after messing around with the character skin for a bit I found a way to fit two arms and legs into the original size without rotating or splitting up any limbs,(but it does require that that the two body sides are rotated). Unfortunately, i have no idea how to upload images
Ok seeing your map with the leg all mixed up and sideways reminds me of (Obscure reference coming up here)
In Yu-Gi-Oh when they played that game with the dice where you had to unfold the die into a shape and fit it onto a grid to summon a monster and at the very end of the episode he managed to get one last die to fit perfectly into that little sideways space and win.
Please tell me someone else remembers this too so I don't sound like a wierd crazy person
I don't even know how I remember that, it has been years since I even thought about that show... whatever, I would totally support having a bigger image to work with.
I am overjoyed that my thread got necro'ed with such positive support (thank you!). I re-examined my proposal and agree, that there has to be a better solution than rotating the leg bits they way I originally suggested. I started playing around with a bigger texture size, but that ultimately went down a path that didn't look promising. Trying to keep everything together got complicated when I had to factor in using the same mapping for armors. Lots of wasted space came out of that.
So I played Tetris with the player skin some more and came up with this beauty. Behold!
Pros:
+ independently textured arms and legs
+ maintains the original 64x32 size
+ keeps all sections grouped together in a fairly intuitive way.
+ sub sections follow a mapping VERY similar to the original (with only some parts slightly shifted)
+ NO HACKY SUB SECTION ROTATION (improvement over my first attempt)
+ can still apply to zombie, pigmen, and zombie pigmen plus all armors
+ ONLY code change to minecraft is the class which translates the mapping onto the models.
Cons:
- just one... actually converting existing files over to the new format.
I gotta say, I'm really surprised that I didn't see this earlier... the mapping here is near perfect. It even manages to keep a lot of the same arrangements of each section as the original mapping. The head for example, follows the same pattern of right, front, left, back, as do the arms/legs and the body is actually completely unchanged as well! Each body part section manages to remain grouped together for the most part.
As stated, I'm pretty confident in this latest version. So now it falls back to the one remaining downside to this entire suggestion: conversion form old to new formats with existing textures.
I am still confident that this will work given enough early warning, a long period of preview exposure from snapshots, and a utility to automatically convert old skin files to the new format.
[Edit]
I also decided against changing the texture size for another reason... Consider that I am proposing a change to an aspect of minecraft that is already well established. In my opinion, for this to have any chance of being considered by Mojang, the change must be as simple to implement as possible - ideally with few actual changes required. You know the saying, if ant ain't broke... well, technically this ain't broke - but it could be improved. An yet it is all too easy to go overboard and let a simple suggestion grow until it is too complex to be considered "simple" anymore. At which point it runs the risk of being dismissed out of hand.
This suggestion at is core only has one requirement - changing the single part in the code that controls where the mapping is applied to the player/zombie/armor model. Everything else is upon the community to update their skins. You really can't get more simple than that.
[Edit2] Pikmin Red pointed out a typo on the image labeling (though it seems the forums update ate that post). Corrected now!
I was talking to someone who I know who has a bit of Java knowledge, and while I can't verify what he said as fact he thinks it is possible for Java to automatically update skins to the new version.
Starting by looking at the skin, it would check to see if it is an old MC skin by checking specific sections for image data. For example, it could check the leftmost side to see if there is a 10x32 set of transparent pixels (which all current skins should have, provided someone didn't fill in all the extra blank space for no reason), and only initiate the skin change if the section it looked at matches what should be there.
He also said that Java could also edit the image directly. It can move specific parts of an image into new sections and duplicate parts of it to other areas. Therefore it could in theory update any new skin to this new format on it's own. I know this isn't exactly the same but I made an action in Photoshop CS5 that can do this as well. I have a starting skin, a finished skin, and the action as a proof of concept. You should be able to install it into any version of Photoshop CS3 or newer. (Tutorial)
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
I do think this would be good but it would break TONS of skins out there.
Do you know how many sites are based around uploading and sharing skins?
Their entire database of skins would go to complete waste.
Not to mention the millions of players. Plenty of people don't edit their own skins and just download them off the sites. They won't be too happy about having a broken skin, especially if they're a super newb that has no clue how to fix it.
Yes, updating the skins to make them work right would be easy but it just isn't worth having the entire library of skins on the net going to waste.
it really isn't much different than having a future MC update break every single world and instead of having some auto-converter, you'd have to manually go into MCEdit to update the worlds.
If they have someway to get around this then I'll support this idea, but otherwise no.
If they have someway to get around this huge flaw then I'll support this idea, but otherwise no.
I'm a step ahead of you lol. Java could potentially check to see if a skin is an old one by checking specific parts of the image and update only the old ones automatically by systematically moving, copying, and pasting specific parts of an image. I can't verify that it would work but seeing how we've already seen Minecraft automatically assign specific parts of a file to specific sections of a 3d object and we've seen things such as animations in Minecraft it seems highly possible.
