Korg Forums

I have just purchased masterbits mellotrax mellotron samples in WAV format.
I can read the wav files on the PC but I have no idea how to load them into the extremes sampler. I have a 2Gb compact flash card fitted and have connected the correct USB cable. Have been into Media function on the extreme but it is not at all clear what Im supposed to next. I have consulted the manual but it is of no help at all.
By the way each mellotron sample has its own wav file for each key on the keyboard it is assigned to, does this makes things more complicate to load it in?

By the way each mellotron sample has its own wav file for each key on the keyboard it is assigned to.

Supposedly those waves are organized in folders, where each folder would represent one instrument, you need to copy those folders from PC to CF (when in MEDIA mode on Extreme, go to Tab Utility - upper right dropping menu, select USB storage mode - then Extreme will connect with PC, you'll see it as external media on PC, do the copying).

MAKE SURE THOSE WAVES ARE IN 48 kHz, 16 bit, mono (or stereo, if you prefer) - that's maximum quality Extreme can support - if they are not, convert them before doing anything else.

When you have copied folders with wave samples, disengage USB storage mode, go to tab Load, open one folder with waves, and in upper right dropping menu choose "load selected" - do it for each wave sample. If it is really "one sample per key" you should have 61 wave in each folder.

After this, go to SAMPLING mode. You'll see MS is set to New_MS. Go to upper right dropping menu and choose Rename MS - rename it so you can manage multiple multisamples later.

Then in Samples section you see "no assign". Go to dropping arrow and choose first wave sample, it should be C2 (on 61 key Extreme). Choose this wave and in key range, just below this, choose C2 for base and top key.

Then go to the third tab, don't recall it's name right now because I don't have the keyboard in front of me, you should go to menu which will allow you to define indexes - choose indexes to cover only one key. Default option will be 12 keys, so it's hard to miss.

Return to the first tab. Now, press "Create" in the right portion of the screen. This will create next index - it should be C#2 (base and top key). Now assign wave sample C#2 to this index.

Do the same thing for each key. Next index, next sample - see how it goes?

Of course, you can stretch samples so one sample can cover up several tones, meaning original tones and few tones below and above original one, but if you have samples for each tone, leave it this way.

When you are done, loading all samples in this multisample, you should have all your keys covered.

Now, go to MEDIA mode, create a new directory (from Utility tab), then open that directory, go to tab Save, choose Save sampling data, choose option to save one multisample, choose multisample you made and save.
You should have one yourinstrument.kmp file and one yourinstrument folder (yourinstrument referring to multisample name you selected in Rename MS step).

Do the same with every other mellotron folder.

When you have several multisamples, load them one after the other - just open newly created folders and load those .kmp files.

When all multisamples are loaded, go to MEDIA mode, make new directory, go to Tab save, choose Save sampling data and then choose Saved all.

This will make one .ksc file (which will contain information about all multisamples), several .kmp files (multisamples) and exactly the same number of folders, named the same as .kmp files. This is because each multisample - .kmp stores its samples - .ksf in dedicated folders.

When you are done, go to Program mode, make a new program, or choose preset one and instead of preset multisample assign your newly created multisample or multisamples.

All the samples and multisamples arrangement can be done in Awave studio software, but if you are not familiar with ti, then it's all the same to you.

And everything I wrote is in the Parameter Guide. So read more carefully.

Regards, shrike

shrike wrote:

By the way each mellotron sample has its own wav file for each key on the keyboard it is assigned to.

Supposedly those waves are organized in folders, where each folder would represent one instrument, you need to copy those folders from PC to CF (when in MEDIA mode on Extreme, go to Tab Utility - upper right dropping menu, select USB storage mode - then Extreme will connect with PC, you'll see it as external media on PC, do the copying).

MAKE SURE THOSE WAVES ARE IN 48 kHz, 16 bit, mono (or stereo, if you prefer) - that's maximum quality Extreme can support - if they are not, convert them before doing anything else.

When you have copied folders with wave samples, disengage USB storage mode, go to tab Load, open one folder with waves, and in upper right dropping menu choose "load selected" - do it for each wave sample. If it is really "one sample per key" you should have 61 wave in each folder.

After this, go to SAMPLING mode. You'll see MS is set to New_MS. Go to upper right dropping menu and choose Rename MS - rename it so you can manage multiple multisamples later.

Then in Samples section you see "no assign". Go to dropping arrow and choose first wave sample, it should be C2 (on 61 key Extreme). Choose this wave and in key range, just below this, choose C2 for base and top key.

Then go to the third tab, don't recall it's name right now because I don't have the keyboard in front of me, you should go to menu which will allow you to define indexes - choose indexes to cover only one key. Default option will be 12 keys, so it's hard to miss.

