Uku (Romhack)

[Canvas]

This romhack is named Uku (float) Umihara Kawase, I just slapped a Japanese water related name that I thought looked good and fit thematically on there. It will feature; completely optional bosses, no true branches (only extra doors are boss/challenge doors) and no respawning enemies. There are 24 fields including 2 bosses and 2 challenge maps, with a grand total of 8 backpacks (though some are so hidden I might as well say 4), in general the maps are harder than the average Umihara Kawase map though none require pixel perfect maneuvers and even the more challenging physics puzzles have leniency. So while the maps are designed with deeper game knowledge in mind, if you can beat level 0 you can probably beat the next 10 ez. You will not however want to play this if you are not familiar with the movement in Umihara Kawase

1.5 FINAL is the last update, it makes many considerable changes such as layout changes for f15, the removal of an entire platform on f17, and more slice of life fixes such as better optimization on f21 and more fun swinging on f1. Atop this are easter egg fixes, it isn't really a secret anymore. In 1.4 and before if you played for 30 minutes straight you were sent to a hidden ending field, which wasn't intended. This was only to be reached if you managed to beat the boss on the final level (this confused the heck out of them), so with direction of Alc who credited Naulahauta and Princess Sevenleaf (tripfag) this feature has been removed. Another easter egg has also been added

1.5 - https://drive.google.com/file/d/1yOimtRbnPfJmRbZiYKqQi_fz769xakCY/view?usp=sharing

To use this .ips file you will need both the original Umihara Kawase (Japan) ROM (usually comes in a .zip file, you will need to uncompress it to get the needed .sfc file, I recommend the program 7-zip) and the program LunarIPS. You just open LunarIPS, click "Apply patch", select "Uku Umihara Kawase 1.X.ips", and lastly select "Umihara Kawase (Japan).sfc".

PREVIOUS VERSIONS:
1.4 - https://drive.google.com/file/d/1IKGdITAT6VmNZyE4Ki_ixBDPUhd3fnQb/view?usp=drivesdk
1.3 - https://drive.google.com/file/d/13SDRMLNUGqbp3N7vNL7w6CoShv1Gj29C/view?usp=drivesdk
1.2 - https://drive.google.com/file/d/1abJuDZT8jwYMObpaLpReN_LUOF8u3xfv/view?usp=drivesdk
1.1 - https://drive.google.com/file/d/168BJMRgtLz7JJ7sw3fBoi48wemhNr3MQ/view?usp=drivesdk
1.0 - https://drive.google.com/file/d/1AE1tqO-QSMuqPm1FxffkT16kOCxsE4Pe/view?usp=drivesdk
Huge thanks to;
Alc (tester)
Commando215 (tech support, tester)
KawaseFan (tester)
Texh (mapper, tester)
Uku wouldn't have been as good as it is without you guys

###The team and their levels
texh - 4, 28
Canvas - Rest

[Canvas]

Here is the latest guide I have compiled for the mapping software Riverback. Updated 02/15/2018, added more information such as how to remove 30 minute timer and improved accuracy of information
https://pastebin.com/LS8tjD3v

There is now a tutorial video too
https://www.youtube.com/watch?v=xYihG2_Ooac

See the original thread for even more cool information;
http://kawasefan.net/forum/index.php?PHPSESSID=b28472e38246cecb2bedacb5beb8957b&topic=13.0

[Commando125]

Make sure to back up your levels and rom when you use the editor. I think there is a bug in it that saves palette number indexes as 0 when exporting levels, crashing the editor.

The editor really does need a rewrite; it is rushed and not very well planned at the time.

I have attached old notes I have while making the editor and figuring out data for the game. It probably won't make sense at first since I compiled the information for myself. Maybe it will help in some way.
https://1drv.ms/u/s!Au1RLDIZCdk8g_MlxYQcnipOVb5sKg

[Commando125]

Riverback was actually named the same way: just finding a Japanese water related name and using it. I am interested in becoming a tester. I am pretty good at the game, and the series in general (all endings).

