Kuribo64
Views: 19,851,770 Home | Forums | Uploader | Wiki | Object databases | IRC
Rules/FAQ | Memberlist | Calendar | Stats | Online users | Last posts | Search
03-28-24 06:21 PM
Guest:

0 users reading Shadows on Custom Models in SMG 1/2 | 1 bot

Main - Archived forums - General SMG hacking - Shadows on Custom Models in SMG 1/2 Hide post layouts | New reply

Pages: 1 2
blank
Posted on 08-27-13 06:27 PM Link | #31014
Posted by SuperMario64DS
Have you considered that not much may be going on because of the lack of a proper editor or importer?


It is possible that we have a sort of chicken and egg situation. If that is the case I guess SMG hacking is dead and there is not really anything to do about it.

But if you look at the first custom Mario Kart Wii tracks they look like crap. When MKW hacking first started out, the tools wasn't advanced enough to make tracks that didn't look like crap. But there were still people who made custom tracks, because they passionate about custom tracks.

You can easily make custom SMG levels of a lot higher quality than the original custom MKW tracks with the tools available at the moment, yet no levels are being made. I think this is because there is simply are no capable people who is really passionate about custom SMG levels.

If that is the case, then a better model converter wouldn't really make any difference.

Posted by SuperMario64DS
I'm not entirely sure myself where you'd start with a project like that though, there's still a lack a proper documentation.


As I've already told you, almost all of the BDL file format is documented in the source code of bmdview2, except for the MDL3 section, which lacks a written documentation, but anybody interested can just ask me.

SuperMario64DS
Posted on 08-27-13 08:21 PM (rev. 4 of 08-27-13 08:49 PM) Link | #31028
Posted by blank

It is possible that we have a sort of chicken and egg situation. If that is the case I guess SMG hacking is dead and there is not really anything to do about it.

But if you look at the first custom Mario Kart Wii tracks they look like crap. When MKW hacking first started out, the tools wasn't advanced enough to make tracks that didn't look like crap. But there were still people who made custom tracks, because they passionate about custom tracks.

You can easily make custom SMG levels of a lot higher quality than the original custom MKW tracks with the tools available at the moment, yet no levels are being made. I think this is because there is simply are no capable people who is really passionate about custom SMG levels.

If that is the case, then a better model converter wouldn't really make any difference.


It's this type of attitude that's doesn't help either, Blank. You cannot grow fruit without first planting a seed.

While everyone and their Grandmother may not suddenly jump to Kuribo64 the moment an importer like this is released, it wouldn't bad to have at all. Would Brawl, a game that is similar in quality to Super Mario Galaxy, ever had gotten anywhere in the whole hacking business if BrawlBox had now been released?

There was absolutely no passion to speak of whatsoever when Mario Kart custom tracks were new. You were limited to a small number of polygons, along with few items. People tried it because it was new and easy.

When Super Mario Galaxy first came out, it was quite the opposite. The game works entirely different. You now have gravity controllers, different sections, no importer, more complications. The most likely to try galaxy editing would be those in Similar communities, which was, at the time, the Mario Kart Community, which had already moved into it's unlimited object and model size phase. Then there were new people who had no idea, not knowing what a co-ordinate even was, and those were the people who stuck around, it would be all they've ever known.

Firstly, the people who started the Mario Kart Wii community didn't leave it in a state of mediocrity. They went one step further, then one step further, then one step further, then one step further. They didn't leave us with just the SZS Modifier, they instead advanced and gave us nice tools like CTools. With CTools, you could now add whatever object you wanted, and, you could import nearly an unlimited amount of polygons. Not only was it easier, and more efficient that the previous methods and tools, but this also excited the few who were around then. Pretty soon we were given nice tools like Wiimms SZS Tools, KMP Modifier, KMP Cloud, and several others. The communtiy began to grow to such a length, that BlackJax, creator of BrawlBox, made his tools compatible with Mario Kart Wii.

Secondly, Mario Kart Wii is of much lower quality than Super Mario galaxy 1 or 2. The first few OBJ tracks were first acceptable because Mario Kart Wii wasn't exactly praised for it's high quality or pretty graphics. Instead, shadeless and bright tracks were accepted because Mario Kart Wii wasn't too different itself. There was no greater example of a nice course than, say, Daisy Circuit.

Thirdly, Mario Kart Wii and Brawl are a different genre than the Galaxy games. They're much older, feature competitive & multiplayer modes, and have online play. Time Trials, World Records, VR, competitiveness, all things that these games feature. If you recall, Chadderz released a video of a multiplayer hack in SMG2, which would atleast introduce to co-operative mode that New Super Mario Bros. Wii fed off of.

