haitex@pnet01.cts.com (Wade Bickel) (01/29/88)
In a letter crash!eniac.seas.upenn.edu!ranjit (Ranjit Bhatnagar) writes me: > A couple of questions about your 3d package... > > - any estimates on approximate cost of package w/one set of glasses? > I'm not really sure what the price will be but they should be under $100 with software. How much under will depend partly on how U.S. customs classifies them. > - why no intuition or multitasking? Speed isn't really a factor > here, because the fastest possible swap between eyes is 30 swaps/second. > I'm no expert, but I can think of several ways to do this under multitasking > and intuition that at least seem feasible. For instance... > Essentially, speed is the factor. In order to create reason- able 3-D the highest possible switching rate must be maintained. On most (but not all) Amiga's this is constrained to 60 switches per second. Since Intuition's screen update options all require re-merging of the copper lists, they are out of the question be- cause they take too long. I have experimented with finding the ViewPort intuition is using and using it for my LoadView() call to no avail. It may be possible to get around this, but intuition doesn't have enough to offer to make it worthwhile. Needed intuition functions are always available by opening a window underneath a graphics screen. > + install a VBI interrupt, modeled after the one used on interlace > screens, which changes the bitmap location of your stereo screen > at every frame, and also (probably writing straight to hardware, > for speed - I concede this intuition-stopper) sends the "change-eyes" > signal to the glasses. > Actually, I have a little bit more complicated system sitting on the VBlank chain. I also tried using the copper to generate the interrupt with some strange results (though it generally worked). Basically you have the right idea. As noted above, intuition's screen re-draw functions are indeed too slow. Multi-tasking is stopped here because if another task tries to take the screen this server (VBlank) will try to take it back. I think I see a work around to this problem. I sent C= and a few others a suggestion that an addition be made to the Task Block structure to eliminate this problem but have gotten no response. > + use a normal interlace screen, and write left-eye pictures on even > lines and right-eye pictures on odd lines. This has the two dis- > advantages of not being able to use standard line/area calls to > draw into the screen, and when you pull down this screen to expose > the workbench, the workbench will be interlaced too - causing > ugly effects if the user keeps the glasses on. The first trick > has the nice property that you can pull down the screen and see > the workbench with both eyes (unless it's an interlaced workbench). > This is out of the question for a number of reasons (some of which you point out. In particular, it messes up the vertical orientation of things just a bit, and of course the drawing functions are non-functional. > + do whatever the people who make ANIM and MOVIE players do. They > reach 30 frames/sec easily, and compatibly, while doing horrendous > huge decompression calculations too. > What do they do? I'm always open to suggestions. > Though I am quite interested in the use of LCD glasses for 3D, I would > not buy a package with these limitations, and I suspect that many others > > would feel the same way - preferring to wait for a competitor to do the > same thing with multi-tasking and intuition compatibility. Unless it can > be shown to be impossible, which I doubt. > There comes a point of diminishing returns for trying to support these features. It is my hope that I will be able to perform a work around for the multi-tasking problem that will be almost transparent, or maybe someone will offer a solution I've not seen? I hope people can live without intuition as it carries too much inappropriate overhead to really use. I've seen a suprising number of graphics aplications and games which do not use intuition. > By the way, it would be nice if you could post this to comp.sys.amiga, > if what I said isn't completely stupid, so I can hear what others think. > (I can't post from here.) > OK. I'm posting it with my responses to avoid redundancy. Thanks for your intrest, Wade W. Bickel Haitex Resources. UUCP: {cbosgd, hplabs!hp-sdd, sdcsvax, nosc}!crash!pnet01!haitex ARPA: crash!pnet01!haitex@nosc.mil INET: haitex@pnet01.CTS.COM
