Hero of Allacrost Forums

You don't think that's emulating Chrono Trigger a bit too much?

gorzuate wrote:You don't think that's emulating Chrono Trigger a bit too much?

I don't think that's really an issue. I mean our original battle system was a close "emulation" of Final Fantasy. The important thing is that we have some very different features of our battle system; that's what makes our game unique.

Moving this discussion right along, let me reiterate why we're even discussing this:

Roots wrote:THE PROBLEM
This proposal was made because our current battle system is very taxing on our artists. The sprites are very large and doing animations with them using traditional 2D methods has proved to be fruitless due to the complexity.

Basically, it's taking way too long, so we want to speed up development as much as possible. Of the original 2 methods proposed, I believe the 3D models are out of the question. Leaving Jetryl's "CT" method, and some combination of map sprites and enemy battle sprites with the current battle system.

Jetryl's "CT" method is less taxing on artists, but is probably more taxing on programmers.
The other method is more taxing on artists, nothing has to change code-wise.

Over time we have come to learn that finding coders is a piece of cake, and finding artists is PITA. Given this knowledge, we can implement Jetryl's "CT" method and push out the first release of the actual game faster than any other method. Coders abound, find some and sic 'em on battle mode, and the code can be done before Jetryl (and any help he may get) can complete all the required artwork.

Of course, split this off into a separate branch of the SVN repository. That way we'll still have our current battle system, and when (not if ) the game gets popular enough to attract hordes of artists, we could always revert back to the current battle system and put it in as a "new" feature.

Or not. Maybe we'd have a big revolt on our hands. But nothing has to be permanent. This could just be a stop-gap solution for now that could end up becoming a mainstay of the game or end up being replaced by an "improved" battle system.

Seriously, let's get something moving here.

(I am partial to the (to-be-) animated battle backgrounds we've got. )

Jetryl wrote:Here's the picture, and this is what I'm suggesting. As you can see, there's tons of space for different characters; easily enough for the 4 we were slated to have in the current design. This should give you an idea of what I'm gunning for.

Tons of space? You have all the character and enemy sprites squished together in a very small section in the center of the screen, and all that other area (walls, pits, etc) where the sprites can not go is for the most part wasted space. Not to mention that in that particular example we will either have to restrict who can target who or will we have to allow for sprites to overlap one another. Not an idea solution IMHO.

gorzuate wrote:Over time we have come to learn that finding coders is a piece of cake,

Correction: finding coders to apply to be on our team is a piece of cake. Finding competent programmers who make significant and regular process on the code is a completely different story. Those people are hard to find, and to date I'd say we only have had maybe a total of 10 of these people on this team. I don't want to invest in something that is very taxing on the programmers and the overall battle system design, since we're already far behind right now and we only have a small portion of our programming team who are actively making commits on a regular basis. We don't have the luxury of dedicating resources to "experimentation" with the different designs.



Back to the major subject at hand, yes I think we need to start casting our votes here like gorzuate suggests. My vote for each design point (along with a short explanation supporting my decision) is below.

Character sprites
Vote: #1. Use the character's map sprites, enlarged 2x in size, for battles

Reason: We want to have a nice handful of animations for our characters, and its been shown that they are far too complex to do for 64x128 pixel sprites. Furthermore, the 3D to 2D proposal to ameliorate this process seems to not do us any good, since we didn't really receive any serious 3D artist applications in the past several months.


Enemy sprites
Vote: #2. Use separate, detailed battle sprites for enemies along with damage frame blending.

Reason: Despite claims otherwise, I am not convinced that doing numerous animations for each enemy sprite would be significantly less taxing than doing a single detailed sprite frame and 3 derivative damage frames. Furthermore, comparing the likes of FFVI and CT as a basis, I don't see much merit in CT's extremely simplistic and basic animations. To me personally, I would rather be battling vicious-looking, detailed enemies than the rather cartoon-ish looking map sprites. Also I have an affinity for the damage blending feature.


Battle environment
Vote: #2. Battles occur on detailed background images

Reason: Two reasons: 1) I enjoy have different (and gorgeous) scenery to look at in battle rather than being "stuck on a map" the entire game. 2) As I pointed out before, there are severe complications for both programmers and map designers by having battles occur on the map.