Return to the first tab. Now, press "Create" in the right portion of the screen. This will create next index - it should be C#2 (base and top key). Now assign wave sample C#2 to this index.

Do the same thing for each key. Next index, next sample - see how it goes?

Of course, you can stretch samples so one sample can cover up several tones, meaning original tones and few tones below and above original one, but if you have samples for each tone, leave it this way.

When you are done, loading all samples in this multisample, you should have all your keys covered.

Now, go to MEDIA mode, create a new directory (from Utility tab), then open that directory, go to tab Save, choose Save sampling data, choose option to save one multisample, choose multisample you made and save.
You should have one yourinstrument.kmp file and one yourinstrument folder (yourinstrument referring to multisample name you selected in Rename MS step).

Do the same with every other mellotron folder.

When you have several multisamples, load them one after the other - just open newly created folders and load those .kmp files.

When all multisamples are loaded, go to MEDIA mode, make new directory, go to Tab save, choose Save sampling data and then choose Saved all.

This will make one .ksc file (which will contain information about all multisamples), several .kmp files (multisamples) and exactly the same number of folders, named the same as .kmp files. This is because each multisample - .kmp stores its samples - .ksf in dedicated folders.

When you are done, go to Program mode, make a new program, or choose preset one and instead of preset multisample assign your newly created multisample or multisamples.

All the samples and multisamples arrangement can be done in Awave studio software, but if you are not familiar with ti, then it's all the same to you.

And everything I wrote is in the Parameter Guide. So read more carefully.

Regards, shrike

cheers for this info shrike. Ive attempted to do what you say but the individual loading of each key sample seems very long winded and hard to keep track of. Also each set of samples are around 15 to 25mb each and theres 26 of them in total (if I want to use them all) so does this mean because of the extremes 96mb limit I can only limit myself to 5 or 6 samples?
Also the keys the samples are assigned to are from either G3 to F6 or G3 to G5 so only 3 octaves. Not C2 as you thought. Im struggling with this I have to say I wish it were simpler......welcome to sampling

Also the keys the samples are assigned to are from either G3 to F6 or G3 to G5 so only 3 octaves.

Depending on instrument samples represent, this may not be the problem. Most of the instruments doesn't even sound good in too low or too high frequencies and musicians don't even use such frequencies. This is where Extreme will calculate those frequencies for you. You should assign G3 sample to G3 key and allow Extreme to calculate every other key below G3. To do this assign G3 to G3 as base key and select G3 for top key also - when you do this, F#3, F3, E3 etc etc will be calculated by Extreme. When in upper portion, if your top sample available is F6, assign F6 sample to F6 key and select C7 (on 61 keybed) for top key. This way you could stretch 3 or 4 octaves of samples to 5 or more octaves keybed.

With this method you can even use smaller number of samples, but we wrote about this before, try searching, the basic thing is too assign one sample and let the sampler calculate two keys below and up from the original sample and so on.

Ive attempted to do what you say but the individual loading of each key sample seems very long winded and hard to keep track of.

This indicates your samples aren't looped and are long enough to sustain usual play duration. But this isn't proper way to use samples as tone generators. You should loop them, allow them to start and then repeat specific portion of sample to infinite.

You can loop them in Extreme, but it's way easier to do it with some software. I use Adobe Audition to find perfect loop points and Awave Studio to assign those loop points into sample (can be done with Audition, but it's easier with Awave). When this is done, those samples can be chopped beyond the loop end point and loop information is stored within them, so when you load them to Extreme, everything is already set up. Also, this way you make shorter samples and shrink their size. This will allow you to load all those samples.

Sharp posted one thread about looping with software named Seamless Looper or something like that, I don't do with that software, but it seems OK, do your searching.

Im struggling with this I have to say I wish it were simpler......welcome to sampling.

Sampling isn't so hard when you understand the concept, get skilled with some software and know your workstation. Read Extreme's Parameter Guide, learn to use some audio editing software and create some samples. There is good chance first samples will be useless, but you will learn on your mistakes.

Try reading older topics on this forum also.

Regards, shrike

thanks for this further info although I cant see myself delving much further into sampling other than this mellotron and decent acoustic piano samples (where the extreme is lacking). Im thinking of getting K sounds piano 2 for this.
You didnt answer my 96mb limit question. Does this mean I cannot load all of these mellotron samples (@ between 15 and 25mb x 26) into the extremes sampler, or is this limit just what I can load at one time?, so I can load it all in just in separate segments. This could be very limiting and may mean I cannot get this K sounds piano 2 sample aswell as the mellotron into the sampler?

On a side note, how are those mellotron samples? Are they any good? I have purchased M-Tron vsti and I have to admit I'm rather dispapointed at the quality of the recording.

