What's left to be done?

  • Thread starter Thread starter Drahcir
  • Start date Start date
Status
Not open for further replies.
Writing up a tutorial for reverse engineering a game is 1000x harder then it sounds.
Is that so?? If I can't even make a new custom model then I can't help Akari. Damn I'm dumb. :-\
 
Added exporter for effect's mesh from RAM.

00016y27
 
Writing up a tutorial for reverse engineering a game is 1000x harder then it sounds.
In real most part of reversing is easy =)
I just translate asm into pseudocode and then analize it. I have description of most addreses already so it's not that hard.
 
I agree with seeing Sl's scene in FF7 (if possible at current state) that'd be an amazing teaser haha ;D
 
I agree with seeing Sl's scene in FF7 (if possible at current state) that'd be an amazing teaser haha ;D
Yup. I create this teaser as soon as i can =)

My current plans are:
2 step:
- Particle system (in progress)
- Resource management and organization (in progress)
- Battle models exporter (in progress)
3 step:
- basic field and battle moduls
- loaders for field and mattle moduls
- walkmesh and character controller work together
4 step:
- unified lua scripting system
- animation commands
- battle and field scripting
 
I agree with seeing Sl's scene in FF7 (if possible at current state) that'd be an amazing teaser haha ;D
Yup. I create this teaser as soon as i can =)

My current plans are:
2 step:
- Particle system (in progress)
- Resource management and organization (in progress)
- Battle models exporter (in progress)
3 step:
- basic field and battle moduls
- loaders for field and mattle moduls
- walkmesh and character controller work together
4 step:
- unified lua scripting system
- animation commands
- battle and field scripting
So, on a theoretical note. Since this is basically a whole new game engine. Could one say.. Ohh... Add HDR and And bump mapping? just curious. No need to implement it of course
 
I agree with seeing Sl's scene in FF7 (if possible at current state) that'd be an amazing teaser haha ;D
Yup. I create this teaser as soon as i can =)

My current plans are:
2 step:
- Particle system (in progress)
- Resource management and organization (in progress)
- Battle models exporter (in progress)
3 step:
- basic field and battle moduls
- loaders for field and mattle moduls
- walkmesh and character controller work together
4 step:
- unified lua scripting system
- animation commands
- battle and field scripting
So, on a theoretical note. Since this is basically a whole new game engine. Could one say.. Ohh... Add HDR and And bump mapping? just curious. No need to implement it of course
Everything is posible.
 
Now that's interesting, how is the poly count on this model?
Too much on my opinion. It can be a lot of simpler and it doesn't reduce quality. For render it's ok. Though it's harder to skin it =)
 
I agree, but as I previously mentioned, if you're able to implement measures that can be seen as better than the current FF7 getup, i.e dynamic lighting, reflect maps, dynamic cameras etc, only then will it be viable to create realtime-friendly models for the scenes. For now, I think, using the more detailed models and high render settings will produce an overall better looking game.
 
I agree, but as I previously mentioned, if you're able to implement measures that can be seen as better than the current FF7 getup, i.e dynamic lighting, reflect maps, dynamic cameras etc, only then will it be viable to create realtime-friendly models for the scenes. For now, I think, using the more detailed models and high render settings will produce an overall better looking game.
If someone here can create 3d like here on this image then I gladly spend all my time to add this features. But I don't believe in this.
991396_20100312_790screen001.jpg

This is scene from game we developing now (Battle vs Chess). Whole scene takes 3 weeks for an artist to finish. Scene contains twice less polygons than this scene from ffvii including all particles.

ps: I really hope I mistakes and we can produce ffvii levels at such quality.
 
I'm sure the creative and skilled minds within Team Avalanche could eventually adapt to creating such great looking grapics. However, the only problem I see here is the time scale of it all.

Of course, scenes such as the one depicted in that image require more complicated 3D techniques. This means the Team Avalanche graphic artists including myself would have to adapt to the new techniques, and amost all previous work on the scenes would need to be started from scratch.
What we'd need if we were going to switch from traditional style to Q-Gears style is to know exactly what we'd have to do (obviously modelling, texturing, lighting maps?), and what has to be done via the engine (i.e lighting) so that once these updates are implemented we wouldn't need to update work we'd already done. Is it possible to place a theoretical ETA on implementing the basics of a few of these awesome features? Because if it's going to be over the course of a few months/ even years, it could delay the release of the bombing mission demo and even future TA releases if we were to switch over

But as good and as epic sounding as all this is, it's still a little conflicting with the main aims of TA, which is "stick to the original as much as possible", and hence 2D field scenes. Personally, I wouldn't mind switching over to 3D at all if it was up to me, but it's most definately not.
 
I agree, but as I previously mentioned, if you're able to implement measures that can be seen as better than the current FF7 getup, i.e dynamic lighting, reflect maps, dynamic cameras etc, only then will it be viable to create realtime-friendly models for the scenes. For now, I think, using the more detailed models and high render settings will produce an overall better looking game.
If someone here can create 3d like here on this image then I gladly spend all my time to add this features. But I don't believe in this.
991396_20100312_790screen001.jpg

This is scene from game we developing now (Battle vs Chess). Whole scene takes 3 weeks for an artist to finish. Scene contains twice less polygons than this scene from ffvii including all particles.

ps: I really hope I mistakes and we can produce ffvii levels at such quality.
I think Square-Enix will have remade Final Fantasy VII for the PS3 or PS4 by then.
 
Actually, 3 weeks is less that the average it's taking us right now the way we're currently doing it xD
 
Now that's interesting, how is the poly count on this model?
Too much on my opinion. It can be a lot of simpler and it doesn't reduce quality. For render it's ok. Though it's harder to skin it =)
I agree, but if we're going to give this a go as a proof of concept we need the poly-budget.

I'm sure the creative and skilled minds within Team Avalanche could eventually adapt to creating such great looking grapics. However, the only problem I see here is the time scale of it all.

Of course, scenes such as the one depicted in that image require more complicated 3D techniques. This means the Team Avalanche graphic artists including myself would have to adapt to the new techniques, and amost all previous work on the scenes would need to be started from scratch.
I don't think you'll have to start from scratch...

If someone here can create 3d like here on this image then I gladly spend all my time to add this features. But I don't believe in this.

ps: I really hope I mistakes and we can produce ffvii levels at such quality.
How tempting.....:-D Is it real time rendered?
 
If someone here can create 3d like here on this image then I gladly spend all my time to add this features. But I don't believe in this.

ps: I really hope I mistakes and we can produce ffvii levels at such quality.
How tempting.....:-D Is it real time rendered?
Yup. This is real time rendering. More screens at http://www.ag.ru/screenshots/battle_vs_chess. Game will be released by the end of summer.

In static of this scene is 50000 polygons (but this is square scene and you can look all way around). Models has aproximately 20000 polygons each. Whole scene takes 300000 polygons one pass and about 1000000 polygons to render as whole (including all particles, shadows, depth-pass and so on). Models are the most heavy here.

I think nmkin5 scene could be redused to 5000-10000 without losing quality at all. And needs to be reduced to 1000-2000 in real. Then for game, i think, nmkin5 nmkin4 nmkin3 needs to be created as one model. They use same material and same geometry. In 3d it will be posible look around (if we don't deside lock camera to certain angle).

Ok for now I return to my battle exporter. There is so much left to be done =)
 
Last edited:
I think camera angles should remain close top the smae, except maybe for the ability to SLIGHTLY move it, I mean slightly(like holding down in super mario)
 
Status
Not open for further replies.
Back
Top