reading sample maps on another computer

Maybe this is an FAQ, but I couldn't find the answer in the archives.

I have uploaded an ensemble, Karmic_Entanglement, to the user library. It uses 4 sampler modules, and I used sample maps to store the wav samples (a total of 70 samples). Works fine on my computer, but others have problems.

As far as I can tell, the sample map contains both the complete pathname of the individual audio files as well as the wav data itself. If the pathname does not actually exist (as would be the case if someone tries to run the ensemble on a machine other than mine) Reaktor complains about not being able to find the sample data, even though it is right there in the sample map!!

So is this a bug? Or am I missing some obscure interpretation of how sample maps are supposed to be used. It seems very pointless to have the ability to store wav data in the sample map if you still have to have the original wav file present and in the exact same location.

And one more thing - is there a limit on the size of uploads one can make to the user library? I tried to upload a version of my ensemble that has the sample map data embedded in the ensemble, which makes for an 8MB+ size zip file. Every time I try to upload this the upload just times out and I get a Site not found page on the web browser.

Thanks in advancee for shedding some light on these mysteries!

Jim

 

Sample maps are just that... they map your samples on your HD so the sampler modules can find them when they need them (in reality the only time they need to know where they are on the HD is when they are first loaded into RAM). It makes sense that you still need the original samples when you upload the ensemble.

You should of realized something was off when your samples totaled 8+ MB and the ensemble you uploaded was a couple hundred kilobytes at most. How, under such circumstances, could you be surprised when other users tried to run your ensemble but were notified that they were missing the wav files?

The only way they will not need the individual files is if you embed the wav data in the ensemble (which is an option in the module's properties). And yes, as you discovered, the library does not like large file sizes. I'm not sure what the limit is. It used to be 2MB, but I believe this was increased.

What you should do is post your ensemble without the map, and include a link in the description to the sample map which is hosted elsewhere. Ideally you should cut down the size/number of samples so you can upload the whole thing, but if this is not something you want to do, you should follow my previous suggestion.

 

>are first loaded into RAM). It makes sense that you still need the >original samples when you upload the ensemble.

Then why is there an option to store the wav data in the sample map itself? When I saw this option I thought that this was so that one could portably package up a whole map of samples and make a map to MIDI note numbers and so on.

>What you should do is post your ensemble without the map, and >include a link in the description to the sample map which is hosted >elsewhere. Ideally you should cut down the size/number of samples so >you can upload the whole thing, but if this is not something you >want to do, you should follow my previous suggestion.

But John, this is exactly what I did! I posted the small ensemble without the maps embedded, and then I posted the sample maps separately.

The problem is that in order for the separate sample maps to be read by the ensemble, they have to be in exactly the same location (directory) as on the computer the ensemble was originally made on.

This, to me, is a bug, as it prevents the portability of sample maps. Currently, as far as I can tell, the only remedy is to upload the samples themselves and have the users recreate the sample maps themselves by laboriously adding the files themselves.

 

If you store the data in the sample map itself, you no longer need the original files (although I highly advise that you keep them).

 

>If you store the data in the sample map itself, you no longer need >the original files (although I highly advise that you keep them).

Ah, but this is the whole point that I am trying to make - it seems that if you delete the files (or put them someplace else) Reaktor will complain, even if you store the data in the sample map.

This is what I am complaining about. I think it is a bug.

People can't seem to get my ensemble to work, even though they have my sample maps, chock full of the samples. It seems that to get it to work they have to have the samples themselves located in the exact same path that is indicated in the sample map.

 

JJ, your ensemble works. The sample maps load into it fine, you just need to include clearer instructions. When Reaktor encounters an ensemble with missing samples, it asks you to locate them. This is because the sample *files* are missing from the map. You can save other users this inconvenience by deleting the sample map data before uploading the ensemble. Having got past this stage, the users will need to open up the structure and find the sampler map modules, open them, and load up the separately stored sample maps. For convenience they should then check the box marked 'store map with ensemble' in each module's properties 'cogwheel icon page'. You might want to consider re-upping the ensemble having saved it in a state with the modules showing, so inexperienced users don't have to dig about.

Here are those instructions in note form. You can skip number one if you strip the maps out before you upload your ens.


1. On loading, reply to the request to locate wav by pressing 'Cancel'

2. Open up the structure and locate the sampler fm modules.

3. Open up their property sheets, and click the sample map icon on the 'cog-wheel' page.

4. Select 'load map' on the sample map page.

5. Load the appropriate map.

6. Select the box on the property sheet marked 'store map with ensemble'

7. Repeat for all four sampler modules.

8. SAVE ENSEMBLE

9. thats it. There are no bugs involved in this. Its really a pretty adequate system, except if you are working in the User Library.

Actually, How about a separate browser window for loading stuff into modules that take samples? Maybe in a future update... (doesn't hold breath).

Hope that helps you out,

Wuntun

 

i think it is a bug. i have saved a maps with the samples but not
with the ensemble (so save on memory) and when i load up the ensemble it asks me for the location of the samples. then i have to find the samples which is a pain.


james, maybe you could do an upload with individual samples?

mosca did that with his drum machines.

sowari

 

Sowari,

You are right, there is something funny going on here after all (although the solution to the karmic problem should still work).

I can't quite work out what it is...

I did discover the following, however:

The button "store map with ensemble" appears to be an 'on/off' type of thing, but actually the state of the button is irrelevant. Pressing it simply changes the map that will be saved with the ensemble. Later unchecking it makes no difference, the map remains part of the ensemble unless a new map is loaded, and the "store map with ensemble" button checked again. This is a little surprising. I don't have time to think about it right now, but my feeling is that this is the key to the strange behaviour you have been experiencing.

Wuntun

 

Wuntun, your procedure will work, of course, but not for people who are running Reaktor Session, as they can't edit the ensemble.
Even for people running regular REaktor it is definitely a pain to have to edit the ensemble. It shouldn't be necessary if the sample map interface was working properly.

As for the business with the "Store Map with Ensemble" button, setting this certainly does make a difference. If you don't set it the ensemble size is much smaller than if it is set. It is true that the "Map" (meaning the pointers to the files) is stored no matter whether the "Store Map with Ensemble" is set or not. The manual actually calls this button the "Store Sample with Ensemble" button, leading me to believe that it determines whether the samples are stored, and NOT whether the map is stored.

Anyways, the way I would expect this all to work is that if the samples were stored with the ensemble, then the user shouldn't need to have the actual samples stored anywhere on his computer. The fact that it doesn't work this way right now is to me a bug.

Maybe it is fixed in v4.1?
In the meantime I will upload the actual samples (not the map) for my ensemble in the hopes that the REaktor Session users can use them to get things running.