experiment: LCD monitor
Mar. 1st, 2004 04:33 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
cellioDang. That was quick.
This afternoon I asked if I could borrow one of the LCD monitors from a test lab for a few days. I have the impression that they're easier on the eyes, which bears investigation, but since I seem to have so many stupid finicky vision issues with monitors, I want to use one for an extended period before deciding anything.
That experiment lasted about 30 minutes, as it turned out. :-( I could not find any combination of settings that wasn't harsh, and the fonts were fuzzy besides. It looked like the fonts got better if I increased the resolution (to 1280x1024) and switched the OS to "large fonts", but I would have also had to make changes in every application, so I didn't go down that path too far. And things were still fuzzy even with those settings.
It's possible that my current machine and/or graphics card is just too old for this kind of monitor; I don't know. I believe this machine is about three years old. (The video card is an NVida Riva TNT2, according to the device manager.)
On the CRT monitor, I have maximum contrast (100) and low brightness (20). It makes some colors a little wonky (photos usually don't look good), but this is my work machine and I'm not doing graphic design. At home, where the lighting properties are a bit different, the settings are both closer to the midpoint. That might be due to lighting diffferences or different monitors (I have a 19" flat-panel CRT at home) or something about the machines; I don't know how to evaluate that without shlepping equipment. So it's possible that an LCD monitor at home would actually work out just fine even though it didn't at work, but if I try it I should buy one locally from a place that takes returns. (Returns with mail-order are way too much of a PITA.)
I guess I was hoping for a more conclusive answer, preferably in the positive direction. Oh well.
This afternoon I asked if I could borrow one of the LCD monitors from a test lab for a few days. I have the impression that they're easier on the eyes, which bears investigation, but since I seem to have so many stupid finicky vision issues with monitors, I want to use one for an extended period before deciding anything.
That experiment lasted about 30 minutes, as it turned out. :-( I could not find any combination of settings that wasn't harsh, and the fonts were fuzzy besides. It looked like the fonts got better if I increased the resolution (to 1280x1024) and switched the OS to "large fonts", but I would have also had to make changes in every application, so I didn't go down that path too far. And things were still fuzzy even with those settings.
It's possible that my current machine and/or graphics card is just too old for this kind of monitor; I don't know. I believe this machine is about three years old. (The video card is an NVida Riva TNT2, according to the device manager.)
On the CRT monitor, I have maximum contrast (100) and low brightness (20). It makes some colors a little wonky (photos usually don't look good), but this is my work machine and I'm not doing graphic design. At home, where the lighting properties are a bit different, the settings are both closer to the midpoint. That might be due to lighting diffferences or different monitors (I have a 19" flat-panel CRT at home) or something about the machines; I don't know how to evaluate that without shlepping equipment. So it's possible that an LCD monitor at home would actually work out just fine even though it didn't at work, but if I try it I should buy one locally from a place that takes returns. (Returns with mail-order are way too much of a PITA.)
I guess I was hoping for a more conclusive answer, preferably in the positive direction. Oh well.
(no subject)
Date: 2004-03-01 03:30 pm (UTC)(no subject)
Date: 2004-03-01 04:09 pm (UTC)(no subject)
Date: 2004-03-01 04:29 pm (UTC)What OS? It's possible that it's doing antialiasing on the text, and 1) some people just don't like antialiasing, and 2) it could be configured wrong, e.g. it still has the idea it's driving a CRT. (For XP, the word to google for is "ClearType".)
Finally, you mentioned changing the resolution -- definitely figure out what the LCD's native resolution is, and use that, because any other resolution really just means "stretch or squash so as to fit the true pixel layout", and looks dreadful.
(no subject)
Date: 2004-03-02 08:17 am (UTC)OS is Win2k Professional in both places, by the way. (I wonder if I'll still be able to buy that on a new machine. Hmm.)
definitely figure out what the LCD's native resolution is, and use that, because any other resolution really just means "stretch or squash so as to fit the true pixel layout", and looks dreadful.
Ah, that could be it. We presumably have documentation for these monitors somewhere... :-)
(no subject)
Date: 2004-03-01 05:11 pm (UTC)Also have you looked into gamma settings, Forest is very fussy aboutthose for his photoshop work and the difference a slight change in the colour balance can make to the overall effect beyond the changes in refresh rate was a real surprise to me. Oh goodie, I just threw you another variable, aren't I sweet? :>
(no subject)
Date: 2004-03-02 08:18 am (UTC)Didn't do anything with gamma settings (don't actually know what that means). Sounds like I should investigate, if it makes that much of a difference for Forest!