I tried to make a boss level based on field 61, but once I started making the field from scratch, I had trouble triggering the boss for field 61. I tried editing the original field 61 piece by piece and still couldn't figure out how the mechanics work. All I know is that object type 6 (out of 7) must be 0x28 (to create the barrier blocking the door), enemy types must be 0x39, 0x3A, 0, 0, 0, 0; 0x02 and 0x01 phystiles on the level spawn the boss, and phystile 0x45 is used as a way to trigger the boss. Even with all of this in the level, I can't trigger the boss. If anyone has advice on how the level should be designed to get the boss to trigger, let me know.

[Canvas]

I will keep a copy of your notes though I can't understand much of it. Will backup and reupload if for some reason the information is lost. Hopefully soon I will have at least 6 playable levels I am willing to show and will give you a copy of those puppies but I am redoing several levels as well as my original as I didn't like them at all and haven't even gotten to making them pretty. I also still haven't decided if giving them graphics should be a priority at all right now as it takes the longest. I really wanted to remake that boss so I will try and figure it out too

[Canvas]

I haven't tried remaking it yet so cannot answer your question but I think I have discovered another problem. Unlike the tadpole which will traverse new levels, the crab seems to follow one strict path. I have been testing by going to field 62 and just changing small things at a time. He will spawn anywhere on that level and proceed as if he spawned on the right. He will always start with breaking the blocks on the left and making his way down toward the door. However, if you completely remove breakable blocks, textures and all, he will break the "block" at exactly the same place as if it were still there. If you add a slope for him to traverse, he walks right fucking through it. I just tested by completely removing the floor on one side and he just walked right over it too. If you spawn him higher, that will work as a new platform for him, but it appears the crab will still only break blocks in the exact same place. Unless there are undiscovered variables, remaking this boss fight is probably a lost cause

[Commando125]

The crab boss is hardcoded specifically for that field. You can change the stuff around the breaking platforms. It just needs some creativity.

[Canvas]

I think I got an idea for one and it involves a lot of moving platforms and spikes, someone better make one and fast before I manage to make something even more infuriating than the original boss

[Canvas]

How do people feel about time limits? The base game is pretty lenient about it, but did you dislike that or did you like the amount of time you had to explore? Would you prefer levels that had longer or shorter times based on the level obstacles?

[Canvas]

This past few months I have been messing around with the levels I have already made. They are more difficult, which is so fun to map for that I really can't see myself making levels like field 0 anytime so Uku will probably be much more difficult in the end. I have decided I want to utilize the time limit more strictly because I feel this was underutilized in the original. Anyone reading, give your input and I will consider it or reconsider. Also should I remove enemies spawning in like the Sayonara Umihara Kawase did? This is a really easy way to add difficulty, but I know easy does not always equal more fun so it is probably worth the extra effort to remove

[KawaseFan]

If you want to make Uku difficult, it would make sense to use the timer as part of that goal.  If you are going to do that, though, it's my personal opinion that enemy spawning should be kept to a minimum.

[Commando125]

I feel that the timer only adds artificial difficulty. None of the games really utilized the timer well.

It is more useful to have a timer going upwards than downwards. I know a japanese ips patch to do that. It allows you to field select too. I can probably remove the field select part at some point and just have the timer. This is untested and might break the editor (it did open when I tried the patch on a rom), and the patch is not reversible on a rom, so backup like always. https://www.dropbox.com/s/s3h9yirwjt9rka4/umihara_fieldselect_jp.ips?dl=0

[Canvas]

I am not intent on an inherent challenge of time, but just less time to mess around. More than enough time to run around and collect everything but enough to still look at the timer and get that anxiety. Instead of 400 seconds maybe 80 with more depending on the type of challenge. They wouldn't all be split second kind of maps but now that you've given me the idea I might surprise the player at some point. I could create separate versions of the hack with the different timers toward the end. Do you think there would be any necessary changes for this style of play, such as completely removing enemy spawns? How would I patch a game with an .ips patch?

[Commando125]

80 seconds is a bit harsh. I don't think removing enemy spawns is needed for this style of play. It is more important on where the enemies spawn, which is controllable by physmap tiles (0x03 0x07 on ground). Setting spawn rates to 0 disables spawns for that enemy.

