Because its Halloween or something...

Discuss the Wing Commander Series and find the latest information on the Wing Commander Universe privateer mod as well as the standalone mod Wasteland Incident project.
Post Reply
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

klauss wrote:A) No fresnel. But it's easy to add if you need it - just count on it.
Great! :D
B) No automatic scaling. I thought of adding it, but since the artist can pretty easily do it... why bother?
But the artist can't increase the highlight specularity without increasing the environment mapping specularity. That's the problem.
It could be an option though... in fact, yap, it's really easy. Ha! Remember that lookup table? You can code the desired intensity scaling on that lookup table - violá
You mean that look-up 'texture'? Yes, let's do more than that with it: Like, make it square, and throw some perlin noise in its alpha channel, for use as shininess detail texture.
And the screenshot... I don't know if it'll help you from what you say, but since it worked for me...
Nope; no luck; when I add the Spec button it modulates my specularity, but doesn't even touch my shininess. My shininess slider I put it at 100, btw; I should be able to see differences...

Here's my shininess texture:

Image

and this is what I get:

Image
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Haha, sorry Klauss, it was working; just my shades weren't right. With this shininess map:

Image

I get,

Image

BINGO!

But you see the problem, right? The moment I lowered the shininess, the whole ship glows brighter in specular. Blender should know better...

Anyways, Klauss, don't forget, this scaling of the specular intensity by the shininess should be ONLY for highlights; NOT for environment mapping...

Hey, btw, you could throw a fresnel table into that same texture.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Post by klauss »

Ya, I was going to tell you that you usually have to have much stronger contrast.
And about the ship glowing with specularity, that's why I made it modulate both on that screenshot.

And the "table" (it's a lookup table embedded in a texture - ok, lookup texture if you will) can't have the fresnel effect in it because that would make it 3D. Right now, it gets indexed by two independent values: RdotV and Shininess. You'd have to add NdotV for fresnel, and that's three coordinates - 3D. The thing gets hairy with 3D textures, so I prefer a computation for fresnel effects. That table only affects direct specularity, and not environment specularity, so it's exactly what you want.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Ah, good :D

I meant a separate table within the same texture; I didn't know these things formed a 3D continuum. What about my earlier suggestion of making the texture square and putting perlin noise in the alpha channel, for detail texture modulation of shininess? Not sure if you've seen my posts. I was saying that shininess modulation would make a perfect substitute for normal map detail texturing, and if it's blended additively, it would affect materials with less shininess more than it would affect materials with more shininess, which is what we'd want in every case, and could in fact apply it universally, rather than have to produce detail textures for every ship. Not that we couldn't reserve the option...
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Post by klauss »

I don't see why it would have to be with the lookup texture. Why not use the ordinary detail textures (when added)?
But ya... shininess can also be detailed, it was always my intention.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Well, my idea was to try and save a texture unit, but I guess you'd need separate interpolants anyways?
Okay, so make that a monochrome, 256 texture, then, that is applied universally to shininess modulation. I don't think we'll need detailed anything else, if we have that. But we'll see.

OT:
Klauss, wouldn't it be a good idea to have jpeg decompression and dds compression built into the engine? I mean, for optimizing game download size, jpeg is the best; but then we'd probably want to convert to .dds for the gpu's sake, no?
/OT
charlieg
Elite Mercenary
Elite Mercenary
Posts: 1329
Joined: Thu Mar 27, 2003 11:51 pm
Location: Manchester, UK
Contact:

Post by charlieg »

Sexy. 8)
Free Gamer - free software games compendium and commentary!
FreeGameDev forum - open source game development community
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Thanks! Long time no see; how's the going?

Here's an update, just details:

Image
snow_Cat
Confed Special Operative
Confed Special Operative
Posts: 349
Joined: Thu Jan 05, 2006 12:43 am
Location: /stray/
Contact:

Stunned silience

Post by snow_Cat »

^ -.- ^ *awed*
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Thanks, Snow Cat; but wait till you see it in-game... With Klauss' shaders I bet it will look better than in renders. ;-)
Kangaroo
Venturer
Venturer
Posts: 543
Joined: Sun Jan 22, 2006 9:55 am
Location: Baltic States
Contact:

Post by Kangaroo »

It's excelent! I bet it would look even better with a HDR backgroud in the render.
Btw, can you send me the model to my email? I want to see how's everything looking there (settings, UV layout, etc.)
There are no stupid people on Earth; they are only alternatively thinking.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Thank you. Here you go:

The blend file with the textures setup:
http://deeplayer.com/dan_w/WCUships/Dem ... ting.blend
The normal map:
http://deeplayer.com/dan_w/WCUships/Dem ... t_norm.png
Diffuse:
http://deeplayer.com/dan_w/WCUships/Dem ... use512.jpg
Specular:
http://deeplayer.com/dan_w/WCUships/Dem ... lar512.jpg
Lightmap:
http://deeplayer.com/dan_w/WCUships/Dem ... map512.jpg
Shininess map:
http://deeplayer.com/dan_w/WCUships/Dem ... ess512.jpg


Not finished yet, though; I still have to darken the rivets in diffuse, add front yellow lights and a bit of green glow to the cockpit, and a couple of greebles, plus a couple of rounds of color matching to the original, and baking PRT's. Should be done this weekend.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Post by klauss »

When you're done, try to send me the XCFs so that I can archive them.
I imagine they'll be rather heavy... hence "try".
BTW: I've been toying with a different way to compute PRTs, a much easier one, but one that does not compute radiosity. It's with Blender's AO baker: it can be set up to take the color from an environment texture, and that texture could have the required color coding. It would have a much better spatial resolution, but no radiosity. A tradeoff, I guess.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Hahaha, I was planning to do exactly that. Only thing I needed to find out what kind of format Blender's cube textures use. xNormal uses a format that the sides are arranged vertically:
-Z
+Z
-Y
+Y
-X
+X
But xNormal uses env maps for display, not for baking. I was trying to convince Santiago to allow baking with env maps, but he started asking too many good questions about PRT's, so I gave him your email. Hasn't he written to you?

Yes, I'll post the xcf's, but I have to warn you, there's many of them. I'm using like a tree or graph of xcf's: Bumpmap, Normal (because I have to blend the normalmapped bumpmap with xNormal's normal map), Ambient Occlusion (which includes baked occlusion and bump occlusion), I have a paint xcf, as well as diffuse, specular and lightmap xcf's.

Yep, the radiosity is a problem. I was trying to convince Santiago of that, too. If he implemented that in xNormal it would be so much easier.
xNormal can produce better quality bakings than Blender, already; I think with Blender the limit is 256 rays per point ("16 samples"). xNormal will do tens of thousands of rays, if you tell it to.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Update; very close to final now; just the round of color matching left.

Image

Just in case your eyes don't get it right away, the greenness on the cockpit is not diffuse surface color; it's added glow, to represent illumination from inside. Best thing would be to have a transparent cockpit and an interior; but any transparencies require switching shaders, and a change of shader takes as much time as displaying about 10k triangles; plus I'm too lazy to add an interior. So, next best thing to transparency is at least a hint of transucency.

The rivets...

I trust the rivets will look better in-game than they do in a render. Reason being, they are like chrome-plated, but that's hard to appreciate in a still image. They are bumpmapped, but this is also hard to apreciate statically.
Klauss' shaders for the coming VS engine apply texture offset, self-occlusion and self-shadowing to bumpmaps, which Blender doesn't.
Another thing that messes them rivets up is that my bumpmap ambient occlusion is a bit too strong. Each rivet has a bit of a halo of darkness around it, which doesn't necessarily look unnatural, but it looks a bit more like dirt than like shadowing.

I've also added a tiny bit of yellow light inside the spaces at the front of the arms.

Next is color matching; but I gotta catch some sleep first.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Done!

Image

Well, the perspective is not exactly the same; and the light positions are impossible to guess exactly; but in any case, enough to show the errors. Biggest one is the cockpit, which is not really an error; just laziness. It was so hard to get a bottom line that looked nice and round, that after I realized it was too big I said "hell with it".

So, I'm going to organize my files and make an upload for Klauss.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Here we go, 100 megs:
http://deeplayer.com/dan_w/WCUships/Demon/DEMON.7z
That should be everything that's current or relevant, including all the xcf's, and the coarse and fine meshes, both .blend and .obj files. The problem with .obj files is that they don't retain crease and hard edge attributes.
I haven't done PRT's yet; I'll see if I can find out how blender expects cube maps to be represented.

And a few more shots:

Image

Image

Image

Image

Image

Image

Image

Just to clarify, no fancy features set up in the renderer, like ambient occlusion or mirror effects. There's not even environment mapping here.
And all the textures used (diffuse, specular, light and shininess maps) are 512; only the normal map is 1024.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Post by klauss »

That's an amazing job Chuck.
Congratulations.
It's not only your first (real) texture job, it's pretty much one the best ones I've seen in a long time, along with the Hornet (and I might venture, even, it looks a bit cleaner than the hornet). Granted... it took you some time... but I can't say it wasn't worth it.