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) (01/31/88)
In article <2444@crash.cts.com> haitex@pnet01.cts.com (Wade Bickel) writes: >In a letter crash!eniac.seas.upenn.edu!ranjit (Ranjit Bhatnagar) writes me: >> - why no intuition or multitasking? Speed isn't really a factor >> here, because the fastest possible swap between eyes is 30 swaps/second. >> I'm no expert, but I can think of several ways to do this under multitasking >> and intuition that at least seem feasible. For instance... >> > Essentially, speed is the factor. In order to create reason- > able 3-D the highest possible switching rate must be maintained. > On most (but not all) Amiga's this is constrained to 60 switches > per second. Since Intuition's screen update options all require > re-merging of the copper lists, they are out of the question be- > cause they take too long. > Sez you! :-) > but intuition doesn't have enough to offer to make it worthwhile. [ ... ] Oh, I dunno. Screen dragging is pretty nice... > Multi-tasking is stopped here because if another task tries to > take the screen this server (VBlank) will try to take it back. [ ... ] There is no, repeat, NO excuse for disabling multitasking for this application. >> + use a normal interlace screen, and write left-eye pictures on even >> lines and right-eye pictures on odd lines. [ ... ] > > This is out of the question for a number of reasons (some of > which you point out. In particular, it messes up the vertical > orientation of things just a bit, and of course the drawing functions > are non-functional. > They ARE??? Gee, I guess my 'box' program is working by black magic then.... >> Though I am quite interested in the use of LCD glasses for 3D, I would >> not buy a package with these limitations, and I'm sure others >> would feel the same way - preferring to wait for a competitor to do the >> same thing with multi-tasking and intuition compatibility. [ ... ] I guess we have a sale right here.... > There comes a point of diminishing returns for trying to support >these features. [ ... ] On the other hand, if you don't support multitasking or Intuition, you're going to get flamed to a crisp on the Net and elsewhere for it. I've written a number of programs utilizing our 3D technology. The most esoteric I've gotten is to use raw ViewPorts and pingpong between them using an interrupt routine. This in no way impedes multitasking; you simply can't bring the workbench to the front in this case. I've written a program that bounces a ball around in a 3D box. It uses an Intuition screen that can be dragged, and uses standard graphics calls. I've written a 3D viewer for Videoscape 3D objects. It uses Intuition screens, windows, gadgets, layers, and standard graphics calls. If the application demanded it, I could have used menus, too. It multitasks. It works. Mind you, you have to pull some coordinated tricks, but it's doable. We'll be using the Sega glasses, but I'm very interested to see yours. Ours should be out Real Soon Now. See you on the shelves.... :-) (Consider this article the friendly posting that it is. I'm not attacking you. I'm sure you do stuff we don't.) _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Leo L. Schwab -- The Guy in The Cape ihnp4!ptsfa -\ \_ -_ Recumbent Bikes: dual ---> !{well,unicom}!ewhac O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor
blgardne@esunix.UUCP (Blaine Gardner) (02/01/88)
in article <2444@crash.cts.com>, haitex@pnet01.cts.com (Wade Bickel) says: > In a letter crash!eniac.seas.upenn.edu!ranjit (Ranjit Bhatnagar) writes me: >> A couple of questions about your 3d package... >> - any estimates on approximate cost of package w/one set of glasses? > I'm not really sure what the price will be but they should > be under $100 with software. How much under will depend partly > on how U.S. customs classifies them. Just curious, are you using Tektronix glasses, Stereo Graphics visors, or... ? -- Blaine Gardner @ Evans & Sutherland 540 Arapeen Drive, SLC, Utah 84108 UUCP Addresses: {ihnp4,ucbvax,allegra,decvax}!decwrl!esunix!blgardne ihnp4!utah-cs!esunix!blgardne usna!esunix!blgardne "Nobody will ever need more than 64K." "Nobody needs multitasking on a PC."
