denbeste@etnibsd.UUCP (Steven Den Beste) (12/11/90)
First off, I don't know anything about v0.9 - I ordered after 1.0 was released and that's what they shipped me. Comparisons to "Silver": As best I can determine, Imagine replaces Silver, it doesn't augment it. There doesn't appear to be anything Silver does that Imagine doesn't do better, faster and much, much easier! With that out of the way: The "detail editor" is very nice, a pleasure to use. Just for the hell of it, I tried duplicating an object which I created with Silver: A miniature "Acropolis" made out of Au. With Silver, this object took me the better part of two evenings. With Imagine, I had it in less than an hour - and that included doing part of it twice because I did something stupid and blew away part of it. However, there are problems. Here are two: The math to handle an object which is simultaneously reflective and transparent (that is, filter>0) averages the light intensity derived from each to determine the actual light intensity of a given pixel. This isn't right - they should add. This leads to the odd result that if you've got a clear sphere of glass on a surface with a dark sky, then though it should be 100% transparent it nonetheless gets darker as the reflectivity is raised. It begins to look like it is made out of smoked glass. (My first reaction once I figured out what was going on was "They've perfected the Dark Emitting Diode!". Just how do they figure that adding darkness and light results in less light?) The other problem I'm not yet clear on. With the 1000-to-2000 upgrade last year I managed to make a deal with my dealer to retain my 1000 minus its keyboard. Since then I found someone with a 1000 which was collecting dust, so I bought it from him for small $, and am now using its keyboard. The upshot is that in addition to my 2000 with 9 megabytes of RAM, I've got my old 1000 with 2.5M and its 68020/68881 board. So I've been running the floating point version of Imagine on it, mostly to compare calculation times. With a 16MHZ 68881, the floating point version appears to be about twice as fast as the integer version running on a stock 68000. To get this comparison, I had to duplicate a scene setup. [Here's a trick to help this: Imagine stores file paths in its setup files; if you use an "assign" device path, it will use that. You can then copy a directory structure to another machine and make the same assign, and it will work just dandy.] Anyway, the image I transferred was of two wineglasses (one "spun", one "swept" - I can't figure out what the difference is, and the documentation doesn't say) on a "check" ground. For no reason which is at all apparent, the floating point version mangles the direct view of the ground very badly. Strangely, everywhere you see the ground THROUGH THE WINEGLASSES looks just fine. Very strange. I'm putting together a disk with these two problems and example traces to demonstrate them and will be mailing them in. However, I'm completely satisfied. My first really big project is going to be a car's-eye-view of a roller coaster - should be pretty good once I figure out how to make the camera follow complex paths, which doesn't look too hard to do. Folks, this is one hell of a package. It is now tops on my list of "things to show someone who says 'So what's so great about an Amiga?'". Oh, and for those of you who've been struggling with Silver's documentation (or lack thereof), the Imagine documentation is a hell of a lot better.
her@compel.UUCP (Helge Egelund Rasmussen) (12/12/90)
denbeste@etnibsd.UUCP (Steven Den Beste) writes: >Comparisons to "Silver": As best I can determine, Imagine replaces Silver, >it doesn't augment it. There doesn't appear to be anything Silver does that >Imagine doesn't do better, faster and much, much easier! As far as I can see the "FOG" feature is missing from TS, I've never used it so it may not be a big loss... Also the 3D possibilities found in TS 3.01 SV is missing. I agree that Imagine has a much better "detail editor" than TS. The stage editor with its action screen is MUCH better than the film strip found in TS. I haven't yet found any uses for the "forms editor", any help/good ideas would be greatly appreciated. In general I am very satisfied with the user interface for Imagine. HOWEVER: I have until now tried to create two pictures with Imagine, and NEITHER did work. The first one was the last example from the tutorial manual, which didn't look correct in trace mode. The other was a much simpler picture: I wanted to create a reference picture showing how specular reflection and hardness worked in different combinations, so I created a scene consisting of 50 perfect spheres with different settings for these parameters. This scene couldn't run because of lack of ram on my 3mb B2000. I then tried with 40 and 30 spheres and 3 mb was still not enough! Finally I reduced the scene to 18 spheres grouped together into one object; this scene could render (at least in scanline mode), but only half of the spheres did show up (one of the spheres was not rendered fully, only the central part of it was visible). The number of visible spheres is dependent on the order that I group the spheres together!! All this was in scanline mode, I haven't yet had time to try it in trace mode (no 68881/68882 you know :-( ) Has anyone experienced this kind of problem?? Can it really be correct, that it is impossible to trace 30 perfect spheres on a machine with 3mb RAM????? Helge --- Helge E. Rasmussen . PHONE + 45 31 37 11 00 . E-mail: her@compel.dk Compel A/S . FAX + 45 31 37 06 44 . Copenhagen, Denmark
RIDOUT@ddnvx1.afwl.af.mil (12/12/90)
