Are you a bad enough dude to increase FFR's classical library?

[jyris1]

Re: Are you a bad enough dude to increase FFR's classical library?

Right now even with DDReam syncing classical is still pretty difficult. Plus, the program is not integrated with SM so for people that are very effective file makers with the SM editor like myself DDReam becomes a burden. I tried syncing "Cast Down the Heretic" with DDReam and it was more effective to just sync it manually using F7/F8. Also, since I don't use Windows I have to log on to a Windows machine every time I want to use DDReam.

Well, I hope youre not the only one who is going to sync stuff on this project.... For people who unwaveringly and exclusively use the SM editor, and moreover those who don't even have ready access to windows, I can definitely see DDReam being a burden. For others who pretty much exclusively use DDReam, or use both programs, it's not a burden. EDIT: That's why I was offering ddream help to anyone who wants to sync stuff for this project.

The two programs use totally different timing algorithms, so complete integration with the SM editor is (practically) out of the question. Each program finds strengths in different areas, assuming a person runs windows and is comfortable with both programs.

The most effective feature DDReam could have is a feature that turns every tapped note into a BPM change closest to the primary BPM, so for example if I tap out every note as the way it actually is (so that you'd have lots of 64ths/48ths/32nds/24ths/16ths/etc due to the rapidly varying sync) on some specified BPM the feature would quantize every note into 4ths using BPM changes. However, I think this feature could be implemented into the SM editor.

I want to know more about what you are talking about. The subbeat feature might address your problem but I am not sure.

In the interest of keeping this thread on topic, I will stop here. I would love to hear more about this via PM or the DDream thread, though

[sleeplessdragn]

Re: Are you a bad enough dude to increase FFR's classical library?

I don't think I understand how DDReam works exactly, although I'm assuming that its positive aspects lie in how it overlays the waveform directly beneath the editor. I kind of wonder why stepping software hasn't been developed yet that simply reads midi data. This should account for perfect sync to the microsecond, and would allow stepfiles to be quickly dumped even from midi files that were recorded live and don't contain quantized steps. Tempo changes can easily be read the same way, and even pitch-matching steps could probably be easily accounted for. This would also allow artists to export their music to midi format to give to steppers for easy stepping.

[Arch0wl]

Re: Are you a bad enough dude to increase FFR's classical library?

I don't think I understand how DDReam works exactly, although I'm assuming that its positive aspects lie in how it overlays the waveform directly beneath the editor. I kind of wonder why stepping software hasn't been developed yet that simply reads midi data. This should account for perfect sync to the microsecond, and would allow stepfiles to be quickly dumped even from midi files that were recorded live and don't contain quantized steps. Tempo changes can easily be read the same way, and even pitch-matching steps could probably be easily accounted for. This would also allow artists to export their music to midi format to give to steppers for easy stepping.

I've wondered that as well and think it would be a very good idea, however I have absolutely no knowledge of programming so I don't know how difficult or easy it would be to make.

[dore]

Re: Are you a bad enough dude to increase FFR's classical library?

Syncing classical music isn't that hard, it's just tedious. Very very tedious.

[jyris1]

Re: Are you a bad enough dude to increase FFR's classical library?

I don't think I understand how DDReam works exactly, although I'm assuming that its positive aspects lie in how it overlays the waveform directly beneath the editor.

That, and how f7, f8, f11, and f12 is essentially replaced by dragging beat markers to line up with the wave.

I kind of wonder why stepping software hasn't been developed yet that simply reads midi data. This should account for perfect sync to the microsecond, and would allow stepfiles to be quickly dumped even from midi files that were recorded live and don't contain quantized steps. Tempo changes can easily be read the same way, and even pitch-matching steps could probably be easily accounted for. This would also allow artists to export their music to midi format to give to steppers for easy stepping.

I actually had a prototype of (almost) this exact idea working about 4 years ago. I scrapped it because there's a monopoly called 'the SM format'

EDIT: ddream supports perfect sync to the actual bytes of the sound file itself (even a midi file cannot easily do that, if at all), and because the SM format is less precise, ddream downgrades when you save to SM format.

EDIT2: deleted
EDIT3:
"microsecond" level syncing is technically overkill and will make zero difference with the sampling frequency of conventional audio files. Assuming 44.1Khz, the maximum resolution of any sync is +-1 sample (1/44.1Khz) = 1/44 milliseconds. One blip of an audio file is thus much larger than a microsecond.

