The Dark Mod Great Summer Vertical FM Contest!

[nbohr1more]

Thanks for the answers Baddcog. Is there any way to force AI not to path-find areas since monsterclip doesn't have this function?

[Komag]

got three beta-testers, that's enough, thanks! Hopefully will be able to release tomorrow, we'll see

[Baddcog]

Thanks for the answers Baddcog. Is there any way to force AI not to path-find areas since monsterclip doesn't have this function?

 

 

I don't think so, I think pathfind is calculated for everywhere an AI could be. So even a 'blue room' that isn't connected physically to the rest of the map is still sealed from void and thus calculated.

 

It basically just lays out the floor area, and AI have info in their def files that say the max height they can reach above that floor (for stairs and whatnot). Then they have a bounding box (aas size) that tells which areas they can fit into (but those areas still have to be calculated so they know).

 

And I don't think that would help performance at all, it would only maybe reduce the size of the cm or proc files.

 

As for ai seeing you flying, don't fly cheater lol.

[Fidcal]

Thanks for the answers Baddcog. Is there any way to force AI not to path-find areas since monsterclip doesn't have this function?

There's some confusion here. Have you studied http://wiki.thedarkmod.com/index.php?title=Pathfinding

 

Perhaps you don't mean pathfinding but 'hide'. You want something that is invisible to the player but blocks the AI from seeing you? But that would mean you can be in bright light in full view of the AI but he doesn't see you. Is that what you want? Perhaps for long distance? I don't think there is such a thing.

[nbohr1more]

I was a way to keep the AI from even investigating an area to reduce CPU load. It looks like func_static geometry is the trick from my brief review of your article.

 

 

Edit:

Very good article BTW!

Edited August 24, 2010 by nbohr1more

[Fidcal]

To stop AI from pathfinding into a whole area just fill it with monsterclip. Vertically, Monsterclip is a barrier to them, a solid wall. You should extract lots of FMs from their pk4 and load the map file into Dark Radiant. Make sure the clip filter is OFF and examine how and where the monster clip is used. Look at Builder's Influence for instance.

 

You can filter out patches and all entities and visportals and then you see virtually what the AI 'see' - just a simple solid maze of worldspawn that they navigate.

[Fidcal]

This is one street area in Builder's Influence. All that vertical monster clip is blocking entities like barrels or street lamps or func_static posts etc. You see the ramp up the steps to make it easier for the AI to walk up? And the huge monsterclip high over everything is to cover all the roofs so pathfinding is not calculated for the roofs. The AI can't get there so why waste pathfinding.

 

[nbohr1more]

Yep.

 

I may not have worded things correctly but placing monsterclip on ceilings seems to be a standard practice and should reduce CPU load.

 

I can't really celebrate being accidentally correct though

[DeTeEff]

Fieldmedic said:I do not know much in computer programming

 

No I in no way is good at computer programming...Should I take that citation as a mockery? Am I wrong if I state that optimization allways should be done?

Edited August 24, 2010 by Fieldmedic

[nbohr1more]

You have broken two of the primary commandments here:

 

1) Thou shalt not mention an engine change

2) Thou shalt not talk about switching from TDM to Thief 4

 

Your penitence is five hail Mary's and another 2 Fan Missions (at least).

[DeTeEff]

You have broken two of the primary commandments here:

 

1) Thou shalt not mention an engine change

2) Thou shalt not talk about switching from TDM to Thief 4

 

Your penitence is five hail Mary's and another 2 Fan Missions (at least).

 

I knew I was throdding on sensitive ground when bringing up the topic of the possibilities of a thief-game on a different engine. Therefore I clearly pointed out that it was theoretical and added jokes and humbly admitted my scarce skills in programming. I understand if the dev team gets pretty damn annoyed on people who degrade all the work on the mod. It wasn't meant as an insult when I said we'll have to wait for Thief 4 to be able to fill areas with AI and lights...

I am in pure awe of the team that has worked hard on this for 5 years. It's a solid mod with no connection to the original game. But I find it quite sad that the topic cannot be discussed in a mature way.

 

But then again, this is a forum and I might very well misunderstood Dave and if that's so I appologize.

Edited August 24, 2010 by Fieldmedic

[Baddcog]

It's not that it can't be discussed it's that it is pointless to discuss it. It just won't happen.

 

Took us 5 years to get to this point and maybe it's not the most optimized game ever, but it is pretty nice imo. Swapping to a different engine and you're going to have just as many issues to work out, but it'll have to start from scratch.

And nothing's as easy to 'port' as it sounds.

 

Scripts are in different languages so that probably all needs rewritten (not only is that TONS of work programmers are the hardest people to find for mods and ours are all probably at wits end by now, or they had a kid and don't have the time anymore...).

Textures themselves are pretty universal, but not the shaders which make them appear in game as is, so that all needs rewritten. Not fun, very time consuming.