You didnt answer my 96mb limit question.

I did.

When this is done, those samples can be chopped beyond the loop end point and loop information is stored within them, so when you load them to Extreme, everything is already set up. Also, this way you make shorter samples and shrink their size.

I'll explain further: when you loop samples, you don't have need for them to be as long as they were before looping. For example, if original sample was 2 seconds long and 500 kB, you can loop it (loop start at 0,5 seconds and loop end at 1 second) and chop the rest of the sample. So, you'll get infinite sample because of looping and will cut the sample in half - it won't be 2 seconds long and 500 kB any more, but 1 second long and 250 kB. And even this would be too much, you can get excellent samples in 50 - 100 kB.

So, when you loop those mellotron samples, you don't need them to be as big as they were. Cut everything beyond the "loop end" point in each sample. This way original samples that were 15 - 25 MB will now be 2 - 3 MB. They are probably stereo as well, so if you make mono samples of them, this will reduce their size in half too.

96 MB of Extreme's sample memory should be enough for what you want to do.

Dniss wrote:On a side note, how are those mellotron samples? Are they any good? I have purchased M-Tron vsti and I have to admit I'm rather dispapointed at the quality of the recording.

I couldnt tell you for sure as it sounds like its going to be a while before Ive learnt how to loop these samples and then got them into the extreme, however Ive heard some examples of the masterbits mellotron 16 bit and they sound quite impressive. Whether that then translates after its all been loaded and Im playing these samples on the keyboard remains to be heard...

shrike wrote:

You didnt answer my 96mb limit question.

I did.

When this is done, those samples can be chopped beyond the loop end point and loop information is stored within them, so when you load them to Extreme, everything is already set up. Also, this way you make shorter samples and shrink their size.

I'll explain further: when you loop samples, you don't have need for them to be as long as they were before looping. For example, if original sample was 2 seconds long and 500 kB, you can loop it (loop start at 0,5 seconds and loop end at 1 second) and chop the rest of the sample. So, you'll get infinite sample because of looping and will cut the sample in half - it won't be 2 seconds long and 500 kB any more, but 1 second long and 250 kB. And even this would be too much, you can get excellent samples in 50 - 100 kB.

So, when you loop those mellotron samples, you don't need them to be as big as they were. Cut everything beyond the "loop end" point in each sample. This way original samples that were 15 - 25 MB will now be 2 - 3 MB. They are probably stereo as well, so if you make mono samples of them, this will reduce their size in half too.

96 MB of Extreme's sample memory should be enough for what you want to do.

So how come these samples were not supplied in loop form on the CD if what your saying is correct? Is looping them going to affect the quality and sound?

shrike wrote:

Also the keys the samples are assigned to are from either G3 to F6 or G3 to G5 so only 3 octaves.

Depending on instrument samples represent, this may not be the problem. Most of the instruments doesn't even sound good in too low or too high frequencies and musicians don't even use such frequencies. This is where Extreme will calculate those frequencies for you. You should assign G3 sample to G3 key and allow Extreme to calculate every other key below G3. To do this assign G3 to G3 as base key and select G3 for top key also - when you do this, F#3, F3, E3 etc etc will be calculated by Extreme. When in upper portion, if your top sample available is F6, assign F6 sample to F6 key and select C7 (on 61 keybed) for top key. This way you could stretch 3 or 4 octaves of samples to 5 or more octaves keybed.

With this method you can even use smaller number of samples, but we wrote about this before, try searching, the basic thing is too assign one sample and let the sampler calculate two keys below and up from the original sample and so on.

Ive attempted to do what you say but the individual loading of each key sample seems very long winded and hard to keep track of.

This indicates your samples aren't looped and are long enough to sustain usual play duration. But this isn't proper way to use samples as tone generators. You should loop them, allow them to start and then repeat specific portion of sample to infinite.

You can loop them in Extreme, but it's way easier to do it with some software. I use Adobe Audition to find perfect loop points and Awave Studio to assign those loop points into sample (can be done with Audition, but it's easier with Awave[/b]). When this is done, those samples can be chopped beyond the loop end point and loop information is stored within them, so when you load them to Extreme, everything is already set up. Also, this way you make shorter samples and shrink their size. This will allow you to load all those samples.

Sharp posted one thread about looping with software named Seamless Looper or something like that, I don't do with that software, but it seems OK, do your searching.

Im struggling with this I have to say I wish it were simpler......welcome to sampling.

Sampling isn't so hard when you understand the concept, get skilled with some software and know your workstation. Read Extreme's Parameter Guide, learn to use some audio editing software and create some samples. There is good chance first samples will be useless, but you will learn on your mistakes.

Try reading older topics on this forum also.