[Arch0wl]

Re: Are you a bad enough dude to increase FFR's classical library?

I actually had a prototype of (almost) this exact idea working about 4 years ago. I scrapped it because there's a monopoly called 'the SM format'

I think he was talking about converting from MIDI --> SM... at least I was.

[jyris1]

Re: Are you a bad enough dude to increase FFR's classical library?

I think he was talking about converting from MIDI --> SM... at least I was.

Yep, and vice versa.

[sleeplessdragn]

Re: Are you a bad enough dude to increase FFR's classical library?

Reason (IMO) seems to do a great job at converting and remastering midi, because you just import the .mid, delete/reset the tempo changes, and rewire the sounds.

FL does pretty much the same thing, but I think there is a large concern with midi files that aren't quantized nicely; happens when people record live. Doing a simple midi dump into a sequencer would still result in offsync notes, probably moreso offsync than the original midi because every sequencer has a minimum possible beat length.
EDIT:

I actually had a prototype of (almost) this exact idea working about 4 years ago. I scrapped it because there's a monopoly called 'the SM format'

I would be interested in taking a look at your source, if you didn't mind of course. I admit the only abstract concepts of midi that I understand I learned a few nights ago, but I'm an active CS major. ^_~

[Ice wolf]

Re: Are you a bad enough dude to increase FFR's classical library?

The only problems I had with it were the ritardandos near the beginning and the end. I have not ever looked at the sheet music, but I have never heard it played that way. Other than that, there was way too much echo.

Overall it was amazing. Good job.

I have found some more unnatural "ritardandos" (possible lags?) at 2:08 and 3:45.

[sleeplessdragn]

Re: Are you a bad enough dude to increase FFR's classical library?

These "unnatural" tempo changes are because I added tempo changes from information directly contained in the midi files. Since midi files can't -really- emulate natural tempo transition, rather its just a series of what the tempo is at a certain time, the tempo changes are bound to sound weird and stoic. In other words, the tempo is changing instantly at certain, small intervals, rather than continuously over one interval. I could have placed the changes on an LFO, which would have sounded a lot more natural, but then it would not have matched the midi file's tempo events, which I thought was going to be used by steppers.

[Arch0wl]

Re: Are you a bad enough dude to increase FFR's classical library?

Speaking of which, could you give the updated BPM changes to La Campanella from your last version?

[Tcjunior18]

Re: Are you a bad enough dude to increase FFR's classical library?

http://www.youtube.com/watch?v=9XTpwihpddI

I don't know if ya'll except classical videos for songs to step, if not then I apologize in regards to sending this video. I was checking to see if this song could be steped and if not just delete this post, If so I have one more called "Grand Etude" very fast piano song. Anyways just let me know.

[Ice wolf]

Re: Are you a bad enough dude to increase FFR's classical library?

http://www.youtube.com/watch?v=9XTpwihpddI

I don't know if ya'll except classical videos for songs to step, if not then I apologize in regards to sending this video. I was checking to see if this song could be steped and if not just delete this post, If so I have one more called "Grand Etude" very fast piano song. Anyways just let me know.

FFR does not use videos or backgrounds anymore, so we would convert it into an mp3. I am assuming, however, that we will need permission from Boris (awesome name, is it not?) to use one of his performances in FFR. You can still use the performance for a stepfile, just not for FFR.

PM me if you want an mp3 of the performance without the applause.

[Bynary Fission]

Re: Are you a bad enough dude to increase FFR's classical library?

Shashakiro used http://www.classicalarchives.com/ to get Grand Gallop Chromatique, I'm sure if you got this site's permission, you could use classical music from there.

And I have stepped a bunch of files for Stepmania, I'd certainly be willing to step classical music, considering it's one of my many favorite genres of music. (Antonio Vivaldi's The Four Seasons ftw).


~Bynary Fission

[sleeplessdragn]

Re: Are you a bad enough dude to increase FFR's classical library?

Since midi determines tempo changes as time offsets from prior changes, I'm a bit confused as to whether you want the exact time, i.e 2:15, or the tempo event time offsets as listed in the midi. I'll update this post with which ever it is.