(no subject)
Date: 2004-03-02 07:32 pm (UTC)Hi!![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
dmnsqrl pointed me at this entry. I'm a researcher in a computer graphics lab, and, although I don't work in technologies for displays, we certainly use them a lot and know something about them :-) so it's possible I might be able to help. Particularly since I have a particular love of good quality displays. Unfortunately, I'm not quite clear, from reading this entry, on what the qualities are that you're looking for. In particular, I don't know exactly what "finicky vision issues" you have, and, since from a quick scan of journal entries (http://www.livejournal.com/users/cellio/289408.html) it looks like you have an actual vision problem, there probably are special issues I can't judge without having them or the vision problem described. Also, I'd be interested in more description of what you found "harsh" about the monitor settings.
I don't want to intrude or anything; since you mention a "test lab", for instance, it's possible you're already well stocked with people to advise you. If you are interested, drop me a follow-up with anything you think I might find helpful and perhaps I can figure out whether I can contribute anything.
(no subject)
Date: 2004-03-02 09:00 pm (UTC)My vision is somewhat wonky. I was born with cataracts, so I have no lenses in my eyes. Best correction is around 20/40 (better eye), with bifocals. (Can't wear contacts.) There's also some astigmatism in there. Net effect is the glasses are thick and only partially successful, but I've learned to cope.
(This shouldn't be relevant for monitor technologies, but my eyes also don't track together. My depth perception is learned, not real.)
I am sensitive to extremes of light, particularly brightness. I don't have whatever it is that causes some people to have to wear rose-colored glasses; it's just a quirk, probably because the fluid in my eyes acts as a pseudo-lens but disperses the light more. Dunno; I'm not an optician. Anyway, bright lights hurt my eyes; I squint outdoors on sunny days and can get whacked by the glare from new-fallen snow. With respect to monitors, this seems to mean that the standard "black text on white background" approach used by many applications and most web pages hurts my eyes unless the brightness of that white is turned down. (Reading a book under a bright light is no problem; I assume this is the difference between reflected light and phosphors.)
I see flicker, in monitors with low refresh rates and in fluorescent lights. Gives me a headache.
Finally (in the catalog of weirdnesses), my vision seems to be a little bit color-shifted (toward blue), particularly if I take my glasses off. You know those "black lights" that were popular in the 70s? They're deep purple through my glasses and bright cobalt blue to my naked eye. When I go camping in the summer and there's a full moon, I can find my way to the privy as well with my glasses off as on. The sky just lights everything up in a way that I'm pretty sure others don't see.
Ok, that was kind of long-winded... So I don't see very well from a visual-acuity perspective, and I'm sensitive to high quantities of light. I'm currently using a CRT and could continue to do so, but I've heard such good things about the LCD displays (and you can't beat the weight!) that I'd be interested in checking them out. They're supposed to eliminate the flicker problem and have a nice crisp display.
Oh, just checked: on the home computer, brightness is 0 and contrast is 100. The room (about 15x15) is lit by a pole lamp (incandescant) with a 200-watt bulb. (I love yellow light...) Colors look more normal on my home machine than at work -- that is, photos look better. Dunno what accounts for that. (Home: 19" flat Phillips CRT, 1024x768, 85Hz.)
Any advice on what characteristics/specs I should be paying attention to in monitors?
(no subject)
Date: 2004-03-02 10:23 pm (UTC)You can see ultraviolet light! The retina responds to UV, but the lens absorbs it. Glass also absorbs it to some extent, which I think explains the color shift with your glasses. This mildly confused article (http://education.guardian.co.uk/higher/medicalscience/story/0,9837,724257,00.html) says it usually appears blue-white, which fits.
These days, people whose cataracts are removed usually get an IOL that blocks UV, so you have a superpower that not many share.
(no subject)
Date: 2004-03-03 05:53 am (UTC)Yeah, if I were having the cataract surgery now, they'd replace the lenses with implants. Every few years I ask my opthamologist if I can have implants yet and she says ask again in a few years. (The problem is that the eyes matured without anything in there, so it's likely that things are shaped a little differently with no lens in there to hold a place.)
Long winded? I'll show you long winded :-) Part 1 of 2
Date: 2004-03-03 07:30 pm (UTC)Well, my reply has turned out to be pretty long, I'm afraid...I'm going to have to break it into multiple pieces because LJ has a limit on comment size...Hope you're not bothered by length...