Some model formats MAY work in other engines, but not very likely. So all models need at very least rexported. Possible scale changes, shadow and collision mesh removal or fixes. A few of my models I made for T2 have made their way into DarkMod but it was almost a total redo.

AI would basically be toast, rigs should be fine but would need all models and anims re-exported.

maybe attachments are different so maybe bone names ned to change...

Plus it sounds like T4 is going to use modelled enviros, which means it's very hard for most people to customize, they have to have specific software, etc...

Right now all you need for TDM is Doom3, the mod and DR. Free programs like Gimp and Blender can do just about anything you need to do to customize.

 

Redoing all that would basically take about the same amount of time it took the first time around, so do we want a highly optimized TDM (maybe) in 5 years or do we want to continue working on improving TDM like we are at the moment?

 

IF TDM was ported to another engine the only way to go would be the Ogre renderer IMO. Great renderer (Torchlight and Trine are well known games using this renderer) that is open source, and about any plugin you'd possibly need for free to make a great game from scratch. truly open source. No need for anyone to own games, etc...

[nbohr1more]

I think Greebo cleverly expressed his frustration at yet another member bringing up this topic (yet again) so the Dev's may not be quite as grumpy as we imagine.

 

I was just trying to take advantage of the guilt factor and get you to make some more missions

[SneaksieDave]

I find it quite sad that the topic cannot be discussed in a mature way.

 

Your suggestion has been made several times over the years. Do a forum search. It has been discussed and basically shown to be impractical. And not just for TDM; it happens even for professionally published titles at times. It is nothing new. It is however unfortunately a very, very shortsighted and inexperienced view of the amount of work that goes into a project like TDM. The fact that it took more than 5 years to produce -- and that there is still very, very little help being offered (almost none), particularly in the programming department -- should give some perspective on the scope of the project and the reception at the suggestion.

 

At the same time, nothing stops you from developing such a mod yourself. Heck, you can even use the open TDM source as your codebase.

 

Neither quoting you in a response nor underlining logical weakness in your statement should be viewed as mocking or immature, I think. Don't read hostility into it.

 

 

 

 

 

 

 

Much.

 

Disclaimer: Joking should not be viewed as mocking or immature.

 

Ugh. I even included a compulsory winkie to be extra clear. Pitiful when it comes to that.

[DeTeEff]

It's not that it can't be discussed it's that it is pointless to discuss it. It just won't happen.

 

Took us 5 years to get to this point and maybe it's not the most optimized game ever, but it is pretty nice imo. Swapping to a different engine and you're going to have just as many issues to work out, but it'll have to start from scratch.

And nothing's as easy to 'port' as it sounds.

 

Scripts are in different languages so that probably all needs rewritten (not only is that TONS of work programmers are the hardest people to find for mods and ours are all probably at wits end by now, or they had a kid and don't have the time anymore...).

Textures themselves are pretty universal, but not the shaders which make them appear in game as is, so that all needs rewritten. Not fun, very time consuming.

Some model formats MAY work in other engines, but not very likely. So all models need at very least rexported. Possible scale changes, shadow and collision mesh removal or fixes. A few of my models I made for T2 have made their way into DarkMod but it was almost a total redo.

AI would basically be toast, rigs should be fine but would need all models and anims re-exported.

maybe attachments are different so maybe bone names ned to change...

Plus it sounds like T4 is going to use modelled enviros, which means it's very hard for most people to customize, they have to have specific software, etc...

Right now all you need for TDM is Doom3, the mod and DR. Free programs like Gimp and Blender can do just about anything you need to do to customize.

 

Redoing all that would basically take about the same amount of time it took the first time around, so do we want a highly optimized TDM (maybe) in 5 years or do we want to continue working on improving TDM like we are at the moment?

 

IF TDM was ported to another engine the only way to go would be the Ogre renderer IMO. Great renderer (Torchlight and Trine are well known games using this renderer) that is open source, and about any plugin you'd possibly need for free to make a great game from scratch. truly open source. No need for anyone to own games, etc...

 

Thank you for a nice reply. I feel quite assaulted here, just by a THEORETICAL asking of IF it would be much work to move the game to another engine... It wasn't meant to say that we SHOULD move the game. It wasn't a "suggestion to move the game". It wasn't a demand or anything. It was more of a "I-just-would-like-to-know" basis (for me who doesn't know so much in computer programming )...I know the topic is inflammatory but it would be enough for me if you just said "No, too much programming and shit" and I'd be happy with that, but you took your time and wrote a whole bunch of text

[DeTeEff]

I think Greebo cleverly expressed his frustration at yet another member bringing up this topic (yet again) so the Dev's may not be quite as grumpy as we imagine.

 

I was just trying to take advantage of the guilt factor and get you to make some more missions

 

It feels nice that you like my mission. Maybe there will be more...

[Baddcog]

Yeah, I was bored. But other people can reference that now that it's been written done.

 

Should it be Wikified? lol

[nbohr1more]

Yes!

 

Then a post with the link will be the reply to all future occurrences.