haitex@pnet01.cts.com (Wade Bickel) (02/02/88)
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: [ > = Leo, >> = my earlier posting. >>> = Ranjit Bhatnager) ] > >> Multi-tasking is stopped here because if another task tries to >> take the screen this server (VBlank) will try to take it back. [ ... ] > > There is no, repeat, NO excuse for disabling multitasking for this >application. > I'm not "disabling" multi-tasking, and if the user wishes to bring a task forward, I can deal with that. The problem occures if some unknown task takes the console. Short of examining the EXEC lists prior to a screen flip, which I may try to do, I can see no reasonable way around this. I could go to an interlace screen but this would cause problems should people which to pop back to CLI, which would then be in interlace. >>> + use a normal interlace screen, and write left-eye pictures on even >>> lines and right-eye pictures on odd lines. [ ... ] >> >> This is out of the question for a number of reasons (some of >> which you point out. In particular, it messes up the vertical >> orientation of things just a bit, and of course the drawing functions >> are non-functional. >> > They ARE??? Gee, I guess my 'box' program is working by black magic >then.... > Thanks Leo, I guess it is worth a bit more experimentation. I've seen your box demo. Have you seen mine? I have a program which spins 3-D objects (wire frames, soon filled polygons) in 3-space with parallax. Just how calculation intesive are your real-time demo's? We are also working on a game which uses tabulated data as opposed to calculated perspectives, which is how I suspect your 'box' program works (which I have, by the way). > >> There comes a point of diminishing returns for trying to support >>these features. [ ... ] > > On the other hand, if you don't support multitasking or Intuition, >you're going to get flamed to a crisp on the Net and elsewhere for it. > Do people on the Net buy games? I suppose some do. I don't intend to use intuition simply because I don't need it. Eventually I may re- consider, but for now the graphics system works well enough. Most multi-tasking applications will work, but some user attention may be required. > I've written a number of programs utilizing our 3D technology. The >most esoteric I've gotten is to use raw ViewPorts and pingpong between them >using an interrupt routine. This in no way impedes multitasking; you >simply can't bring the workbench to the front in this case. > How do you keep your interrupt routine from interfereing with other tasks which might also want to claim the sceen? > I've written a program that bounces a ball around in a 3D box. It >uses an Intuition screen that can be dragged, and uses standard graphics >calls. > > I've written a 3D viewer for Videoscape 3D objects. It uses >Intuition screens, windows, gadgets, layers, and standard graphics calls. >If the application demanded it, I could have used menus, too. It multitasks. >It works. > > Mind you, you have to pull some coordinated tricks, but it's >doable. Care to be a bit more enlightening? > > We'll be using the Sega glasses, but I'm very interested to see >yours. Ours should be out Real Soon Now. > > See you on the shelves.... :-) > Good. Our glasses are much better than the Sega glasses. We had looked at buying 4 different types of glasses, including the Sega glasses and decided on the ones we're releasing because they're better. If you'd like a pair let me know. I can (probably) send you a pair for free. > (Consider this article the friendly posting that it is. I'm not >attacking you. I'm sure you do stuff we don't.) > On the contrary Leo, because of your posting, I will re-examine the possiblity of using interlace screens again. When I did this before I had much less Amiga experiance than I now do, so I may have a bit better luck this time. Clearly if I can do this then supporting both intuition and Multi-tasking becomes trivial. Any clues as to how you use the draw() function to draw lines into only half an interlace screen? Leo, when you say "We" do you mean UmiCorp or someone else? As I understood it you'd taken the "activision" (??) contract to convert some programs. Are these 3-D? Also, I heard a rumor that Nintendo's 3-D glasses products were to be indefinitly delayed, particularly due to a lack of good software. Have you or anyone else heard anything along these lines? Thanks, Wade. UUCP: {cbosgd, hplabs!hp-sdd, sdcsvax, nosc}!crash!pnet01!haitex ARPA: crash!pnet01!haitex@nosc.mil INET: haitex@pnet01.CTS.COM
yuan@uhccux.UUCP (Yuan Chang) (02/03/88)
In article <2467@crash.cts.com> haitex@pnet01.cts.com (Wade Bickel) writes: > > Also, I heard a rumor that Nintendo's 3-D glasses products were to > be indefinitly delayed, particularly due to a lack of good software. > Have you or anyone else heard anything along these lines? I'm not too sure about American version of the Nintendo game machines, but the Japanese version has 3-D glasses for quite a while now. Several games that I've seen in my friend's collection has a 3-D logo on them, so I'm pretty sure they're still being produced (in Japan, that is. According to my friend, softwares for American version of the Nintendo still lags the Japanese version by about two years [the Japanese version uses a cartridge about half as small as the American version]). Anyway, the glasses that my friend has use those ancient blue/red filters to accomplish the 3-D effects. And the results are abysmal, understandably. I remember hearing my friend mentioning that there's LCD-shutter type glasses out now, but I'll ask him more about that before I make a fool of myself... 8) > > > Thanks, > > > Wade. > > > >UUCP: {cbosgd, hplabs!hp-sdd, sdcsvax, nosc}!crash!pnet01!haitex >ARPA: crash!pnet01!haitex@nosc.mil >INET: haitex@pnet01.CTS.COM -- Yuan Chang UUCP: {ihnp4,uunet,ucbvax,dcdwest}!sdcsvax!nosc!uhccux!yuan ARPA: uhccux!yuan@nosc.MIL "Wouldn't you like to INTERNET: yuan@uhccux.UHCC.HAWAII.EDU be an _A_m_i_g_o_i_d too?!?"