Applying an IPS patch isn't too difficult. Download Lunar IPS (http://fusoya.eludevisibility.org/lips/) or another IPS patcher. Click "Apply IPS Patch." Find your .IPS file. Then find your ROM file. I am not sure if the patch will work on an expanded Umihara Kawase rom (it auto expands to fit more levels in because the levels are compressed tightly in the game data). What I would do is patch a clean Umihara Kawase rom (1024kb), then save in editor to expand the rom to 2048kb, then create an IPS patch of it using Lunar IPS and apply it to the hacked ROM you are working on. Don't forget to have a spare ROM copy.

[Alc]

Pardon me for being a bit blunt here: I hate random enemy spawns. They add nothing positive to the game, they're just pure "RNG" frustration. For years I've been meaning to scrub up on 65C816 and poke around in RAM just to find the trigger and get rid of them - the game would be substantially improved, IMO.

Personally, I'd also just scrap the giant enemy crab boss fight. Most of the bosses in SFC UK are pretty badly designed, but that one in particular really irks me. It ramps up the difficulty right at the end, you need to know exactly where and when to fire your lure to get to the exit, and this after you've spent a few dull minutes dodging the damn thing and its spawns. (You do get to mess around on the swing on the next and final level, but I'm usually too grumpy from that fucking crab to get any fun out of it.)

On a more positive note, I just want to say thanks - it's great that you guys are making modifications and new levels and such, really cool that it's finally happening.

[Canvas]

Thank you for the posts, I will discuss this with texh and see what he thinks. It wouldn't be hard to turn off enemy spawns itself, the challenge would come from creating meaningful encounters, and if we decide on clock going up it would be best to start early. When we have decided on something I will post about it here and possibly edit the OP

Also I have considered making a mod of the original without random enemy spawns. It would definitely be possible in Riverback. I hate the shit out of some of the bosses too which you can also skip in riverback. If I made a mod like this it would definitely not JUST be enemy spawns but all the tedium. If not cutting out many of the bosses (fuck the crab boss but FUCK THE SECOND TADPOLE, i love school of fish though), find some way to make them completely optional, perhaps with new doors. In Uku I will try my best to make bosses avoidable with no content explicitly locked beyond it. My idea is to make them shortcuts instead. I could also make it to where there is actually 1 ending in the game which might make taking those shortcuts more meaningful. How would you all feel about this?

Edit; I think it would be best if the bosses were completely optional in Uku. Right now I have it set up so that if you beat the boss you just go straight to the level you would have went to through the other route. Shortcuts which skip entire levels in a well designed game don't seem like such a great idea, reminds me of how while playing Doom I outright skip secret levels because I know it would skip the next normal level, since I love playing those maps too, and therefore never play the secret maps even though I like them too. I don't want that to happen

[Alc]

fuck the crab boss but FUCK THE SECOND TADPOLE

Oh god yes. I'd actually forgotten that fight, even with the strats that people came up with, it still felt like a roll of the dice. Urgh. Thankfully it doesn't take you to anywhere you can't otherwise reach.

Personally I'm not too bothered about the timer - like most here I'm sure, I just zip through levels, so as long as the timer isn't aggressively short I wouldn't have a problem with it per se, but I'm not sure that it would add much, and sometimes it's fun to screw around on a level. An increasing timer (as per the old level select romhack) would probably be my pick. Or maybe two patches - one as a hard time attack, the other as a general release.

[Canvas]

This has been constructive. We will reduce if not remove enemy spawns, change the timer to go up a-la Sayonara Umihara Kawase, and make bosses optional but I still definitely want them to be included. With the new Riverback release I guess I will get back to work

[Canvas]

I am trying the patch now, I don't know if field select will be a problem, it depends on if I end up using only 1 ending. About how hard do you think it would be to patch out field select?

[Commando125]

I just mentioned the field select thing to help with testing. I didn't look into what the patch does yet, so I don't know how easy it will be to remove field select and only keep the upwards timer.

[Commando125]

Is your rom hack going alright?

