[ReaperAA]

Phredreeke, on 04 February 2020 - 12:43 PM, said:

I suppose he's talking about some changes done to the way lighting is done in software renderer

Oh okay.

But still, doom's software like looks can be achieved by using the "palette" tonemap + "vanilla" sector light mode while using hardware renderer

[Jimmy]

Photonic, on 04 February 2020 - 12:18 PM, said:

I love how easily he dropped classic. Dead weight. Interested to see how it goes once polymost is out too, a true polygon based renderer will probably cause compatibility issues, but should handle the majority of the vanilla games ok, they didn't do anything too crazy like what I've seen with TROR exploits in polymer only maps.

>Duke Nukem 3D has some non-euclidean 4 dimensional environments
>true polygon renderer

oh boy


As an addendum to oasiz' great posts, I think people vastly underestimate how much of an undertaking BUILD is compared to id Tech 1. Being upset about minor bugs is one thing, but expecting something so massive to be done quickly is absurd. I was never convinced that Polymer would be completed anyway. (Nothing against the team, its just a massive undertaking.) The expansion of Polymost has been delightful.

I say, KEEP CLASSIC GREAT! AND LETS MAKE POLYMOST GREAT AGAIN!

[Kyanos]

I agree with you entirely, yet the world apparently has this "void" where polymer had once shed some light. I wonder at what GZ see's in those shadows.

[Kyanos]

Your face when he updates the model format and Tea Monster ports years of work over.

jk. I know your opinion on the HRP

[Blzut3]

Radar 100 Watts, on 04 February 2020 - 12:51 PM, said:

Source

I think you may have extrapolated what was said too far. There is still id Software and Randi code in GZDoom's software renderer. There has been a lot of refactoring but fundamentally the way the "Doom software renderer" renderer works is still the same as the original.

It is worth noting though that GZDoom has a software renderer because GZDoom has the man power (i.e. dpJudas) to maintain it. If this wasn't the case, I have no doubts the software renderer would have been thrown out shortly after Randi quit. I don't think Graf would necessarily have an issue with having the software renderer return in Raze, but it would require someone willing to clean up and ultimately maintain it long term.

[Sangman]

I've played around with it a bit. While I hope it's eventually succesful in its endeavor, currently it doesn't match the quality of either EDuke32 or Rednukem for me - despite what kind of inner architecture cleanup may have been done.
All the same I'll be sure to make a bug thread on ZDoom forums from time to time

[Jimmy]

Photonic, on 04 February 2020 - 03:03 PM, said:

I agree with you entirely, yet the world apparently has this "void" where polymer had once shed some light. I wonder at what GZ see's in those shadows.

I wouldn't really say there's a void. Modding/mapping has massively slowed down in the past ten years and people who use full Polymer features to mod are ten or less very loud and vocal people.

Photonic, on 04 February 2020 - 03:11 PM, said:

Your face when he updates the model format and Tea Monster ports years of work over.

jk. I know your opinion on the HRP

Tea Monster does great work. My problem with the HRP has always been that it's been poorly maintained, has no art director resulting in a mish mash appearance, and ultimately it only aimed to be an asset replacement. In order to truly be a great mod, it needs programming to take advantage of everything 3D models could do. I'm not opposed to an HRP in theory, but only in it's historical practice. I beg someone to prove me wrong. I think Tea Monster and I are ultimately on the same page, but we got there from different ends of the same book.

[Zaxx]

Blood seems to run okay but with a bunch of bugs, performance is of course way lower than what you get in NBlood though:

Duke is playable too, Ion Fury is not, didn't try Redneck.

[Kyanos]

Jimmy, on 04 February 2020 - 04:01 PM, said:

...people who use full Polymer features to mod are ten or less very loud and vocal people...

Perhaps because it's broken consistently and outdated.

[Striker]

Photonic, on 04 February 2020 - 06:24 PM, said:

Perhaps because it's broken consistently and outdated.

Unless someone plans on finding us a programmer experienced in modern rendering who's willing to put up with Build engine's quirks, who's also a big enough fan of Duke3D/Build Engine games to give a shit, then nothing is going to be done about that.

To be honest, at this rate we're better off improving Polymost. Down the road, if nobody beats me to it, I'd like to take a crack at implementing basic point lights (no bumpmapping/spec, etc) in Polymost, so that projectile dynamic lights and colored point lights in WT maps (both official and custom) would work, but that's really low on my priority list right now.

