flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) (04/21/88)
In article <3467@gryphon.CTS.COM> richard@gryphon.CTS.COM (Richard Sexton) writes: >In article <549@zehntel.UUCP> donw@beaujolais (Don White) writes: >> >> Wouldn't it be nice if ALL the current Amy NTSC graphics modes were kept >> as a SUBSET of the overall graphics capability? That way, it would still >> be THE desktop video machine, AND it could service those Hi Res-ophiles! >> (1000x800 grin, or 2000x2000 grin grin, or 1Meg x 1Meg grin of wooly >> mammoth proportions) > >The problem is, sport, that monitors in this range cost more than >a used porsche. You know, I never could understand why people want such high resolution color monitors. The *ONLY* valid application I can see for them is CAD, and very few people need to do this at home. And I think workstations should have more Umph anyhow (I'm spoiled with a Sun 3/260, which is still too slow). Think about it: The times you need high resolution is when you are working with several windows of text, or want to preview some laser printer output, or similar applications when color isn't necessary. High resolution mono- chrome monitors are significantly cheaper than color ones. The times you need color is when you are playing games, working on anima- tions, or presenting information (such as maps) where color provides impor- tant information. When was the last time you saw a game where very high resolution was a necessity? With any reasonable update rate and memory requirements, it's difficult to even fully utilize the capabilities of a low resolution screen. Similarly for animations, we still have a long way to go in providing realism even at low resolutions without spending days of high power CPU time, why strive for higher resolutions which require even more calculating time? Besides, most good animations will eventually target video tape, so 640x480 resolution is all you can use anyway (of course I am assuming good filtering routines and a full 24 bit planes of color). Why not just provide two different displays, a high resolution monochrome (with maybe 4-bitplanes) display and a low resolution (640x480 with 24 bit planes) color display? I wouldn't even mind having the color display interlaced, that way you can get a better idea of how animations will look on video tape. If it flickers, you need a better filtering routine. Arguments (no flames) welcome. ______________________________________________________________________________ ___ , , ,;,;;;, / Y /| /| Charles Flaig ;/@-@\; | |/ __, ,__ |/ flaig@csvax.caltech.edu | ^ | | /^\ / | | | / /\ /\ \=/ \____/| \_/|_/\_/ \_/ \_\/_/_/_/ "What, you think they PAY me for this?"
dsill@nswc-oas.arpa (Dave Sill) (04/22/88)
>You know, I never could understand why people want such high resolution color >monitors. Why do people want high-res color vision? >The *ONLY* valid application I can see for them is CAD, and very >few people need to do this at home. Perhaps you've just become conditioned to expect only what the current technology can deliver. I remember there was some joker about 25 years ago who said a dozen Univacs would supply enough computing power to meet the entire world's needs. And about 10 years ago there were many people who considered ludicrous the idea of a personal computer with a megabyte of memory. Personally, I'd like to see a display with the resolution and color capability of 35mm Ektachrome (something like 4k x 3k x 25 million colors). But even that has it limitations. It's only 2-dimensional, for one. The ultimate output (and input) device, of course, is the Direct Neural Interface; sound, sight, touch, smell, taste - all limited only by what our brains can handle. The blind could see, the deaf could hear... Whoa, buddy, take a Reality Check. That's science fiction. My point is that it's laughable to suggest that the average user doesn't need a hi-res color display. ========= The opinions expressed above are mine. "The limits of my language mean the limits of my world." -- Ludwig Wittgenstein
pds@quintus.UUCP (Peter Schachte) (04/22/88)
