169928
> DO NOT LISTEN ON MU50
Ok. It happens that I don't have this. Very nice tune, impressive!
*favourited"
Ok. It happens that I don't have this. Very nice tune, impressive!
*favourited"
tags
Level 21 Mixist
Yamaha XG
4th/6
springiness
114th Σ4.834
equinox
116th Σ4.846
germinations
104th Σ5.061
flower power
123rd Σ4.783
spring in pants
114th Σ4.947
Living On Time's Edge
********************************************************
DO NOT LISTEN ON MU50 HARDWARE!!
********************************************************
This song will not play back properly on MU50 hardware due to the vast difference in polyphony limits between S-YXG50 and MU50. The MU50 is hard-capped at 32-voice polyphony and this song uses between 50 and 117 voices on average by taking advantadge of the variable limit in the VST version's settings. You may be able to play it back on an MU128 or later but that's not valid for hardware renders. (There are some differences between MU80 and MU50 that are carried into the later models, and I am currently working on documenting these. However, my attention is divided between a bunch of different projects.)
********************************************************
END OF SPOOKY NOTICE, BEGIN DESCRIPTION...
********************************************************
This song is my attempt at making something half as impressive as the XGroove demo. I started with a chord and from it a song was born. A song that is so 80s, yet also so 90s, you could say it lives on the edge of time...
Made and rendered in SOL2
58
5
19
7
21
previous entry
nextious entry
169951
I'm no expert but, there's a parameter in the s-yxg50 sound source definition in Domino for XG-voice map seemingly being part of the XG-standard. It seems to behave similar to Roland SOUNDCanvas compability-modes, where you're able to choose "XG Basic" which, according to what's described in the definition, uses "old timbres of the MU50 era".
Maybe it's flawed in the same way on MU100 and newer units, as the SC-55 compability mode is on the SC-88 units?
I thought that would've made it possible to hardware render with newer units that's got more polyphony.
Maybe it's flawed in the same way on MU100 and newer units, as the SC-55 compability mode is on the SC-88 units?
I thought that would've made it possible to hardware render with newer units that's got more polyphony.
170257
I believe you're trying to explain the XG Native vs XG Basic map situation.
Fuckle your seatbelts, this is one wild ride.
Before I get too carried away into esoteric MIDI talk, let's discuss terminology for XG devices. Each of the 600+ sounds you can select via program change / bank select LSB / bank select MSB is called a "Tone". Each is comprised of two layers, each called an "Element". Each element is comprised of collections of "samples" that are mapped to specific "regions" of the keyboard, akin to how multisampling works in, say, ImpulseTracker.
To clarify: Starting with the MU100, the map for MU-series modules changed the default tone for a handful of the XG "basic" tones (or those that do not require a bank select to access) to reflect some of the additions to the ROM. However, all of the old versions of those tones still exist at Bank Select LSB 127, and the new version is duplicated at Bank Select LSB 126.
To recap: LSB 126 is the newly added GM-compatible tone for the MU100, whereas LSB 127 is the original version that would play on the MU80 and MU90.
All that the Basic map does is make it so that the GM tones pull from LSB 127 instead of LSB 126, making the instrument map function less like the MU100 and more like the MU80.
These are not the changes that are problematic for playback. Be warned: we're going full throttle MIDI past this point.
Unlike the MU90->MU100 changes, the MU80->MU50 transition is unique in that the MU80 was released first, and the MU50 was a stripped down version of it.
As such, it lacks some alternate tones entirely. The clearest example of this is Synth Strings 3 on the MU80 and MU90+, which is completely absent on the MU50. However, Synth Strings 4 and 5 are still there, and their names were not changed. Another classic example of this is that the reverse cymbal sound was changed for the dance drumkits from RevCym2 to RevCym1.
I hope you're still with me because this isn't the only change that occured between the MU80 and MU50. The MU50 contains additional sample regions for a good number of elements. Mostly found on the high notes, these don't add much to the elements, but it is still a change that breaks playback compatibility.
