Since the beginning, before even the contest part started I was aware of the high potential work needed. Even that nobody in the Staff Team was originally interested in leading the contest, specially in the compilation ROM part. Nimono originally offered himself to lead the project, but not for the actual ROM compilation phrase. Jack wasn't interested either. During that time, I was the head administrator and worried that the expectations over VLDCX would end up ruined due of lack of people needed to run the project, I ended up suggesting myself for both contest organization and ROM compilation. Fun fact, I was gonna lead and organize VLDC9 originally in 2016 (even that I did the preliminary discussion thread), but I ended up dropping because I started studying Computer Engineering and I was afraid that I wouldn't be able to handle the whole leading contest because I had no idea how hard the course would be. Anyways, I offered myself to lead VLDCX because I had more free time compared to the last semester and was more confident that it would lead to good results and I believed that it was possible to make a so big contest and collaboration ROM project. I don't remember anyone against me at that time (it was in January, so I'm not 100% sure).
The very first thing needed was setting up the rules and the judges to make the contest possible to run. In said time, there were the whole polemic discussion about palette and tile remapping usage. People wanted to VLDCX have stricter rules while others preferred the current VLDC8+ styled rules. Another thing to worry was about the possible contest limitations to the collab don't even up impossible and I had to be sure that the rules over ExGFX, music, levels/sub levels, map16, secondary entrances, etc., had enough freedom to let the level designer do the best work possible and at the same time still possible to do the building ROM. Yet, disallow gigantic levels (so the timer limit was there for that).
My biggest fears were:
- Getting out of Map16 space. We had ~60 pages. The rules currently allowed a half map16 page per entry. So that limited us to ~120 entries.
- Getting out of levels. We had 512 levels. The rules allowed uh, 1 main level + 5 sub levels? That was up to 6 levels on worst case. That meant around ~85 levels. Let's say the average was 4 levels per entry. ~128 entries. Still not much, I was expecting from 150 to 200 entries. And don't forget X World, hubs and judge levels....
- Getting out of overworld space. Because if we had over 150 entries, certainly the overworld would get too many levels per submap. While that wouldn't make the compilation impossible, the overworld part definitely would get very dense in levels, having little room for decoration and design.
- Getting out of music. Because AMK allowed around 192 musics. We allowed four musics per submission. Yeah.
- Secondary entrances. We had only 512 secondary entrances. On the recent VLDCs, people already commented that we had few secondary entrance space and that might be a problem.
And many many more. I had no ideas about the possible Dual ROM system, so I had to rely on the only person that could break the logistics limits: FuSoYa, the LM creator.
I exchanged around 10 e-mails with him, exchanging ideas, possible solutions and risks.
The first obvious thing I wanted to talk is about BPS. I know you probably already heard of the whole BPS changes now and all (heck, I was supposed to send that e-mail on january 3, but I always ended up forgetting to!), but a fact is the main reason that SMW Central changed their hack patch standards to BPS to allow newcomers to patch hacks for both headered and headerless ROMs, specially since if you download a SMW ROM today, the first ones you will find will be headerless.
Since LM comes with a IPS patch generator and people seem to prefer patching and creating via the program directly, I think it would be good if you added BPS support too. FLIPS is open-source so you can easily add the algorithm to your program. Alternatively, if for some reason you don't want to implement the BPS algorithm yourself, just allow the user to program a 3rd party tool in LM just like how you can do that with the emulators.
Of course people will call the emulator much more times than a patcher, but that would be an easy (no manual custom toolbar hax) and interesting alternative if you cannot bring BPS right now.
However, that is not the most important thing and the main reason I wanted to talk with you. Well, I suppose you already heard about VLDC, right? The "Vanilla" Level Design Contest. Basically it's a contest where people must make a vanilla level, that's it, closer to the original SMW and pretty much only using Lunar Magic. Since the 7th edition of this contest, we're getting all contest entries and merging into a single ROM, so people can enjoy playing all of the entries together without having to patch each one separately and with the additional feature of being a single hack with own overworld, special levels and all. It's been going pretty well, with every year getting a new ROM hack that usually is the most downloaded of the SMW Central during the period.
However one issue that is getting common is the default SMW/LM limits. The last year (9th) edition had 115 entries, so in total 115 levels to be accessed to the overworld, while the limit, you know, is 96. Fortunately, I convinced to the contest organizer, Jack, to take the effort and create a patch that allows you accessing more than the 96 levels to the overworld and the patch worked quite well (also would you be interested in getting that patch included in LM? I never seen the .asm file myself, but I got it recently since I'm the new VLDC host and I can send it you).
On the other side, the ROM was pretty tight regarding other aspects, like map16, secondary entrances and level numbers. ROM space wasn't a problem since we're using SA-1 and VLDC9 used the 6MB SA-1 ROM expansion method. How challenging would it be to expand the amount of available map16, secondary entrances and level numbers? I never really studied much about SMW's level system and since they're strongly connected with Lunar Magic, I thought you would be the best person to ask with.