In article <6210@cit-vax.Caltech.Edu>, flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) writes: > The times you need color is when you are playing games, working on anima- > tions, or presenting information (such as maps) where color provides impor- > tant information. > > Arguments (no flames) welcome. Ok, an argument. Color allows the presentation of more information in the same space. For example, wouldn't it sometimes be nice, when you edit C code, to see which things are macros and which are procedures? Or to see which calls are to library procedures? Or to see which calls are to functions defined in THIS file? Or for comments to stand out? Or to see which parts of the file have been modified recently, or by other people? (Imagine looking at souce code where the red printing was the code that had just been changed, and the black printing hadn't been changed since the file was first written, and shades in between were various ages in between.) Or to see variable scoping? Or ... should I go on? These are only suggestions for C code. One can easily think of similar bits of information that would be useful for other applications. All this can be nicely represented with color (but not at the same time :-). And besides, color is prettier to look at! Now, should I start arguing for high resolution, or do you see how useful that is? -- -Peter Schachte pds@quintus.uucp ...!sun!quintus!pds
flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) (04/23/88)
In article <2213@louie.udel.EDU> dsill@nswc-oas.arpa (Dave Sill) writes: >>You know, I never could understand why people want such high resolution color >>monitors. > >>The *ONLY* valid application I can see for them is CAD, and very >>few people need to do this at home. > >Perhaps you've just become conditioned to expect only what the current >technology can deliver. [Talks about CPU and memory views 10 years ago.] But that's exactly what I base my reasoning on: what the current technology (ie. Amiga 3000) can handle. I'm not talking about 10 or 20 years in the future, and strongly agree that higher resolution displays will be useful then. They will even become useful after we have an HDTV standard we can base them on. But *right now* we don't have the CPU power available on a home machine to do justice to a high resolution color display. If you give me a 2Kx2K color display, I darned well want you to give me a 256-node Ametek 2010 to run it with, too! For under $10K since we are talking about the home market! :-) See my point? [ Mouth watering dreams for the *future* deleted. ] ______________________________________________________________________________ ___ , , ,;,;;;, / Y /| /| Charles Flaig ;/@-@\; | |/ __, ,__ |/ flaig@csvax.caltech.edu | ^ | | /^\ / | | | / /\ /\ \=/ \____/| \_/|_/\_/ \_/ \_\/_/_/_/ "What, you think they PAY me for this?"
flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) (04/23/88)
In article <890@sandino.quintus.UUCP> pds@quintus.UUCP (Peter Schachte) writes: >In article <6210@cit-vax.Caltech.Edu>, flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) writes: >> The times you need color is when you are playing games, working on anima- >> tions, or presenting information (such as maps) where color provides impor- >> tant information. >> >> Arguments (no flames) welcome. > >Ok, an argument. Color allows the presentation of more information in >the same space. For example, wouldn't it sometimes be nice, when you >edit C code, to see which things are macros and which are procedures? >Or to see which calls are to library procedures? Or to see which calls >are to functions defined in THIS file? Or for comments to stand out? [lines deleted listing other ways to use color to highlight text] Now, I agree with you that it is often usefull to have things highlighted, but that is what flashing, underlining, and shading do very well too, and all of these work just as well on the high resolution monochrome monitor I mentioned. I personally dislike colored text (flashing too :-), and suspect that most people agree with me since we don't have such things now even with a color monitor for an Amiga. If they were usefull, someone would have done them, and if someone does them and they catch on, I'll update my views. >And besides, color is prettier to look at! Again, I think colored text is garish. But color pictures are wonderful. >Now, should I start arguing for high resolution, or do you see how >useful that is? It was actually the combination of *high resolution* and *color* that I was discussing. I agree that each is necessary *separately* but we don't have the CPU to make good use of them *together* (except for some CAD) so why waste the money and memory? Any other ideas? ______________________________________________________________________________ ___ , , ,;,;;;, / Y /| /| Charles Flaig ;/@-@\; | |/ __, ,__ |/ flaig@csvax.caltech.edu | ^ | | /^\ / | | | / /\ /\ \=/ \____/| \_/|_/\_/ \_/ \_\/_/_/_/ "What, you think they PAY me for this?"
