Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - insidious611

Pages: 1 2 [3] 4 5 6
51
A new beta has just been released

This is a hotfix (long time coming, sorry for the delay!) that should both allow the downloader to failover to one of 2 alternate mirrors should it be unable to download a file from the main one, and properly return an error instead of freezing if it's unable to download a file at all.

Check the linked releases page for "20170918" or if you're lazy and on windows click here.

If you've got a working install of Dancing Mad, don't bother with this one. No changes have been made to the patch. :P

My apologies for the fact the installer has somehow grown from 15MB to 60MB. I'm not sure why this is other than it seems to have something to do with a massive difference in the way cx_Freeze packaged PyQt5 this time. I haven't changed anything of my process so I have to assume either PyQt5 or cx_Freeze changed.

This installer is still using Python 3.5 (specifically 3.5.4) for now.

52
YukiS: Unfortunately your mirror, as well as your pycurl binaries, appear to be down at the moment.

I'm switching my development environment over to Python 3.5 temporarily until the PyCurl binaries for 3.6 from either YukiS or the actual PyCurl people resurface, just so that I can get the hotfixed installer out. It'll contain the failover to YukiS's server even though that's not currently up atm (I assume it will be put back up eventually/soon), as well as it will contain a failover to another server I have access to which has multiple people maintaining it instead of just my busy self.

53
Thanks again for your help. I'm going to try to buckle down and implement these fixes and get out a hotfix release ASAP. My apologies to everyone for not having the time to devote to this recently.

That said, and I'll be sure to modify the first post to reflect this as well, development has now moved from BitBucket to GitHub. I'm hoping this will encourage PRs from people willing to help with the project, as well as simplify organization for myself as all of my other projects are on github.

https://github.com/Insidious611/DancingMadFF6

For those just wanting the current binaries

https://github.com/Insidious611/DancingMadFF6/releases

54
Thanks! I've made some modifications so that it properly errors out if it can't connect (I had the code in there before but forgot to use it when checking for the *size* of the file prior to starting to download it. duhie >.>) and also it *should* use your mirror as failover...

The bad part? I'm using Python 3.6 locally atm and there are *no* win32/win64 pyCurl binaries for 3.6 available despite pyCurl saying it supports 3.6. So I have no means of testing or deploying right now, so that will have to be put on hold until pyCurl has built working python 3.6 win32/win64 binaries. I've submitted an issue to the pyCurl tracker so hopefully this gets resolved soon.

I'm going to give it a month and if I haven't heard anything back from them I'll move my system back to Python 3.5 for this project just so I can get this change out.

As this is imo more vital (being an actual potential freezebug on the installer) I have not yet completed work on allowing one to choose a preset on the custom track selection page, but that will be my next task.

55
Hi! I did not realize my site was down. I'll get that back up and running asap. I'm slightly confused by YukiS saying he contacted me as I've not gotten any contact from anyone, or else I would have fixed this issue right away :P

EDIT: And up! Silly enough, apache had crashed.

My apologies, again, for the downtime.

56
Unfortunately due to a new job I have next to no time to work on this lately. I will likely try to come out with a new Installer soon to incorporate Hal_Emmerich's suggestion (Why didn't I think of that?) though this may take a while as it'll involve editing the GUI which is a bit more work than changing the back-end functions of the installer (*shakes fist at Qt5*). I have *not* had any time to edit the ASM lately though, because this is an extremely time consuming process at this point due to how much time testing takes. I apologize for this.

I'll also look at the provided new set of music. Thank you for putting so much time into that. I might add that as a new option at some point.

57
Just an update, the DNS issue should be fixed and the installer should be working properly now. Sorry for any inconvenience.

58
Just wanted to say a couple things, one, I am still working on this when I have time, I've had a bunch of new obligations come into my life since the last time I posted which is why things went a bit quiet for a while. Still working on the fading code based on Madsiur's work above, and still appreciate any help people can bring in on the ASM side. :)

The second is, the installer will currently fail to download songs due to issues with my domain provider, this will be fixed within the week but if anyone is trying to install right now and wondering why it won't work, that's why.

59
Beta 2 is out

This is a minor bugfix release. Changes include:

* Removal of unnecessary requests to Track 0 for MSU-1.
* Workaround for Figaro->Sabin->Figaro bug.
* Workaround for Opera track issues (Opera tracks reverted to SPC)
* Credits section in Readme

As always, download here

60
Currently testing an installer that has the Figaro/Sabin workaround (see https://bitbucket.org/insidious611/dancingmad/issues/16/incorrect-track-is-called-in-figaro-castle) and the current Opera workaround (which reverts to SPC for the opera tracks) (see https://bitbucket.org/insidious611/dancingmad/issues/6/opera-tracks-may-not-be-implemented). If this works, this installer will be bundled up as Beta 2.

