Full Games List
Featured Games
New Games
Reviews
Media
Images
Blogs
Game Updates
Images Feedback Summary
Game Design Highlights
Tutorials
Articles
Scripts
Utilities / Plugins
Resources
Engines
Engine Bulletins
What is Makerscore?
Past Events
Achievements
Misaos
Forums
Post History
Chat
New to RMN?
Full Store List
RMN Music Pack (free)
WHY RM2K3 IS TERRIBLE FOR ACTION GAMES
Posts
Darken
3952
Darken
View playlists
Closeauthor=Itajuauthor=LockeZ++++++++++
"It doesn't bother me when the gameplay is shitty as long as <x>" isn't a valid argument. Moving the game to an engine where the gameplay isn't as shitty doesn't hurt any other part of the game.
And your last statement is simply flat out wrong and he explained why, oh my god, did you even read the post you're responding to, have you ever used RM2K3 for anything? You cannot detect pixel coordinates of events in RM2K3, you cannot detect if an event is currently in the process of moving between two tiles, it is not within the engine's capabilities. You fail at arguing.
I am surprised that this topic comes up shortly after Darken saw and commented on the video of my Rm2k3 Action based RPG.
Invalid YouTube URLhttp://www.youtube.com/watch?feature=player_embedded&v=o35eAsrmjjE
OMG and you wouldn't believe it when I say this game is tile based AND is a lot of fun (and I just do say this as I am the creator.
yeah ill just take your word for it, and uhhh I enter this argument a lot with people in random topics so i felt maybe some kind of discussion could be made on an internet forum
yeah dude, i agree with you darken. even as an infant user in a rm program you will understand the various limits.
gw days carius and co would breach the rm scene with various new challenges and i thought it was pretty interesting. but ya, basically this though dude:
author=Ocean
All you 5 rm2k3 users please take note!
rm is great for basic design, but coding, gm, and construct is probably the way to go.
rm is great for basic design, but coding, gm, and construct is probably the way to go.
Yeah, I still kinda like RM2k3 but those limitations are too much, I had to move on. I did RMVX/Ace for a bit but I think I'm gonna try Construct more once I get some more free time.
Ghost
787
Real-time action battle systems require precision. Using rm2k3 will automatically limit you or will require you to spend infinite amount of time going around the limits of the engine. Regardless, you can use just about anything else and be better off. RMXP will most likely lag in a heartbeat. With VX / VX Ace I don't know I guess you can script something in, I still haven't seen anything that great done with those engines yet (action battles, I've seen a few...).
Construct, Gamemaker, maybe even Engine001? I don't know, when you want to make something you pick the most suitable engine or create your own.
Construct, Gamemaker, maybe even Engine001? I don't know, when you want to make something you pick the most suitable engine or create your own.
Darken
3952
author=Ocean
All you 5 rm2k3 users please take note!
does that mean like 20-30 active people in this community makes everything here even more redundant??
author=Ghost
RMXP will most likely lag in a heartbeat. With VX / VX Ace I don't know I guess you can script something in, I still haven't seen anything that great done with those engines yet (action battles, I've seen a few...).
That leads to another thing actually. There is a goto script that you can paste into the later RPGMakers called Blizz ABS. But from what I've seen, most games made with it have vague hit boxes (play sword01.png animation when action key is pressed and hope the actual attack connects) and overall just spammy combat which has more to do with the implementation than the capability of the engine. I have not actually used blizz so whatever.
Scalytank
462
I have tried blizz. You can do great things with that, but its not optimized (or it is, but there are too many features), and coded in rmxp... i was lagging a lot.
I had a custom system in rmxp in the past. The missiles used pixel perfect collision with the player. Melee attacks first check for an enemy in front of the player, and if the script couldnt find any, it also checks diagonally. This way, if the player is sorrounded, all the enemies can attack him. Also, we utilized fast tile movement, and good AI. It was lagging a lot, so we moved the project into ika. A few people tested that, and they didnt run into any issues, with " precision ". This is what i talked about in the first place, that one can make both system good, and both system bad.
Now, this is my opinion. Dont take this as an arguement. Just personal preference, and experience.
I had a custom system in rmxp in the past. The missiles used pixel perfect collision with the player. Melee attacks first check for an enemy in front of the player, and if the script couldnt find any, it also checks diagonally. This way, if the player is sorrounded, all the enemies can attack him. Also, we utilized fast tile movement, and good AI. It was lagging a lot, so we moved the project into ika. A few people tested that, and they didnt run into any issues, with " precision ". This is what i talked about in the first place, that one can make both system good, and both system bad.