pds@quintus.UUCP (Peter Schachte) (04/27/88)
In article <6247@cit-vax.Caltech.Edu>, flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) writes: > >Color allows the presentation of more information in > >the same space. For example, wouldn't it sometimes be nice, when you > >edit C code, to see which things are macros and which are procedures? ... > > Now, I agree with you that it is often usefull to have things highlighted, > but that is what flashing, underlining, and shading do very well too, and > all of these work just as well on the high resolution monochrome monitor I > mentioned. True. But color allows the presentation of MORE of these. How many different ways can you "mark" a character in monochrome? You mention a few. Let me suggest these: boldface, italics, underline, different point sizes. Flashing if you insist. Not bad. But color offers a lot more (remember that color can be used IN ADDITION to these). Color can be used to show multistate properties, and degrees, more readily than can monochrome. > I personally dislike colored text (flashing too :-), and suspect > that most people agree with me since we don't have such things now even with > a color monitor for an Amiga. If they were usefull, someone would have done > them, and if someone does them and they catch on, I'll update my views. I certainly agree about flashing. But I suspect that if you used a system that used color subtly, you might change your mind. As to your argument that if it was useful, it would have been done, I can give you several explainations of why it hasn't been done. Firstly, it's hard to do. You have to have the information you want to present, and most of the kinds of information I suggested in my posting is not readily obtainable from the text of a C program. The editor would have to maintain much extra information besides the text. I don't think such an editor exists for C on the Amiga. I believe such editors are available for IBMs and Macs now, but I don't think they use color. In the case of the Mac, that's probably because they haven't HAD color until recently. On the IBM, the older video cards don't support any subtlety of color. Highlighting with color would look gaudy. And it might be a mistake to target such a program editor to only color IBMs, especially since the older color standards sport pretty unreadable text. Anyway, I'd much sooner have a high resolution monochrome (or grayshade) display than a low-resolution, color display. I don't do much color graphics, but mostly program development. > >And besides, color is prettier to look at! > > Again, I think colored text is garish. But color pictures are wonderful. If all we were talking about was a terminal, I might agree with you. But we're talking about a workstation. With windows, icons, etc. Color makes these things much richer. Even if you don't have colored text, it would still be nice to, e.g., have the active window have a different background shade or different color title bar. Much nicer than graying over title bars, or darkening the outline around the active window (a la Sunview). > > >Now, should I start arguing for high resolution, or do you see how > >useful that is? > > It was actually the combination of *high resolution* and *color* that I > was discussing. I agree that each is necessary *separately* but we don't > have the CPU to make good use of them *together* (except for some CAD) so > why waste the money and memory? Wait a minute. If each is necessary, then both are necessary. Or are you suggesting having separate monitors? I'd still rather have a single 3x4 foot 100dpi (i.e. 3600x4800 pixel) 24 bit/pixel monitor 18-24 inches from my eyes :-), or separate, high-resolution color monitors for each eye :-), but I could settle for that. -- -Peter Schachte pds@quintus.uucp ...!sun!quintus!pds
msl5864@ritcv.UUCP (Michael S. Leibow) (04/27/88)
In article <904@sandino.quintus.UUCP> pds@quintus.UUCP (Peter Schachte) writes: >In article <6247@cit-vax.Caltech.Edu>, flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) writes: >>> and a million other people will write... I don't understand why peoply argue about this stuff. Why doesn't someone just post a survey asking specific questions about what people want in displays and then just post the results. A survey would be much more interesting than arguing about religous subjects. As far as the religion goes: Blinking is ok if it is slow and doesn't distract the user. I find that a 2/3 duty cycle for cursors is much nicer then a 50% duty cycle. The best way to design the displays (as far as I am concerned) is to make something like preferences, and let the user decide what's best. Amiga preferences lets you select color; maybe in future versions, it will let you choose everything. Ever heard of resources and defaults (.Xdefaults) My two cents..... -- Michael S. Leibow UUCP: {allegra,seismo}!rochester!ritcv!msl5864 CSNET: msl5864%rit@csnet-relay.ARPA
doug-merritt@cup.portal.com (04/28/88)
Michael S. Leibow suggests avoiding religious discussion about colors
and resolution (a good idea), and instead taking a survey/vote. But
that wouldn't prove anything; people's actual reactions to user interfaces
are often different than how they *think* they would react.