You are definitely more sensitive on the blue end of the spectrum, probably UV as well, but I don't know the filtering properties of the lens and it would be pretty difficult to tell for sure about UV proper short of getting a monochromatic UV source. Response curves and filter curves are generally pretty gentle, and most people can actually see colors technically in the UV and IR under the proper conditions.
Congratulations and commiserations! Commiserations because the blue end (including UV) is scattered more, and probably more importantly, is focussed slightly off the retina, making blue things slightly more blurry. Ideally, therefore, you'd have glasses that have color filters in them that duplicate the filtering effect of the lenses you don't have. There are filters that will block UV; I don't know whether there's any additional blue filtering that would need to be done for you. My guess is that your eye doctor knows this and has given you glasses that have whatever filters you need in them, as lacking those filters would strike me as being likely to cause increased eyestrain. This is particularly important because the white on the computer monitor not only is unusual in having a high blue emphasis (because of being made up of red, green, and blue, rather than broad spectrum like an incandescent), but also is generally a perceptually rather blue white. If for some reason you don't have filters that cut out enough blue, this might be another reason why you get less eyestrain from a brightly lit page than a monitor, as you're probably lighting the page with a much less blue light. This might even be related to your love of yellow light (though many people have a similar love, so it might not).
You might try experimenting with changing the white point of the monitor you're using. The monitor will probably refer to it as color temperature, and it may be represented by three circles arranged in the shape of a triangle. Most monitors seem to be coming with default settings around 9500 K or 10000 K. Daylight (and the white point in the standard for images on the web, when people bother to follow the standard) is taken as 6500 K, though real daylight actually varies; this should be another preset available on the monitor and should appear noticably less blue. For comparison, incandescent lighting is around 2800 K, a warm-white fluorescent is around 3000 K, and a cool-white fluorescent is 3400 K. (Explaining what these numbers actually mean is a discussion I'll only get into if you're interested :-) If you do change the color setting, things on the monitor may look the wrong color because you're used to seeing them more blue, but after a short while it should look much more natural. Vision is like that, fortunately :-) If you use someone else's monitor, beware that they may have changed the color balance settings from the default (this has happened surprisingly frequently in the lab here) and you may need to use the factory reset facility to undo that to see the original presets.
In case you're interested, nighttime "black & white" vision is even more shifted in sensitivity towards the blue end of the spectrum than the blue photoreceptors used in "daytime" vision.
Re: Long winded? I'll show you long winded :-) Part 1 of 2
Date: 2004-03-03 10:36 pm (UTC)Argh, I got that last paragraph wrong too. It's more towards the blue than "daytime" vision overall, not than the blue receptors.
Re: Long winded? I'll show you long winded :-) Part 1 of 2
Date: 2004-03-03 11:14 pm (UTC)Ok, so I should read other people's posts and links (http://www.livejournal.com/users/cellio/290782.html?thread=1562334#t1562334) more carefully before posting, too. Sigh. Anyway, yes, definite UV vision. And I found the researcher's web page (http://starklab.slu.edu/humanUV.htm) about it with various publication references and interesting pictures and graphs.
Long winded? I'll show you long winded :-) Part 2 of 3 (oops)
Date: 2004-03-03 07:37 pm (UTC)Unfortunately, it's not true that LCDs don't flicker! I have one at work that does. It probably depends on what they're being backlit by; if a fluorescent is being used, you'll see flicker from that. I don't know if electroluminescent lights flicker or not (I don't see flicker from them); however, my guess is that they're mostly used on laptops because the expense is only justified by their robustness. So, be sure to actually look at the LCD and see if you see flickering.
The brightness and contrast settings may not do what you'd think they would from the name. If you haven't tried this already, you might try flipping your settings (to 100 brightness, 0 contrast) to see whether that's better or worse.
Gamma settings will probably not help you with problems with full brightness white, as gamma correction doesn't change the maximum or minimum values, just the distribution within the same range.
CRTs all use the same phosphors. (This at least did not use to be the case for TVs; in fact, my understanding is that the US broadcast standard NTSC used to be satirically misexpanded as Never The Same Color because TVs were so different from each other and from the cameras, despite the attempt at standardization.) The brightness response of the CRT phosphors to the electron guns is well-understood and follows a very simple curve (parametrized with a gamma, hence gamma correction). Most computer images are designed for them, and although different monitors definitely have different response characteristics due to different gammas (possibly different for the different phosphors), these can generally be corrected fairly simply with gamma correction.
