[engineer]

I'm having an issue- I can't get the Parallax maps to work.

I got them to work once. ONCE out of over 30 tries, but even after overriding and tweaking the scale and bias settings I can't reproduce the effect I was able to pull off sometime last night.

[Jokke_r]

make sure you set the r_pr_parallaxoverride back to 0

[engineer]

I do- But I can't replicate the results of this one instance! That's my current issue.

[Jokke_r]

Are you sure the heightmap is stored correctly as alpha channel for then normal map? Are you sure the game is loading the texture, have you tried deleting the texture cache?

[Roma Loom]

So I'm still havign trouble with gloss with texture #4304 on two machines here at the office. Can anyone test it at their end?

The deflines for #4304:

texture 4304 {
   pal 0 { file "highres/textures/4304.png" specpower 25 specfactor 2 }
   glow { file "highres/textures/4304_glow.png" }
   detail { file "highres/detail/metal.png" xscale 0.15 yscale 0.15 }
   normal { file "highres/textures/4304_normal.png" parallaxbias 0 parallaxscale 0.05 }
   specular { file "highres/textures/4304_gloss.png" }
}

The link for texture #4304 package

The results I'm getting concerning specular (black spots and yellow outline on the letters):


The result I want is the pic Plagman posted above.

[Parkar]

Roma Loom, on May 19 2009, 02:52 PM, said:

So I'm still havign trouble with gloss with texture #4304 on two machines here at the office. Can anyone test it at their end?

The deflines for #4304:

texture 4304 {
   pal 0 { file "highres/textures/4304.png" specpower 25 specfactor 2 }
   glow { file "highres/textures/4304_glow.png" }
   detail { file "highres/detail/metal.png" xscale 0.15 yscale 0.15 }
   normal { file "highres/textures/4304_normal.png" parallaxbias 0 parallaxscale 0.05 }
   specular { file "highres/textures/4304_gloss.png" }
}

The link for texture #4304 package

The results I'm getting concerning specular (black spots and yellow outline on the letters):


The result I want is the pic Plagman posted above.

your specpower seems way to big. This should give you a really big specular with a rather sharp edge that just looks very unnatural, like a big round spot. Try putting it at something like 0.5. I have yet to use anything over 3 I think and thats on stuff like carpet etc that shouldn't have much of a highlight at all.

[Roma Loom]

Looks like I've misread how glossmaps work in polymer. Plagman has already explained it to me. Everything is fine when I use simple RGB greyscaled gloss maps. Problem solved. Thanx.

[Plagman]

The problem here is that he put his specular map in the alpha channel. 0 alpha means a specular exponent of 0, which means an infinitely wide specular spot. If you're trying to achieve no specular highlight, it should be 0 in the RGB channels instead. Basically if you want specular maps to act like gloss maps in most games (like Doom 3) just dump the grayscale of the gloss map in all three RGB and don't put an alpha channel.

[Parkar]

Plagman, on May 19 2009, 03:41 PM, said:

The problem here is that he put his specular map in the alpha channel. 0 alpha means a specular exponent of 0, which means an infinitely wide specular spot. If you're trying to achieve no specular highlight, it should be 0 in the RGB channels instead. Basically if you want specular maps to act like gloss maps in most games (like Doom 3) just dump the grayscale of the gloss map in all three RGB and don't put an alpha channel.

Wouldn't it be an infinitely small specular with alpha (specpower) 0?

In any case only make changes to the alpha if you want varying size of the specular, the RGB controls the strength per colour channel.

[Plagman]

No, an exponent of zero always results in 1, so specpower 0 means full specular highlight no matter what the angle is. A specular exponent of 1 makes a very diffuse and wide specular reflection as the result is directly proportional to the dot product of the perfect reflection vector and the eye vector. As the specular exponent tends towards infinity, the specular reflection looks more and more like a point.

