCDs we've used at work were faded, and the colors looked awful after a copule of years. but I've never never seen any kind of lag like this in any kind of monitor.
My guess is that there is something wrong with the video drivers, or the mouse drivers, or some other part of his computer that's causing these problems.
I can't see the vid because the file is apperantly slashdotted.
autopr0n is like, down and stuff.
rtfa?
the lag is not coming from that.
hell, just read the damn blurb.
here's for the stubborn people:
two monitors, fed from the same computer. other one is some flatty dell and the other one is a crt. now, the movie is about doing something with the mouse that affects both screens, and happens at the same time in the video cards memory, and having observable(with a vid cam..) lag between the two monitors.
world was created 5 seconds before this post as it is.
http://www.vtnetworks.net/CrtLcdComparo.wmv
Site Mirror: Click here.
Video Only: Click here.
If I autoadjust while showing normal windows, the bitmap will usually still have fuzzy areas when I pull it up. If I autoadjust while the bitmap is being displayed, the monitor is able to lock onto it perfectly. The text looks noticeably better with a perfect lock, especially when using sub-pixel sampling on the fonts, which needs pixel-perfect alignment to work properly.
I have a shortcut to this image on my systems because I have a KVM switch, so I need to autoadjust a lot. No two systems have the exact same video timings.
It's also a Dell, so what can you expect? "Dude, you're going to Hell!"
That Dell monitor is probably a rebadged Samsung or LG.
It would be cool if it didn't suck.
UT2004 specifically has two options:
1: Smooth Mouse
2: Reduce Mouse Lag
The normal usage of USB mice should be fine without lag, but when the computer is using all of its resources, USB doesn't get updated as quickly as it should, thus causing the mouse lag.
PS/2 mice have better access to Windows resources and the mouse position gets updated properly and on time.
I am defenseless. Use your button. Mod me down with all of your hatred.
sub-pixel sampling on fonts does not work very good at all unless you use a DVI connector for your LCD. I changed from a VGA to a DVI connector on my LCD panel at work and the difference is astounding. I'm still amazed that 90% of consumers are completely oblivious to this difference. Not to single you out specifically, but I'm tired of the average consumer being ignorant of the differences between video connection standards.
BTW, companies now make excellent DVI/USB KVM switches, so there is no execuse to use a VGA connection on a LCD panel anymore.
For some reason, all of my games ran like crap after picking up the display... Game after game simply ran like a slug after the LCD was added to the mix and I couldn't figure out what the problem was.
I finally noticed that if I took my hand off the mouse, things ran smoother.. After some trial and error I discovered my first generation optical Intellimouse Explorer didn't like the USB hub on the Dell monitor (I plugged it into the 2001FP's USB ports to add some slack on the mouse cable). While the problems were not readily apparent on the 2D apps, they were incredibly apparent in the games.
So after moving the mouse back to the PC's main USB ports, everything improved dramatically. It gave me an excuse to pick up that new fancy Logitech laser deal.
www.lonseidman.com
You can probably do the same thing with Gimp, but it's not immediately obvious to me how to do it.
Since everyone is skeptical, I would like to chime in and say that I'm having the exact same problem (same Dell monitor, too). Perhaps the addition of my specs will shed some light on the culprit.
So far I've tried two different video card setups (both MacOS X on a dual 1GHz g4 power mac). The first was the GeForce 4MX card that shipped with the computer. I was using analog output to analog monitor input. Thinking the lag could be the result of analog to digital conversion, I purchased the ATI Radeon 9000 with digital output.
I'm currently using the digital video output to digital monitor input. The problem is still there. Both cards are AGP, and I never experienced a lag before buying the Dell.
Hopefully this helps. If I've left out something important, let me know.
It displays Pascals triangle.
No it doesn't. It's not Pascal's triangle. It's Sierpinski's triangle. Pascal's triangle is such that the ith row gives the binomial coefficients for the expansion of (a+b)^i. Sierpinski's triangle is a made by drawing a triangle and recursively joining the midpoints of its sides. Pascal's triangle is chiefly an algebraic entity. Sierpinski's triangle is chiefly a geometric/fractal entity.
Time flies like an arrow. Fruit flies like a banana.
This was a big problem with passive matrix screens. I've had old-school (black and white) PowerBooks that did this, and my first ThinkPad (365X) did this too.
However, I have *zero* problems with this on any active-matrix screens I've ever worked with. ThinkPad 600E: lovely, crisp screen, no lag, cursor right there where you want it. PowerBook G3: the most awesome LCD I've ever seen this side of a Cinema Display. I even have a cheapy Taiwanese 15" LCD panel, Envision is the brand, and it's splendid. No lag, no lost cursors, nice and crisp.
