Deleting Custom Voices

[musicalmemoriesuk]

Ive tried deleting custom voices from UVN Files to make space for others and im still getting not enough space.

[tyrosaurus]

On a Tyros4, deleting expansion voices from slots in the Library Edit screen does not automatically delete any wave samples that they use that are stored on the expansion memory module!

The wave samples have to be deleted manually from the module using the Expansion Module display.

This is easy if a voice only uses a few samples and they are named in such a manner that it is obvious which voice that they are used with but this is not always the case!

However in order to recover the memory space that they use, they must be deleted.  Tip: I delete the samples before the voice data. Then after I have deleted the samples I select and play the voice that I want to delete. If there is any sound it means that I have missed one or more samples that it used, and need to identify and delete those too.

Unfortunately, even after deleting the waves from the expansion module, the space that they used to occupy may not become available for new voices!

This is because the OS only loads new voice samples into contiguous blocks of memory which must be large enough for all the samples used by a voice. In addition it appears to only load them into the remaining empty memory above that already in use.

This means that unless the samples that you have deleted happened the latest ones loaded to the expansion module, the space that they used to occupy will not be at the end of the memory used, and therefore still not available for new voice samples!

The solution is to 'Defrag' the expansion memory after you have deleted all the voices and samples that you want to, and there is a built in function to do this on the 'Property' screen opened by pressing 'Property' on the 'Expansion Module' screen.

'Defrag' will move all the existing wave samples into a contiguous block of memory starting at the beginning of the expansion memory, meaning that all the empty 'holes' in the memory are filled, and all of the empty memory is now located at the end of the module, making it available for loading new samples.

'Defrag' is effective in recovering memory space formerly used by deleted samples, but it can take a very long time depending on how full your expansion module is, and the location of the 'holes' in the memory that remain after deleting samples.

In fact in some cases, it is quicker to make a Library of all the voices that you want to keep, then delete everything from the Custom Voice tables, and Format the expansion module.  Finally, reload the Library to restore everything that you want.

Some users of keyboard models that use YEM to manage the expansion memory complain about how complicated and time consuming it is to use! 

Perhaps they should try the method used by the Tyros4, where although it is relatively simple to just keep adding new voices up to the limits of the available memory (although there are other limits on T4 too!), once you want/need to delete existing custom voice samples, things get decidedly more awkward! 


Regards

Ian

[musicalmemoriesuk]

Thanks for that. Thought I was missing something.

[musicalmemoriesuk]

I managed to remove the sounds in the expansion module thank you.