Sprite orientations
Vote: #2. Sprites battle in pseudo-2D (characters on left, enemies on right)

Reason: It only makes sense to have our battle in full 2D movement if we are going to actually take advantage of it for our design. For example, allowing the player to control the position of their characters, or area of effect skills. Otherwise, it adds little to the game, and most importantly it requires TWICE the amount of artwork for the game than would be needed otherwise! (refer back to the original purpose of this whole discussion)


So that's it. Those are the design points that I am voting for, and I don't think I can be convinced otherwise anymore. I've weighed the pros and cons of both sides, analyzed criticism about my own proposals, and made my decisions based on what I felt would be best for this game. Everyone: please take the time now to seriously consider what design decisions you feel would be best for Allacrost.

gorzuate wrote:You don't think that's emulating Chrono Trigger a bit too much?

Or "Secret of Mana." Visually, it doesn't look too different from warcraft II, or five hojillion other games. About the only defining things about it are that: the map sprites are "randomly" arranged (not in neat rows on each side of the screen), and the map sprites are facing in different directions, rather than all being in single file, facing each other. Practically any top-down RPG like this, with it's battle characters randomly arrayed, would look like this.



Since we're on the subject, here's a direct comparison of how similar/different Allacrost would look compared to Chrono Trigger, Seiken Densetsu 3, and Final Fantasy 3 (6 in jp). Allacrost's map mode, in my suggested system, would view-wise be identical to its battle mode, just sans all the HUD stuff. Even -with- the zooming that I/rain/others have suggested, we'd still be about 1/2 - 2/3 the size of the sprites in these other games, and with about twice the visual detail (640x480 of game graphics, 1280x960 of HUD). It's enough to be a visual improvement in clarity over them, and yet share some of the same benefits of screen proportion.

Images abound:
[img:640:512]http://www.allacrost.org/staff/user/jetryl/Misc/zoomTests2/FF6-screen-size-battle.png[/img]
[img:640:512]http://www.allacrost.org/staff/user/jetryl/Misc/zoomTests2/FF6-screen-size-map.png[/img]
[img:640:512]http://www.allacrost.org/staff/user/jetryl/Misc/zoomTests2/allacrost-screen-size2.png[/img]
[img:640:480]http://www.allacrost.org/staff/user/jetryl/Misc/zoomTests2/allacrost-battle-mockup-small.png[/img]
[img:640:512]http://www.allacrost.org/staff/user/jetryl/Misc/zoomTests2/chrono-screen-size.png[/img]
[img:640:512]http://www.allacrost.org/staff/user/jetryl/Misc/zoomTests2/seiken-densetsu-screen-size.png[/img]

gorzuate wrote:Over time we have come to learn that finding coders is a piece of cake, and finding artists is PITA.

Roots is right that, currently, with our game in its unfinished, unproven state, it's hard to attract great coders. Experience from wesnoth and other OSS projects has taught me, though, that if the rest of the project is in order, good coders will come to help with ancillary things. They usually need the core of the existing game's code to be sane (they're not going to reimplement the video engine like roots just did), but they can still do some major things, and eliminate a lot of grunt work.

Gorzuate is right, however, that of all the different "human assets" we have as an OSS project, programmers are ultimately, strangely, in easiest supply. It might be part of the culture; that programmers have a major incentive to subscribe to, and have exposure to+knowledge of open source, whereas other usually know as much about this "open source" thing as they do about nuclear physics. It's unfortunate, too, because open-source seems like it would have such an affinity with the core values of art, but so many artists have art as their money-maker, rather than having it as art for its own sake.

Even now, wesnoth really doesn't have a lot of people working on its art. As far as regulars go, we've got about 7-9 people working on the code. On the art, we've got _maybe_ 3, including myself. Eventually, good artists will come, but you usually have to have the game already established to have that happen.

Honestly, the strangest thing Allacrost has is a dedicated musician. Wesnoth didn't have one of those for the longest time, and even now, Alexis/Timothy are kinda crapping out; West is really showing up their quantity of work.

gorzuate wrote:Of course, split this off into a separate branch of the SVN repository.