That sort of thing shouldn't happen with a modern TFT active matrix screen. There is something very wrong with it.
Knowledge is power. Knowledge shared is power multiplied.
For Windows users, Windows will easily bring up such an image on the desktop; it's under Display Properties, Background, Pattern, 50% Gray, (at least under Win2k).
Megane said:
> I'll bet the monitor in question is connected
> with a VGA plug
And Zorilla responded:
> That Dell monitor is probably a rebadged Samsung
> or LG.
Megane,
I have one of the Dell 2001FPs connected via a VGA cable (it's on a machine that doesn't get used for much gaming so it's connected to a slightly older video card) and I haven't notice a lag when moving the mouse (although I'm in front of my Hercules right now, so I can't actually test to see if the Dell shows the symptoms displayed in his video).
Zorilla,
You're partially correct. The Dell 2001FP contains a LG.Philips LM201U04 panel. The rest of the monitor is Dell designed; although not Dell built.
The following sequence seems to do the trick w/ GIMP 1.2.x:
That should get you a checkerboard pattern on a 1-pixel increment. I haven't seen what this does for an LCD monitor's ability to fine tune an analog signal (since I don't own such a display), but I think it's the pattern you're using. It's the same fill pattern the old monochrome Macs used for their desktops. LOTS of edges to sync on, on every line! :-)
--JoeProgram Intellivision!
I just don't get it. This is the most basic thing you check with the LCD display.
Some people out there still look for higher vertical refresh rate at LCD. *sigh*
Common features:
Diagonal
Color quality/reliablity [1]
GFX input capablity. (VGA/DVI/S-Video etc)
No missing (dark) pixels.
Important with CRT:
Maximum resolution
Maximum Vertical refresh rate at resolution you most frequently use.[2]
Image sharpness
Black pitch [3]
Flatscreen/Trinitron(cyllinder)/Sphere screen.
Important with LCD:
Default (non-interpolated) resolution [4]
<b>Pixel switch-on time</b> (display lag)
Pixel switch-off time (ghosts)
Vieving polarization angle[5]
Maximum brightness
Working temperature range
backlight LED lifetime [6]
[1] These ARE different. LCDs have sugar-sweet beautiful colors, that can't be repeated in print, that's why LCDs are the worst choice for a graphician, while your average end user will enjoy the more-than-lifelike graphics immensely
[2] On CRT image at 25HZ hurts your eyes badly. On LCD you can freely read books at 25HZ, the refresh rate doesn't mean cycles between switching the image on and off, but between changes to constant content.
[3] Is black really black or just a shade of grey?
[4] LCDs have one fixed resolution at which they look great, all the other resolutions suck as computer output pixels don't match display pixels.
[5] If you don't look straight ahead at the screen, some colors just go dark on some screens.
[6] LCD doesn't shine. LCD switches half-transparent pixels on and off, masking the white backlight LEDs off. Without backlight you'll see hardly anything. It's the backlight that eats up most of your batteries too. And it's the LEDs that die first if the screen doesn't get broken/scratched etc first.
Anagram("United States of America") == "Dine out, taste a Mac, fries"
Umm. The pixels on a LCD are always perfectly square. What you mean is that if you used the 1280x1024 mode on a standard 4:3 CRT, the image would be slightly distorted.
And the mode caught on because it's the largest mode using standard pixel numbers that fits into a 4MB framebuffer at a depth of 24 bits. It's been a standard for a lot of Unix workstations (which used fixed frequency 5:4 CRT monitors for this).
Did you know you can fertilize your lawn with used motor oil?
It isn't valid in C89 (ANSI C), because main must either take no parameters, or take one of type (int) and one of type (char **). However, ANSI C compilers are allowed to silently compile this program (and perhaps fail at runtime).
Calling a function without prototype assumes the function was declared as:
int f();
(NOT int f(...) as another poster mentioned, this isn't even a valid declaration as there must be at least one non-variadic parameter)
which means that the number and type of the parameters are not yet known, and it is undefined behaviour if any call to f doesn't match[%] the actual definition of f(), wherever that might be.
Calling a variadic function without prototype is specifically undefined behaviour (for example, many compilers use a different calling convention for variadic functions to non-variadic functions, as another poster mentioned).
However if the convention is the same (eg. gcc on IA32) then it's likely to work correctly. (But still non-portable, obviously).