I'd like to hear what Brad has to say about it.
It's clear to me you proved your point about excessive mindless detailing: this clean look certainly looks wow without it.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Thanks, Klauss! :D
Yep, both Brad and Howard had nice things to say at wcplanet, yesterday.
I think I've proven another point, if I may say so. 512 is pretty damn enough resolution for a small ship. I didn't even use dual UV's and mirroring, here.

One argument I put to Brad at wcplanet that I never got around to post here is that if we avoid the mindset of looking at a ship in isolation, and adopt instead a mindset of thinking how a ship looks within the artistic ecology of a whole game environment, 512 looks better than 1024 on a small ship; because 1024 would make it look discordant, next to a larger ship or station. One needs to have some kind of overall game-plan, not just strive for the best looks possible for each individual unit.

And I believe also that part of a good WOW factor is for the amount of information in a texture to far exceed its representation capacity: The fact that the working textures and xcf's are 2048 kind of overwhelms the 512 reductions. I'm sure very few texels side by side have the exact same value. And our brains are used to seeing this, subliminally, in photography. It suggests to us that there's more there than meets the eye, which is true; though it's limited; but we don't know where the limit is so we assume none. If I were to leave the textures at 2048, the extra detail would be there for our eyes to see, but it would give away the limited extent of information present. Call it a "religious" belief, if you will; but I do strongly feel it's true.

I hope I haven't missed anything in that zip file; it's 100 megs, and compression doesn't do much for it, but my whole working folder is over 800 megs...

By the way, part of the reason it took so long is the nesting of the xcf's.
And since you may need to work with them to make color, specularity and shininess adjustments, I should tell you how this nesting works.
One thing that Gimp's xcf's lack (afaik) is a way to represent filtering operations, so they need to be done manually and documented separately.

So, let me describe it via an example:

Suppose you were adding a greeble somewhere. Let's assume it's a square of paint framed by a groove in the bump map. The steps to follow. to obtain a set of final textures, would be:

1) Draw the groove: Open great_bump3.xcf, there's a layer called grooves. Use the brush tool, fuzzy circle #9 for consistency with the other grooves, and draw the rectangle. There's also a layer called "paint" in there, that's NOT used by the xcf (always turned off) but is there for convenience of generating the paint layer for the paint xcf. So you would switch to the paint layer and paint your rectangle there. Now you can export that layer to a temporary PNG, then turn it back off, invisible; making all the other layers visible. Except the bottom layers, that is; I usually have one or two layers at the bottom of the xcf stacks that are there for tracing reference convenience only. I usually click the chain icon on them, which I'm not sure what it does, but helps me remember which layers should NOT be on when exporting. So, now, export the bump map by going to File -> Save a copy; --I give it the same name as the xcf, but change the extension to .png.

2) Open the new .png (first item in File -> Open Recent), and produce an updated bumpmap ambient occlusion, by going to Filters -> Line Detect -> Difference of Gaussians; with radiuses 5 and 2. File -> Save Copy, and give it a name of your choice .png. Don't close the window yet; instead, Ctrl-Z to get back to the bumpmap.

3) Make a normal map: Filters -> Map -> Normal, then Dewitt 5x5 I think gets the best results. File -> Save a Copy; name of choice. Close window without saving.

4a) Updating the paint mask. Open the paint.xcf and temporarily slap on the paint layer you exported from great_bump3.xcf. Transfer the new info to the paint layer of the paint.xcf, but adjusting colors as necessary. The paint layer of the bumpmap doesn't have color authority; it's just for boundary information and tracing convenience. Delete the temporary layer when done. Save a copy .png. Open the new png, select background by color, make black; invert selection, make white. Color threshold for selection has to be very precise not to grab things it shouldn't or break up the lines, yet not to fatten the lines too much. The magic number is 26. Save copy that as paint_mask.png for later use. Ctrl-Z back; don't close yet. Get back to where the background is selected.

4b) Updating paint: With the background selected, go to color curves, but switch from Value to Alpha curve, grab the upper right corner of the line and bring it down to zero. If there is no alpha curve, go first to Layer -> Transparency -> Add Alpha Channel. Save copy as paint_diffuse.png. Don't close yet. Open paint_specular.png as reference and manually change colors to match paint_specular.png, using select regions by color, and the paint bucket tool. Sorry; I don't know of any other way to do this. Save copy over the old paint_specular.png. Close without saving.