Unfortunately, LCDs are totally strange. Not only does the brightness intensity not vary with anything resembling the same curve as phosphors, but it also varies quite wildly with viewing angle. I don't know if any LCD displays properly correct for the different response characteristics. It's definitely possible for them to do so; I just don't know if they do. (A simple gamma correction is not sufficient.) Display manufacturers do attempt to correct for viewing angle issues, with varying amounts of success. If the monitor is particularly bad this can affect viewing even if you don't move your head around at all, just because the same nominal color will look different at the top of the screen than at the bottom. If the monitor is really bad you can see a shimmering effect because the colors will look different in each eye. Monitors with good viewing angle behavior in one direction may not necessarily have it in another. To test viewing angle issues by looking at an actual monitor, try moving your head up and down and left and right while looking at the screen to see if the brightness or color of anything displayed changes. Be sure to have a variety of different colors on the screen while doing this.
I don't get the impression you do much animation or gaming, but LCDs all have much slower response times than CRTs, which can interfere with animation.
You might also be interested to know that the stereoptic depth cue you're lacking (from not having your eyes track together) is only a small part of the depth perception story anyway. It's invaluable for threading needles, useful for general stuff within arm's length or so, and worthless past I think 30 feet or so (forget the exact number here; might be a lot less). But even in cases where stereopsis is helpful, there are plenty of other depth cues that people use normally—accomodation (eye focusing), depth of field, motion, movement parallax (either yours or the object's), shading, shadows, textures, and on and on. There's even a little debate as to how useful stereopsis really is for general structure understanding (other than obvious things like threading needles) in the computer graphics community. So it's not like stereopsis is the whole story and lacking it you lack an enormous range of depth perception that other people have (which is the impression I kind of thought you had from your description about your depth perception being "learned, not real").
Long winded? I'll show you long winded :-) Part 3 of 3 (oops)
Date: 2004-03-03 07:43 pm (UTC)So, in summary: check your glasses have the right filters, try changing the monitor's color temperature, beware of flickering LCDs, beware of funny contrast and brightness controls, gamma is probably not helpful for your problems, be slightly ware of LCD color problems, and beware of LCD viewing angle issues. Those in addition to the usual considerations (glare, crispness, resolution, etc.), of course.
Hm. Well, this has been a long reply! It really wasn't intended to be long winded; it just turned out that way... Hope it's useful...
Re: Long winded? I'll show you long winded :-) Part 3 of 3 (oops)
Date: 2004-03-04 07:56 am (UTC)Hope you're not bothered by length...
Nope! Thanks for all the info!
Ideally, therefore, you'd have glasses that have color filters in them that duplicate the filtering effect of the lenses you don't have.
Interesting. I assume I do, especially as I only get the blacklight/moonlight effects I described if I take my glasses off. (Sunlight, snow, and bright monitors bother me even through the glasses, but no solution is perfect.) I'll ask my opthamologist if I have filters next time I see her. (I see an opthamologist, not optemetetrist, periodically because I also have glaucoma; I forgot to mention that before but I'm not aware of it having any direct effect on quality of vision anyway. If you treat it, I mean. :-) )
Color temperature: thank you! I found the settings on my monitor at work, and this is fascinating. There are four preset color modes, each of which has a "K" number and a set of RGB numbers that appear to be on a scale of 0-100. (No triangles, and no way to directly edit these RGB values.) Here's what I found:
9300K (default): 70, 70, 70
7500K: 78, 70, 70
6500K: 78, 70, 65
5000K: 86, 72, 60
Choosing 7500K rather than the default makes a big difference. You'd think that 6500K would be the winner, as it has lower blow without outrageously-unbalanced red/green, but it looks a little strange to me. (5000K makes me think my monitor has jaundice. :-) ) It could be, however, that what I need is a gentler transition, and that after I work with 7500K for a while 6500K will look better. I'll come back to it in a few days. At 7500K I've been able to raise brightness to 30 without apparent problems.
I'm looking forward to seeing whether my home monitor has these settings too, and if so what it's currently set at. Too bad there's no interface for directly twiddling the RGB values; I can see why they don't want to give consumers that much flexibility (I remember those old TVs), but it's inconvenient.
In case you're interested, nighttime "black & white" vision is even more shifted in sensitivity towards the blue end of the spectrum than ["daytime" vision].
That's interesting, and consistent with what I've seen about moonlight. I'm used to thinking of my night vision kind of sucking, but that's in the city. When the moon is allowed to do its thing (and it's close to full), and I'm willing to take my glasses off (a discovery made out of necessity once on a camping trip), things are remarkably bright. This came as a shock to me when I discovered it.
