.Map :: MoHAA Mapping

I would like to apologize to everyone for the recent lack of activity on my behalf. I got myself a summer job (well, more of an unpaid apprenticeship) at an electronics manufacturing company and during its timespan (until July 31st) I'm living at another place where I have no internet access or even a computer, and only come home on weekends.

I know this is going to disappoint a lot of people and undermine the project greatly, but I do not plan to continue any work on OpenMoHAA. The potential suitability of the engine for modern game projects was the main driving force of the development for me - I wanted to base other games on it. It no longer is, though.

You have the source code, it can be done, you may say. Well, yes and no. I conducted some experiments with the original MoHAA and the ioquake3 engines to explore the potential for rendering scenes comparable in complexity to modern game engines (remember my huge Middle-Eastern city map project? it was supposed to be one of them). The results are very disappointing, to say the least.

Crysis, for example, easily pushes 1-2 million (sic!) triangles down the rendering pipeline each frame (not to mention the shaders and other stuff). My system (P4 3GHz, 1.5GB of RAM, GF8600GT) pulls off 30FPS at high quality settings. Now, an average MoHAA map has about 10-20 thousand triangles at peak. My system renders the game at 200-400FPS. I tried pushing as many triangles down the renderer as possible. It turned out that the engine is already struggling at 30FPS with as few (compared to Crysis) as 100 thousand triangles, even if all it was drawing was a number of simple cubes with single-stage shaders.

This is far from an acceptable limitation for a modern game engine, and the reason lies deep in the video renderer's very architecture. One of the problems is that too much data has to be sent from the system RAM to video RAM. The transfer must be done by the CPU, is limited by the speed of the system bus and a couple of other factors, effectively creating a bottleneck. The cause of that is the shader system - the feature that made Quake III so visually attractive and mesmerizing in its days of glory is the burden that prevents the engine from being modernized, because all the wacky stuff that those shaders do - scrolling, twisting and deforming the textures and vertices, blending transparent textures, etc. - creates more and more overhead.

Back in 1999 this was an affordable architecture, because the 3D accelerator chipsets' performance was poor and with the numbers of triangles and vertices involved, sending that data from one memory to another was an optimal solution. Now, with the more powerful video cards, it's just slowing everything down. Modern engines tend to precompile as much stuff as possible and store it in the video memory in advance instead.

Rewriting the renderer from scratch is out of the question, as all of the shader features would have to be rooted.

Considering all of the above, I hereby withdraw from the OpenMoHAA team and leave Wombat in charge of it. Whether he chooses to continue or abandon the project - I leave the decision up to him.

I would like to thank all of you for all the support and enthusiasm you showed over those months, and extend my sincere apologies for disappointing you in such a way. I feel it's time to move on with things - I'm working as a technical artist, music composer and voice actor for the team responsible for adding the Polish Army to the Project Reality mod, I have also began work my own game engine, with some more modern design goals in mind; hopefully it will see the daylight some day.

Will you still help with the OpenAssault project if it ever gets started?

Oh man! This hurts my ears. I hope I'm sooo drunk I can't read right.

Oi..

Well, it is what it is. I hope you could still help out if we have any issues, and maybe someday you could reconsider in the future when you have more time. Building an engine isn't easy, but we have had alot of success with what we got done.

I am a bit disappointed, because this is the 2nd project that I have taken part in and unfortunately never got released with all the hard work that was put into it.

I know it's hard, but I hope that wombat will still continue with this and that we can find some more knowledgeable programmers as willing to help out. Despite the fact that this rendering engine can't compete to modern day games, I still see alot of potential in this game especially if you consider the fact that we took a half assed, half finished exploited game (MOHAA), and turned it into something that's half way playable. I know there is still alot of work that needs to be done, but there really is alot potential for this and I hope we can get more support with this from others.

Well enough of my rant, inequation, you can say what you will, but of all the people I have met your one of the few most knowledgeable people I have ever met. I hope the best for you and your future endeavors. Wish you the best of luck in your apprenticeship\internship and any other future jobs that you may pursue in the future. I am very much looking forward to your work that you plan on releasing and hope to continue seeing more from you.