5) Update ambient occlusion: Open finemesh_ao_bumped.xcf. There's only two layers to it: finemesh_ao_padd, which is the manually fixed and background padded ambient occlusion output from Blender; and the bumpmap ambient occlusion layer. So just turn off visibility for the second, but highlight it in the layers dialog. Then go to File -> Open as Layer and give it the name of the bumpmap difference of gaussians we saved. Make sure the blending type and ratio are the same. Then delete the old layer. Save, then Save Copy as ambient_occlusion.png.

6) Generate diffuse and specular multiplier masks: File -> Open Recent... first item. (Open also the existing ambient_occlusion_diff_mult.png, for visual reference.) Go to Tools -> Color -> Curves. Make a square root curve with the value line; horizontal parabola. Should brighten the picture a lot. Now go to the red and blue component curves and tweak the lines a little bit. I do this to decouple quantization between color components. I just put a point at a quarter of the way, and another at 3/4 of the way; make the darker part of the red line a tiny bit darker, and the brighter part a tiny bit brighter. Same tweak in blue, but reversed. Make sure it looks similar to the old ambient_occlusion_diff_mult.png, and then save copy over it. Don't close the window. Just Ctrl-Z back. Do another color curves but this time a square function, upright parabola, and similar tweaks for red and blue but reversed. Try and get a match to the existing ambient_occlusion_spec_mult.png, and save a copy over it. Close without saving.

7) Diffuse: Open diffuse.xcf and update the ambient_occlusion_diff_mult.png and paint_diffuse.png layers, making sure you have same locations in the layer stack, same blending mode, and same blend ratio. There may be a grooves layer there too, can't remember, which may need updating too, by opening great_bump3.xcf, singling out the grooves, and saving copy. Yes, I seem to remember that was the case, and unfortunately there's one other step, changing the color of them to make the grooves a little brownish.
Then delete the old layers. When done with diffuse.xcf, save it, then save copy as diffuse.png. Close.

8 ) Lightmap: Open lightmap.xcf and update the ambient_occlusion.png and diffuse.png layers. Save, then Save Copy as lightmap.png.

9) Specular: Same thing, open specular.xcf, update the ambient_occlusion_spec_mult.png and paint_specular.png layers, save, save copy as specular.png.

10) Shininess: Just transfer the new square to it and color to taste. Save and save copy as shininess.png.

11) Normal map: Open the normal map we made from the bump map, then open as layer the paint mask. Select by color the background of the paint mask, turn visibility of the paint mask off; click on the normal map layer to make it the working layer, Filters -> Noise -> Scatter RGB, 3 for all components, uncorrelated, independent. Will add a bit of noise to the bumpmap, except the painted areas. Delete the paint mask layer. Select All. Filters -> Map -> Normal Map, but change function to Normalize Only.
Save copy some new name.png.
Open great_norm.xcf and update the bumpmap normals layer with the png above, same blend mode and ratio. Save. Save copy as great_norm.png. Close.
Now File -> Open Recent, first item, and again Filters -> Map -> Normalmap, with Normalize Only. Save.

12) Grab each of diffuse.png, specular.png, lightmap.png and shininess.png, and for each of them, Image -> Scale Image 512; then Filter -> Enhance -> Sharpen 33.

Done.
mkruer
Site Administrator
Site Administrator
Posts: 1089
Joined: Thu Jan 02, 2003 10:07 am
Contact:

Post by mkruer »

I don't say this much, but ... I am impressed. I think I can safely say this is the best model and texture job posted to WCU to date. What do you plan on doing next?

-Matt-
I know you believe you understand what you think I said.
But I am not sure you realize that what you heard is not what I meant.

Wing Commander Universe Forum | Wiki
Wing Commander: The Wasteland Incident
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

Thanks, MK. As for what to do next, that's a million dollar question. Got so many things I'd like to do...
* Finish the Bengal
* Texture the Hellcat
* Do another WC1 ship
* Work on DL3D
* Finish the Space Elevator Station
But what I should be doing first of all is a pre-WC1 carrier and Alexandria station, so that WC0 can get started.
And what I should really, really do first is write software to automate much of the texturing work... It's absolutely insane.
And what I should really, really, really be doing is work on my software project for my day job, before I end up getting fired and panhandling to pay for my internet connection...