More to the point, there have been ZILLIONS of studies on such issues
(see SIGCHI for instance), and they are at least moderately well understood.
I'm sufficiently out of touch with that literature that I'm not going
to argue the points, but if you go check it out, you'll find that
judicious use of colors has highly desirable effects, depending on the
particular application. Same goes for resolution. Poor use of colors
is actively deleterious, but when used in a smart way, there simply is
no longer any question but that you're better off with more colors and
more resolution. Expense is a whole different question, but disregarding
that, the opposition hasn't got a leg to stand on.
Doug Merritt ucbvax!sun.com!cup.portal.com!doug-merritt
or ucbvax!eris!doug (doug@eris.berkeley.edu)
or ucbvax!unisoft!certes!dougflaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) (04/28/88)
In article <904@sandino.quintus.UUCP> pds@quintus.UUCP (Peter Schachte) writes: >In article <6247@cit-vax.Caltech.Edu>, flaig@cit-vlsi.Caltech.Edu (Charles M. Flaig) writes: >> It was actually the combination of *high resolution* and *color* that I >> was discussing. I agree that each is necessary *separately* but we don't >> have the CPU to make good use of them *together* (except for some CAD) so >> why waste the money and memory? > >Wait a minute. If each is necessary, then both are necessary. Or are >you suggesting having separate monitors? Sorry I didn't make that clear. Yes, I *do* suggest separate monitors. A high resolution monochrome (with 4 bits/pixel?) and a low resolution (NTSC) color monitor (with 24 bits/pixel). Use the monochrome for text/schematics/ etc. and the color for games/animations/etc. You even get more total display area. Probably at less cost. With fewer alignment problems (our color SUN monitors almost always have some color fringing along sharp edges). Allowing faster updates to *both* displays. > I'd still rather have a >single 3x4 foot 100dpi (i.e. 3600x4800 pixel) 24 bit/pixel monitor >18-24 inches from my eyes :-), or separate, high-resolution color >monitors for each eye :-), but I could settle for that. Yeahhhh.... (Dreamy smile :-). But I would rather buy a house at this point. It's cheaper (someone said $60K for 2Kx2K color resolution?). Budget considerations also apply.... ______________________________________________________________________________ ___ , , ,;,;;;, / Y /| /| Charles Flaig ;/@-@\; | |/ __, ,__ |/ flaig@csvax.caltech.edu | ^ | | /^\ / | | | / /\ /\ \=/ \____/| \_/|_/\_/ \_/ \_\/_/_/_/ "What, you think they PAY me for this?"
peter@sugar.UUCP (Peter da Silva) (04/29/88)
The point isn't "Color displays are useful", or "color displays are not useful". The question is whether color displays are useful enough to offset the factor of three or so premium you have to pay to get a color display with the same resolution *and generally less clarity* as any given monochrome display. For most tasks, for most people, the answer is: Yes: for CGA resolution displays. (including Amiga) Maybe: for EGA resolution displays. (including ST monocrome) No: for workstation resolution displays.(Hedley monitor?) PS: 1. clarity is a subjective impression. If you don't agree, I'll not argue. 2. The ST mono display is the one advantage the ST has over the Amiga. Anyone know any way to get Amy to drive 70Mhz displays? :->. -- -- Peter da Silva `-_-' ...!hoptoad!academ!uhnix1!sugar!peter -- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter -- Disclaimer: These aren't mere opinions, these are *values*.
foy@aero.ARPA (Richard Foy) (05/02/88)
In article <4856@cup.portal.com> doug-merritt@cup.portal.com writes: >to argue the points, but if you go check it out, you'll find that >judicious use of colors has highly desirable effects, depending on the >particular application. Same goes for resolution. Poor use of colors I use Txed a lot for creating and changing data files. I use TxEd, less and search for examining the information in these files. It would be very helpful to me if I could use color in each of these programs. When I am typing stuff into TxEd a would like to be able to change the drawing ped for a letter, word or sentance or what ever, to any one of the three colors avialable in workbench, so that I would see the colors in TxEd, less, and search. Does anyone know of a simple editor that does this? Richard Foy