Absolutely. It'd be crazy not to, and it would give leave to try a few other experimental things.

In essence, Jetryl..

3 Years ago, there was a dicussion.

1) Chrono Trigger style?
2) Final Fantasy style?

the developers chose option number 2, THREE years ago. And, as such, the game has been designed around that idea.

Three years in, you, decide to further your own goals, and suggest the same choice again.

A choice that everyone who has signed onto the project in the past 3 years, has already *chosen*. They chose the Final Fantasy style.

3 years in, changing the core gameplay type, is insanity.

3 years.. in which a (j)crpg has been developed. To change the game to an action (j)crpg... is insanity. The change would virtually involve a rethink of the game from the ground up.

And, in spite of what you say, it would require *more* work for the artists. I should know. Before you say something about me knowing nothing about art.. I'm an artist, myself.

This discussion is for the purpose of simplifying the workload, not about a vast increase of it.

3 years in, mate. You don't do this kind of change 3 years in.

I'm feel pretty much the same way as Roots, I've been playing a bit of both CT and FF7 recently (FF7 basically is our current battle, just a bit more in 3D), and I just plain prefer that style (separate battle mode to fighting on the map) over the CT style. Using map mode character sprites would lessen the artwork cost of this mode as well.

KaelisEbonrai, your reasoning may have some merits to it, but it's worthwhile to note that 3 years ago we had no idea what the art requirements would be and whether they would even be feasible. If we set everything in stone from the beginning it's safe to say we would no longer be around.

KaelisEbonrai wrote:In essence, Jetryl..

3 Years ago, there was a dicussion.

1) Chrono Trigger style?
2) Final Fantasy style?

the developers chose option number 2, THREE years ago. And, as such, the game has been designed around that idea.

Three years in, you, decide to further your own goals, and suggest the same choice again.

A choice that everyone who has signed onto the project in the past 3 years, has already *chosen*. They chose the Final Fantasy style.

3 years in, changing the core gameplay type, is insanity.

3 years.. in which a (j)crpg has been developed. To change the game to an action (j)crpg... is insanity. The change would virtually involve a rethink of the game from the ground up.

And, in spite of what you say, it would require *more* work for the artists. I should know. Before you say something about me knowing nothing about art.. I'm an artist, myself.

This discussion is for the purpose of simplifying the workload, not about a vast increase of it.

3 years in, mate. You don't do this kind of change 3 years in.

Personally, I'm still on the fence with the idea of making such a drastic change to the system. That said, I think all of the developers here, regardless of feeling one way or another, are agreeing that it's at least up for discussion.

As for it being more work for the artists, I would be interested in your reasons. Jetryl has made a strong argument that it would required less work, and I, for one, would like to hear any counterpoints you have.

Enough.

I should not have discussed this until I was done with the code and images to make it. It has spawned a bunch of discussion which would have been annulled by that being done. I described this topic as "bullshit speculation" before, and I did so because nothing we decide here, or on any other matter, has any meaning. The only decisions that have meaning are those which get acted on - those which can get finished. Remember the fable of the mice discussing the idea of putting a bell around the cat's neck - their idea is useless unless one of them actually -does- the job. Thus far, we have had a failure to act on the existing design decision. We have had three years to complete it, and our accomplishments on battle mode have been pretty pathetic. This is in no small part due to the fact that I have not been helping with that (since I judged it to be unfeasible).

I don't like going against a team's decision, but when the team has decided to do something unfeasible, their decision isn't going to magically make it happen. I'm especially unwilling to do it when the team's decision is something that _I_ have to accomplish the bulk of the work on, and which I don't even like, and where my only payment for it is having a result I'm happy with.


I'm going to continue working on the other 90% of our game's content, and in the meantime, I'm going to churn out my own battle mode, perhaps with a few people's help, perhaps all on my own. When it's done, we can compare it, and the progress made on the current battle mode. That should serve to indicate which plan is more feasible. I want to do this, and it would be good practice for me, even if it doesn't become a final module of the game.

And, in spite of what you say, it would require *more* work for the artists. I should know. Before you say something about me knowing nothing about art.. I'm an artist, myself.