[Canvas]

Heck no I have gotten lazy, playing nothing but Team Fortress 2, its addicting AND I CANT STOP. I want to make some maps with moving platforms but I can't get them to work for some reason so have been making shells for maps but nothing I can even spawn into

Also since the timer hack also included a level select it wont really work for this hack. I still want there to be branches so going through a door not moving you to the next level is a big problem. I will just keep the timer for now

[Commando125]

Did you look at other levels with moving platforms?

[Canvas]

Yeah, I have even tried adding additional platforms to levels where it already works like slot 16, it just won't load. I cannot figure it out, picture related. I tried both 06 and 07 platforms on the bottom middle, highlighted in red, without changing any header variables and whatnot, but it wont load with this addition. I tried deleting another section of single platforms with the thought that maybe it was overloading or something I don't know, didn't do a thing either. What could be wrong? I know 06 and 07 are the slots where the object values of the platforms are and I figure I shouldn't need 0307 on empty spots for my usage but I plugged it into there anyways

[Commando125]

Did a bit of investigating. You cannot have more than 3 of that platform (0x06 or 0x07). I don't think you can stack them in the same vertical space either.

[Canvas]

But here is the map I am trying to make. I thought it was a conflict with escalator and platforms, so I made it only the blue single platform to free up the 6th object slot but it just isn't working. I can spawn into the map if I remove the single platforms

[Commando125]

All of the single platforms are linked together. When you step on one platform, all of the platforms of that type fall down as well (see field #11). A requirement of this is for all of the platforms to have the same vertical length (the same # of 0x07's vertically). If one of the platforms has an incorrect vertical length from the other platforms, the field will crash. To overcome some of these limitations, use the other platform type: 0x20. The platforms for type 0x20 also link together. It might be possible to move 0x21 to object slot 5 and 0x20 to object slot 4. That should let you keep the escalator object type. As far as I know, only graphics bank 0 (the one you are using) has two graphic types for a single platform.

P.S. I think that level is impossible without ground speed-boosting, probably with that too.

[Canvas]

YOU CHANGU BACKGROUND NANI?? The level definitely needs a lot of work but it at least spawns me now thanks a lot, I will add this to the guide

That said the level is completable without that exploit, but not in the way I intended it to be, there will be much change to it, especially now that I know the limitations of the platforms

[Canvas]

So I have been being very lazy, I apologize, I do not want anybody to think this project will die. I might be going away for a few months for work and don't know if I want to bring my computer as I think I will use that time as a vacation from my normal NEET life (haven't heard a word back, probably isn't going to happen). I promise you guys I will commit more work to it, I find mapping fun and I think I am getting better every day. I have recently switched fulltime to Linux (at the time, Linux Mint, then Ubuntu, now Xubuntu somebody save me) and Riverback runs fine in WINE (it will crash if you try to turn on grid, and if it crashes on saving you might need to move the "romwriterdata.xml" file to your home folder ~, though this was a new issue for some reason and might not happen to you). I have remade that map into something I think is a lot of fun, but I wont spoil it. This level has replaced slot two entirely

[Canvas]

Spill the beans on this background changing please I want to try putting this bad boy in there

[Commando125]

The graphics for the backgrounds are uncompressed. It is easily viewed with a graphics tile editor near the first half of the ROM file. I might try adding that background at some point.

[Canvas]

If anyone has advice on how the level should be designed to get the boss to trigger, let me know.

I have managed to make the flying fish spawn. Try using byte 47 as the spawn trigger, it worked for both flounder and tiger fish for me

Edit; You cheeky fuck I just realized what you said

[Canvas]

I have been wanting to do this for awhile now. Our level zero is servicable, but we can do better. How about a competition for the first map in Uku, with the glory of headlining the first megahack for Umihara Kawase being the sweet reward, your name bolted before even mine in the OP for the lifecycle of this website. Interested? You should be maggot, because I am talking to YOU. I feel like I am unbiased, though a volunteer judge would be noice. Until then though I will judge, with my critique being flow, style, and difficulty. I want something which is unique, with as much character and love as level zero in Umihara Kawase itself. Nothing too hard is the only rule, let's define that as 'a level where a new player wont likely die more than 3 times first run'. It doesn't have to be on graphics bank zero or even be too familiar, though I do want consistent theme so what you make will likely influence the next three levels, you are so powerful. The cutoff will be December 23, 2017, but submissions can be made after that for a spot in the hack of course. I will be competing as well. Post it here or message me or whatever else. Anyone interested should begin posthaste, you only have 10 months.

