TODO(ADX) by hcs at 2:37 PM EDT on October 19, 2007
1) fix all decoders I have control over to use scale+1 (in_cube, Rockbox)
2) have adxloop generate a valid ADX (I made modifications to do this for someone so they could actually put the modified files back in a game, but I don't know what happened to it..)
by dag at 7:03 AM EDT on October 21, 2007
While you are at it, I think the "base volume" (fixed in in_cube) is wrong. Seems each sampling rates need different values (48khz sound well with ~0x4000, 24khz with ~0x2000, 16khz with ~0xC00, etc), or they get too amplified/distorted. Not exactly these values, just "lower sampling rate needs lower volume" (16khz with ~0x4000 sounds terrible and so on).
by hcs at 8:00 AM EDT on October 21, 2007
Ah, that's an interesting thought, I'll look into some other decoders and see if there's anything to that.
by dag at 11:38 AM EDT on October 21, 2007
Oh, and could you include some option to play a file until the end rather than fading? A few ADXs have a proper ending after their loop end, but in_cube will always try to loop+fade out, or just stop if the fade is set to 0.

edited 11:39 AM EDT October 21, 2007
by hcs at 5:34 PM EDT on October 21, 2007
Indeed, another good point, thanks.
by hcs at 7:46 PM EST on November 26, 2007
dag, if you're still around, do you have an example of a file that plays past the loop end as you say?

Regarding distortion, I've made a modification to bero's adx2wav to make it use scale+1, that fixed issues that someone else was complaining about, does that sound better?
by dag at 7:00 PM EST on November 29, 2007
Several of God Hand's ADXs do (boss2_3ningumi_ver6.adx is a good example I think) but I can upload a more examples if you need.

The mod doesn't sound much different to me but I'll test a bit more thoroughly. Should I upload a few ADXs of different Hz?
by hcs at 1:36 PM EST on December 12, 2007
Note to self:
Cube Media Player gets ADX volume right.
by hcs at 7:49 PM EST on January 29, 2008
adx2wavmod2.
It's a bit complicated to calculate the coefficients, but I think this is now finally correct.

Still get more clipping than I'd like, maybe some more work is in order, but it is certainly better.

edited 8:40 PM EST January 29, 2008
by hcs at 8:29 PM EST on January 30, 2008
Alright, I think we're golden now. Had a few math mistakes and thus had to fudge the constants, it's now working well with the 500 Hz cutoff it was meant to have.

adx2wavmod3

cleaned up the math a little
adx2wavmod3v2

edited 10:55 PM EST January 30, 2008

For comparison purposes, check the level of suck difference between this any any other decoder for this Soul Calibur III track.

edited 11:45 PM EST January 30, 2008


Go to Page 0

Search this thread

Show all threads

Reply to this thread:

User Name Tags:

bold: [b]bold[/b]
italics: [i]italics[/i]
emphasis: [em]emphasis[/em]
underline: [u]underline[/u]
small: [small]small[/small]
Link: [url=http://www.google.com]Link[/url]

[img=https://www.hcs64.com/images/mm1.png]
Password
Subject
Message

HCS Forum Index
Halley's Comet Software
forum source