61
Thanks to the flexible nature of the MSU-1, you should be able to continue by removing the opera tracks from your folder... Give me a second to look up which ones those are...

EDIT: assuming you're using Higan, delete

track-38.pcm
track-39.pcm
track-65.pcm
track-66.pcm
track-67.pcm
track-68.pcm
track-69.pcm

from Emulation\Super Famicom\ff3msu.sfc\

If you're using the SD2SNES iirc they'll be named either ff3-38.pcm etc or ff3msu-38.pcm etc. Same numbers, delete them.

This will trigger the MSU-1 to return "badtrack" for these files, which will make it default to SPC music.

I'll be modifying the installer soon to temporarily cause the "Other" versions of these tracks to not be downloadable/selectable, until a solution to this issue can be found.

62
Well, the thing is, changing the songs requires redownloading them, so pretty much guaranteed you're not going to cut much off the install time.

That said, I do plan on releasing an updater, which will also function as a way to change which songs you have selected.

Please report bugs on the issues tracker in future. https://bitbucket.org/insidious611/dancingmad/issues/

It's really good to have someone who's actually tested the opera scenes though. I was expecting trouble. I'll probably just use the SPC for those in the next version of the patch, because it's probably a timing bug and frankly the amount of fiddling that'd be necessary to get the timing right in that scene... I'd basically have to rewrite the whole thing from scratch I'm pretty sure.


EDIT: To elaborate and get into some technical details: The bit where it's freezing is likely where Celes is supposed to "sing" the words, which probably relies on direct calls to the SPC to do its dirty work, and thus relies on the proper track to be playing on the SPC. The only workaround I could think of that would allow the MSU-1 stuff to play without me having to significantly change the game's behavior would be to mute or significantly quiet the SPC but still have it playing, on top of the MSU-1 music, during that bit. But you wouldn't actually be able to hear Celes singing.


EDIT EDIT: Also, good to know that it works with the retranslation patch!

63
Hey, always glad to have another eye on the code. And thanks so much for the commented disassembly, this may help, even if just in implementing a single-case workaround.  This project has been literally my first time doing anything serious in ASM, and my first time doing anything with the SNES, so the more the merrier as far as eyeballs. Feel free to submit a patch if you've got a really clever idea.

I've already tried messing with the NMI a little, and yeah, it's *really* tight, and I feel like I'm nowhere near good enough at optimization to touch that atm. I've tried with a variation of mlarouches code and got massive corruption, so clearly CT is much looser as far as how much "extra time" you have during the NMI.

RetroRGB just put up a video he did with an interview with RetroDan, great stuff, really intriguing to watch. It mentioned this project in a couple spots and I'd like to clarify one thing. RetroRGB asked Dan when we were planning for release, and Dan (correctly) stated there's no real planned release date. But I'd like to explain in more detail my thought process as regards a "release".

I kind of cut my teeth programming wise in the open source community. In this community, the concept of the "perpetual beta" is an accepted if sometimes joked about thing. The idea that it'll reach 1.0 when it's perfect, but it's never perfect.

Basically what I'm saying is, don't hold out for a "final release" to try this puppy out, because you might be in for the long haul. I've still got fading, properly timing the the Opera, and even implementing the PSX FMVs to tackle (Something I have to admit I'm terrified about because I've read bits on different wikis about how SNES graphics programming works 3 times and still don't have a clue what's going on). None of these are what I would consider 100% important to enjoying this hack right now, but all of which I would consider necessary for a "1.0".


64
The bugs are partially SD2SNES firmware bugs. Ikari is helping a great deal in this regard and I am very grateful for the assistance. A new version of the patch with basic workarounds in place is available as an IPS here. SD2SNES users, apply this patch to an *unmodified, original* FF3 US v1.0 rom, and replace ff3msu.sfc with it. Other users, the known bugs are simply not applicable to you. Continue to use the official beta. Whether or not this IPS patch becomes part of a second beta release depends on how quickly new versions of the SD2SNES firmware come out and whether or not these workarounds are necessary or beneficial on the new firmware. IOW, consider this a "stopgap" solution.

65
Just so people are aware, there's currently a bug on the SD2SNES that we're attempting to fix. Head to https://bitbucket.org/insidious611/dancingmad/issues/13 if you want to see the gory details but in general just be aware that I know of this bug and am doing my best to track down the cause and fix it.

66
WE HAVE HIT OPEN BETA

Assuming I've got everything properly working, the open beta should now be publically available from https://bitbucket.org/insidious611/dancingmad/downloads/DancingMadInstaller.exe

Linux/OSX users please download from https://bitbucket.org/insidious611/dancingmad/downloads/DancingMadInstaller.tar.xz . Linux/OSX users please remember that PyQt5, PyCurl, and python-ips are necessary dependencies for this installer on your platform. Windows users just grab the exe, you don't need anything special installed.

