Doom 3: Phobos or Doom 4 1st?

[Face23785]

Which will come out first and why? And what happened to the poll function?

[DoomUK]

How long is a piece of string?

Doom 4 is an id game. That means it will be "done when it's done" (sic). We've yet to even see an official screenshot of the game (discounting those "leaked" concept screenshots that cropped up on here a few months ago). No one except the team who are working on it knows anything about it for definite. Most of us are hoping there will be some news at this year's QuakeCon, but who knows.

As for Phobos, I can tell you that the mod is being worked on tirelessly and fastidiously. Though I'm not as active a member as I used to be, I'm constantly seeing what's being cooked up by the other members and I can promise you it will be worth the wait. The ethos of Team Future has long been to give people something of the absolute highest quality. The media that has been released over the years should be a testament to that. But, being a very small team, it takes a great deal of time to reach that goal. You'll just have to be patient. We want the mod to be finished and released just as eagerly as you do :)

[Shaviro]

| | . . . | . . . | . . . | . . . | . . . | . . . | . . . | . . o . | . . . |

[printz]

^^^ Development timeline?

[DooM_RO]

I know Phobos has been in development for a very long time but I am quite sure they will release it before Doom 4, they are close to completion I think.

I know you guys have decided not to make the rest of the episodes but will you change your mind if the feedback is outstanding? I would really like to see the story come to an end, it doesn't necessarily have to be two more episodes, just a few maps finishing the story. Besides, if I am correct, it shouldn't take as long as the first episode because you are much more experienced and wouldn't have to scrap so many things. Just sayin'.

Speaking of quakecon, if any of the TF team members go to the event, it would be cool it you offered Id a DVD with the mod(if not this year, maybe the next?) to hear their opinion about it!

Btw, do you still think you could finish it this year?Also, will there be a gameplay trailer? Oh, and it would be cool if you made a filmed documentary about all those years invested in making the mod and maybe even talk about what you thought about Doom 3 and RAGE and how you think Doom 4 will turn out to be based on these games!

[printz]

There should definitely be a tool that does the work faster than DoomEdit and co. I read posts and even blog entries by Shaviro for example, complaining how tiresome it is to map for Doom 3.

What are the things the Doom 3 build-in editor needs? Has any Doom 3 port addressed them?

[geX]

We are pretty close to completing phase 1. You can read about the phases here

We can still not tell if we are releasing this year. Its done when its done. Like Doomuk mentions we are aiming at a really high quality, and quality takes time. We rather have it take a little longer than rush (it can hardly be called a rush anymore) anything out the door we arent happy with.

Right now I can't really see myself doing anything other than this episode for idtech4. Like mentioned we have bitched about the engine and the tools for a long time. Shaviro can be more specific if you want. Sure if the feedback is out of this world, I would reconsider. Right now I just want to finish this thing.

We want to make a gameplay video at some point, but it propably won't be before somewhere halfway through phase 2. When we have something final we can show.

About something post-mortem like a documentary. We have talked about making something like that, but maybe not a movie. Maybe a paper of some kind with the development years and advice for others etc. etc.

[printz]

geX said:

Like mentioned we have bitched about the engine and the tools for a long time. Shaviro can be more specific if you want. Sure if the feedback is out of this world, I would reconsider.

So, which are the main problems? Too buggy, unresponsive, too many operations to do to accomplish low-level crap? All this can be solved (I hope), Doom 3 is open-source now.

[DooM_RO]

Yeah, now that the source code has been released, what do you think about modifying the doom 3 editor to be more convenient?

[Koko Ricky]

Either way, the screenshots look promising as hell. Some seriously impressive stuff and the only Doom 3 mod I can honestly say I give a shit about.

[Shaviro]

printz said:

So, which are the main problems? Too buggy, unresponsive, too many operations to do to accomplish low-level crap? All this can be solved (I hope), Doom 3 is open-source now.

Most of our problems originate around the editor and the hard-coded limitations of the engine (sound familiar? ;) ). The editor is kinda half-baked and it puzzles me that they never corrected some of these errors/design flaws. Simple things like rotating part of a map is a hassle. Sure, you can rotate brushes and patches no problem. If you have entities (lights, models, monsters, etc.) selected as well, it'll seriously fuck up.

The reason for this is simple. When rotating, the editor finds a pivot point which is most likely the center of gravity, rounded down to the nearest grid point. It then subtracts the pivot point from the individual brushes and patches, rotates them and adds it back in to the geometry. That's fine. Works like a charm. The problem is that the editor skips the whole pivot point thing for the entities, so they rotate around their own axis (They use a rotation matrix instead of actually rotating the coordinates of the brush planes etc.).

Our solution for this was that I wrote a map-loader in C# (XNA) which could parse the .map file, load the entities and objects into their respective object representations, rotate the map properly and save the file. Writing the code for rotating entities took 1½ minutes. Seriously. I can't believe they didn't fix this themselves. Did they never have to rotate anything other than brushes? :P

That said, the bug hasn't even been fixed in GTK Radiant or Dark Radiant, so maybe there's some sort of fucked up structure in the editor code. Either that or there is a setting hidden away somewhere that we never found ;)

We've used the map-loader tool to circumvent quite a few problems, errors and slave tasks since then. Like searching for specific brush patterns to find a lot of similar structures and make them func_static, part of the worldspawn or just add/delete keyvals from the entities.