[Jimmy]

Polymer is broken and consistently outdated because no one wants to do it, and honestly, what is even the point? The features a few people clamor for could be done MUCH better in an engine like Unreal or something anyway. Beefing up what Duke Nukem 3D is good at is the right direction to go for right now, because stability, compatibility, and expansion of capability would benefit every mapper and modder currently active or thinking about becoming active again. I've personally felt struck to restart/replan some old projects because EDuke32 is finally cleaning things up that have been lackluster for a long time. The new era is here, but it doesn't have to have a dynamically lit bump mapped plastic sheen. I think the team is going the right direction now that Ion Fury is out of the way.

[Radar]

Blzut3, on 04 February 2020 - 03:21 PM, said:

I think you may have extrapolated what was said too far. There is still id Software and Randi code in GZDoom's software renderer. There has been a lot of refactoring but fundamentally the way the "Doom software renderer" renderer works is still the same as the original.

It is worth noting though that GZDoom has a software renderer because GZDoom has the man power (i.e. dpJudas) to maintain it. If this wasn't the case, I have no doubts the software renderer would have been thrown out shortly after Randi quit. I don't think Graf would necessarily have an issue with having the software renderer return in Raze, but it would require someone willing to clean up and ultimately maintain it long term.

In Graf's own words, "it had eventually became unusable and had to be replaced in its entirety". I'm not saying you're wrong, but if Graf doesn't mean what he says then I guess Graf doesn't mean he says. Not an issue with my ability to infer.

[Mark]

Speaking only for myself, my Polymer want list is do something, anything, to speed up rendering in large areas and fix the long standing transparency bug. In fact, I'd be happy with only the transparency bug fixed. Through creative mapping and con control of lighting its been possible for me run all my Polymer projects on a now 10 year old computer with a 4-5 year old video card. Yes, normal,spec,height maps need improvement but they are not game breakers for me.

[Blzut3]

Radar 100 Watts, on 04 February 2020 - 07:04 PM, said:

In Graf's own words, "it had eventually became unusable and had to be replaced in its entirety". I'm not saying you're wrong, but if Graf doesn't mean what he says then I guess Graf doesn't mean he says. Not an issue with my ability to infer.

Honestly it feels like something got dropped in his sentence since knowing what I know everything makes sense individually but the whole sentence doesn't parse right.

Anyway not that you asked for proof but in case anyone wants it.
RenderOpaquePass::RenderSubsector -> R_Subsector
RenderOpaquePass::AddSprites -> R_AddSprites

At this point it's hard to find a good 1:1 example off the top of my head at least, but if you look at the much simpler Chocolate Doom code and look at the function calls it makes you should find identical calls in roughly the same place in the GZDoom code. And of course most of the comments made it over in tact. There's just a ton more code around it for all the features ZDoom got over the years. Stuff moved around a lot, and things got more structured to support multithreading, but in the end it's doing the same basic thing.

[Perro Seco]

NightFright, on 02 February 2020 - 11:37 AM, said:

Stuff like Tek War can be ignored, Capstone games were nothing more like utter garbage, anyway. Civvie's Tek War video is all you need to watch. No coherent game design to be found there at all. But it was enough for a few good jokes about the most ridiculous aspects.

Utter garbage are all those reviewers that try to be funny instead of descriptive. If you want to know what TekWar is, here's the most fair review I read about it.

So I don't need to listen to a youtuber barking me what should I like and what not, because I played TekWar when I was a kid and it impressed me even more than Duke 3D. Just look at the subway, the hospital, the park or the midtown levels; there's a lot of effort there, and they're by far the most realistic city levels I ever seen before Duke came out.

Looking at the game's promo screenshots and videos you can see how ambitious the project was. Capstone may have failed with TekWar, but at least they tried to do something different than all those unoriginal Doom clones. I think they deserve some credit for that.

Sorry for the offtopic. TekWar is one of my favourite games, so this had to be said.

[Tea Monster]

Jimmy, on 04 February 2020 - 04:01 PM, said:

I wouldn't really say there's a void. Modding/mapping has massively slowed down in the past ten years and people who use full Polymer features to mod are ten or less very loud and vocal people.

There used to be a thriving modelling community when the HRP was at it's height. That's all gone now. I'm pretty sure it is down to less than two people who use Polymer now. There was a mod that tried to recreate Slick Willy from the early version of DNF. That, AFAIK, has been shelved due to Polymer's multitude of problems. I'll reiterate that I think Polymer had a significant role in driving away the Duke modelling community. There's lots of other stuff going on, but Polymer had a big part in it.

Jimmy, on 04 February 2020 - 04:01 PM, said:

Tea Monster does great work.

Cheers!

Jimmy, on 04 February 2020 - 04:01 PM, said:

My problem with the HRP has always been that it's been poorly maintained, has no art director resulting in a mish mash appearance, and ultimately it only aimed to be an asset replacement. In order to truly be a great mod, it needs programming to take advantage of everything 3D models could do. I'm not opposed to an HRP in theory, but only in it's historical practice. I beg someone to prove me wrong. I think Tea Monster and I are ultimately on the same page, but we got there from different ends of the same book.