[Parkar]

Don't listen to me, not sure how I was thinking about the specular power. Infact I have several textures defined with a power that high.

[Tea Monster]

At this point, I'd like to second a call for some kind of list of what is done and what isn't.

[Danukem]

Tea Monster, on May 19 2009, 10:36 AM, said:

At this point, I'd like to second a call for some kind of list of what is done and what isn't.

Or more precisely, a list of what is being worked on, since I don't think anything is final yet.

[Plagman]

MusicallyInspired: I don't remember if I asked; does disabling shadows help with your sprite issue? I.e. set r_pr_shadows 0 and restart the game.

[Plagman]

You can see what Parkar already did for Polymer by getting this file. It's a HRP package with normal and specular maps defined for certain textures (most of them in E1L1, I think), some tiles replaced by diffuse maps and lighting for E1L1. Keep in mind running E1L1 will be slow as hell using these maphacks as there's plenty of them and I haven't written the code to avoid recalculating all the lights with each frame yet.

[Richard Shead]

Plagman, on May 19 2009, 12:10 PM, said:

You can see what Parkar already did for Polymer by getting this file. It's a HRP package with normal and specular maps defined for certain textures (most of them in E1L1, I think), some tiles replaced by diffuse maps and lighting for E1L1. Keep in mind running E1L1 will be slow as hell using these maphacks as there's plenty of them and I haven't written the code to avoid recalculating all the lights with each frame yet.

Awwwwwww, thanks Plagman!!!! Gonna try this out right now.....

Keep up the amazing work, everybody.......I can't wait to see where all this goes next!


Is there a limit how awesome we can get Duke 3D to look? I think not.....

[Hellbound]

Thanks! That's what I was looking for. Downloading.

Anyway, I think we need some kind of repository to store all normals/specs anyone did to avoid doubles.

[supergoofy]

Ok I want to write one thing: The new polymer is amazing

I compiled the source from the svn repository with MinGW 4.3.3 and MSYS 1.0.11.

I downloaded the polymer_hrp.zip, unpacked it and repacked it. why Parkar included the files in polymer_hrp folder inside the polymer_hrp.zip ? The HRP is not loaded that way, polymer_hrp folder is not needed. Anyway I repacked the hrp, thus it worked.

Great work guys (TX and Plagman and Parkar).


If you want I can upload a full pack of the compiling environment, if someone wants to compile the source. Is that ok?
(edit: check here http://forums.duke4....?showtopic=980)

I cannot upload the compiled binaries (eduke32.exe and mapster.exe) or the source. Else TX and Plagman will kill me.

[Benichon]

Thanks for the file, i saw how manipulate Maspter, defs and con and maybe it will help me to see how put dynamic lights etc...

[MusicallyInspired]

Plagman, on May 19 2009, 01:58 PM, said:

MusicallyInspired: I don't remember if I asked; does disabling shadows help with your sprite issue? I.e. set r_pr_shadows 0 and restart the game.

Yes, you mentioned that before. No it does not help, unfortunately.

A little update. I didn't try any older Catalyst drivers yet, but I did install the Omega ATI drivers and it automatically removes sprites from view so it runs smoothly...but still no sprites (but the model replacements are there). But also the lighting is all messed up as now you can't see any lights unless you get within a few feet of them and when you do see it they're way too bright. I'll try older Catalyst drivers later on...

[TerminX]

God, all of these ATI issues irritate the shit out of me. Why can't they come up with a driver that just works? It's not like Plagman's code is bad or anything; on the contrary, considering his job and employer it's probably some of the cleanest rendering code around. Didn't ATI already rewrite their OpenGL driver on account of it sucking once before? I can't even imagine what this thing would do on their previous GL driver...

Oh, btw, we're pretty close to having something to release. Plagman actually wanted to do it today, but I'm an ass and refuse to let a public release happen until the remaining performance issues are fixed. I don't want the first reaction from the actual public to be "damn, this thing runs like ass."