In article <1300@compel.UUCP>, her@compel.UUCP (Helge Egelund Rasmussen) writes: > denbeste@etnibsd.UUCP (Steven Den Beste) writes: > >[Stuff deleted] > Finally I reduced the scene to 18 spheres grouped together into one object; > this scene could render (at least in scanline mode), but only half of the > spheres did show up (one of the spheres was not rendered fully, only the > central part of it was visible). > > The number of visible spheres is dependent on the order that I group the > spheres together!! >[Stuff deleted] I tried to do a simple picture of three chess pieces. A King, Queen and a pawn. Each rendered seperatly in the same scene work fine. However when I put all three in the scene I get a very dark picture with only part of one piece visible. I haven't figured it out yet. Sounds like a bug! I don't get any memory complaints. > > > Helge > --- > Helge E. Rasmussen . PHONE + 45 31 37 11 00 . E-mail: her@compel.dk > Compel A/S . FAX + 45 31 37 06 44 . > Copenhagen, Denmark -- Brian **************************************************************************** * Brian Ridout Internet: ridout@ddnvx1.afwl.af.mil * * wl/scev * * Kirtland AFB NM 87117 My Apple is better than your Orange. * ****************************************************************************
her@compel.UUCP (Helge Egelund Rasmussen) (12/13/90)
>>[Stuff deleted] >> Finally I reduced the scene to 18 spheres grouped together into one object; >> this scene could render (at least in scanline mode), but only half of the >> spheres did show up (one of the spheres was not rendered fully, only the >> central part of it was visible). >> >> The number of visible spheres is dependent on the order that I group the >> spheres together!! >>[Stuff deleted] Latest news: The scene works PERFECTLY in trace mode, so the bug must be in the scanline code. It is possible that I can render the 40 sphere scene in trace mode; I haven't tried that one yet... Until now, I have one scene which work in trace mode but not in scanline mode, and another scene which work in scanline mode but not in trace mode.. I think that I'll have to write to Impulse soon... It looks like the real error is in the version numbering, the current version must be 0.95 and not 1.0 :-) :-) RIDOUT@ddnvx1.afwl.af.mil writes: >I tried to do a simple picture of three chess pieces. A King, Queen and a >pawn. Each rendered seperatly in the same scene work fine. However when >I put all three in the scene I get a very dark picture with only part of one >piece visible. I haven't figured it out yet. Sounds like a bug! I don't get >any memory complaints. Was this in trace mode or what? I haven't seen any changes in the background colors. The only errors I have seen is missing or distorted objects. Well, apart from the errors, I still think that Imagine is a very good product, I'm looking forward to trying the F/X stuff. Helge ---- Helge E. Rasmussen . PHONE + 45 31 37 11 00 . E-mail: her@compel.dk Compel A/S . FAX + 45 31 37 06 44 . Copenhagen, Denmark
RIDOUT@ddnvx1.afwl.af.mil (12/14/90)
In article <1301@compel.UUCP>, her@compel.UUCP (Helge Egelund Rasmussen) writes: >>>[Stuff deleted] > Until now, I have one scene which work in trace mode but not in scanline > mode, and another scene which work in scanline mode but not in trace mode.. > I think that I'll have to write to Impulse soon... > It looks like the real error is in the version numbering, the current version > must be 0.95 and not 1.0 :-) :-) > > RIDOUT@ddnvx1.afwl.af.mil writes: >>I tried to do a simple picture of three chess pieces. A King, Queen and a >>pawn. Each rendered seperatly in the same scene work fine. However when >>I put all three in the scene I get a very dark picture with only part of one >>piece visible. I haven't figured it out yet. Sounds like a bug! I don't get >>any memory complaints. > > Was this in trace mode or what? I haven't seen any changes in the background > colors. The only errors I have seen is missing or distorted objects. I was using trace mode and no background. I was going to use the starfield for the background (which worked for a single object) I will try the full trace leter. I just figured I would use scanline to test and trace when I liked it but I guess that won't always work. > > Well, apart from the errors, I still think that Imagine is a very good > product, I'm looking forward to trying the F/X stuff. I have to agree. I was never able to do anything in turbo. I can create a impressive but simple object and begin the trace in about 30 min. now. > > Helge > ---- > Helge E. Rasmussen . PHONE + 45 31 37 11 00 . E-mail: her@compel.dk > Compel A/S . FAX + 45 31 37 06 44 . > Copenhagen, Denmark -- later Brian **************************************************************************** * Brian Ridout Internet: ridout@ddnvx1.afwl.af.mil * * wl/scev * * Kirtland AFB NM 87117 My Apple is better than your Orange. * ****************************************************************************
amigan@cup.portal.com (R Michael Medwid) (12/15/90)
I posted this Imagine 1.0 complaint earlier in another thread..but here it is again..it appears that 1.0 Imagine is not fully compatible w/ the Amiga 1000..that is you can not render to anim format animations..altho you can render to imagine format..but then you can't have sound or mix it into things like the director, DVideo III or other. I get a message "only two screens to work with" no matter what kind of anim format I render to.. but at least the imagine format does go and render anyway..but with rendering to anim format I get the message "can't create anim file". If I try to render the same anim on a housemate's Amy 2000 with the ECS/1meg chip ram NO problem..the point is you should be able to render anim format animations with 512k of chip ram..this would be the first animation program that I've seen on the Amy that could not. If anyone has any work-around ideas for this..I'm all ears (eyes). Right now the only way I can think to get an anim is to render all frames of the animation and then employ page-flipper fx or director to string the frames together into an anim. Othe ideas?
davidm@uunet.UU.NET (David S. Masterson) (12/15/90)
>>>>> On 12 Dec 90 12:51:34 GMT, RIDOUT@ddnvx1.afwl.af.mil said:
RIDOUT> I tried to do a simple picture of three chess pieces. A King, Queen
RIDOUT> and a pawn. Each rendered seperatly in the same scene work fine.