EDIT:
Another thing we're going to need for WC0 is the Concordia class carrier (NOT to confuse with the Confederation class TCS Concordia). From what I remember, the Concordia we have is horrible. Anybody here has reference material on it? All I could find is this horrible image:

Image

They must have been really concerned about polygon count if they had to paint a tunnel on it...
mkruer
Site Administrator
Site Administrator
Posts: 1089
Joined: Thu Jan 02, 2003 10:07 am
Contact:

Post by mkruer »

If I may make a suggestion. The order I would like to see things accomplished in.
  1. Texture the Hellcat: I think that you could finish the hellcat inside of a week providing that you don’t have to do a lot of poly editing.
  2. Finish the Bengal: I didn’t care the way you were going with the design, I would rather have the romantic, yet inaccurate, version then one that is trying to be realistic. I think you will also save a lot of time because the Romantic version is simpler to develop. I also think that most people would enjoy romantic version, because its reminiscent of the original game without be cartoony.
  3. Finish the Space Elevator Station
  4. Work on DL3D
  5. Do another WC1 ship: This would be the last thing on my list, but if you decide to do it, use the models I made, it should greatly speed the development, the models should be accurate, but simplistic do to the poly restrictions at the time.

And Yes the poly counts were very low, and the image front end did help the speed, you have to remember that this was jast as 3d was starting I don’t even think poly clipping was invented yet.
I know you believe you understand what you think I said.
But I am not sure you realize that what you heard is not what I meant.

Wing Commander Universe Forum | Wiki
Wing Commander: The Wasteland Incident
micheal_andreas_stahl
Elite Hunter
Elite Hunter
Posts: 1030
Joined: Mon Apr 10, 2006 10:02 am
Location: Gemini, Troy, Helen

Post by micheal_andreas_stahl »

chuck_starchaser wrote:And what I should really, really, really be doing is work on my software project for my day job, before I end up getting fired and panhandling to pay for my internet connection...
Is DL3D your job??? Well then, that should be number one.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Post by chuck_starchaser »

mkruer wrote:If I may make a suggestion. The order I would like to see things accomplished in.
  1. Texture the Hellcat: I think that you could finish the hellcat inside of a week providing that you don’t have to do a lot of poly editing.

It would take far longer than a week, even without moving any vertices. First I would have to organize smoothing groups and creases, then UV unwrap it, then create a high poly version to diff the normal map against, then do the normal and light map bakings, before I can even start working on the texture. And the big problem is that we don't need the hellcat for anything in any urgency that I know of.
[*]Finish the Bengal: I didn’t care the way you were going with the design, I would rather have the romantic, yet inaccurate, version then one that is trying to be realistic. I think you will also save a lot of time because the Romantic version is simpler to develop. I also think that most people would enjoy romantic version, because its reminiscent of the original game without be cartoony.
Not me. I like being able to tell the relative sizes of things, in a flying game. Thing I most hated about the current wcu version is the asteroids, which one has no idea if they are a few meters in size, or thousands of kilometers. The point of having small details in the carrier is simply that: anchor spots for the eyes, that one can assume the smallest features there are roughly the size of the smallest features in the ship you are flying. Wish I could do the same thing texture-wise...
[*]Finish the Space Elevator Station
I'm very tempted to; tho that's yet another thing not urgently needed; unless we're going to use the SES in WC0, for McAuliffe. There's a reference to a carrier being lost in the battle of McAuliff by a falling "skyhook" (I think it was from Action Stations).
[*]Work on DL3D
Yep, absolutely.
[*]Do another WC1 ship: This would be the last thing on my list, but if you decide to do it, use the models I made, it should greatly speed the development, the models should be accurate, but simplistic do to the poly restrictions at the time.[/list]
What about WC0? Actually, we haven't heard from starlord for a few days; hope he's not MIA. But the idea is good enough I might take it and run with it, if he's gone. But I'm with you, anyhow; most if not all WC1 ships might appear during WC0, one way or another.

EDIT:
Never mind; I just got two emails from starlord, he's doing good research work. Custer's Carnival is indeed in 2649, as the KS manual and Claw Marks say; but it wasn't the Tiger's Claw maiden voyage; her maiden voyage and first kitty encounter happens in 2644. Yet another mistake in Voices of War and the CIC's Encyclopedia.
And Yes the poly counts were very low, and the image front end did help the speed, you have to remember that this was jast as 3d was starting I don’t even think poly clipping was invented yet.
True, there used to be many issues with 3D shapes having to be "convex", back in those days, I remember.

@Klauss:
You think you could get the Demon into the old engine first, as a patch? Just so that Priv Remake players can download it; and then we could see how it looks with the new shaders. Or else I guess I could do it.
Post Reply