[MusicallyInspired]

Good to hear!

[Thndr]

Would it be ironic to say that I'd rather wait for a quality release so it doesn't run like shit VS something that runs like shit but grants immediate satisfaction?

[Hendricks266]

Help me..?

[Jason S.]

Hi! imma noob!

Any who, I was wondering if there is now a problem with my 6200LE. The normal maps do not show, but I notice that the new defuses do, however. Is there anything wrong with my card (about to upgrade to a 7300GT BTW), or is there a setting in a def/cfg I need to tweek?

EDIT: It's mostly to do with my eduke32 version isn't it. It's of course not 2.0, however, I didn't see a dld link for this ver on the eduke32 page.

Most likely this is for people that were given 2.0 and it's not out yet?

[eddym4814]

Jason S., on May 19 2009, 11:03 PM, said:

Hi! imma noob!

Any who, I was wondering if there is now a problem with my 6200LE. The normal maps do not show, but I notice that the new defuses do, however. Is there anything wrong with my card (about to upgrade to a 7300GT BTW), or is there a setting in a def/cfg I need to tweek?

EDIT: It's mostly to do with my eduke32 version isn't it. It's of course not 2.0, however, I didn't see a dld link for this ver on the eduke32 page.

Most likely this is for people that were given 2.0 and it's not out yet?

It has not been released yet.

"TX-Oh, btw, we're pretty close to having something to release. Plagman actually wanted to do it today, but I'm an ass and refuse to let a public release happen until the remaining performance issues are fixed. I don't want the first reaction from the actual public to be "damn, this thing runs like ass."

That quote is from post #310 right above.

[Jason S.]

eddym4814, on May 19 2009, 09:24 PM, said:

It has not been released yet.

"TX-Oh, btw, we're pretty close to having something to release. Plagman actually wanted to do it today, but I'm an ass and refuse to let a public release happen until the remaining performance issues are fixed. I don't want the first reaction from the actual public to be "damn, this thing runs like ass."

That quote is from post #310 right above.

Yeah, I saw it later as I thought it was some sort of beta thing when I dlded. It's why I quickly edited my post.

[Danukem]

Plagman, on May 19 2009, 12:10 PM, said:

Keep in mind running E1L1 will be slow as hell using these maphacks as there's plenty of them and I haven't written the code to avoid recalculating all the lights with each frame yet.

Yeah, I stopped working with polymer a while ago because it is just too frustrating to have low fps when the spotlights are casting shadows and I can't afford a better rig right now.

[Parkar]

Hellbound, on May 20 2009, 12:14 AM, said:

Thanks! That's what I was looking for. Downloading.

Anyway, I think we need some kind of repository to store all normals/specs anyone did to avoid doubles.

We have already moved the private HRP svn repository to a public server so that is coming. This will be used both for Polymer and none Polymer HRP.

It will be publicly available for people to checkout. It's also browsable with any web browser. We will give give the regular artists write access to it so they can update the contents directly.

There are a few things structurally I want to sort out before I let anyone else mess with it. I am also going to write a small guide to how I want it to be used and how commits should be commented and such before I opne it up as well. I think we will ramp the number of users up slowly so we can see how it works out to. Don't want a bunch of people learning how svn works on the HRP repo at the same time .

I am kinda busy at the moment though so it may take a week or two until I get the time to get it ready.

@Plagman, feel free to post the url to the repository.

supergoofy, on May 20 2009, 12:25 AM, said:

I downloaded the polymer_hrp.zip, unpacked it and repacked it. why Parkar included the files in polymer_hrp folder inside the polymer_hrp.zip ? The HRP is not loaded that way, polymer_hrp folder is not needed. Anyway I repacked the hrp, thus it worked.

That was not intended to be a drop in zip file, just something Plagman created to make the svn contents available without going through svn.

[Roma Loom]

He has zipped it for me because I'm a loser who has https disabled