I know several of my points are incomplete or in pieces, but I wasn't sure exactly what to say. Super Mario 64 Hacking never really got anywhere far, until Super Mario 64: Star Road made by Skelux. He was a one man band, he did everything but create a level editor. And it turned out to be quite popular, despite not much interest before, and it's difference from commonly hacked games like Racing, Fighting, or Action titles.

You never know what will happen after you release an editor or importer like that. It may be a total failure. It may cause someone to want to do something as simply as replacing Mario with someone like Wario or Sonic, then it may set the whole community on fire.

Ehh, what I'm trying to say I guess, is that game is difficult to edit, and still is. It isn't worth the time it takes to add fake shadows to your model, nor is it worth the time to set up levels with ugly models. Sonic Generations is modding is pretty tough, level layouts are difficult, as are the Models themselves. But people do it. You may argue that people have had more interest in that game, so clearly they'll be more likely to hack it, and it's doing well. Sonic is considered more 'Grown up' than Mario, attracting older, and smarter people. However, you should also consider that 'hackers' in the Sonic community are plentiful, they've been there for ages and hack nearly every game that comes through the door.

Perhaps Super Mario Galaxy hacking isn't dead afterall, but rather, it was never alive to start with.

Now feel free to point out the inconsistencies in this post, and tell me once more why there are thousands of reasons to not do the reasonable.

And I'm going to quote this again:

Posted by blank
You can easily make custom SMG levels of a lot higher quality than the original custom MKW tracks with the tools available at the moment, yet no levels are being made.


There was almost no passion in the 'original' Mario kart Wii custom tracks. The community did not prosper until it was pushed further. regardless of how object editing works.

Levels of higher quality than the 'original' Mario kart Wii tracks can be made for any game, easily. But in this particular case, it's not worth the level editing time for a bight, ugly model, nor is it worth the time to set up fake shading with huge textures that don't really look so good to start with. This game has high standards in terms of quality and capability.

---

I'm starting to realize that this is a lost cause, however. It's never actually going to happen, and there's always going to be some reason why it shouldn't happen. Lack of activity, users, interest, whatever. Once a community's in ruins, it's in ruins. Just let us know now, in big, bold letters, that you're not going to, ever, make such an importer or editor. Please. I'd rather not have a reason to stick around anymore, the thought of it has given me motivation to check here every so often, but not much anymore. Just dismiss the subject.

blank
Posted on 08-27-13 09:09 PM Link | #31041
If I have understood you correctly, you are basically saying that are interested in making new levels but don't because it is hard and time consuming. At the same time you are asking me to create a model editor, a considerably harder and more time consuming task. I fail to see why I'm the only one who have to put an effort.

If you truly care about SMG hacking, make a level. If SMG hacking dies it is as much your fault as it is mine.

shibboleet
Posted on 08-28-13 09:15 PM (rev. 3 of 08-28-13 09:16 PM) Link | #31241
Posted by SuperMario64DS
Have you considered that not much may be going on because of the lack of a proper editor or importer?

Make one then, Mr.Genius.

We're even lucky to have a model importer at all. Be grateful we even have one.



____________________
a

SuperMario64DS
Posted on 08-28-13 10:59 PM (rev. 4 of 08-29-13 01:42 AM) Link | #31248
Posted by blank
If I have understood you correctly, you are basically saying that are interested in making new levels but don't because it is hard and time consuming. At the same time you are asking me to create a model editor, a considerably harder and more time consuming task. I fail to see why I'm the only one who have to put an effort.

If you truly care about SMG hacking, make a level. If SMG hacking dies it is as much your fault as it is mine.


That is incorrect.

I do not want to give the time needed to implement fake shading, it takes too long to get right and results in a large file size. At the same time, I do not want to take the time to make a level with bright, ugly models. I never asked you, specifically, to make a model importer. Rather, I was questioning your thinking behind the reason why the community is not doing well, you say it's because there was never any interest at all, when it could very well be the lack of advanced tools. The OP seems to agree that the lack of vertex colouring is a nuisance. I may have said you in several sections of my post, but that was really towards anyone in general, whether they agree with your statement(s), or simply decide to read my post.

I will admit, at the end of the post, I did write that paragraph towards you. Out of everyone, I was most hopeful that you would be the one to make it happen, you had the motivation to make the OBJ importer, and you're capable. But seeing that even you, as with everyone else capable lacking motivation to do so, I'll use inductive reasoning and say it will never happen.

Posted by Luigi
Make one then, Mr.Genius.

We're even lucky to have a model importer at all. Be grateful we even have one.



You seem to have misunderstood the context of that comment. Previously, he stated that the community has not, and will not, do well, despite the existence of any other tool. I was trying to convince him that this may not be the case. Model importing is indeed something to be grateful, however, in it's current state it's not capable of matching the game's quality.