(Night-lighting aside: you know those orange-pink street lights they like to use on interstates near major exits, that I assume are cheaper to operate? Hate 'em; they cancel out the effects of my headlights in lighting up the road in front of me. I'd rather have no lights (other than my headlights) than those. All my friends think I'm weird. :-) )
[Continued]
Re: Long winded? I'll show you long winded :-) Part 3 of 3 (oops)
Date: 2004-03-04 05:54 pm (UTC)I did a little looking around, and I'm now pretty convinced that you must at least have UV-blocking filters, because it would probably be harmful for your retina if you didn't. There's a quick and dirty way to see if your lenses also block the additional blue wavelengths your eye lets in, though: ask someone with normal color vision to look at your lenses against a white background and see if they look slightly yellow, about the same color as the bottom set of these lenses removed from human eyes (http://starklab.slu.edu/lens.JPG). (According to the linking page (http://starklab.slu.edu/humanUV.htm), the top ones are from a 79 year old and the bottom ones from a 39 year old.)
Color temperature: I'm glad that helped! I've not yet encountered a monitor that had color temperature adjustment without also some way to either use a user setting or modify one of the presets, so it seems likely any other monitor you got would let you play around with them.
Since the color temperature change helped you, you might also be interested in investigating additional filtering just in general, even if your lenses already do some. Some ski goggles are tinted to block out blue to reduce glare, so there may be clip-on "sunglasses" that do the same. Presumably you'd want something that didn't cut down on the light much other than in the blue region, though, so not real sunglasses, and I don't really know anything about this area or what's available.
Interesting and too bad about the street lights. I don't like them much either, but they don't cancel out headlights or anything. I'm not sure what would cause that.
Part 2
Date: 2004-03-04 07:58 am (UTC)Aha. I didn't know that there were different lighting technologies within the LCD bucket. I have no idea what the monitor I tested uses, but the fact that someone shelled out for 30 or 40 of them for a test lab should provide a hint about quality. (Test lab, not people's personal desktops.)
The brightness and contrast settings may not do what you'd think they would from the name. If you haven't tried this already, you might try flipping your settings (to 100 brightness, 0 contrast) to see whether that's better or worse.
This made for an unhappy CRT experience, but I haven't yet tried it on an LCD.
If the monitor is particularly bad this can affect viewing even if you don't move your head around at all, just because the same nominal color will look different at the top of the screen than at the bottom.
Wow. I never thought of that. I know that viewing angle can affect what you see, but I was thinking in terms of looking at the thing side-on, not differences just from sitting in front of it. :-) (My monitor is usually about 8 inches from my eyes (bifocal range), so I would be affected by this kind of thing.)
I don't get the impression you do much animation or gaming, but LCDs all have much slower response times than CRTs, which can interfere with animation.
You're right; this wouldn't be much of an issue for me. My husband is the Diablo player, not me. :-)
Depth perception: interesting. And here I thought that most people didn't really rely on things like shadows for cues because, well, things just worked. :-) Makes sense that the brain would use all available inputs, though. (And that's why I can't thread needles without help? I thought it was an acuity issue. Might be that too, of course...)
Thanks again for all the good advice and information. I've learned a lot from you, and I really appreciate you taking the time to write all this.
Re: Part 2
Date: 2004-03-04 05:56 pm (UTC)You're welcome, and I'm glad I could help!
(no subject)
Date: 2004-03-03 08:54 pm (UTC)Oh, just realized I should look again at your original entry, not just your follow up to me: Your brightness/contrast settings seem to be inverted between your home and work machines. That may be why the colors appear more normal on one than the other. Other people have already commented on resolution mismatching and DVI/analog. (BTW, in case you didn't already know, you can tell whether a plug on an LCD is DVI by comparing it to the plug on a PC CRT. If they're the same, it's analog, otherwise it's probably DVI. Especially if it has weird flat pins along with the round ones. If you were able to plug the LCD into the same plug you used for the CRT, you weren't using DVI.) There are plenty of new cards that don't have DVI, btw, so I wouldn't assume that.
(no subject)
Date: 2004-03-03 09:01 pm (UTC)Contrast is 100 on both; brightness at home is 0 and at work is 20. (And still the colors look more normal at home, even though that would seem to be the more extreme configuration.) I think I confused you by flipping them textually. :-)
(no subject)
Date: 2004-03-03 09:04 pm (UTC)Argh, you're right, sorry. Then I don't know why they'd look different, unless it's due to very different lighting conditions or the monitor at work having some settings (possibly not user-accessible) messed up.