Also just a reminder, I am still looking for testers. I just released another pack with the latest levels to my two current testers and would love to get more. Give me your honest to goodness brutal thoughts in exchange for some sick maps. Even the biggest change possible isn't so hard to implement so don't hold back, I will consider anything.

[Canvas]

It has been awhile, I feel like long enough to give an update. The past three months I have outlined only two maps I like, polished others a bit more (still haven't bothered giving them pretty objects like big vegetables and bikewheels and whatnot), and made a few gimmicky levels that honestly suck but with a few cool ideas I will end up re-purposing. That is about it, there wont be another pack for a while, but so far we have a good 14 levels. Most people wouldn't know what I mean, but it was a common complaint that the tadpole boss was extremely easy to cheese which is boring, so if I have done my job right expect him to require a strategy to beat next time muhahahaha. Don't let that deter players though, it is still optional. Attached is a concept for the most recent map made, an early level which I think is coming along nicely. There was a lot of inspiration stolen from level 0 in the base game (if you flip it horizontally it almost looks just like the dang thing now that I have tried it out). Now that I am much more familiar with the editor I am also attempting to make the editor tutorial video, but the more I actually record the less important I feel like a lot of the information I am spitting out is. So it will probably just be an advanced tutorial and not focus too much on for example navigating the GUI as that is actually pretty straightforward. That isn't too relevant to Uku though and is also just another thing on the backburner. I will post that in the "rom hacking" thread whenever that is finished.

[Canvas]

But here is the map I am trying to make. I thought it was a conflict with escalator and platforms, so I made it only the blue single platform to free up the 6th object slot but it just isn't working. I can spawn into the map if I remove the single platforms

I really like where this level went

[Canvas]

This is an EXCLUSIVE screenshot of codename tunapalooza, a stress test map from several months ago. The goal of the map was to have as many enemies on screen without it lagging, to which I deduced that the game only seems to lag with many enemies when I try to use my hook. So I designed a map where it was not necessary, or even really possible, to use the hook at all, made entirely out of glass. It was a failure because I had intended the tuna to slide down but they didn't, as well as a very rare crash which happens when I can only assume so many enemies are onscreen at once. I had taken a screenshot when it crashed because the situation was funny, however cannot find it so have recreated the screenshot sans the crash which I cannot replicate. If I can find that dang thing I will post it too

[Commando125]

I think the game can only track up to 8 enemies at once. I have no evidence to back that up.

[Canvas]

I found the initial screenshot, I don't know how it ended up in THAT folder. How you explain must be it, because as soon as the 9th fish tried to move in the view it froze. Note to mappers, if you wanted to make a laggy as FUCK map spread them out a bit

[Canvas]

There is now a Riverback tutorial video (second post in thread) but other than that haven't done a darn thing in half a month. It is already nearing the end of August, the clock is ticking on those map competition submissions (in hindsight this site doesn't really get enough traffic for it to have made sense anyways so I don't really expect any at this point). Level 0 has at least been redone and is serviceable for now. Right now my focus is to redo the remaining maps to be more interesting and preferably have made another two maps before next wave. Thus far we have 17 maps and many of which fully textured and at most need touch ups. ITS COMING TOGETHER LADS

[KawaseFan]

There is now a Riverback tutorial video (second post in thread)

Just watched the video; good stuff!

I have updated the level editing info page with a link to that post.

