Hi! I decided to batch convert Orchestral Tools Berlin Strings library from .nkx to .ncw.
Why? Well that library is old, with updates added, and thought it might be nice to save the library in an "all updated" form. (And it did save some space for the one i've done so far)
Kontakt said it was successful and I have the "Instruments" folder and the "Samples" folder in a new location (compressed/mirrored folders). I copied the other stuff over and added the library and it seems to work.

The question I have has to do with the new folder structure of the "Samples" folder.
NOW there are many subfolders, whereas before there were NO subfolders in the "Samples" folder.
Considering the problem samples not being found due to "too long file paths" I am concerned about this.
I have had to rename Kontakt library folders to correct that (Orchestral Tools Berlin.Strings.Exp.B.Special.Bows.v2.1 had that problem & another I can't recall some time back)

Right now I have a short path to this Kontakt library folder, and I am using Kontakt portable to test them, but in future the path might be a bit longer and depending on the library it seems it could be a problem.

So, is it going to be a problem? If so, what is the best procedure here?
Can the .ncw samples all be moved from subfolders into the main "Samples' folder and batch resaved?
Thanks for any info!

 

mac or windows?

 

Ahh! Thanks for explaining. I do not want to edit the samples, and I understand now what it is doing.
Oh I don't mean Berlin strings are very old ... i only mean there is an update.
And I am using Kontakt Portable for testing this library on another drive, so I do not worry about destroying
I also have many backups all the time.

Well, thanks... I will not care to batch convert anything, I will just archive the Woodwind library like it is, all updated, working and batch resaved. Woodwind took longer and a little thinking to install the updates

And...the long names with the dots is just my RAR source folders. Here is how i named my libraries:

 

It is Windows but TinTin old me why I do not care to convert anyway
I thought it would just convert the .nkx to a newer "version".

 

librarys gets updated , but if there are newer content or samples developer providing additional nkx sample container , most of the time instruments and resource files gets updated . thats all .

i also dont know why people in general not learning first how the folder structure should be for Kontakt Librarys . and if its needed to convert Librarys to ncw files make sure that there are no files which have same sample names and so on . and its always better too keep the original Folder Structure .

i could go on and write a book , but thats too much .

 

Actually I'd add few comments to this.
Unpacking the structure makes the library take more space, because each file has to round up to the cluster size of said partition, but this also makes the library load much faster as larger cluster sizes, especially on SSD are faster for random access.

The library won't get destroyed in the process (rather strange way of putting it), but you have to be aware that nkr are always saved with absolute path, meaning you have to remember to copy the nkrs separately, the batch process won't do it. NKRs are also much more difficult to unpack.

In rare cases, the library converted to ncw will take less space compared to encrypted nkx containers, simply because it included content/samples that were unused, not linked to from nki/nkm. SF libraries are great example of this.

 

learn to use.. "Total Commander" with "inNKX.wcx" the best way for me

 

i using NAS Systems , and i can not see much differents within loading times if it is unpacked samples and or within the container , its near same you dont notice the differents much .

there are a few older libs out there which use the same sample names for different mic positions and so on , there you needs to care of this otherwise kontakt use the samples which are found first , you know what i mean with this .

 

That might be because most NAS systems are limited by 1Gbps bi-directional bandwidth, which for modern SSD is already bottleneck at 4K clusters. SATA III does 6Gbps, and NVMe much more than that. Moving to 8K clusters, random access gets boosted by about 70% as long as there isn't bandwidth bottleneck elsewhere in the system.

Yes, that remains a possibility, but I have never seen it happen - that is to have two containers with samples bearing the same filename.

 

yes sure within Theoery , do you buy really 4-5 one terrabyte ssds to loading orchestral templates for one PC ? i also using 2 pcs conected via VEP so i would needs a few other ssd´s for the second PC . with NAS Systems i have access to all librarys on both systems at the same time and copy only a few libs to ssds of both PC .

if you have and use an Professional Studio and Recording much and writing commercial stuff for Games and Movies than it would be correct , but if you dont make much money with it its too expensive .

dont get me wrong , just saying it only .

 

Last edited: Jan 5, 2019

There are other approaches, such as using one primary machine with all native SATA and NVMe connected, and share the sample resources from there with other PCs when needed. BTW, I'm fine with just one 1TB SSD for samples, and even that is almost half empty.