No criticism meant towards blank, he didn't invent the OBJ format, and plenty of games use OBJ for importing.

If I were Mr.Genius, I would totally make one. Too bad I'm not, my first's name Dan and my last name's not genius.

If you want to make assumptions about any comment I have, or ever will, make in the past, present, or future...

Make on then, Mr.Rean. (And joking, of course. Apparently I'm Mr.Genius, You're MrRean, harhar :P)

Phantom Wings
Posted on 08-30-13 06:51 AM Link | #31378
If you're looking for a simple solution to the coloring problem blank, then an easy way to do it would be to tweak obj2bdl to take 2 models as input. Process the second mesh in parallel with the first, and for each vertex in the first mesh, assign the texture color of the corresponding vertex in the second mesh to be its vertex color. It's a bit of a crude solution, but I don't think it would be too hard to implement. (of course, you may have a bit of extra work to do if obj2bdl doesn't currently export the vertex color info...)

As far as workflow for modellers is concerned, all you'd need to do is develop your model like usual, texture it, then create a vertex color texture. Then, just export the model once with regular textures and once with vertex color textures. Put both models into the tweaked version of obj2bdl and you're good to go.

In anycase, SM64DS, try to remember that all hacking communities start out small and often don't immediately have access to all the "nice stuff" you've come to expect. The Smash hacking community - as big as it is - originally started with a small texture modding group called ]Syntax Error[. The first texture mods were manually hexed and in order to preview the texture on the model you'd need to burn a whole new DVD of the game. Despite these limitations (couldn't modify shaders, couldn't modify normals, couldn't modify lighitng), they still managed to make some pretty badass mods such as the original Fierce Diety Link and Phazon Samus textures. It would be nearly a full year before the SD based file patching was released and another year after that before anything remotely resembling any of the current existing SMG tools would be released.

blank
Posted on 09-02-13 04:14 PM Link | #31664
There is definitely ways to get vertex colors without going all the way and make a COLLADA importer. One way I have considered is to modify the OBJ exporter of blender to include vertex colors in the exported file, which would be a fairly simple modification. Then changing obj2bdl to include vertex colors wouldn't be hard at all.

But vertex colors wouldn't really change much. Everything you can achieve with vertex colors you can also achieve with textures alone. You can bake lighting into textures just as well as you can bake lighting into vertex colors. The only problem with the texture approach is the files size, but as far as I know nobody have actually had levels crash due to model file size.

NWPlayer123
Posted on 09-02-13 04:20 PM (rev. 2 of 09-02-13 08:03 PM) Link | #31665
Yeah, because we had to downscale all the textures so that it DOESN'T crash. The most recent Sky High model with textures baked on amounted to a good 33MB which would more than easily crash the game, whereas using the same texture over and over again and using vertex colors to shade it would make download sizes much more practical in the large scale. But until then I'll keep making lowres textures to keep the filesize manageable.
EDIT: lol, the sky high planet I'm importing now, without changing any texture resolutions, is 64.1MB. Let's overload the memory :D

____________________
"I hate playing musical chats" ~ Quote of the month

blank
Posted on 09-02-13 04:35 PM Link | #31667
Well that changes things a bit. I might try and see if I can implement my idea then.

On a related note: It would be really helpful if people actually tells me stuff like this. You can't just expect me to magically know.

Bluma
Posted on 09-10-13 07:58 PM (rev. 2 of 09-10-13 08:12 PM) Link | #32233
Posted by blank
Well that changes things a bit. I might try and see if I can implement my idea then.

On a related note: It would be really helpful if people actually tell me stuff like this. You can't just expect me to magically know.


Excuse me for posting for a week old thread (I see you haven't been online in three days, I assume you only log on when you see that a message has been posted).

I've read through the thread. I see that you're currently not willing to do anything with DAE Importing, which is perfectly acceptable and okay.

However, on PhantomWing's subject of Brawl:

As you may already know, Brawl Hacking has advanced to a point where it can now import collada Dae, and convert it the MDL0. MDL0 is considerably easier to understand than Collada.

I'm surprised this hasn't been suggested already...

What would you think of instead MDL0 to BDL conversion, as opposed to the OBJ Idea? I see several benefits, the Brawl editor, BrawlBox, would handle the Dae for you and convert it to an easier format (MDL0). Through BrawlBox, you can edit material properties, colours, shaders, all of that nice stuff. The TEX0 format used by Brawl is more or less the same as BTI used by Galaxy. The shader format is nearly identical.