This is only true if we assume that the map-mode sprites would have very limited animations. I would be unsatisfied with that; they're going to have full movement in every direction, and will have animated standing frames to boot. This, in addition to other one-off animations of monsters performing various "idling" actions (hopping up and down, rooting in the ground, etc). That's what's needed to make map-mode interesting to look at.

About all we need on top of this, to make the battle mode I've described, is a single animation of them striking in each of three directions (N,S,Side). Additional animations would make things look nicer, but that's all we need, and that's already considerably more than our current battle mode has. That's pretty easy to do.

3 years.. in which a (j)crpg has been developed. To change the game to an action (j)crpg... is insanity. The change would virtually involve a rethink of the game from the ground up.

What we lose out of our existing content/code is:
- about nine still images of monsters standing.
- about 1-2 background images of battle scenes.
- one code module, which we can actually salvage about 30-40% of.

Hardly a "remake of the whole game".

I should not have discussed this until I was done with the code and images to make it

No, you shouldn't. The whole point of trying the new method is to see in the end which battle system people prefer. And by people I mean the general public who will be playing, not us. We already knew weeks ago who objected to this approach and why, so I'm not sure why this is even being discussed. Jetryl wants to try this so let him. Everyone else, go back to doing what you were doing and let this thread die.

one code module, which we can actually salvage about 30-40% of.

I doubt we'd lose even that much. We could still use the same time meter progression functionality, battle menu navigation, skill execution, item usage, etc. If anything, what the new style would be is the old way, just with more features piled on top of the existing code.

Jetryl wrote:

And, in spite of what you say, it would require *more* work for the artists. I should know. Before you say something about me knowing nothing about art.. I'm an artist, myself.

This is only true if we assume that the map-mode sprites would have very limited animations. I would be unsatisfied with that; they're going to have full movement in every direction, and will have animated standing frames to boot. This, in addition to other one-off animations of monsters performing various "idling" actions (hopping up and down, rooting in the ground, etc). That's what's needed to make map-mode interesting to look at.

About all we need on top of this, to make the battle mode I've described, is a single animation of them striking in each of three directions (N,S,Side). Additional animations would make things look nicer, but that's all we need, and that's already considerably more than our current battle mode has. That's pretty easy to do.

And, for that matter, considerably more art to produce.

1 battle-mode piece, perhaps with 3 or so damage-frames = 4 frames at max, per enemy.

compared to..

3-directional, 6-frame (or there abouts, give or take one or two) striking animation = 18 frames, not including possible stance animation, magic casting animations... per enemy, per npc that fights at all, per character

3 years.. in which a (j)crpg has been developed. To change the game to an action (j)crpg... is insanity. The change would virtually involve a rethink of the game from the ground up.

Jetryl wrote:
What we lose out of our existing content/code is:
- about nine still images of monsters standing.
- about 1-2 background images of battle scenes.
- one code module, which we can actually salvage about 30-40% of.

Hardly a "remake of the whole game".

You forgot a somewhat important piece.

Game Design, level design, and gameplay.

Action-(J)CRPGs don't play the same, and don't have the same design considerations as (J)CRPGs, the level design needs to, in essence, allow for combat within it. You /will/ need to rethink the design, action-(j)crpgs don't port directly over to (j)crpgs, without design changes.

gorzuate wrote:KaelisEbonrai, your reasoning may have some merits to it, but it's worthwhile to note that 3 years ago we had no idea what the art requirements would be and whether they would even be feasible. If we set everything in stone from the beginning it's safe to say we would no longer be around.

This is true, but I do believe that there are changes being discussed here that do not really serve the true purpose of this discussion (to reduce the artwork load to a manageable level). If we think about our past experiences with art difficulties in battle mode...

- We could not get decent animations for character sprites
- We had very little trouble getting enough enemy battle sprites
- We had little problem getting battle backgrounds available to us in time

Only the item in bold, I feel, is what really brought us to this discussion. Somehow it seemed like things were blown out of proportion, and suddenly battle mode wasn't allowed to have any original art at all, and we have these totally unnecessary (IMO) design change proposals that do not serve to address the main problem, but rather seem to serve an interest in a complete redesign of a major portion of this game.