[%] The number of arguments must be the same, and they must have the same types after the default argument promotions: float->double, and (integer-type-smaller-than-int)->int, (other-types)->(stay-the-same)
If you got the sort of random flickering I got, then don't worry. This is because the monitor isn't precisely sync'ed to the pixel clock of the display card. Doesn't happen on DVI systems where it's an all-digital connection. What you're seeing is effectively a close-up of a moiré pattern between the two slightly-different scan rates. Although it's ugly as sin, it'll do no harm to your monitor. If you auto-recalibrate the monitor with the black-and-white pixel image tiled to fill the screen, you should see this problem reduced to almost nothing, as the monitor resyncs. If your monitor doesn't auto-calibrate, and you have to tweak by hand, remember that the goal state is one in which all the black and white pixels are visible at once, with no bands of all-black or all-white. For the best results, you *must* do this in the monitor's native resolution - anything higher won't work as it's downscaled, and anything lower will produce an inferior result due to scaling (assuming your monitor is set to scale the image - should be OK if not!). Incidentally, if there are any Windows users out there with nVidia graphics cards, you might be interested to know that their current drivers include a calibration screen for LCD panels which is ideal for this, and includes other items such as large colour blocks for the monitor to calibrate to. It leads to quite an improvement!
-- What goes up must come down. Ask any SysAdmin.
I'm at work right now, and I have two Dell 2001FPs running dual-monitor. I was able to replicate *exactly* what's shown in the video--when dragging a window that spans both displays, the window moves faster on the primary display (on the left) than on the secondary display (on the right).
It's not the monitor. It's not CRT vs. LCD. It looks like that's the way Windows deals with multi-monitors.
I humbly suggest that the article submitter swap his displays and use the LCD as primary, and see if the CRT then displays the lag. Bet you dollars to donuts that it will.
If you're running games at the native resolution of your display (1600x1200), the most probable reason for the lag you're seeing is that your video card simply can't keep up. It takes a pretty beefy video card to push that many pixels per frame. Try cutting the resolution to 800x600 and see if your results improve.
Another thing to try would be toggling the "vertical sync" option in your video card's advanced properties. This option specifies whether your video card synchronizes frames with the monitor's refresh. Your CRT probably refreshed at 100Hz, and your LCD is probably just 60Hz, so vertical sync could be slowing you down even if you haven't increased your display resolution.
So, of course, it got opened.
Specifically, I was curious about the fact that I was able to plug the thing directly into my (very) old graphics card which was built before there were such things as desk top flat screens, and actually have it work.
The signal being output by a graphics card is designed to be understandable by the average computer CRT. --Which, (when I've opened those in the past), don't contain a whole lot of extra electronics beyond on-off switches and very basic control systems. That is, with a standard CRT, the signal from the graphics card in my compy pretty much feeds directly into the electron gun and magnetics control system of the CRT monitor with very little intermediary electronics in between. All the really clever electronics is done by the graphics card back in the tower case.
So. .
Since TFT monitors work on a radically different principal than CRT technology, this means that the output signal from my old graphics card, (which I'm guessing is analog), must be translated into a very different type of signal which can be interpreted by the TFT screen electronics, which I am guessing is a digital signal.
This would mean. .
The original image dreamed up by the computer is digital, then converted to analog by the graphics card so that the CRT can apply it, and then because there is no CRT, it is converted back again into a digital signal for the TFT.
Oh yeah. Now that's efficiency!
And it worried me, actually. When I was shopping for my flatscreen, I was bugging sales people, "So are you SURE I don't need some kind of proprietary graphics card to run this thing? If that's the case, then I'm no going to get a flatscreen. I need a GOOD graphics card. Not some hunk of standardized junk made by the flatscreen manufacturer!"
The sales guys always just shook their heads. "No sir. You just plug it in."
"Oh. .
But what do you know? I plugged it in, and no problem. It worked like a charm. So, like I said, I had to open it up.
When unscrewed and pulled apart, voila! Unlike the guts of a standard CRT, there before me inisde the TFT was a whole LOT of extra circuit board and chip set confusion sitting between the monitor cable plug and the flexi-cable which feeds into the actual screen system. So there is some serious signal in interpretation going on! --And none of it, I imagine, would be industry standard; each CRT to TFT signal converter is probably designed and built by whoever happens to be making the flatscreen. This extra engineering necessity provides a whole pile of room to make bad decisions and crappy electronics.
My guess is that this is where the lag you are experiencing is coming from.
For my part, I was fortunate in that Samsung did the job well. I ended up with a system which works invisibly, with no perceivable lag between any input and screen output. Perhaps you can sell your screen off on Ebay and get a better monitor.
Of course, the problem may be something else entirely, but that's my two cents. Hope it helped!
-FL