For Map16, I remember you quadruped the limit on LM 1.70 and I believe it's now the maximum fixed limit without having to use even more additional bytes to index them. So one thing I thought was making one page specific per level. Let's say, page 3 and page 43 can be configured to change depending on the level number. How hard would it be, in your personal opinion? I know page 2 used to do something similar on the act likes depending on the tileset used, but what about making the tilemap and act like level dependent? If that's not too much complicated, I could set up some ASM to make its contents change based on the level number, but still it would require to manually updating it to the level get displayable on Lunar Magic.
Secondary entrances on the other hand, probably are easier to expand, I believe? Unless if they're sticked with level numbers. But for that option I don't have much choice since unlike Map16, I can't export them as an e.g. .bin file and therefore it would be obligatory to make Lunar Magic support more of them.
Levels are the most complicated ones. While there's some unused bits to make the level number safely range up to, maybe, 0x10000/3 levels, probably there's a lot of pointers that needs to be updated, expanded and therefore repointered. It's not impossible (you know, ASM-wise there's nothing really impossible unless if there's some hardware imposing limit, either by the CPU power or by the PPU, for example), but how hard would it be to allow, let's say, 2048 (0x1000) levels? Of course inserting levels manually would be overkill, so a LM implementation would be needed just like for the secondary entrances. But it depends on you.
The main reason I'm asking that is simply because the contest is very special and possibly one of the most recent appreciated hacks over SMW Central and Talkhaus on the recent years. Expanding those limits would not just allow more people participating but also make decrease some annoying restrictions, like the amount of secondary entrances and Map16 available per participant. While I could simply split in two or more ROMs, it would not have the same feeling of seeing so many levels beaten and later unlocking more levels once you played all of the entries. So yeah, I'd love if you could help me out either with an expanded LM that surpasses these limits or simply with some hints and a special LM version that would work with expanded limits. The contest will gonna start 9 days from this e-mail date, so please let me know if this is a possibility or not, so I can adjust the rules before I get it too late.
And of course, why not you give a try to the contest too?
I bet you haven't designed a level for a lot of time and I think it would be a nice opportunity to you make a simple but cool vanilla level for the contest, even if you don't have the chance at winning or anything.
That's all. Thanks for reading.
With the conversation with him that Lunar Magic 2.50 became a thing and doubled FG, BG space and 16x secondary entrances count. And as well the Dual ROM system idea.
What is the Dual ROM system? It's basically two ROMs in one. The contest would be split into two ROMs, but since the compilation is done the ROMs would end up merged into a single one and when the game boots up, the SA-1 chip would be capable of exchanging ROMs depending on the overworld the player was. Let's say that the ROM 1 had a ice, fire and grassland maps while ROM 2 had desert, forest and x world. When the player goes a pipe in the ROM 1, a special code checked to what overworld the player was going in and in case he was going to e.g. forest map, the code would tell the SA-1 chip to change the ROM internally and then everything, including levels, music, overworld, data, etc., would get exchanged as well. So the principle is not hard. Even that Medic got the system working perfectly. Once the ROM is exchanged internally, you don't have to worry about anything level: all levels, all data, all music, everything is automatically exchanged too. There were just a single limitation: you couldn't e.g. access a sublevel in ROM 2 when you were in the ROM 1. Because the ROM 2 is unloaded. The same applies for ExGFX, map16, music, etc. So the compilation would only be possible to start when the results were out, because otherwise there would be a risk of a couple of levels (not 5 or 10, things like 50 levels) ended up having to be in the ROM 2 instead of ROM 1 or vice-versa. Or even ROM 1 or ROM 2 getting out of space for logistics reasons. So to not have the risk of having to compile the ROM multiple times, it was made clear since the beginning that the compilation would only be possible when the results were out.... Still, considering that the overworld contest would run in parallel and the automatization tools, this wasn't a big issue for me. For me.
Now that I had one thing settled up, I had to settle the judges. The PR Team, which was lead by Koopster at that time, gave me a list of potential judges. Some of the names I didn't know fully, others I did but I was in fear if they would do well or how the judging policy would go. At the same time, some people (e.g. MrDeePay) wanted to judge but the staff was against it. And in parallel, I had a few personal names in mind. And still, considering the incidents from the VLDC9, I wanted to be sure that the same would not happen during VLDCX. So I went ahead and made judging interviews. Basically, I got all people that were interested in judging and had a private and a group conversation with them, along the other potential judges in an IRC channel. The questions asked during the process were this. During the conversations, I made clear my expectations that we would have from 150 to 200 entries and it'd be a challenging task from all judges. The intention from the judging interview was to learn more about the person and let know all possible details that could determine the success or not of the VLDC judging.