I think we're all in agreement for using character map sprites in battles from now on. That should have been the end of the discussion, because now that we are decided on that (I think), we can continue working as usual. Anyway...

Jetryl wrote:I should not have discussed this until I was done with the code and images to make it. It has spawned a bunch of discussion which would have been annulled by that being done.

I'm sorry, but that is bullshit. The [url="http://allacrost.sourceforge.net/wiki/index.php/New_Staff_Guide#Staff_Policies"]principals that we work with[/ul] on this team are to collectively decide on a direction to take and head towards that goal as a team (read the link, it really does say that). What you said here suggests that each and every one of us so inclined should go off and implement there own ideas/design in secret, then come back with a finished result and say to everyone "this is now how this part of the game will be done". Its insulting to the rest of the people on this team to do something like that. Now, in this specific case Jetryl has already said what he plans to do completely on his own initiative and even though we haven't decided on this as a team yet, there's no reason why we should tell him "no". All we can say is "okay, but we can't guarantee anything you do will be implemented because the team hasn't come to a decision about this".

A little bit off topic, but this whole "team" mentality we have going is one of the reasons why many people stick around here for years. Working on a disorganized and incoherent team is incredibly frustrating, and that negative experience should be mitigated as much as possible. After all, we're here to design a game and have fun doing it.

Jetryl wrote:Thus far, we have had a failure to act on the existing design decision. We have had three years to complete it, and our accomplishments on battle mode have been pretty pathetic. This is in no small part due to the fact that I have not been helping with that (since I judged it to be unfeasible).

Three years. Did you know that we didn't even have battle mode working in the code until about after 1.5 years into the project? The fact that there was no battle code was sure as hell a good reason why the artists weren't motivated to produce content for it.

Jetryl wrote: I don't like going against a team's decision, but when the team has decided to do something unfeasible, their decision isn't going to magically make it happen. I'm especially unwilling to do it when the team's decision is something that _I_ have to accomplish the bulk of the work on, and which I don't even like, and where my only payment for it is having a result I'm happy with.

You seem to think that you will now and forever be the only artist on this team. That is false, and you know it. I won't ask you to work on anything you don't want to do, but you do not represent the entire art potential of this team. Hell, we have a new artist joining us this week. If you don't like the battle artwork, you don't have to work on it. Ever. End of story. But don't think that just because you're the only active artist we have at the moment gives you the supreme power to say "I will only make this type of artwork, and hence the design must work in the way I want it to". I'm the god damn dictator and not even I have that power (and I shouldn't!).


And about continuing to label everyone else's ideas as "unfeasible", you have not given much support for that at all. I agree with you now (as most others do I believe) that our original plan of large, animated character battle sprites was unfeasible, yes. But I strongly, strongly disagree with you that: our current enemy battle sprites are unfeasible to produce, that our battle backgrounds are unfeasible to produce, and that continuing a psuedo-2D battle environment is unfeasible to produce (hell, its a lot damn easier than the full 2D motion you've been pushing). Pardon my speculation, but I have become increasingly concerned over the past week that you are continuing this crusade because you personally do not enjoy the game's battle system and want to modify it to your own likings. If you can search your heart and assure me otherwise, that would be great.

Jetyrl wrote:What we lose out of our existing content/code is:

Its not really about what we lose here (which I admit will not be much). It will be about the additional work that will be required. Including:

- More work involved in designing maps with encounters
- Adding full 2D motion support into battles
- Complex pathfinding and collision detection for moving sprites (so they don't appear to walk over cave walls, for instance)
- Ensuring that sprites are not placed over pits or other unpassable terrain
- Doing an series of animation for every enemy sprite in the entire game as opposed to one larger sprite frame with a few damage variants.


Once again, I'll state my belief that you are vastly underestimating the complexity and implications of your designs here. Its not a "plug and play" change (although re-using some of the battle code will be that easy...writing the new code absolutely will not).



* I am not angry and meant no malice or harmful intentions in this post, Jetryl et. al. I am calmly and objectively stating what I believe to be correct, so please don't misinterpret the meanings of my writings here.