Yep. When Polymer came along, a lot more could have been done by the maintainers to integrate the older, Polymost stuff with the newer renderer. None of the Polymer models even have specular definitions. Even though no specular maps were ever made for them, they should at least have had a baseline amount of shine defined in the def file so that they didn't appear fully glossy. None of this was done, in fact, Nightfright referred to this as "Nerd stuff". As a result, all the old models built for Polymost look like Christmas tree ornaments under polymer. I personally think that there should have been two HRPs. One for Polymer specifically for the newer models created with next-gen workflow and an entirely separate one for Polymost with all the pre-Polymer stuff in it.

Your suggestion for the role of the HRP is an interesting one. Years ago I was considering faking particles using animated MD3 models with alpha planes for the HRP. You could recreate cool fire, lightning and other effects this way. It all comes down to how much do you want to do with the HRP. Do you want a straight asset replacement, or do you want to do more? If you do want to do more, where do you stop? In HHR we tried replacing various sector creations with 3d models. It makes it look more modern, but I don't think that purists will accept that for the HRP. Once you go down that route, at some point you may realise that you want to make something more than the engine can (or even should) support. Then you are looking at using another engine for your mod, which brings me to your later post.

You are right in that you have to have solid art direction to such a project, preferably by someone who knows something about how this shit works.

[Player Lin]

Perro Seco, on 05 February 2020 - 02:43 AM, said:

Utter garbage are all those reviewers that try to be funny instead of descriptive. If you want to know what TekWar is, here's the most fair review I read about it.

So I don't need to listen to a youtuber barking me what should I like and what not, because I played TekWar when I was a kid and it impressed me even more than Duke 3D. Just look at the subway, the hospital, the park or the midtown levels; there's a lot of effort there, and they're by far the most realistic city levels I ever seen before Duke came out.

Looking at the game's promo screenshots and videos you can see how ambitious the project was. Capstone may have failed with TekWar, but at least they tried to do something different than all those unoriginal Doom clones. I think they deserve some credit for that.

Sorry for the offtopic. TekWar is one of my favourite games, so this had to be said.

I never played this game. And looks like a good idea for a FPS game but ruined by bad executions with many flaws...too bad.

Capstone games are still in copyright hell, whatever they're good game or not, so I hope at least the hell should be fixed first.

[DisgustedGrunt]

Player Lin, on 05 February 2020 - 03:35 AM, said:

Capstone games are still in copyright hell, whatever they're good game or not, so I hope at least the hell should be fixed first.

I doubt many people would want to open that can of worms.

[NY00123]

Can probably just wish good luck with the act of experimenting with the codebases of the games and related libraries.

I maybe weren't contributing to Build Engine ports as much as others (like TerminX or Hendricks266) had been, and also didn't get to release that much user-made contents.
But, I've seen enough of the code, especially Duke3D and the differing libraries, that I can say that it is often very far from easy to get something decent and new done, especially if you want it to be done well. The differing game codes being quite distinct (even if eventually derived from the same Ken-Build test game) obviously add to the above.

[Jimmy]

Perro Seco, on 05 February 2020 - 02:43 AM, said:

Utter garbage are all those reviewers that try to be funny instead of descriptive. If you want to know what TekWar is, here's the most fair review I read about it.

This is literally the end of the first paragraph; "This is a game that kills empires." Civvie's review is spot on.

Perro Seco, on 05 February 2020 - 02:43 AM, said:

So I don't need to listen to a youtuber barking me what should I like and what not, because I played TekWar when I was a kid and it impressed me even more than Duke 3D. Just look at the subway, the hospital, the park or the midtown levels; there's a lot of effort there, and they're by far the most realistic city levels I ever seen before Duke came out.

Yeah and I'm sure people were impressed with Suburbs too. Times change. But TekWar was even pretty bad back then.

Perro Seco, on 05 February 2020 - 02:43 AM, said:

Looking at the game's promo screenshots and videos you can see how ambitious the project was. Capstone may have failed with TekWar, but at least they tried to do something different than all those unoriginal Doom clones. I think they deserve some credit for that.

Doing something different doesn't create an increase in quality.

Tea Monster, on 05 February 2020 - 02:50 AM, said:

There used to be a thriving modelling community when the HRP was at it's height. That's all gone now. ..... I'll reiterate that I think Polymer had a significant role in driving away the Duke modelling community. There's lots of other stuff going on, but Polymer had a big part in it.