- Did I tell one of the judges that he wasn't gonna be part of the judging team (originally five) but after S.N.N. suggesting six judges I accepted him? Well, true and false. What happened was during the judging interview, I have told Deputy BS while he had potential to be a very good judge but the other ones were in front of him, but he would still be part of the judging team because S.N.N. convinced me that six judges would be a better go just in case one or another ended up leaving the judging team. It wasn't my intention to disqualify him nor make him sad. Even that I only heard about his complaints one month later and then I made it clear to him again that he was a qualified judge just like the other ones.
- What personal names you had for the judging team? I don't remember all, sorry if I added or forgot someone or other one, but I remember asking worldpeace (actually I even interviewed him), Hinalyte, Mirann, Wakana and Eevee.
- Would those names have priority over the other judges suggested by PR? No.
Okay, after the whole progress, you know what happened. The six judges were picked, you saw the meet the judges thing, I have prepared everything minor for the contest (like getting the banner from Magi, the scheme from Skewer, etc.) and running the contest.
- Okay, but did you held to yourself everything? All plannings, information, etc.? Not exactly. Like the previous VLDCs, there were a forum that only the leader + judges had access to it. In that forum, I pretty much tossed all ideas, thoughts and concerns about the contest. The judges knew about everything. The planning phrase was done in the PR Office forum, so all staff members had access to it, including periodic reports about the contest after the planning phrase was done. And of course, the preliminary discussion I tossed over 15 ideas about VLDCX to be discussed with everyone and you all should remember how it was. The forum currently looks like this. As you might have noticed, there were almost no activity aside from two threads made by idol. I suppose the recent VLDC discussions were either moved to the PR Office or vanished completely, or moved to the VLDC judge discord server.
- Why did you hold so much to announce the judging changes? For who don't remember, when Deputy BS dropped, he decided to leave judging without notifying me and made a public thread in this forum and it lead to somewhat dramatic level. It wasn't a very good thing, so I told the other judges to not do that and in case they wanted to leave, to let me make the proper announcements to avoid similar cases again (see Deputy BS thread name in the image). Plus: Torchkas was unsure whenever to leave or not judging and idol/me were looking for potential backup judges. Only after everything was settled up that the actual changes were announced, all at once with proper information and status.
- What is the story of you putting measurement metrics over the judges, to pressure them to do more work? When the judging started, Deputy BS made an online sheet with all judges and me, so everyone could put their scores and the sheet dynamically updates with the submission id score plus other information such as if the level should get disqualified or not or other comments. Since the sheet was dynamic and all judges were using it to put their scores, I added some dynamic rows that measured how many levels were judged, the average score, how many to left and an estimated judging finish time (ETA), plus a speed factor row to account how fast or slow you are considering the normal, expected judging deadline. It looks like this. When I made it, nobody complained about it, rather found it impressive and interesting to know if they are ahead time or not. Even that I remember lolyoshi was already going faster than usual while Noivern and Koopster always kept in the normal rhythm to accompany the deadline. Eventually, the judges started losing motivation or rhythm for several reasons and the values started looking much worse than usual.
- But you didn't motivate them properly? Actually if you look in the editing history of the judges sheet, their progress stopped and went to almost null after I left the administration team and the site. Probably some will argue that would happen regardless if I stayed in the team or not, but that's a fact.....
- Did idol take care of the replacement judges herself alone? At that time, idol was one of the PR Office members and she offered to help getting potential backup judges since I had no clue in who to pick. She brought me Rameau's and Ryrir (which wasn't added to the judging team) and Sinc-X I don't remember if it was her idea, mine or from Sinc-X himself (he was staff at that time). Still, after idol's picks I made sure to talk and context them about all details needed for the judging, so it's isn't really true that idol was taking of everything alone.
- What about the history of you firing staff members that didn't agree with you or something? That episode happened in the SMWC Discord Server. I remember I was very pissed off in that day and there was going in either in the PR or Forum departments. I was tired of some complaints (specially staff feedback) and I went wild on some staff members in that day (particularly leod and part of PR iirc). Still, that had nothing to do with VLDC anyways.
In the end, what made VLDCX collab pretty much die were:
1) I resigned from my admin status in June, including everything I was going due of internal issues.
2) Getting a job and having to handle it with university, adminship and SMWC.
3) Personal problems like my mom/dad breaking up.
4) The complex engineering behind the collab ROM which would require me waiting for the results and relying on my summer break to build the ROM.
5) No other ASMer in the site offer to handle the full compilation ROM.
6) Demotivation from SMW Hacking...
Either way, the reason I made this post was to bring up my viewpoints over the concerns presented over me and show some facts for the contest which not many people here probably know or remember. I don't have any intentions of keeping a discussion over this, so it will be up to you to decide if I'm the guilty over it, or if it's deserved, etc. If anything, feel free to drop me a message.
Once again, I'm sorry if I disappointed or pissed off you guys. VLDCX was gonna my last collaboration project and I don't think I will ever have another opportunity of running another one, so I just want to lament that it didn't work well...
Edit: Also everything from X World was planned around in July by me. These were just ideas but they were all applicable. read it by your own risk, if it ever gets applied it's super major spoilers.
GitHub - Twitter - YouTube - SnesLab Discord