But where it gets MONDO fucked up is that YAMAHA decided that the MU80 should be the basis for the MU90, but instead of using the MU80's elements, they used the MU50's elements. This means that the MU80 can't play MU50 music 100% accurately, the MU50 can't play MU80 music 100% accurately, and the later models can't play MU50 OR MU80 music 100% accurately.
S-YXG50 is 100% true to the MU50, with the exception of the GS mode being licensed, DOC and Performance mode and the UI being absent and the adjustable polyphony limit.
Currently working on a compatibility table for XG device playback.
Fuckle your seatbelts, this is one wild ride.
Before I get too carried away into esoteric MIDI talk, let's discuss terminology for XG devices. Each of the 600+ sounds you can select via program change / bank select LSB / bank select MSB is called a "Tone". Each is comprised of two layers, each called an "Element". Each element is comprised of collections of "samples" that are mapped to specific "regions" of the keyboard, akin to how multisampling works in, say, ImpulseTracker.
To clarify: Starting with the MU100, the map for MU-series modules changed the default tone for a handful of the XG "basic" tones (or those that do not require a bank select to access) to reflect some of the additions to the ROM. However, all of the old versions of those tones still exist at Bank Select LSB 127, and the new version is duplicated at Bank Select LSB 126.
To recap: LSB 126 is the newly added GM-compatible tone for the MU100, whereas LSB 127 is the original version that would play on the MU80 and MU90.
All that the Basic map does is make it so that the GM tones pull from LSB 127 instead of LSB 126, making the instrument map function less like the MU100 and more like the MU80.
These are not the changes that are problematic for playback. Be warned: we're going full throttle MIDI past this point.
Unlike the MU90->MU100 changes, the MU80->MU50 transition is unique in that the MU80 was released first, and the MU50 was a stripped down version of it.
As such, it lacks some alternate tones entirely. The clearest example of this is Synth Strings 3 on the MU80 and MU90+, which is completely absent on the MU50. However, Synth Strings 4 and 5 are still there, and their names were not changed. Another classic example of this is that the reverse cymbal sound was changed for the dance drumkits from RevCym2 to RevCym1.
I hope you're still with me because this isn't the only change that occured between the MU80 and MU50. The MU50 contains additional sample regions for a good number of elements. Mostly found on the high notes, these don't add much to the elements, but it is still a change that breaks playback compatibility.
But where it gets MONDO fucked up is that YAMAHA decided that the MU80 should be the basis for the MU90, but instead of using the MU80's elements, they used the MU50's elements. This means that the MU80 can't play MU50 music 100% accurately, the MU50 can't play MU80 music 100% accurately, and the later models can't play MU50 OR MU80 music 100% accurately.
S-YXG50 is 100% true to the MU50, with the exception of the GS mode being licensed, DOC and Performance mode and the UI being absent and the adjustable polyphony limit.
Currently working on a compatibility table for XG device playback.
170258
If I had a nickel for every time I had to explain the playback differences between different devices on this site I'd have one nickel but believe me that pile is only going to get larger as time goes on
172621
me, an eyeball who enjoys the music with no technical knowhow: 👁️
174319
I love your ramblings about XG, I'm glad to see someone carrying on with researching XG-related MIDI tech, an undertaking that i once started.
I'm actually really interested where did you manage to find such detailed information about the sample layers and elements that make up each of the patches. Is that mentioned in the official documentation?
I'm actually really interested where did you manage to find such detailed information about the sample layers and elements that make up each of the patches. Is that mentioned in the official documentation?
174332
Information goldmine:
shingo45endo's tone-browser
This was mentioned by now its dark in the discord at one point or another. This developer has been slowly dumping the ROMs for these GM+Friends ROMpler's tone data and listing the parameters in extreme detail. If it weren't for this single, random person's efforts, I wouldn't be nearly as invested in XG as I am, and I would be far less knowledgeable about the inner workings of it all.
shingo45endo's tone-browser
This was mentioned by now its dark in the discord at one point or another. This developer has been slowly dumping the ROMs for these GM+Friends ROMpler's tone data and listing the parameters in extreme detail. If it weren't for this single, random person's efforts, I wouldn't be nearly as invested in XG as I am, and I would be far less knowledgeable about the inner workings of it all.