Best of luck man! Hope to stay in touch.

I know it has potential, but unfortunately not in the field I was hoping for.

Of course I'll be here if you need any advice. During my experiments I got quite an insight into the various parts engine, so if you need anything explained, I'll be happy to help.

Very helpfull post. If you have the time I'd love to see a more in depth of the process you went through to come to this conclusion: I am curious.

Tam

Lol... I just announced my retirement from the project, and you say that it was helpful.

Anyway, answering Tourist-tam's request, I've elaborated on the subject in this thread.

Rookie One.pl wrote:Lol... I just announced my retirement from the project, and you say that it was helpful. ....

Yep, 'cause you could just have said: "F*** it, guys; I'm out

Tam

At least you have learned a lot of this project and gained new experiences. Didn't you, Leszek? So don't be too disappointed after quiting a project .

Interesting project. Hello !

In answer to the better game engines comment. I personally think that it's the spirit of the game that matters. Crysis is interesting. I love it sometimes, but it just does not have the atmosphere that the Medal of Honour series has IMO. It's counterintuitive as well. Look up "Uncanny Valley". Ultra realistic things can actually end up seeming LESS involving. This is why a lot of the older game engines are far MORE involving IMO and are often better games. I'm grappling with the same issue with my Open Battlefield project (early stages).

Hi

Does this mean that Open MOHAA is now dead, or are there people still working on it and if so, when can we expect to see it released?

Despite rookie's withdraw from the project he still shows his support for it and is still helping out with some of the coding. Thanks rookie. The project is NOT dead. I have talked to Wombat and he will be taking on the project from where rookie left off. He is as much motivated in finishing up what was started as everyone else on the team.

As far as the release date, nothing is official, and unfortunately, I can't guarantee a official release date for a few reasons.

1. Projects that are followed by a certain date tend to be rushed and come out crappy.
2. Everyone is very busy with other things (School, work, personal life), this project is merely something a few of us took on for 1 of 2 reasons. The love of the game and as a hobby\challenge.
3. Despite the fact that Open MOHAA is built off of the ioq3 engine, it still requires alot of modifications to get everything working. Right now Wombat & Rookie_One are the top coders we have, so we just need to show them our support and wait to see what happens. So far we already have textures, maps and the terrain running seamlessly, right now the hard stuff is the models (because ioq3 uses a completly different model structure system than MOHAA), UI & you can't forget about the TIKI and scripting Systems. On top of that we still need to work out the networking, but the features above are what will slow the progress of the project done.

Wombat will have some more time to work on it towards the end of this month and beginning of September.

Thanks for that info m0d hippY. I was beginning to get worried that it might never happen.

I wasn't expecting an exact date and something more along the lines of a year, or an approximation of when we might hope to get a glimpse of a beta or something.

I can't believe that there are so many programmers out there who play mohaa, who aren't helping with this project. Damn shame that. I only wish I had bothered to learn to code properly.

Anyways, I understand and appreciate the reasons given for the delay (life goes on and all that) so I just want to say thanks for sticking at it and good luck.

don't worry
currently, there are only rookie and me working on the code (-> any help welcome). i will be leading the project from now since rookie resigned, but i still hope for his support.
i had to prepare for exams in the last weeks and rookie had a job too so this is why there was no news recently. but now i'l have more time.
additional coders would definitely speed up the development, but we are still progressing

Okay for a while I've been wanting to learn a programming language. First off I would like to know what language OMoHAA is being written in, second I want to know how long it take to learn it, and any good guides out there?

Open MOHAA is coded using C (Not C++)

Learning a language is extremly time consuming. It takes alot of practice, alot of motivation and a whole lot of patience for trial and error. Another thing that I lack, that is helpful in coding is math and computation and understanding how things work. The thing the Open MOHAA team has is that we are all pretty knowledgeable in the game. We know how things work to a certain aspect and we know what needs to be done to get it up and running. As far as the coding is concerned Wombat & rookie are the guys with the brains, everyone else atm is more or less there for moral support & minor helping with what we can. I can tell you though that heiko & I have also been working on some fixes that we plan on implementing into the engine once everything else is finished up.