Update: 10Mar06 - LEISURE SUIT LARRY 3 DIGITAL - UPDATE!

[Tom]

shad0wfax has completely re-worked the Fantom XR translation of the Leisure Suit Larry 3 soundtrack, from scratch ... and it's now available for download on the LSL3 Digital Soundtrack page.  It's warmer -- it's brighter -- it's NICER!  Check it out!  

http://queststudios.com/quest/lsl3cd/lsl3cd.html

[Ari]

Listening to it right at this moment. Sounds Great!! You sure have come a long way, Shad0wfax.

I especially like your rendition of "Bambi's Video Workout". Very nice work!

[Alistair]

THe first track doesn't sound better (or good) still, yet.. But I remember liking lots of pieces in the second rework, so hopefully this third rework is even better.

I wonder why I don't like the Fantom XR's sound in parts. My ears just don't seem to like various patches. Maybe I'm too used to the MT-32/SC-55?

- Alistair

[Ari]

Could very well be.

What parts don't you like?

[shad0wfax]

As I told Tom, the LSL3 conversion was the first one I made, when I only had the Fantom for a period of two months or less. I didn't mastered the module very well and, like many of you, I was never very satisfied with the results.

Now I know the module much better and I think there's been a big improvement in warmness and balancing of the overall sound. I think the LSL3 soundtrack to be one of the best ones and I wanted to do more justice to Mike Dana's great work. I suspect there's still a long way to go until I'm able to take the most out of the Fantom, because it's a really capable and wonderful although complex machine. Alistair: regarding the sound itself (the patches), I simply can do nothing if you like more MT-32's or SC-55's patches over Fantom's  IMO, Fantom's patches are a lot better, but as long as it's a matter of taste, every opinion is OK.

Thanks for your comments. Please feel free to express your opinions (it doesn't matter they are negative!).

[Alistair]

My opinion is known, I'm not that fond of Shadowfax's work as a whole. That said I think he should keep doing it, because not only will he get better, there's also a market for it.

And I never said I liked the SC-55's patches over the XR's, that'd be ridiculous. I would say almost every XR patch would be better than the SC-55 equivalent, assuming one existed. I'm just perhaps more used to it (the SC-55).

I also think Shadowfax has a way to go with his conversions, for reasons I've spelled out clearly in the past. That said, keep up the good work, there's lots of it.

Actually ,one more thing. It's unfortunate Shadowfax has to work with sometimes poor MIDI recordings, such as the PQ1VGA GM score which although at most times is well done has lapses, like Hoffman's Arrest, and also has about 10 minutes of missing game music. Of course the PQ1 score was composed for the MT-32 which also presented other problems like some tracks being woefully converted by Sierra audio dept. themselves (i.e. Sweet Cheeks Marie's theme and one or two others), and thus sounds bad however you'd record it.

One final final thing, actually- please use the latest LAME VBR MP3, Shadowfax. If you insist on using the format, at least use the best codecs that the format currently provides for.

And take my posts with a grain of salt. I usually post the things only that concern me. Don't assume thast's all I think, I love some of your conversions, Feral Pig from LSL3, Ivana Skates from LSL5..

If I could say one thing? Try and start more with the original piece conversion before you do your own tweaking. Some tracks don't resemble the originals in a good way. (Hard Disk Cafe tracks from LSL5 spring to mind.)

- Alistair

[shad0wfax]

One final final thing, actually- please use the latest LAME VBR MP3, Shadowfax. If you insist on using the format, at least use the best codecs that the format currently provides for.

There's a simple reason why I'm using MP3 instead of other formats: both my car cd player and my portable audio player play mp3 files but not .ogg or even wma (in the case of my car's cd player). I don't know if they are able to reproduce VBR mp3. I'll check it, and if it's OK, I'll record the soundtracks using it. Anyway, it seems to me that 256kbps mp3 gives already an excellent audio quality.

I'd love to work with the best possible midi file sources, but sometimes all I have is the midi data that's in QS. If you or someone else have got a better source, I'd be very pleased to work with it.

Regarding "resemblance to original", the point is that sometimes I simply want to "make my own version" instead of getting as close as possible to the original track. While some people sometimes would like the results, it's OK that other people not. For instance, the two examples of tracks you like of my conversions are not among my favorites Anyway, I really appreciate your comments!

[Alistair]

I didn't mean to say you should necessarily match the original bona fide and do a straight conversion, I LIKE the fact that you're being creative- for example, Nontoonyt Hotel = excellent.

I just mean you should get the conversion right and make it like the original first, so that part is all well and good ,and then take in another direction for the tracks you want.

LAME VBR MP3 or 128 kbps MP3, shouldn't matter- any MP3file should be fine for the devices. Some CD burning program has issues with VBR MP3 though.

Regardless, it's better than 256 kbps MP3.

- Alistair

[Ari]

2 things I don't quite understand, Alistair:
1. how is 128kbps VBR better than 256kbps? I mean, were talking quality here. Compression is less of an issue, is it not?