BrawlBox is also open source. It comes with a viewer, which in basic terms, breaks the model down to an even simpler format, one that model ripping programs can understand, a simple display model. Skeleton data is also presented the same way, and could possibly be converted.

I'm pretty sure the Brawl people would be willing to co-operate if you could reverse that method, BDL to MDL0. This would eliminate the need for things like a Material editor.

If it's done well, support for this format may be added to BrawlBox. By doing so, you'll also catch the interest of Brawl hackers. As mentioned before, once BrawlBox introduced support for Mario Kart Wii formats, both Mario Kart and Brawl hackers began to work in both communities. Of course, it's only an idea. Animation wise, I do believe they work similar. It would be a fairly simple solution to a rather complicated problem, and in the end it would be beneficial to other beyond this community.

http://wiki.tockdom.com/wiki/MDL0

NWPlayer123
Posted on 09-10-13 08:13 PM Link | #32239
My question is: How is one format any better than any other? You still have to do major edits to a program to read and convert one file into another. It's still the same amount of work, maybe even more.

____________________
"I hate playing musical chats" ~ Quote of the month

Bluma
Posted on 09-10-13 08:30 PM Link | #32241
Posted by NWPlayer123
My question is: How is one format any better than any other? You still have to do major edits to a program to read and convert one file into another. It's still the same amount of work, maybe even more.


MDL0 is less complicated than DAE.

Consider the amount of work it took to originally compile a BDL. BrawlBox had the ability to both compile and de-compile MDL0. Blank's program reads the vertex and UV data from the OBJ, converts the images, and gives a BDL. If you already have another program which can read vertex data, UVs, Colours, and whatever else for another format, then that data should be easier to convert. If one understands that every A in one format is equivalent to a 33 in another, then it should be capable of making the conversion, especially if both programs are capable of fully (Or nearly) understanding the format they were designed to know.

BrawlBox features the ability to export each set of vertices as an OBJ file. UV Mapping should be relatively simply, as should colours, and images are already in a nearly similar format, and that's everything you'd need for a basic model. Most material settings seem to be fully understood in both formats, so whether you'd be looking to make your own material editor or port whatever useful data one has from the other, you could do so.

Skeletons are optional. I don't know enough about the subject, so I'm not entirely sure how easy that would be.

I understand that a Dae module for python already exists, but it's not designed with the task given to BrawlBox.

blank
Posted on 09-15-13 03:49 PM Link | #32546
A MDL 0 to BDL converter is something I have considered, and it would definitely be easier to make than a DAE to BDL converter. The reason that I haven't gone in that direction is that it would still be more complicated than a OBJ to BDL converter and I don't think it's worth the effort at the moment.

Bluma
Posted on 09-15-13 05:42 PM (rev. 3 of 09-15-13 05:45 PM) Link | #32556
Posted by blank
A MDL 0 to BDL converter is something I have considered, and it would definitely be easier to make than a DAE to BDL converter. The reason that I haven't gone in that direction is that it would still be more complicated than a OBJ to BDL converter and I don't think it's worth the effort at the moment.


It's certainly interesting that you refer to it as MDL 0, not something I've thought of before regarding the MDL format. Anything would be more difficult than OBJ importing, I agree. Your Vertex coloured OBJ idea is genius enough to hold people over for now.

I'm actually surprised that you've considered that as an option. I agree that given the current state of the community, it may not seem worth it at the moment, though it may be a good idea to help kick start the community in the right direction, dorky model hacks like 'Wario in Super Mario Galaxy' would be made available. Binding the Brawl and Galaxy communities could be beneficial as well.

If you're looking to test drive your OBJ idea, go for it, I'll definitely be pleased with just that at the moment. Keep in mind that Blender is rather un common within most modding communities, most use 3DS Max as a primary choice, and Maya as a second choice, so this will likely please an internal crowd more than an external one.

Can't wait to see what you do!

PaperplateismGuy
Posted on 09-23-13 05:08 AM Link | #32900
Please eventually make a model importer that can do some fancier stuff. It would be a great step in SMG hacking. If you already are making one, that's great! For those who don't want to put the time into contributing to the whole hack, start putting actual time into it if you can. I'm in the middle of making a dash Yoshi level. Good job on your work so far blank.

____________________
Remember, every user has to start out somewhere,- Every time I look back at my old posts, I cringe.


Pages: 1 2

Main - Archived forums - General SMG hacking - Shadows on Custom Models in SMG 1/2 Hide post layouts | New reply

Page rendered in 0.068 seconds. (2048KB of memory used)
MySQL - queries: 27, rows: 227/227, time: 0.015 seconds.
[powered by Acmlm] Acmlmboard 2.064 (2018-07-20)
© 2005-2008 Acmlm, Xkeeper, blackhole89 et al.