motorsep

There ya go: http://mrelusive.com/publications/papers/The-DOOM-III-Network-Architecture.pdf

Does it mean that AMD 7850 just a crappy GPU and R9 line is better ?

Someone did read the whole thread?

I didn't. I said "whole level of insanity" not because of their views, but because of the fact that if I pay for it, it's awesome and accepted, but if TDM team needs to open their wallets, or even run crowdfunding - it's a no-no. Sounds quite hypocritical to me. If no money involved period, then it wouldn't be ok for me to pay for coding work, and TDM team accepting it. Or even pulling fixes/features for a repo where commercial work was contributed.

@Bikerdude: Perhaps you don't understand how GPL license works. Anything added to GPL code becomes GPL. A programmer can not apply any other license to the code added to GPL codebase. He also has no say-so how the code is used once it's in the codebase. It's clean slate, especially with agreement signed prior to executing work. There can be issues with art and assets, but not with GPL code. Anyway, as I already said, point taken. Good luck.

I am sorry, I don't. And chances are no one will, outside TDM tight circle and similar projects who would accept donations, no matter how it came to be, but would refuse to open up their own wallets. Here is basic example - if someone would port TDM to BFG engine, where all you need to do is drop new .exe and TDM would run without any content modifications, you would without a doubt accept that gift, even if it cost money to make such port. Except that if someone would approach you with such offer, you will say N-O, just like you did right now for help with DR. I will never understand that, because I don't operate on the same principles. If I participate in a free project I absolutely love to work on because it's a hobby for me, I will not be going against paid help and won't be questioning why I am not being paid also. If I wanted to be paid while participating in a free project, that would be my initial condition.

Why would I do that ? So you aren't willing to donate, but you are ok with someone else donating and using that work? Now that is whole new level of insanity o.O

You logic guys is skewed. The guy has a skill and experience with Radiant / idTech 4. He has a day job, and no interest burn his evenings for free. He can help, but would like to be compensated. If you guys live in a different world, by all means, keep living in it. There aren't whole a lot of people out there nowadays with a lot of free time and solid skill DR development requires. So if you reject help solely on a basis of .. I don't even have words how to describe it ("OMG, someone will get paid working on DR?! I want to be compensated for my past work on TDM?!" f@$king kindergarten) .. then it fine with me. No need to further comment in this thread. Personally I think I will fork DR and have _paid_ professional to fix what needs to be fixed (if it will be a reasonable fee).

You voluntarily chose to spend time, for free, working on TDM as a hobby. You don't get paid. I wish TDM team luck finding a programmer (only a volunteer, non the less!) as talented as greebo to dig DR and maintain it at the same level it was before.

I am failing to understand your logic. If I work on a project, at my own free will, free of charge and then one day I pretty much quit, why should I be concerned / upset if the team hires someone else to finish/fix the project and pay for it ?

Why do we pay contributors who willingly donated their time to the project ?

$20 per hr (he evaluates the task, comes up with hrs, and fixed amount is calculated). Well, nothing is certain in life. So, nowadays it's impossible to say for how long someone will be poking at something. I suppose as long as the money is on the table, he could work on it. If there are only like 5 people who needs DR, then yeah, it won't work. Too bad 2.0.3 was never finished as far as bug fixes :/

In the light of the recent post by greebo, I propose to do some crowdfunding for future development of DarkRadiant, if anyone is interested that is. I have a programmer, who can take on DR development (fixes and new features), but not for free. Please let me know what you think about this proposal.

Ouch. Thanks for the heads up. Hopefully you will get through whatever you are going through these days.

So AMD getting consistently worse when it comes to not-so-old hardware support and OpenGL. I don't quite understand what Bikerdude is hoping for with AMD, but of course there is always a place for a miracle

Differently how? Better? Worse? Same'ish ?

I have a bug with textures to report (happens when duplicating brushes), but not sure if this vacation is going to last for eternity or maybe another month.

It seems like a looooong break

Smooth shading, not smooth vertices.

Thanks SteveL, I'll give it a spin. Polygon throughput on modern GPUs is very high, so it doesn't really hurt performance on high poly scenes. So basically blendlight decals won't be affected by lighting ? If so, then it's kinda useless I was thinking of decaling maps with grime and dirt using blend lights and it will look out if place if lights won't affect blend light decals. I guess it would be nice to have a special light type that supports normal/spec maps and only lights up a surface with some special flag (like that black light surfaces that are only visible with flagged lights). This way the decal light would not affect surrounding surfaces and the decal would have bump and specular and fit nicely with the environment. On the other hand it would be nice to have some autoprojecting decal entity that projects standard decals on any surface, as if it was a blend light.

I don't since I haven't even tried it. I just thought maybe TDM team tried it and have an insight. I am interested in this approach is due to flexible and fast workflow - projecting decal on a complex surface is always faster (and should look better) than placing patches and aligning them to the surface.

I am wondering if decals can be made using blend lights and if it would be equally fast or faster (performance wise). If blend light decals are possible and practical, are there any limitations compare to regular decals? Thanks!

There is nothing like that in Doom 3. So I suppose there are only 2 ways for Doom 3.

What's the other way? I know that running script from worldspawn with "call" wouldn't allow such thing - I have to literally either hardcode $entity_name or find an entity, and get it's name.

Ah, cool, thanks! Makes it much easier. Just to verify, it only works when running script on an entity via "scriptobject" spawnarg, correct ?