RIDOUT> However when I put all three in the scene I get a very dark picture
RIDOUT> with only part of one piece visible. I haven't figured it out yet.
RIDOUT> Sounds like a bug! I don't get any memory complaints.
I've never used Imagine, but, just making a guess, did you put one of the
chess pieces right in front of the light source thereby effectively blocking
the light from the other pieces?
--
====================================================================
David Masterson Consilium, Inc.
(415) 691-6311 640 Clyde Ct.
uunet!cimshop!davidm Mtn. View, CA 94043
====================================================================
"If someone thinks they know what I said, then I didn't say it!"walrus@wam.umd.edu (Udo K Schuermann) (12/17/90)
In article <36933@cup.portal.com> amigan@cup.portal.com (R Michael Medwid) writes: >I get a message "only two >screens to work with" no matter what kind of anim format I render to.. I don't have a chance to try this myself, but if you change the NUMS parameter in the Imagine.configs file from 3 down to 2, Imagine will try to open only 2 screens and thus will stop complaining. ._. Udo Schuermann "How is American beer similar to making love in ( ) walrus@wam.umd.edu a canoe?" -- "Both are f***ing close to water."
davis@soomee.zso.dec.com (Mark W. Davis) (12/18/90)
I have been unable to render a "glass" object in Imagine or Imaginefp. I first tried imitating the techniques I use in Silver to create glass then used various combinations of attributes with no success. When rendered in "trace" the object looks EXACTLY like the same object rendered in "scanline". Any recommendations? =++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++= Mark Davis Internet: davis@decwet.enet.dec.com UUCP: decwrl!decwet.enet!davis =++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++=
RIDOUT@ddnvx1.afwl.af.mil (12/18/90)
In article <1375@rust.zso.dec.com>, davis@soomee.zso.dec.com (Mark W. Davis) writes: > I have been unable to render a "glass" object in Imagine or Imaginefp. I first > tried imitating > the techniques I use in Silver to create glass then used various combinations of > attributes > with no success. When rendered in "trace" the object looks EXACTLY like the > same object > rendered in "scanline". Any recommendations? > > > =++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++= > > Mark Davis > > Internet: davis@decwet.enet.dec.com > UUCP: decwrl!decwet.enet!davis > > =++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++= I was trying this for a couple of days and I finaly fugured out how to do it. I set reflect color up a little, the filter up a lot (like 190 or so). And finally the most important. I found that you should not touch the shiny slider. It seams to negate anything with reflect and filter. While we are talking about benchmarks. I made an sphere object half red and half blue with the atributes as above. It is sitting on a ground with the dots texture, and two light sources casting shadows. I rendered it in interlaced HAM. On a stock Amiga 2000 1 MEG. It only took 23 hours and 58 min. Fast huh :-) -- **************************************************************************** * Brian Ridout Internet: ridout@ddnvx1.afwl.af.mil * * wl/scev * * Kirtland AFB NM 87117 My Apple is better than your Orange. * ****************************************************************************
walrus@wam.umd.edu (Udo K Schuermann) (12/19/90)
In article <1375@rust.zso.dec.com> davis@soomee.zso.dec.com (Mark W. Davis) writes: >I have been unable to render a "glass" object in Imagine or Imaginefp. >I first tried imitating the techniques I use in Silver to create glass >then used various combinations of attributes with no success. When >rendered in "trace" the object looks EXACTLY like the same object >rendered in "scanline". Any recommendations? What are the attributes of the glass object? Do you have an appropriate background to show refraction, and foreground to reflect in the glass? What exactly is the problem with the object? I had a chrome and a glass sphere once and both looked almost exactly alike because of inappropriately chosen backgrounds and lighting conditions. The only difference between Trace and Scanline is: CAST shadow, refraction, and reflection. That and speed, of course. ._. Udo Schuermann "How is American beer similar to making love in ( ) walrus@wam.umd.edu a canoe?" -- "Both are f***ing close to water."
davis@soomee.zso.dec.com (Mark W. Davis) (12/23/90)
After experimenting I found what Brian Ridout spoke of; no shininess on glass objects. Also, the globals have more of an effect on how objects are rendered in Imagine than in Silver. After mucking around with lighting, background, etc, my blue bottle became as transparent as I envisioned. =+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++= Mark Davis Internet: davis@decwet.enet.dec.com UUCP: decwrl!decwet.enet!davis =+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++=