Now, this is my opinion. Dont take this as an arguement. Just personal preference, and experience.
AznChipmunk
4613
Hi! This is my story.
I was once very interested in trying to make an action-based RPG style game in RPGMaker 2003. You can see some examples of that: here, here, and here. These 3 are only a small selection of a large collection of action-based games/demos that I released or worked on over the years I've been involved in the community. I've attempted to create such an action system at least 15-odd times, each time using different inspirations, ideas, mechanics, controls, etc. I would finish the system, then ditch the project entirely for a new one.
After playing with them for a long time, I realized something: I was wasting my time!
The inherent flaws in the program do not allow you to replicate the accurate game feel that many of our favorite action games do have. There has not been a single RPGMaker 2003 game to replicate the proper "game feel" and smoothness that games like Link to the Past provide. I'm NOT saying your game is supposed to be Link to the Past, or copy its exact mechanics, but almost every single GOOD handheld or console action-game has a tightness of control and feel. This is where games like Link to the Past, or even games like Mega Man, Soul Blazer, and Terranigma, shine and make them fun and playable. Of course, these flaws only appear when you are trying to bend to the program to do something it wasn't meant to do!
The problem itself is not the tile-based movement. The tile-based movement, however, does compound with the real issue, which truly is the collision detection. Every single 2k3 action game out there is either extremely easy, extremely difficult, or extremely tedious. No game has the naturalistic, smooth control of a real action game and battles always feel very awkward and terribly forced. They're either some sort of cheap button mash or some tediously slow crawl. At the point at which you have the ability to create something like a pixel based movement, you'll have this incredibly unwieldy and tedious system. When you're at that step and at that level, why don't you just move onto a higher-level program with real action capabilities? It's fun to bend the limits of 2k3, but some things... just weren't meant to be bent.
Once I realized this fact, I was able to peacefully move on with my life. So go on, you can move on with your lives too.
The end.
-an "ABS veteran"
I was once very interested in trying to make an action-based RPG style game in RPGMaker 2003. You can see some examples of that: here, here, and here. These 3 are only a small selection of a large collection of action-based games/demos that I released or worked on over the years I've been involved in the community. I've attempted to create such an action system at least 15-odd times, each time using different inspirations, ideas, mechanics, controls, etc. I would finish the system, then ditch the project entirely for a new one.
After playing with them for a long time, I realized something: I was wasting my time!
The inherent flaws in the program do not allow you to replicate the accurate game feel that many of our favorite action games do have. There has not been a single RPGMaker 2003 game to replicate the proper "game feel" and smoothness that games like Link to the Past provide. I'm NOT saying your game is supposed to be Link to the Past, or copy its exact mechanics, but almost every single GOOD handheld or console action-game has a tightness of control and feel. This is where games like Link to the Past, or even games like Mega Man, Soul Blazer, and Terranigma, shine and make them fun and playable. Of course, these flaws only appear when you are trying to bend to the program to do something it wasn't meant to do!
The problem itself is not the tile-based movement. The tile-based movement, however, does compound with the real issue, which truly is the collision detection. Every single 2k3 action game out there is either extremely easy, extremely difficult, or extremely tedious. No game has the naturalistic, smooth control of a real action game and battles always feel very awkward and terribly forced. They're either some sort of cheap button mash or some tediously slow crawl. At the point at which you have the ability to create something like a pixel based movement, you'll have this incredibly unwieldy and tedious system. When you're at that step and at that level, why don't you just move onto a higher-level program with real action capabilities? It's fun to bend the limits of 2k3, but some things... just weren't meant to be bent.
Once I realized this fact, I was able to peacefully move on with my life. So go on, you can move on with your lives too.
The end.
-an "ABS veteran"
Blitzen
3253
@LockeZ You can do the action game all in ONE loop, like your regular hard-coded program. You don't NEED a parallel process anywhere. Just a 0.0s stall (equivalent of the NES vblank) for the input and movement to take place. Advanced systems in RM2k3 actually run better without any use of parallel process, because the commands function without overlap.
The lag on the move events is ughghg but this works best for picture-based menus/minigames, in my experience.
Start System
Loop/Label
<>Key Input
<>Wait 0.0 (x2?)
<>Execute
<>Move Pictures (No Wait) (They'll actually move in the 0.0s gap the next time the loop is called)
End Loop
End System
The lag on the move events is ughghg but this works best for picture-based menus/minigames, in my experience.
An argument about RM2K3 and action games? I'm going to do Clyve and Karsuman a favor by not saying a thing about action games made in RM2K3.
Is anyone in this topic considering making an action game, and possibly moving to Unity? Learning a professional engine was one of the best decisions I've ever made :P
The learning curve is higher but you can make truly fantastic-feeling action games.