Polymer has it's own issues that it's put upon the community, but I think people just lost interest over time. I was there for the whole development cycle of the HRP. People came and went all the time, eventually everyone just lost interest and dropped off. There was a huge slump in development even before Polymer's first release. I think it slumped because it wasn't a good project, and the more that was added to it, the more obvious this became.

Tea Monster, on 05 February 2020 - 02:50 AM, said:

Yep. When Polymer came along, a lot more could have been done by the maintainers to integrate the older, Polymost stuff with the newer renderer. None of the Polymer models even have specular definitions. Even though no specular maps were ever made for them, they should at least have had a baseline amount of shine defined in the def file so that they didn't appear fully glossy. None of this was done, in fact, Nightfright referred to this as "Nerd stuff". As a result, all the old models built for Polymost look like Christmas tree ornaments under polymer. I personally think that there should have been two HRPs. One for Polymer specifically for the newer models created with next-gen workflow and an entirely separate one for Polymost with all the pre-Polymer stuff in it.

Nothing stopped anyone from doing any of this. Yet no one did it. This is extremely telling.

Tea Monster, on 05 February 2020 - 02:50 AM, said:

Your suggestion for the role of the HRP is an interesting one. Years ago I was considering faking particles using animated MD3 models with alpha planes for the HRP. You could recreate cool fire, lightning and other effects this way. It all comes down to how much do you want to do with the HRP. Do you want a straight asset replacement, or do you want to do more? If you do want to do more, where do you stop? In HHR we tried replacing various sector creations with 3d models. It makes it look more modern, but I don't think that purists will accept that for the HRP. Once you go down that route, at some point you may realise that you want to make something more than the engine can (or even should) support. Then you are looking at using another engine for your mod, which brings me to your later post.

Purists won't play with 3D models anyway.

[Kyanos]

How are the voxels rendered? vox2poly still??
What model format is planned? md3 still??
What lighting effects are planned if any? Dynamic??

[Perro Seco]

Jimmy, on 05 February 2020 - 12:46 PM, said:

TekWar was even pretty bad back then.

Jimmy, on 05 February 2020 - 12:46 PM, said:

Doing something different doesn't create an increase in quality.

You speak as if I had said TekWar is a good game. What I wanted to say: not a good game, not utter garbage either.

[Phredreeke]

3.6, not great, not terrible

[Morpheus Kitami]

The problem with TekWar is that the levels for it were for the kind of game it just wasn't. Trying to have police and civilians reacting to you having your gun out means nothing when there's a guy on the back of a bus trying to kill you. The level design would have been decent if it were 2001 and it was ripping off Deus Ex instead of trying to be an open-ended FPS in 1995. Its possible that the game would've been okay if it had normal level design. I feel like its hard to tell if there's anything interesting going on in the game when the game is basically:
Shatner tells you something > Wander around shooting things for 20 minutes > Shatner yells at you for defending yourself > Repeat 6 times > Die in eye-searing pain as you see the VR levels

[Sanek]

I don't understand.
Is this mostly GZDoom-powered port with some of the EDuke32's code thrown in that allows you to insert .ART, .MAP and .CON formats into Doom but you have to use Doom level edtor in order to make user maps? And there's multiplayer too?

Anyway, I think it's a very good news for Duke's and Build games' gaming community's, becasue it'll bring...more players? On the other side, it's not a good news for modding community. I also agree that it kinda sucks that Doom guys beat us in this area.


@Mark I wouldn't stick to Mapster if I were you. At least if you want more people to know about your work. It took me almost 17 years to become somewhat decent mapper, and by the time I started to make my best maps ever, nobody cares anymore.

[Phredreeke]

It's EDuke32 (and related ports) with a new backend

[Micky C]

Sanek, on 06 February 2020 - 03:34 AM, said:

Anyway, I think it's a very good news for Duke's and Build games' gaming community's, becasue it'll bring...more players? On the other side, it's not a good news for modding community. I also agree that it kinda sucks that Doom guys beat us in this area.

The port's not even properly out yet. If you're implying that Raze will have better modding support than eduke32 before it even gets the basics right, then you're a bit premature.

[Zaxx]

Interestingly though you can load custom content with Raze just like you would with GZDoom: with the "-file" command. For example you want to play Blood with the voxel pack? You just "-file voxels.zip" and it works. That could be a better solution than EDuke32's unrealiable autoload feature.

[Phredreeke]

You can do the same with eduke32, just write -g voxels.zip
In fact that's what I do with the upscale packs.

The big problem with EDuke32's autoload feature is different mods interfering with eachother, something not addressed by Raze.

[Zaxx]

Yeah, that's what I meant when I said that it "could be" better than the autoload: seems like this functionality is planned to work the same way as it does in GZDoom.