Page 1 of 2

Postmortem: “Orient: A Hero’s Heritage”

Posted: Wed Apr 09, 2008 1:05 am
by dlangdev
Gamedev.net posted an article titled...

Postmortem: “Orient: A Hero’s Heritage”

http://www.gamedev.net/reference/busine ... res/pmAOP/

Please post your reactions, insights and comments here.

Posted: Wed Apr 09, 2008 1:25 am
by Halifax
I have to say that it is amazing what they did with Irrlicht for that project, and it looks simply stunning. If I didn't know better I would've thought it was Ogre. :lol: But seriously, I think that was a great post-mortem, and you can in fact learn a lot of things from it.

Posted: Wed Apr 09, 2008 6:55 am
by CuteAlien
20 months development with a professional team using Irrlicht and we haven't even heard of them? It's really sad to hear that project got canceled. Though from the description it sounds to me like they got very far. 20 months with just 3 programmers isn't that much time.

I got very curious about that level and model editors about which he was talking. I wonder what happened to those now...

Posted: Wed Apr 09, 2008 8:04 am
by MasterGod
I knew Irrlicht was good but I didn't know it's That good :shock:. WOW. Some of the screen shots reminds me of Oblivion ^^.
Amazing..

Posted: Wed Apr 09, 2008 8:15 am
by rogerborg
That was just sad. :(
You might question our sanity when considering our budget (110k USD), that we even went for an RPG…and you might be right. I have no rationale to offer for our decision, except that “We wanted to do it with our hearts and souls, because we all loved RPG games!”
== root cause. Lesson: you are not a unique snowflake, and there is no magical way to do more for less. Experience means having the confidence to say "We can't do that within that time and budget."

Posted: Wed Apr 09, 2008 8:42 am
by dlangdev
I wonder if it's possible to get them to make their code and art donated to irrlicht, so that someone here can continue working on them.

A lot of work has gone into it and it would be sad to see them all go down the tube into the that cold dark place.

Posted: Wed Apr 09, 2008 8:56 am
by JP
I would assume they wouldn't want to part with their hard work for free though ;)

Posted: Wed Apr 09, 2008 12:17 pm
by rogerborg
110K USD? I filled my car's tank up this morning, so perhaps we can arrange a trade.

Posted: Wed Apr 09, 2008 2:18 pm
by arras
The risk they took was not because of low bugget. You really do not know what value 110K USD have in their country. Most of game developing companies started somewhere in dad's garage probably.

The main mistake they did is they took on too big project for start. That's why they failed. They should really have started with something smaller ...may be just small.

Actulay they did pretty well and that they accomplished what they accomplished speaks for their skill.

I see it as example of something I see often also here on Project forum. Kid who likes to play computer games suddenly got idea to make game ...his way. Found Free engine and C++ tutorial. He is full of great ideas and big plans.
I started like that some 10 years ago. Up until now I produced nothing just few demos.

Less is more at the start. I think.

Posted: Wed Apr 09, 2008 3:37 pm
by rogerborg
Agreed, they did do well, and the analysis is excellent.

I really hope they - and particularly Babak, the author - get another chance, as they'll do much better next time.

Posted: Wed Apr 16, 2008 2:13 pm
by Yoran
I just hope they will contribute some of their engine changes

"we were in dire need of a good .X exporter."

They should have used my guide! :D

Posted: Thu Apr 17, 2008 3:22 pm
by Vsk
It was amazing, I would like to show those pictures to agi-shi (I know you are over here agi-shi, come on!, show to us!:P;),).

But seriouly, it is realy good and realy sad.

The money problem I think it was NOT the problem at all. (Even with the 20% of Iran inflation!!! holy sh*t!).
For example in my country a programmer salarie goes between $500 and $1500 monthly and a $1000 salarie in my country is realy good!. (And not, stuff doen'st cost equivalent to us :(, specialy computers :().

Even more in my job we manage around this numbers (a little more) for a proyect. (includin all the money spends besides salaries).
(I am talking of course about a 1-1.5 years proyect, no more).

And as you can see in his postmortem, there are a lot of problem not related with that. Again, as I said before in other post, it is always good study some software engeniering, there you learn process manage, including timing, human resource etc. if you look at the postmortem those kind of point were the point that they have problems.

Reading the postmortem I see a problem that I also am having, and that they name: "irrlicht doen't make optimization at all". (Come on!, frustum culilng doen't even count like and optimization :P:)).
And in my case and the case of this guys I have to disagree with that frase: "earlier optimzation is the root of all evils", it is not easy adding optimization at the end, (at least of course you desing it to do this, thing that I didn't do, cause i have not realy experience with 3d big programming). And as this guy said "We have not easy acces to a core Quad in Iran you know!" :D. So in my country.

Anyway, counting even that problem, irrlicht rocks to make in short time programs that are more than a simple render and the graphic part is just ONE single part. I will continue using it cause I think irrlicht will become one of the best 3 engines and IS EASY TO USE !!! :D.

Posted: Thu Apr 17, 2008 10:29 pm
by Halifax
When people say "early optimization is the root of all evils", they are referring to the fact that people are accusing code of bottlenecking, when they don't even know if it is true.

It is truth that some sort of occlusion culling is needed, so that isn't "early optimization", it is needed.

But saying that your FPS dropped by 100, and you are displaying only 20 more meshes, and want to know how to optimize it. That is early optimization.

Posted: Fri Apr 18, 2008 7:34 am
by rogerborg
To be pedantic, premature optimisation is the root of all evil.

Prophylactic design is not premature optimisation. Identifying high order operations, and deciding whether they are significant, is not premature optimisation.

Posted: Fri Apr 18, 2008 11:48 am
by arras
First of all, whole computer programing and game programing specially was, still is and always be full of optimization and shortcuts. You can't dream of producing good software without it.

If you know that you better start to think about it early on. You do not have to implement it immediately but it should be part of your concept.

Second, Irrlicht is just graphic engine and as such it does optimization graphic engine can do: culling, mipmapping and such. Sure you can always think of ways of improving it but you can't expect much more. More optimization can be done only based on specifics of your application and as such can be made only by you.

They should have thought at last about LODs and some sort of static scene node manager for objects like trees and plants. But thats something you have to incorporate in to your design at the beginning.