sdsnatcher73 wrote:The issue with the 'missing' tracks is the following. I used the ripped MCM's from Manuel Pazos' site (
http://msxcartridgeshop.com/VientoSur/) and MCDRV.COM to play them back. Those tracks that are in the original OST and the remastered OST don't exist in the pack of Manuel nor in the BGM replayer in the game. I am not sure they even ended up on the actual disks. I'll see if I can find how MCM data looks like and check if there is somewhere on the disks some more data of those tracks, but my guess is Manuel did already do that.
Ah, I see. Yeah, searching the disk images for the missing tracks sounds like a good idea. They should be there, you should be able to find them. Alternatively, you could check the KSS file.
sdsnatcher73 wrote:I will work on the other Micro Cabin games (in the same manner) once this pack is finished.
Be careful using MCDRV for games other than Illusion City (from which the driver was ripped); e.g. I’ve noticed that when playing the Xak music on MCDRV there is a speed difference compared to the original. You can compare it with the Xak VGM pack which I logged from the game itself. Logging from KSS files may be a safer option.
sdsnatcher73 wrote:I think this time track 2 is good!
Hmm, nope, the “swirl” starts and then restarts after the loop. That doesn’t happen in the raw (or the first time around). I checked more thoroughly, the reason is that the drum track (with the “swirl”) is a short continuous repeating loop (~3.6s), and not an exact multiple of the main melody loop (60.1s), so they run out of sync. Unfortunately, they only sync up again after the melody has looped 5 times (300.4s), when vgmlpfnd finds it.
So the working trim points are:
Loop: 1412409
End: 17308877
I suggest you take that long loop, and use a
VGM Loop Modifier (0x7F) setting of 1/4, which you can set with vgm_ptch. Since it’s the main theme that’s liked by many, I guess people won’t mind if it loops a few times too many. And a good reason for me to add support for loop modifiers in VGMPlay MSX.
Alternatively, could find a loop point in between two swirls and space em a little further apart towards the end of the loop, so that it doesn’t exactly match the original, having a slightly longer pause in between, but at least doesn’t audibly repeat it.
sdsnatcher73 wrote:2 new tries at track 10
Mmh, I still hear a small hiccup in track 10. Seems to be a similar case to track 02, where they take 206s to sync up again. Vgmlpfnd finds:
Loop: 29249
End: 9113051
Suggest to take the same approach with loop modifier as track 02.
sdsnatcher73 wrote:and 30
As for track 30… doesn’t seem like there is a proper loop, but listening to the raw, the melody only plays once, in the first 60 seconds orso. The last loop in your capture still plays a part of the ending of the melody, so the loop point is a bit too soon.
The part that plays after the melody seems difficult to loop, muting different channels it sounds like there’s four sequences of different length playing (the whirr, the sweeping noise, the ticks, and a very soft swirl), and they don’t converge, so looping there in some place that isn’t too obvious is probably the best we can do.
Here’s some repeat points for the ticks: 2429385 (00:55.09), 2694323 (01:01.10), 2959267 (01:07.10), 3224205 (01:13.11), 3489140 (01:19.12)
And here’s the first couple of repeat points for the noise sweep: 2385203 (00:54.09), 2951883 (01:06.94), 3518563 (01:19.79)
As you can see two tick loops are pretty close in duration to one noise sweep loop, and the noise sweep has a short silence after it, so I was able to find the following good 12-second loop:
Loop: 2385203
End: 2915085
Captures two rhythm and one noise sweep loops (short silence between noise sweeps is slightly reduced), has a soft swirl in it, and the whirr actually just seems to run continuously by itself so doesn’t need any attention. I think this one’s pretty much perfect
.