Rollback Post to RevisionRollBack
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
I support! Recently I tried to make a skin, It had kind of "Danger Shirt" Design. (You know, the Caution/Danger signs? Black and yellow? Thats how the shirt was like) But since the arms were...Well like they are, I had to leave the arm just completely yellow. Which kind of suck, With this, It'd be a lot easyer!
Alright, I did some looking around in Java and it does look like it would be possible to automatically convert old skins to new ones, just copying the limbs and duplicating them. I'm not a programmer at all, but I might try and work on a stand alone converter for this, something that could more or less easily be incorporated into Minecraft itself. This will likely never get finished so no promises there. But this has tweaked my interest quite a bit.
Rollback Post to RevisionRollBack
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
I fully support this. And I agree with Badprenup, I don't have much Java experience, but making a duplicator would not be too hard. Even if that doesn't work, Jeb could just make both skin types compatible.
Thanks for the proof of concept, Badprenup! The only problem I see is in trying to come up with a way to systematically detect old vs new mappings. I have seen many skins use the empty space for logos, fill them in with a solid color, or even have runoff from the main parts. I don't think there is a reliable way to catch all possible cases (from an automated program's point of view).
Yes, updating the skins to make them work right would be easy but it just isn't worth having the entire library of skins on the net going to waste.
Emcitement, I totally get what you're saying. I did mention it as the major remaining problem in my proposal. But let's see if we can try to predict the possible user cases, maybe we can figure out a solution. After all, this forum is a two way street - I'd love more input from the community on how to solve this last issue.
Here is a thought...
If a conversion tool can be put together, Mojang could run it against their databases, automatically converting everyone's old style skins to the new format (when the update goes live). That takes care of everyone currently using a custom skin. (And anyone who updated it for the snapshots should be aware of this impending change and will know to go back and re-apply the correct format)
For anyone who would upload a new skin after the fact, Mojang could enhance their minecraft.net profile interface to support both formats for uploading. It would say something like "Hey, use this new format! But if you have an old style format, check this box and we'll convert it automatically for you!"
That way, all of the current skin repositories won't be totally trashed.
For texture pack creators... well, they have to update any time a new mob/block/item is added, so users should be fairly familiar with seeking out the latest version of their texture packs after each update. No real change there.
Maybe, as time goes on, and people become aware of the new format, the community could gradually go back and update their skin repositories. But with the suggestion above, it at least won't make the old ones unusable.
So now, our main focus should be gaining support while someone works on the java. There's also a good chance that Mojang would remake it themselves if they did decide they liked it. We just need some way to get either a Mojang developer on the case (Jens?) or get some more community response.
Yeah, if this idea were to be accepted by Mojang, I would bet they would develop a conversion tool - at least for the use of mass updating their internal databases.
So at this point, I guess the next thing we can do is to continue to drum up awareness and support of this suggestion. Signature banners are pretty standard for that, aren't they? I can try putting one together - or if someone would like to volunteer, that'd be great too!
Okay I looked at it and I would not be able to figure out in a million years, I'm just not a programmer. But it still is possible in theory, and anyone who likes java should take a crack at it.
Rollback Post to RevisionRollBack
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
In Yu-Gi-Oh when they played that game with the dice where you had to unfold the die into a shape and fit it onto a grid to summon a monster and at the very end of the episode he managed to get one last die to fit perfectly into that little sideways space and win.
Please tell me someone else remembers this too so I don't sound like a wierd crazy person
I don't even know how I remember that, it has been years since I even thought about that show... whatever, I would totally support having a bigger image to work with.
I am overjoyed that my thread got necro'ed with such positive support (thank you!). I re-examined my proposal and agree, that there has to be a better solution than rotating the leg bits they way I originally suggested. I started playing around with a bigger texture size, but that ultimately went down a path that didn't look promising. Trying to keep everything together got complicated when I had to factor in using the same mapping for armors. Lots of wasted space came out of that.
So I played Tetris with the player skin some more and came up with this beauty. Behold!
Pros:
+ independently textured arms and legs
+ maintains the original 64x32 size
+ keeps all sections grouped together in a fairly intuitive way.
+ sub sections follow a mapping VERY similar to the original (with only some parts slightly shifted)
+ NO HACKY SUB SECTION ROTATION (improvement over my first attempt)
+ can still apply to zombie, pigmen, and zombie pigmen plus all armors
+ ONLY code change to minecraft is the class which translates the mapping onto the models.
Cons:
- just one... actually converting existing files over to the new format.
I gotta say, I'm really surprised that I didn't see this earlier... the mapping here is near perfect. It even manages to keep a lot of the same arrangements of each section as the original mapping. The head for example, follows the same pattern of right, front, left, back, as do the arms/legs and the body is actually completely unchanged as well! Each body part section manages to remain grouped together for the most part.
As stated, I'm pretty confident in this latest version. So now it falls back to the one remaining downside to this entire suggestion: conversion form old to new formats with existing textures.