 

It'd be funny to have an editorial style response in the wiki like:

 

Why we're mad at you:

 

 

When I first started the mod... text... text...

 

five years later we are... text...

 

and now after that exhaustive explanation you have asked us this question again... text text

Edited August 25, 2010 by nbohr1more

[SneaksieDave]

For clarification (hope you don't take offense at this, too), here are just a few times this has been brought up that I've found using google for two minutes (the forum search is really ass). We've had Crysis,UDK, and even Wii. Somewhere around here I'm sure Irrlicht, Ogre, and maybe whatever the Penumbra engine is called, have been brought up too. And that's just the ones that come to mind. While "go with different tech" for any variety of reasons is in no way necessarily a "bad" idea, it should be more than obvious that it's way too late to consider that at this stage. One need only look at the time involved to see how practical it is, and that overlooks a lot of the true hurdles.

 

Imagine working for more nearly 6 years on a free project, getting very little help from the community (aside from demanding things and complaining), and throughout, you have people suggesting you trash it all and start over for them. Imagine that when you make a simply awesome editor (and if you don't think it is, you don't know DR except for what it currently lacks (e.g. proper renderer)), they tell you all the ways that it's not as good as sandbox, or hammer, or even friggin DromEd.

 

Yes, when someone asks a question they don't know has been asked before, they deserve some patience. But those repeatedly asked also deserve some as well for the reasons above. Since you haven't responded about the quoting clarification (except in PM which I'm not interested in), I'll assume either it's clarified and closed, or you want to hold a grudge because you were quoted in an unflattering manner. Okay. You shouldn't read more hostility into something than was intended. Either way, we should conclude the derail of this thread that started with the engine switch suggestion.

[DeTeEff]

No, no, I don't take any offense to that. The reason I sent you an PM is that I wanted to clear it out between you and me, and not go so much OT in the thread. I think the thing got overboard by a misunderstandment from the beginning. I am fully aware of the work that has been put into the mod, and when I first brought up the subject, I didn't mean to "suggest" an engine change. It was just a question of HOW much work it is to "port" it to another engine, because I really wanted to know. As I said above, there wasn't a demand or wish. Yes, I could have done a research by myself but as the topic was around performance and lights and optimization I took the chance to ask a friendly question.

 

I edited the post a page back and clarifyed what I said about the optimization. I believe you took it as YOU programmers should optimize the engine constantly no matter how poweful the engine is. That wasn't what I meant. I meant that WE mappers should allways optimize our missions after the engine used. Even if you have an engine that can support massive use of lights and stuff, you should allways see to make it the most "cheapest" for the engine to render. I thought that was obvious as in the previous sentence I talked about cramming in lights and AI in Thief 4 (As I now know, there will probably be difficult to map in Thief 4 but that's not the point).

 

I suspect that you've been ironic and didn't mean to sound sarcastic against me, and hope I got you right now? Maybe you should use the hated winking smiley more, if you're ironic or not so mad as it sounds.

Edited August 25, 2010 by Fieldmedic

[SneaksieDave]

Yes, misunderstanding and both sides blown out of proportion and context. Peace.

 

So what's this about a summer FM contest?

[DeTeEff]

Good thinking my friend Peace

 

Do not know...I've got a feeling that it may be a tough one. Lots of good stuff for this contest indeed...

 

OT

 

 

 

I would like (as people allready has asked about) to have a quickie 4-5 weeks horror contest for Halloween, starting...soon?

 

Fidcal, may we start something after the weekend now? 30/8. No restrictions but must have some kind of horror theme. Are there much maintenance in hosting such a contest? If it is, then we may delay.

 

 

 

[SneaksieDave]

Yes ^

 

I've heard one (sensible) suggestion that it might be good to allow some "regular" mapping time in between contests, so contests don't dominate everything, but I disagree. Contests really keep the community blood flowing, and it's not like everyone participates. If someone has a larger project they're working on instead, they will do so and leave the contest to others.

[stumpy]

I found another use for monster clip, it actually stops moveable items from falling through static furniture, eg moveable plates on a table wont fall into the table when the table is wraped in monster clip when the player picks them up.

Edited August 25, 2010 by stumpy

[Sotha]

Yes ^

 

I've heard one (sensible) suggestion that it might be good to allow some "regular" mapping time in between contests, so contests don't dominate everything, but I disagree. Contests really keep the community blood flowing, and it's not like everyone participates. If someone has a larger project they're working on instead, they will do so and leave the contest to others.

 

-Oh.. I thought it would be fun if we had some standard times for contests. There could be the summer contest and the christmas contest. It might be difficult to get mappers for other times, because the aforementioned times are the ones when people typically have extra time. But if you manage to get mappers to participate, then sure. I don't mind having more missions available, as long as they are high quality.

 

The problem is also that if we have too many contests, we will have people who make only maps for contests, because of the visibility. And contests make it rather dull in the sense that we will have many maps with same aspects. After this vertical contest, I'm not sure I'm gonna like seeing tower or tunnel maps for a while...