Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Buckshot

Aubrey Hodges info

Recommended Posts

jdagenet said:

But EX and VLC both use Fluidsynth, how is this happening?


Possibly different versions of FluidSynth. I did confirm the playback is even more glitchy when you set VLC to dump the file directly into FLAC, and I also determined the timing was off on those last 2 tracks in normal VLC playback. Most of the tracks seemingly play fine, but I'm not going to take any chances and just continue doing my live wave dumps from EX's own playback.

I also want to point out that MUSAMB15 (Alpha Quadrant) is an incomplete dump compared to Aubrey's OST, which is twice as long and unique throughout.

Share this post


Link to post
FirebrandX said:

Edit: Also noticed SynthFont has an interpolation issue that leaves its own artifacts on notes that are played at low sampling rates.

You should mention that to the maker of SynthFont. From past experience reporting bugs to him he's always been very interested in fixing them.

Share this post


Link to post
sector666 said:

You should mention that to the maker of SynthFont. From past experience reporting bugs to him he's always been very interested in fixing them.


It would be difficult to report properly. I'd have to make example segments compared to FluidSynth, upload them on my web site, send him the links, and try to explain what he needs to listen for between the two examples. I'm busy with my own projects first, then maybe if I feel like it after, I might go through all the trouble.

Share this post


Link to post
FirebrandX said:

Possibly different versions of FluidSynth. I did confirm the playback is even more glitchy when you set VLC to dump the file directly into FLAC, and I also determined the timing was off on those last 2 tracks in normal VLC playback. Most of the tracks seemingly play fine, but I'm not going to take any chances and just continue doing my live wave dumps from EX's own playback.

I also want to point out that MUSAMB15 (Alpha Quadrant) is an incomplete dump compared to Aubrey's OST, which is twice as long and unique throughout.

How does foobar2000 compare too EX's output? Disregarding the built-in reverb effects, foobar's output seems to be relatively close to EX's, however, there are some minor sample looping errors in foobar that will cause high pitched samples to abruptly stop when they're supposed to loop and fade out, but other than that, it seems pretty close unless I'm not seeing something.

Share this post


Link to post
jdagenet said:

How does foobar2000 compare too EX's output? Disregarding the built-in reverb effects, foobar's output seems to be relatively close to EX's, however, there are some minor sample looping errors in foobar that will cause high pitched samples to abruptly stop when they're supposed to loop and fade out, but other than that, it seems pretty close unless I'm not seeing something.


Good call so far. I just tested both playback and wave dumping of MUSAMB19 and it passed both tests. In addition, it had incredibly fast dumping to wave. I'm using the midi component by kode54 and have it set to BASSMIDI with sinc interpolation. I'll start testing all the tracks to compare for artifacts, but already I'm liking the results. I'll report back.

Edit: Well not so good now. The sinc interpolation causes audible "clicks", whereas linear interpolation has no clicks, but poor quality.

Share this post


Link to post
jdagenet said:

But EX and VLC both use Fluidsynth, how is this happening?

Different versions of the library, maybe?

Share this post


Link to post

So with the try on foobar being a disappointment, I'm back to having to do manual rips of EX's sound output. Maybe at some point Kaiser can add in a wave-dump option, but it's not a big deal for now. It just means a little extra bit of time doing it manually, but then I can flac the finished waves and call it a done package. Since MUSAMB15 is missing the 2nd half of the song data, I just tacked on the 2nd half from miniUSF output. Best I can do for now.

Share this post


Link to post
Avoozl said:

There's a USF of Doom 64 music but you need winamp and the USF plugin to play them, winamp can dump the music in wav format.

http://www.zophar.net/music/usf/doom-64.html


Already know about those files, and I made better ones with proper tag information and loop points (Josh had incorrectly assumed ID made the game and Bobby Prince was the music artist). As I mentioned before though, miniUSF files emulate the real hardware, and as a result, have a lot of high frequency artifacts. If the USF driver had a smooth sinc interpolation option for playback, then they would be the best playback method. As it is, they are one of many options with their own pros and cons.

Share this post


Link to post

I'm almost certain Kaiser said something about MAP15 once, about the behavior of the song being non-deterministic. Have you tried comparing multiple recordings of the so-called missing second half of the song to see if it is in fact the same every time?

Could very well be the case that the game engine is somehow reading past the end of the song and playing something... else. Kaiser dumped the WAD file from the ROM so I highly doubt his dumper only malfunctioned on a single song lump. POSSIBLY the program he used to translate the song data into standard MIDI had a glitch, of course.

Share this post


Link to post
Quasar said:

I'm almost certain Kaiser said something about MAP15 once, about the behavior of the song being non-deterministic. Have you tried comparing multiple recordings of the so-called missing second half of the song to see if it is in fact the same every time?


Yes I have, and they are the same on both miniUSF, real hardware, and Aubrey's own OST for that matter. They all have the exact same data, and are about 13 minutes long. MUSAMB 15 is just plain missing half the data.

Share this post


Link to post
FirebrandX said:

Yes I have, and they are the same on both miniUSF, real hardware, and Aubrey's own OST for that matter. They all have the exact same data, and are about 13 minutes long. MUSAMB 15 is just plain missing half the data.

I'll see if I can get Kaiser to look into it then.

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×