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) (02/04/88)
In article <2467@crash.cts.com> haitex@pnet01.cts.com (Wade Bickel) writes: > I could go to an interlace screen but this would cause problems > should people which to pop back to CLI, which would then be in interlace. > Why is this a problem? The CLI will only be interlaced if one of the screens is dragged to reveal the other one. If the CLI has the screen all to itself, it won't be interlaced. Apart from making the CLI screen jitter, there are no other adverse effects. There are other compelling (somewhat esoteric) reasons for using interlace, too... >>> ...the drawing functions are non-functional. >>> >> They ARE??? Gee, I guess my 'box' program is working by black magic >>then.... >> > Thanks Leo, I guess it is worth a bit more experimentation. I've seen > your box demo. Have you seen mine? [ ... ] 'Fraid not. > We are also working on a game which uses tabulated data as opposed > to calculated perspectives, which is how I suspect your 'box' program works Nope, I'm calculating everything on the fly. Mind you, I'm cheating a bit... > How do you keep your interrupt routine from interfereing with other > tasks which might also want to claim the sceen? > Perhaps you're thinking farther ahead than I am, but what other task might want to claim the screen? My ViewPort-oriented program is used just for viewing 3D stills. >> Mind you, you have to pull some coordinated tricks, but it's >>doable. > > Care to be a bit more enlightening? > It'll all be in the manual, as part of the released kit. So everyone who buys one will find out how it's done. > Good. Our glasses are much better than the Sega glasses. We had > looked at buying 4 different types of glasses, including the Sega > glasses and decided on the ones we're releasing because they're better. > We liked the Sega glasses because they have built-in neutral grey filters, and they're relatively cheap. Your glasses sound interesting; who makes them? > As I understood it you'd taken the "activision" (??) contract to convert > some programs. Are these 3-D? My Official Employment Status ----------------------------- I am currently working with Reichart Von Wolfsheild (the author of FirePower) on a project. The official straight-line is that we're working on a Superconducting Compact Disk Player :-). Actually, I'm working on something else, but I want it to be a surprise... _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Leo L. Schwab -- The Guy in The Cape ihnp4!ptsfa -\ \_ -_ Recumbent Bikes: dual ---> !{well,unicom}!ewhac O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor
harald@leo.UUCP ( Harald Milne) (02/05/88)
In article <5146@well.UUCP>, ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: > I am currently working with Reichart Von Wolfsheild (the author of > FirePower) on a project. The official straight-line is that we're working > on a Superconducting Compact Disk Player :-). I hope you are not! I was thinking of Superconducting Blue Ether cables! You guessed my game, EtherToys! Oh well. But I do have a few in mind, one of which grew out of necessity. The need to know machine status, before us gurus get harassed to death. Something along the line of Star Trek klaxons going off before being inundated. Ping status, etc. Toys. That, and the fastest vt100 on this side of the planet/Ether. I like it, EtherToys! So, how long before we see you on our shelves? 8^). I can't wait! > Leo L. Schwab -- The Guy in The Cape ihnp4!ptsfa -\ > \_ -_ Recumbent Bikes: dual ---> !{well,unicom}!ewhac > O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") > "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor -- Work: Computer Consoles Inc. (CCI), Advanced Development Group (ADG) Irvine, CA (RISCy business! Home of Regulus and hamiga) UUCP: uunet!ccicpg!leo!harald
haitex@pnet01.cts.com (Wade Bickel) (02/07/88)
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: >In article <2467@crash.cts.com> haitex@pnet01.cts.com (Wade Bickel) writes: >> How do you keep your interrupt routine from interfereing with other >> tasks which might also want to claim the sceen? >> > Perhaps you're thinking farther ahead than I am, but what other task >might want to claim the screen? My ViewPort-oriented program is used just >for viewing 3D stills. > I don't know either, but I have to assume that such a thing could happen and accomodate it or I have to say my system does not multi-task! Thanks, Wade. UUCP: {cbosgd, hplabs!hp-sdd, sdcsvax, nosc}!crash!pnet01!haitex ARPA: crash!pnet01!haitex@nosc.mil INET: haitex@pnet01.CTS.COM