[Canvas]

Coolbeanz tell me if you guys still have trouble with anything, I might be able to help

[Canvas]

It has been another few months so I should update, again little progress. There is now one new map and small changes throughout a few other maps. Soon I will cut(not literally) my internet once again and devote my time to try and finish for the 13th anniversary of the first game as I feel like that is doable. The current goal is still 24 maps, of the 18 we currently have I consider 16 polished. The plan for the upcoming six maps is as follows; one more optional challenge map, one novelty map (if possible, i have a very specific idea but am unsure if I can make it work), and four or more easier(still not necessarily easy) maps at least two being on graphics bank 5 as that has seen neglect from me and I want to use it. Several less important changes include just making all maps have a very high static time limit and utilizing the few objects and enemies that aren't currently in any maps. If everything goes smoothly a week before release (December 16th) I would be sending out a final wave for testers for a thorough testing period so if you wish to be a tester just send me a message and I will put you on the list. This is coming together and that is something to look forward to

[Canvas]

This is a release post, check the OP for link to UKU UMIHARA KAWASE

[Canvas]

Version 1.1 available. Basically it just makes level 0 feel less awkward

[Canvas]

1.2 is available, just a collection of small changes for levels 0 and 1 as focus. Also added are textures for the previously hidden doors in the optional levels

[Canvas]

1.3 available, I hope I got em all this time. Mostly small fixes with minor level alterrations on fields 12 and 15. Fixes very rare crash on field 19

All downloads are .ips patch files now

[mpu]

I've really enjoyed this romhack! Beat it a few times now and I think I'll be coming back to it eventually again. Thanks for your hard work!

[Canvas]

I've really enjoyed this romhack! Beat it a few times now and I think I'll be coming back to it eventually again. Thanks for your hard work!

I LIVE FOR THIS KIND OF ATTENTION, thanks a lot I really am glad you had fun. When you replay be sure to update, it will be EVEN better

1.4 for Uku has been released. It was mainly small changes to f7, 14, and 19. Until I beat Uku again and snapped. So the last level now doesn't have the obnoxious pillar texture corruption on exit

[nathan]

I downloaded it a couple days ago and had a lot of fun. It almost feels like playing a bunch of levels that got cut from the original SFC release of the game.

[Canvas]

I downloaded it a couple days ago and had a lot of fun. It almost feels like playing a bunch of levels that got cut from the original SFC release of the game.

So much changed from conception, levels were originally very lazy. I am really glad the levels developed a more vanilla feeling overtime and that you enjoyed it. I am still working on one final update, 1.5. It is focusing on fixing anything I even question which should have been a focus since the start but I am still learning. 1.5 is final so once released, if any are found, all small things will not be fixed. Only serious problems such as ANY kind of crash whatsoever which I can fix will find a way (though if that does happen and I have to put out a fix anyways I will probably make those minor fixes too). If anyone has any feedback on specific areas of the game which you feel could be reimagined let me know. I think I will wait a month before releasing the update this time as I am sure I am not the only one tired of all the updates

[Alc]

Any interest in physical carts for Uku? If there's enough demand then I'll get it together. Pricing is hard to predict but shouldn't be more than $30.

[nathan]

If it would be a Chinese bootleg cart, note that they often use 3.3v flash chips without any 3.3>5v level shifters, which can cause either the flash cart or the console to break (flash cart due to chips being driven beyond their tolerances, console because of outputs being damaged due to the extra voltage being sent back into the console).

https://www.youtube.com/watch?v=-V2zN2Pj-9A video that explains it better than I do. Note that the end part where he measures the resistor values of that Everdrive cart is irrelevant here because the Aliexpress PCBs straight up connect the 3.3v flash chip to the data bus with no protection.

[Alc]

If it would be a Chinese bootleg cart

No, I'd either be using donor carts or proper repro boards, not cheap Chinese crap. Doesn't sound like there's much interest anyway...

[Commando125]

Whatever cart is used, it has to hold 2048KB (2 megabyte) of ROM instead of 1024KB (1 megabyte) like the original game.

[KawaseFan]

Doesn't sound like there's much interest anyway...

I am interested; I was going to reply earlier, but after nathan's post I decided to wait.

Now I'm wondering about what could be done with the label, but that's probably something to think about once there is enough interest.

[Canvas]

If it would be a Chinese bootleg cart

No, I'd either be using donor carts or proper repro boards, not cheap Chinese crap. Doesn't sound like there's much interest anyway...

I will make a poll

[Canvas]

1.5 final has been released, check it oooout

[Canvas]

Posting texh's run of Uku
https://youtu.be/zW90Yhixxho
I killed him

[KawaseFan]

Nicely done, texh!