Programming isn't for everyone though; I enjoy it but it grates on me at times.
The learning curve is higher but you can make truly fantastic-feeling action games.
Programming isn't for everyone though; I enjoy it but it grates on me at times.
Nightowl
An argument about RM2K3 and action games? I'm going to do Clyve and Karsuman a favor by not saying a thing about action games made in RM2K3.
huh
Killer Wolf
1235
I found myself in an "old school" video game discussion with a friend of mine. He brought up how good he thought Secret of Mana was. I made a point of trying it out, and my first thought during combat was "This feels like it was done in Rm2k3, and worse still, I know how to do this in Rm2k3"
Never played Link to the Past, so I'm light on the gold standard of Action games I guess.
I've used Rm2k/3, Construct, and Multimedia Fusion 2 to make action rpg test demos. Believe it or not, each one has an advantage that the others lack.
MMF2 is great and easy to use (but can become unstable during multiple map changes - though I have a work around for it) but you have to set up your own pathfinding.
Construct's built in path finding is fairly decent, but it still gets dumbstruck every now and then, and you'll feel like you're back in 2k3. (This was a while ago, maybe a year since I last messed with it. Subsequent updates may invalidate this statement.)
Rm2k3 has basic pathfinding, on touch events, the easiest text boxes out of all of the options, but it is still very limited and clunky.
I've built platformers rm2k3, both tile and pixel based movement. I learned that with the latter, it is best (as Blitzen brought up) to run all of your actions out of a central line. I had one pipeline event that triggered everything else from sound to enemy actions and animations. Lag started to become very noticeable with each additional element, meaning you'll hit your maximum onscreen element ceiling pretty quick.
The tile based one I did was real early, and subsequently really crap.
Using PowerMode2k3, I put together a point and click adventure demo, with working inventory, drag-able menus, and a way to take pictures of screens in the game via the detective's palm pilot. It needs a new story, but the base engine is ready to go and very modification friendly.
Lastly, I have a first person dungeon crawler built in 2k3. (I've extrapolated my method to function within VX Ace as well).
There was a quote from my GW days, and I believe Carius was the originator, that had to do with (warning:I'm paraphrasing) "preferring to make a sword out of a loaf of bread instead of buying a new sword."
I know 2k3 is firmly in technological dinosaur territory, but I still enjoy seeing people push it into places it was never intended to go.
Never played Link to the Past, so I'm light on the gold standard of Action games I guess.
I've used Rm2k/3, Construct, and Multimedia Fusion 2 to make action rpg test demos. Believe it or not, each one has an advantage that the others lack.
MMF2 is great and easy to use (but can become unstable during multiple map changes - though I have a work around for it) but you have to set up your own pathfinding.
Construct's built in path finding is fairly decent, but it still gets dumbstruck every now and then, and you'll feel like you're back in 2k3. (This was a while ago, maybe a year since I last messed with it. Subsequent updates may invalidate this statement.)
Rm2k3 has basic pathfinding, on touch events, the easiest text boxes out of all of the options, but it is still very limited and clunky.
I've built platformers rm2k3, both tile and pixel based movement. I learned that with the latter, it is best (as Blitzen brought up) to run all of your actions out of a central line. I had one pipeline event that triggered everything else from sound to enemy actions and animations. Lag started to become very noticeable with each additional element, meaning you'll hit your maximum onscreen element ceiling pretty quick.
The tile based one I did was real early, and subsequently really crap.
Using PowerMode2k3, I put together a point and click adventure demo, with working inventory, drag-able menus, and a way to take pictures of screens in the game via the detective's palm pilot. It needs a new story, but the base engine is ready to go and very modification friendly.
Lastly, I have a first person dungeon crawler built in 2k3. (I've extrapolated my method to function within VX Ace as well).
There was a quote from my GW days, and I believe Carius was the originator, that had to do with (warning:I'm paraphrasing) "preferring to make a sword out of a loaf of bread instead of buying a new sword."
I know 2k3 is firmly in technological dinosaur territory, but I still enjoy seeing people push it into places it was never intended to go.
This is slightly off topic, but I'm sure you're all familiar with IG Maker, which advertises that it can be used to easily code action RPGs:
http://www.rpgmakerweb.com/product/ig-maker
I haven't used this program myself, but I've read mixed reviews. To anyone who has used it; does it solve any of the problems Darken brought up?
http://www.rpgmakerweb.com/product/ig-maker
I haven't used this program myself, but I've read mixed reviews. To anyone who has used it; does it solve any of the problems Darken brought up?