2. Why should Shad0wfax first "make it like the original" and only then play around with it? What does this achieve?

[Alistair]

1. how is 128kbps VBR better than 256kbps? I mean, were talking quality here. Compression is less of an issue, is it not?

I didn't say 128 kbps VBR (since the phrase, "128 kbps VBR" doesn't ean anything). I said LAME VBR MP3, which has been proven to be the best MP3 compression codec, although worse than quality 7 OGG Vorbis. It gives betetr quality and filesize.

2. Why should Shad0wfax first "make it like the original" and only then play around with it? What does this achieve?

Well, from my view that would avoid problems such as conversion errors and other otherwise easily avoidable "mistakes" that I believe Shadowfax does make. Once it's been properly converted then he can tweak it to sound nice on his module and change patches or whatever. Otherwise it's a russian roulette whether a track actually sounds either good, like the original, both, or neither. IMO, naturally.

Regards,
- Alistair

[shad0wfax]

Before making the conversion, I always listen carefully to the original track, so it's quite possible that what Alistair reputes as "mistakes" are in fact a result of a conscious decision of mine (which, on the other hand, it would be even worse from Alistair's point of view ).

[Ari]

I didn't say 128 kbps VBR (since the phrase, "128 kbps VBR" doesn't ean anything). I said LAME VBR MP3, which has been proven to be the best MP3 compression codec, although worse than quality 7 OGG Vorbis. It gives betetr quality and filesize.

128VBR does have a meaning. It indicates the average bitrate you choose your VBR to encode around.
Anyway, my question remains. When you choose to encode in VBR, you select a range of bitrates, so encoding at 256kbps constant bitrate is high enough so it shouldn't really matter quality wise. Granted, VBR is more efficient compression-wise, but how is it better quality-wise if you don't go above 192 or 256 anyway?

Well, from my view that would avoid problems such as conversion errors and other otherwise easily avoidable "mistakes" that I believe Shadowfax does make. Once it's been properly converted then he can tweak it to sound nice on his module and change patches or whatever. Otherwise it's a russian roulette whether a track actually sounds either good, like the original, both, or neither. IMO, naturally.

Regards,
- Alistair

Could you be a little more specific on what kind of conversion errors and mistakes you're talking about? I mean, if you're talking about volume, balance, attack and such, I don't necessarily see the point in adjusting them first to be like the original, when you're about to change them anyway.Also, these things tend to be played by ear as it is, since the same instruments have different characteristics on different synths, no? So you must be referring to other errors. I'd appreciate if you could be more specific then.

[Alistair]

Anyway, my question remains. When you choose to encode in VBR, you select a range of bitrates, so encoding at 256kbps constant bitrate is high enough so it shouldn't really matter quality wise. Granted, VBR is more efficient compression-wise, but how is it better quality-wise if you don't go above 192 or 256 anyway?

If you use LAME VBR, when a bit of the song requires 128 kbps, it'l lget it. Under Shafowfax's recording method you'll simply get wasted space and noise between the 128->256 frequency/quality. COnversely if it needs 320 and gets 256.. you get the idea.

I'm no expert in this field- I'm murdering the jargon. But what I'm saying is true, I'm just mixing up a few phrases.

- Alistair

[shad0wfax]

Under Shafowfax's recording method you'll simply get wasted space and noise between the 128->256 frequency/quality

I assume that recording at 256kpbs can lead to "wasted space" in the sense of "not as good compression ratio", but I don't know why this should also introduce noise, as long as mp3 (as other compression methods) eliminate information, instead of introducing more data. On the other hand, I record the Fantom's output using the S/PDIF coaxial digital out, so it has zero added noise and zero quality loss (before mp3 compression, of course).

[Alistair]

Without wanting to stary TOO offtopic, what's the difference between recording via line-in and spdif? I usually have both, and I have some pro recording devices. Never used it though.

- Alistair

[Ari]

S/PDIF allows transfer of data from file to file without going through converting to analog (i.e. going through line-out and line-in), and therefore doesn't degrade the sound. If you have a module with a "digital-out" and a soundcard with a "digital-in", you can record audio with zero quality loss and no electric interference.

I don't know if there're really any advantages when recording from a module's line-out to a soundcards S/PDIF.

[shad0wfax]

I think Ari has explained this topic very well. In a few words, when you record from synth's line out to soundcard's line in, the signal is converted TWO times: (D/A in the module and A/D in the sound card). There's always a slight loss of quality and electrical interference in the conversion processes. The better the hardware, the lesser the quality loss.

If you use S/PDIF, there's no D/A and A/D conversion, beause digital data are transmitted directly. It's more or less like reading digital data from a hard disk.

Keep in mind that standard RCA audio cables are not recommended for coaxial S/PDIF data transmission. It shoud be a 75 Ohm cable. Some composite video signal cables (they're usually yellow RCA) are suitable for S/PDIF.