As for using the source code to correct the flaws, well. One problem with that is that you'll be forced to use an inferior version of the shadow algorithm because Carmack's reverse was patented by creative. That will put a strain on the performance and our levels are already doing that, so ;)
We also don't want to overextend ourselves. The map-loader tool can fix most of our problems for us.

[Shaviro]

DooM_RO said:

Yeah, now that the source code has been released, what do you think about modifying the doom 3 editor to be more convenient?

Hmm well. If I had to do it, I'd rather start over. Perhaps very inspired by the Doom 3 engine. Also, I'd go for writing a "slightly" slower managed C# version. I'm a sucker for this language and it's all I really know how to code for effiecently ;)

Then it would of course be DirectX instead of OpenGL, which would fix a whole array of problems with ATI cards. As for rendering I would research into deferred rendering for it to allow more lights, a more natural distribution of lighting in the scene and such. Don't get me wrong, I like how Doom3 looks, but it's just WAY too fragile a system. It sucks to set up the lighting in these maps. The good thing is of course that it's real-time, but so is the deferred solution.

As for the editor, I'd start out with the same basic tools and options. More importantly I would change the way the map structure works. Less "stuffing everything into one .map file" and more "object oriented file structure" where you reuse structures by pointing to the same base object and just use properties, position vector and rotation matrix for it. Very much like when writing applications in object oriented programming languages.

But this would take 2-3-4 years just to get into a position where you could start making actual content for it :P

[Shaviro]

GoatLord said:

Either way, the screenshots look promising as hell. Some seriously impressive stuff and the only Doom 3 mod I can honestly say I give a shit about.

Hehe thanks.
Now we just need someone to point id in our direction at quakecon ;D

[DooM_RO]

Well, it's SUCH a shame you won't finish the episodes but such is life. Who knows, maybe they will fix the editor problems in D3 BFG?

I've already posted this in the Doom 3 should have thread but just in case someone misses it, here is some promising news about quakecon http://xbigygames.com/quakecon-2012-will-have-the-revival-of-a-much-loved-classic-pc-shooter/

[Face23785]

Just to be clear, I wasn't posing this question to rag on TF. I think the project looks and sounds outstanding, and I'm fairly certain it will be worth the wait. People think I'm crazy when I say I expect Doom 4 late next year, but honestly they've been working on it quite a while. I don't think it's a case of not having anything to show. I would bet they're further along than most people realize, they're just being more tight lipped about it this time because we got to see too much of D3. If we see it at Quakecon, I won't be surprised if it's within a year of completion. That said, it seems Phobos has a ways to go yet. So I thought it was a legit question, not "Phobos will def be first" like seems to be the consensus here.

I'm sure both will be worth the wait. Keep up the great work on Phobos!

Edit for clarification.

[Caffeine Freak]

Shaviro said:

Most of our problems originate around the editor and the hard-coded limitations of the engine (sound familiar? ;) ).

No Shaviro, most of our problems originate around poor planning, especially in the earlier stages of the project, which has resulted in time and energy spent on re-doing things. ;)

Anyway, I generally don't say much, but I would say it's pointless to try and nail down a release date for Phobos at this point. Yes, the project is moving along quite well right now, and the end is in sight, but there's still a lot of work to do. The fact is that, until you can be absolutely sure of all the tasks you have left on a mod and your ability to finish them within a certain time frame(usually when you have maybe a month or so left of work), there's not much point in imposing a deadline on yourself or trying to fix a time frame in your mind for when you might be done. The only thing you can do is to just keep working as hard as you can, which is exactly what we're doing right now.

[Shaviro]

Caffeine Freak said:

No Shaviro, most of our problems originate around poor planning, especially in the earlier stages of the project, which has resulted in time and energy spent on re-doing things. ;)

Well, he was asking specifically what was wrong with the engine/tools ;)
I actually don't see the planning as the biggest problem we've had. The framework and structure of the project has pretty much remained the same since we started. It's flexible and has evolved, but the red line through the project remains.

I think our biggest problem has been what is usually the biggest problem for this sort of project. We used to work in a "random access" fashion where different members pick different goals and tasks. You take e1m1, I take e1m2 and so forth. Then when the map starts to near a completed status (which is in reality only about 10 to 20%) and the tasks get dull, you take a new map and crunch away on that.

The way we work now is more of a depth first approach where we actually finish (finish as in a preset array of conditions that should be met) things before moving on. This has been so giving and rewarding at the same time.

I'm all for good planning, but to my experience too much or too detailed plans will hurt the project and make the structure way too rigid.

[Caffeine Freak]

Shaviro said:

I think our biggest problem has been what is usually the biggest problem for this sort of project. We used to work in a "random access" fashion where different members pick different goals and tasks. You take e1m1, I take e1m2 and so forth. Then when the map starts to near a completed status (which is in reality only about 10 to 20%) and the tasks get dull, you take a new map and crunch away on that.

Yes, that's called 'poor planning'.

[Shaviro]

No amount of planning will ever be able to save you from the lessons imposed on a mod team for a project like this. The only way to learn for a new team in completely new environments is to throw themselves at the tasks, test, try and adapt. Making a detailed work plan that is able to drive development forward and steer clear of these kind of problems requires knowledge and experience you just don't have initially.

[Caffeine Freak]

Well, experience IS the best teacher.

[999cop]

Face23785 said:

Which will come out first and why? And what happened to the poll function?

Pay the developers 10 million dollars and it will be done by this summer.