I am still confident that this will work given enough early warning, a long period of preview exposure from snapshots, and a utility to automatically convert old skin files to the new format.
[Edit]
I also decided against changing the texture size for another reason... Consider that I am proposing a change to an aspect of minecraft that is already well established. In my opinion, for this to have any chance of being considered by Mojang, the change must be as simple to implement as possible - ideally with few actual changes required. You know the saying, if ant ain't broke... well, technically this ain't broke - but it could be improved. An yet it is all too easy to go overboard and let a simple suggestion grow until it is too complex to be considered "simple" anymore. At which point it runs the risk of being dismissed out of hand.
This suggestion at is core only has one requirement - changing the single part in the code that controls where the mapping is applied to the player/zombie/armor model. Everything else is upon the community to update their skins. You really can't get more simple than that.
[Edit2] Pikmin Red pointed out a typo on the image labeling (though it seems the forums update ate that post). Corrected now!
It would be nice if somebody could show this to some Mojang person as they never really check the forum.
thanks to LucasAntunes998 for making this banner!
Starting by looking at the skin, it would check to see if it is an old MC skin by checking specific sections for image data. For example, it could check the leftmost side to see if there is a 10x32 set of transparent pixels (which all current skins should have, provided someone didn't fill in all the extra blank space for no reason), and only initiate the skin change if the section it looked at matches what should be there.
He also said that Java could also edit the image directly. It can move specific parts of an image into new sections and duplicate parts of it to other areas. Therefore it could in theory update any new skin to this new format on it's own. I know this isn't exactly the same but I made an action in Photoshop CS5 that can do this as well. I have a starting skin, a finished skin, and the action as a proof of concept. You should be able to install it into any version of Photoshop CS3 or newer. (Tutorial)
Starting Image
Finished Image
Photoshop Action (Right click, Save As if just clicking it doesn't work)
https://dl.dropbox.c...SkinUpdater.atn
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
http://www.minecraftforum.net/forums/minecraft-discussion/suggestions/2775557-guidelines-for-the-suggestions-forum
I do think this would be good but it would break TONS of skins out there.
Do you know how many sites are based around uploading and sharing skins?
Their entire database of skins would go to complete waste.
Not to mention the millions of players. Plenty of people don't edit their own skins and just download them off the sites. They won't be too happy about having a broken skin, especially if they're a super newb that has no clue how to fix it.
Yes, updating the skins to make them work right would be easy but it just isn't worth having the entire library of skins on the net going to waste.
it really isn't much different than having a future MC update break every single world and instead of having some auto-converter, you'd have to manually go into MCEdit to update the worlds.
If they have someway to get around this then I'll support this idea, but otherwise no.
I'm a step ahead of you lol. Java could potentially check to see if a skin is an old one by checking specific parts of the image and update only the old ones automatically by systematically moving, copying, and pasting specific parts of an image. I can't verify that it would work but seeing how we've already seen Minecraft automatically assign specific parts of a file to specific sections of a 3d object and we've seen things such as animations in Minecraft it seems highly possible.
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
http://www.minecraftforum.net/forums/minecraft-discussion/suggestions/2775557-guidelines-for-the-suggestions-forum
Check out Port Atdosk!
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
http://www.minecraftforum.net/forums/minecraft-discussion/suggestions/2775557-guidelines-for-the-suggestions-forum
Emcitement, I totally get what you're saying. I did mention it as the major remaining problem in my proposal. But let's see if we can try to predict the possible user cases, maybe we can figure out a solution. After all, this forum is a two way street - I'd love more input from the community on how to solve this last issue.
Here is a thought...
If a conversion tool can be put together, Mojang could run it against their databases, automatically converting everyone's old style skins to the new format (when the update goes live). That takes care of everyone currently using a custom skin. (And anyone who updated it for the snapshots should be aware of this impending change and will know to go back and re-apply the correct format)
For anyone who would upload a new skin after the fact, Mojang could enhance their minecraft.net profile interface to support both formats for uploading. It would say something like "Hey, use this new format! But if you have an old style format, check this box and we'll convert it automatically for you!"
That way, all of the current skin repositories won't be totally trashed.
For texture pack creators... well, they have to update any time a new mob/block/item is added, so users should be fairly familiar with seeking out the latest version of their texture packs after each update. No real change there.
Maybe, as time goes on, and people become aware of the new format, the community could gradually go back and update their skin repositories. But with the suggestion above, it at least won't make the old ones unusable.
So at this point, I guess the next thing we can do is to continue to drum up awareness and support of this suggestion. Signature banners are pretty standard for that, aren't they? I can try putting one together - or if someone would like to volunteer, that'd be great too!
Want some advice on how to thrive in the Suggestions section? Check this handy list of guidelines and tips for posting your ideas and responding to the ideas of others!
http://www.minecraftforum.net/forums/minecraft-discussion/suggestions/2775557-guidelines-for-the-suggestions-forum
[Edit] Now with correct spelling action! (derp)