Regards, shrike

Is this software free and easy to use?(no big manuals please). Another question, to make things even more simpler could I not just loop one sample of one mellotron sample/key and then just load that solitary key sample in to extreme and just assign to every pitch/key on the keyboard?
Or would that seriously affect the quality of the sample?
Its a pity a mellotron/acoustic piano (top quality obviously) simulator does not exist as a rack module for £200 or something because Id be more inclined to purchase that than foray into something technical which takes precious time away from composing/recording. Im sure theres a gap in the market for something like that.

So how come these samples were not supplied in loop form on the CD if what your saying is correct?

I suppose difference between platforms one could use those samples on is the answer. I know how loops work on Korg, but highly doubt they work the same on Roland or Yamaha (I don't own Yamaha or Roland and never had programmed any of them so this is only my speculation) - I'm not sure loops are compatible between different brands. And there are software samplers - completely another field to observe. So those samples of yours aren't looped originally in order to allow you to adjust them to any platform you are using.

Is looping them going to affect the quality and sound?

If you make perfect loop in a sample, it wont affect sound quality. But that's not something you will learn in one day. Or in a week. Seamless Looper, which Sharp described in one of his posts, finds adequate loop points automatically, as I understood it. Awave Studio does the same, but I'm not to happy with results it achieves. I find loop points manually, in Adobe Audition, and the process can take significant amount of time - I need to find loop points with intermediate time long enough to preserve sound character and this varies between different samples (C3 and C6 of some instrument doesn't behave the same when you compare their wave graphs, for example). After this, I cut everything beyond the "loop end" point, go to Awave Studio and assign those loop points to wave itself.

Is this software free and easy to use?

Awave isn't free, but neither is expensive when compared to other audio software, spend some time with google and you will find prices...
Regarding usability, Awave is very simple to use - has very friendly interface, supports bunch of formats, allows you to edit waves (not as good as Adobe, but this could be matter of my personal taste), allows you to save samples in .ksf format and to make .kmp and .ksc structures (native Korg formats - .ksf. as one sample, .kmp as multisample and .ksc as group of multisamples), allows you to assign loop points to samples and much more.

Another question, to make things even more simpler could I not just loop one sample of one mellotron sample/key and then just load that solitary key sample in to extreme and just assign to every pitch/key on the keyboard?

This would seriously affect sound quality in a bad way (there are some sounds that can be stretched from just one sample, but those are exceptions). Original sample would sound as it should, one or two samples below and above the original one would sound OK, but everything above and below those would be useless - try to do it and see for yourself, assign one sample, for example, C4 and stretch it maximally in both directions.

Its a pity a mellotron/acoustic piano (top quality obviously) simulator does not exist as a rack module

I'm not sure does something like that exists, but this site:
http://www.ksounds.com/html/triton.html
could be answer to your needs. Those samples they offer are programmed to work on Tritons, meaning you get .ksc and .pcg data (so they say, I didn't buy any of those packages).

I dont think I have the time or inclination to buy/use this awave software and do the loops. What Im hoping is someone out there with ability to do this looping is interested in some free mellotron samples and we can do a deal for them to send me the looped samples ready for extreme. In return they get the full mellotron samples which Ive paid for

Quote:
Its a pity a mellotron/acoustic piano (top quality obviously) simulator does not exist as a rack module


I'm not sure does something like that exists, but this site:
http://www.ksounds.com/html/triton.html
could be answer to your needs. Those samples they offer are programmed to work on Tritons, meaning you get .ksc and .pcg data (so they say, I didn't buy any of those packages).

yeah I already mentioned this earlier, I am considering purchasing it, but again after enquiring there is a problem with the size of files and fitting them in 96mb , however Im told they do not need any work doing to them they are ready to go.

It may sound like a waste of time learning how to cut, paste, loop, recreate, and insert samples. Fact is, once you have learned how to work with sampling, the sky is the limit. You won't ever be restricted again because you will have a natural feel and approach to any sample libraries you purchase or attain in the future.

I spent countless hours learning my sampler and how to use samples. Sharp put it best in one of his posts by saying, "Sampling is an art."

I couldn't agree more.

Trust me, once you get familiar with the sampling functions on your keyboard, you will feel a certain amount of pride and that "cut above the rest" sensation. It all boils down to becoming the best you can be as an artist and a musician. We all felt your pain and are willing to help you in any way possible. Learn that sampling function and treat it like the b**ch it is!! Your b**ch!!!

thanks all the same but apart from mellotron and maybe piano I am not interested in using the extremes sampler (both these two will fill it up to its limit anyway). Im not prepared to spend an extra £50 on awave, so I did some searching and I found a program called wavosaur: will this create the loop points and do whats required? What about the program Audacity which I have been using anyway?
If these programs are no good then I would rather get someone else to do the job for me