Please read the readme given in the installer thoroughly. Please report any bugs in the issue tracker here: https://bitbucket.org/insidious611/dancingmad/issues but please keep in mind any bugs that have already been reported.

To summarize some bits found in the readme:

You will need Higan v097 or newer, bsnesplus v073+2 or newer, or an SD2SNES to enjoy this romhack. As far as I'm aware, Higan v097 and newer is the only platform which currently has "resume support", ie, the ability to resume the area music after the conclusion of a battle. SD2SNES will likely gain this support soon.

You will need an unmodified, US v1.0 copy of Final Fantasy III for this romhack. Japanese versions and US v1.1 will be supported at some point in the near future. SMC header stripping is included as part of the installer, but for safety's sake I would suggest using unheadered copies of the ROM.

You will need a reliable internet connection to run the installer. It will download a *considerable* amount of raw PCM audio files from a remote server. If this server is down the installer may crash or otherwise fail. If the installer fails, please contact me. I'll guide you on how to retrieve the install log and give it to me for analysis.

Please, please, *please* if you are going to make a bug report make it as detailed as possible, leave nothing out. Please do not report already known bugs (the issue tracker is there for a reason.)

67
The installer has been updated and re-packaged. There was a little bug in the packaging process, which I am doing some testing in order to make sure I've corrected.

68
Absolutely. Though Linux/Mac users should keep in mind that for that, PyQt5, PyCurl, and python-ips are mandatory dependencies. (The last is usually installable via pip, the first two are usually downloadable via a normal package manager or so)

On Windows these are included with the EXE so there's no worries there.

69
As we get closer to Open Beta release I'm going to try to give regular updates on my progress. Today I went over the code for the installer, fixed a few minor bugs, and began to set up my environment so that I can turn it into an exe (It's been a while since I've done this so I did not have the correct python packages installed).

Since the last time I did this, Python 3.5 has come out, which requires Windows Vista or newer. For now, I'm sticking with Python 3.4 for the installer as I want this hack to be installable on Windows XP, but I figured I should let you guys know as eventually I may be forced to switch to 3.5.

70
Alright, I have *finally* got the Dancing Mad parts 1-3, and thus the final battle, 100% working without causing any bugs (that I've found so far anyway) in the rest of the game.

I am going to do some more testing over the next few days, get the Installer 100% working again, test that for a bit, and then at some point after that the open beta should be released. I am not one to give or guarantee release dates, but Soon(TM).

When I do release the open beta, I ask you all to please remember what "beta" means. This will probably have many bugs I and the handful of closed testers have not yet found. It's your job to find them, and give me detailed reports of what caused them. It is not your job to expect a 100% finished product or to complain about this that or the other if it's not a bug, or to give bug reports that are short and unhelpful such as "It doesn't work".

71
I haven't yet fully updated the Installer, therefore I haven't created a new EXE for it.

I'm still working on trying to figure out the weird bugs with trying to get the final boss music working.

72
So anybody figured out yet if this is a proper port or just some sort of emulator?

Either way I've never actually finished FFIX so this might be worth grabbing at some point.

73
My Dancing Mad fix is currently causing a very odd bug wherein a certain random track is being played instead of silence in certain other moments of the game, so I'm trying to track that one down but it's so far eluding me. Figured I'd update you all on progress however.

So far my hunch is it has something to do with how I'm overwriting certain code and it's not quite... clean, and I'm causing certain CPU flags to be set that shouldn't be, etc.

74
I don't want to say too much, in case I'm wrong, but I believe I've found out how to do the subsong switching necessary for Dancing Mad. Which is one of the last blockers before an open beta.

75
Resume support has been committed to the private repository for Alpha testers, and the experimental branch has been folded back into the master branch. The current state of code on the repository should be considered "Closed Alpha 2". I would appreciate someone doing a full playthrough in Higan v097 to make sure that resume support is working right, and same in SD2SNES when resume support is merged into the main firmware for that device (afaict it has not been yet).

Higan v097 uses a different file structure than v094 and while honestly I consider it *easier*, it's also a bit of a pain to convert from. Basically, if you currently have an installed version of this patch with ff3-#.pcm, ff3msu.msu, and program.rom or ff3msu.sfc/ff3.sfc files, you need to rename all the pcms to track-#.pcm instead of ff3-#.pcm (where # is the track number), rename the ff3msu.msu to msu1.rom, make sure the main rom is named program.rom, and delete any manifest.bml file.

*However* this now means that installing for newer higan, once I've finished tweaking the installer, will be almost 100% drop-in, as opposed to the earlier, clunkier method. Basically you'd just have to either install to a directory under %USERPROFILE%\Emulation\Super Famicom, or make one and copy to it, and it should all work fine. Generating manifests for your specific rom are a thing of the past.

Pages: 1 2 [3] 4 5 6