Some decisions to be made #295
Replies: 3 comments 3 replies
-
FWIW, I do believe there is a meaningful workflow benefit to adhering to established spec whenever possible, as it ensures that the range of existing patches and software tools will be fully functional. There are some very well conceived and mature editors/librarians made specifically for 8-part multitimbral DX series synths, which utilize the "legacy" performance parameters. Even if some of the extended params that are unique to MiniDexed aren't present in those tools, it is so useful to be able to see nearly everything going on from one panel. |
Beta Was this translation helpful? Give feedback.
-
@BobanSpasic you are correct that we want to be compatible with the original DX/TX series as much as pragmatically feasible. For example, it is a must that existing DX7 voice editors on the PC can work with MiniDexed. When it comes to the features beyond the DX7 (e.g., storing performances), MiniDexed is currently not following any TX model. This was mainly done for very pragmatic reasons (read: it was most convenient for the persons who have implemented the respective features to implement them in this way). I also happen to think that this is less critical since most editors I know can't edit performances, only voices, anyhow.
@now-its-dark Interesting. Do you have some links handy? Are those open source? Doing it like real TX models would require additional work, and would require that you always need to have the matching voices (in the matching banks?) around if you want to use a performance. Would this a) be a worthwhile tradeoff and b) who would want to do the work? |
Beta Was this translation helpful? Give feedback.
-
I've taken a look at DX/TX SysEx messages for Performances just to find out that every model uses another implementation and the implementations are not compatible with each other. |
Beta Was this translation helpful? Give feedback.
-
Hi @ALL
I would like to ask the developers about couple of points that needs to be decided at some point in time (better sooner than later).
I apologize in advance of being a smart-ass (not being a developer in this project, but coming with such topics anyway), but I think I should share some experience with the others here.
Yamaha SysEx ID
At the moment, MiniDexed is responding to Yamaha SysEx messages. The pros are that you can use any editor for Yamaha DX7 to work with MiniDexed. The cons - you are bound to the existing format and can't change anything about it.
Yamaha's own solution for maintaining the backward compatibility are additions to the base SysEx dumps where each Synth model reads just the part he recognizes (where DX7 reads just the base format and ignores the rest, DX7II reads the base and supplement formats etc., and MiniDexed is expecting just the base format without additional sections)
These additional parts have their own header identifications.
Why I am starting this topic?
Anyway, dunno how Yamaha would react on messing with their property.
Performance files
The current MiniDexed Performance files have nothing in common with DX-series Performance files. Nobody says they need to.
DX-series performance files are not storing the voice parameters. There are just the "links" to the voices with addition of performance parameters. If one edits the voice - the performance will also be affected.
MiniDexed Performance files are sort of Multi-Layer voice file. My vote goes for multi-layer instead of linked performance files.
Why I am starting this topic?
I think that these topics need to be decided soon, before the development of performance filed goes further in undefined direction.
Beta Was this translation helpful? Give feedback.
All reactions