ChopperDave wrote:The whole point of trying the new method is to see in the end which battle system people prefer. And by people I mean the general public who will be playing, not us. We already knew weeks ago who objected to this approach and why, so I'm not sure why this is even being discussed. Jetryl wants to try this so let him. Everyone else, go back to doing what you were doing and let this thread die.

I have to agree with CD here. I thought the purpose of this thread was what design do we use for now, since we're obviously stuck and can't bring the art content for battle mode forward. And since Jetryl is determined to implement his own designs no matter the cost, I propose we pursue the solution that does the best job of ameliorating the artwork production cost while minimizing the amount of additional design and programming work for battle mode. I feel that that is the best way we can keep the battle mode development train moving. Therefore, I propose:

Keep everything in battle mode the same as it is for now, but remove the separate battle sprites for characters (but not enemies) and replace those with the character's map sprites enlarged by 2X.

Can we at least agree on doing that for now?

Roots wrote:

ChopperDave wrote:The whole point of trying the new method is to see in the end which battle system people prefer. And by people I mean the general public who will be playing, not us. We already knew weeks ago who objected to this approach and why, so I'm not sure why this is even being discussed. Jetryl wants to try this so let him. Everyone else, go back to doing what you were doing and let this thread die.

I have to agree with CD here. I thought the purpose of this thread was what design do we use for now, since we're obviously stuck and can't bring the art content for battle mode forward. And since Jetryl is determined to implement his own designs no matter the cost, I propose we pursue the solution that does the best job of ameliorating the artwork production cost while minimizing the amount of additional design and programming work for battle mode. I feel that that is the best way we can keep the battle mode development train moving. Therefore, I propose:

Keep everything in battle mode the same as it is for now, but remove the separate battle sprites for characters (but not enemies) and replace those with the character's map sprites enlarged by 2X.

Can we at least agree on doing that for now?

Agreed!


Just ignore what I'm working on, you can see it when/if I'm done, and then and only then should we discuss it. Until then, just put it out of mind.

KaelisEbonrai wrote:You forgot a somewhat important piece.

Game Design, level design, and gameplay.

Action-(J)CRPGs don't play the same, and don't have the same design considerations as (J)CRPGs, the level design needs to, in essence, allow for combat within it. You /will/ need to rethink the design, action-(j)crpgs don't port directly over to (j)crpgs, without design changes.

I agree with your points that there is a lot to consider here... but a couple of things to keep in mind:

Allacrost isn't becoming an action/RPG. It's an RPG. Chrono Trigger wasn't an action RPG, nor was Breath of Fire 3, nor Shining Force...

As for level design... the levels haven't been designed yet, so I don't see why that's an issue at this point.

Either way, I definitely agree that we should not waste energy fighting over this design decision. Let's concentrate on doing productive stuff, like getting the demo out.

Again some underestimate the time needed to make sprites the small ones can need 2 hours depending how much thought the artist put into it.
For the bigger ones you could actual draw them like a real painting than it would a) get done faster b) looking better but spriting such a big monster is a) a pain in the ass and b) it still looks sprited.

Conclusion you would invest more time in something that doesn't even looks better and could be done faster

Animating for the smaller ones are a piece of cake once you got the different directions right because within this small size the details don't get that difficult and you can reuse many existing parts. While on bigger sprites every little detail is visible.
Example:
[img:368:220]http://www.foolstown.com/pix/progr/dranim.gif[/img]
The man who did this is an expert still the whole animation looks stiff he has changed many small details but for the human eye it is still not enough. Don't mention the time needed just to produce a sprite of this size.

Or you get an 3d artist which are even harder to get than spriters.

The last if you want to compare coding to art than you should note that
while both is art a drawing can't be changed like code or reused with different variables.

Well, the thing is:

1) You've got a 3d Artist. (Hi there!)
2) If its easier and it'll look better painted.. DO IT PAINTED.
3) The comment on it can't be used with different variables is.. not always true.

Materials in 3d stuff can indeed be changed like variables.
POVRAY stuff blurs the line between code and art, and.. certainly can be used like code.

I personally prefer the painted look. This is my favorite example of that:

[img:320:256]http://www.allacrost.org/media/art/sprites_battle_aurum-drakueli.png[/img]