Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Kawai MP11 Recording issue

I'm using a Patriot drive!

I just tried a 64GB ADATA drive, formatted in the MP11, and had the same warning message and delay.

Then I tried a really crappy 4GB Transcend drive, and it worked with no error.

I can't really say whether it's related to the drive size or not, but again, I strongly suspect compatibility can be improved...the drives I've tried that throw the warning messages seem to work just fine. However, I'm subjected to the message, plus a full "drive scan" each and every time I hit record, for seemingly no real reason at all.

Just a side note:
It is amazing to see, how customers of a 2000EUR / 3000USD device are happily working as alpha testers
for a basic feature, such as USB connectivity - in the year 2014 and not even with USB3.0 .
SCNR

What happens ir you create a small fat32 partition at the beginning of the drive and use theremaingspace with another filesystem as ntfs?

Wouldn't even know how to do this, but why should we? It might be a solution, but any working USB drive formatted on the MP11 should work.

So I tried reformatting my 32GB drive in the MP11 and still get the recording messages. My previous success had been with my 4GB, not my 32GB. It appears to be an issue with large drives.

My guess is that the MP11 doesn't know how to properly read the available space for large drives, and is reporting a small amount of available space, thus the messages about recording with an empty drive.

Hopefully James can take this back as a bug report and get this fixed in a future update.

Mine was also a 32Gb drive (Patriot Express). I don't consider 32Gb to be a large drive - not by today's standards anyway. I've been using a 128Gb drive for 2 years, and nowadays we have 256Gb USB thumb drives!

Not at all surprised, and I suspect you may be right. Certainly, this should easily be fixable with a firmware update. No reason we shouldn't be able to use 64GB drives and even more.

I suggested this because I know that, for instance, some phones work properly with SD cards smaller than certain number of Gb...
Anyway these limitations are usually stated on the specs section of user manual.
I used to have a Sony Ericsson W200 that worked slowly with 4Gb memory card (maximum capacity recomended in specs= 2Gb)

By the way Ineedsomeonetorepairmyspacebar

I did some more experimentation on the CA95, using the idea of creating a smaller partition on the thumb drive.

I used the instructions here to create the partition (summary: download a small program called Lexar BootIt, which allows you to assign your thumb drive as "non-removeable" and hence Windows will let you create partitions in its Disk Management utility in the normal way).

I found:

- Creating a 2 GB FAT32 partition instead of the full 16 GB capacity, reduces the freeze time from 10 s to 6 s. Faster but still unsatisfactory.
- Creating a 2 GB FAT partition (which is the maximum size for FAT) reduces the freeze time to about 2 s. In fact, it may just be that the "Connecting USB" message is shown on the screen for 2 seconds and it doesn't freeze at all, because the piano is playable while the message is showing, unlike before.
- The FAT partition can co-exist happily with a FAT32 partition taking up the rest of the drive space (which the CA95 can't see, but you can use it on your computer and use it to store older files).

So there does seem to be a bug in the way that the Kawais cope with large drives and in particular the FAT32 file system, which they need to fix. However if you're happy just to use 2 GB for recording (which is about 3 hours of WAV or much more of MP3) then try creating a 2 GB FAT partition as a workaround.

Thank you for your feedback chaps.

I will pass on your experiences to R&D, and ask them to check access/checking times on larger capacity USB devices.

Kind regards,
James
x

I think I've about had it with the MP11's USB issues.

Assuming that what appears to be a "large flash drive incompatibility issue" may not get resolved quickly (or ever), I went and bought a SanDisk Cruzer Fit 8GB drive. I really wanted to get a meager 4GB version, for testing purposes, but not surprisingly, it's been discontinued.

I was delighted when I put it in the MP11, and it was scanned very quickly. I then formatted it inside the MP11, and I was delighted how quickly it did that. I was even more delighted when I was able to record audio to it, with no warning and no delay from the MP11.

However, I was shocked when I moved the drive to a computer (Windows 7), and it was reported as "Unallocated" (needing formatting) by the disk manager. When I put it back in the MP11, I was able to load and play back the song I had recorded. Then I put it back into a different computer, and it also showed as unallocated.

So then I formatted it as FAT32 within Windows. But then when I moved it to the MP11 and pressed record, it gave me that stupid warning message!

So with this drive at least, it seems to work very well if formatted in the MP11, but then you can never move it to the computer -- not very useful.

Yet my original 64GB Corsair drive does not have this issue (although of course it makes the MP11 throw that error message and delay).

Any my 4GB Transcend drive also does not have this "unallocated" issue.

Is this just bad luck, and I've discovered a bizarrely incompatible drive?

This REALLY shouldn't be so hard.

rungabic, may I ask which version (32-bit/64-bit) of Windows 7 you are using?

I have tested formatting USB sticks on a CA65 (which should use the same process as the MP11), and am still able to read the data on Windows 7. Perhaps 32-bit/64-bit is a factor? I will try to find out.

It may also be possible that some USB sticks don't like being formatted to FAT (not FAT32)?

Cheers,
James
x

Hi James,

Both computers were 64-bit. The issue with formatting the stick in Windows, and then having the MP11 throw that error message...I just chalk that up to the "bug" originally raised in this thread.

But the issue of this particular stick showing as unformatted in Windows, when formatted on the MP11 -- that's just weird.

Maybe it's just a fluke, and bad luck that it happened to me. I don't know what to do other than try yet another one (and I'm trying to get small capacity sticks, since at least until we have a fix, I feel somewhat certain the MP11 can't wrap its head around anything with modern capacity).

@rungabic You are getting the same issue that I reported on the CA95. Maybe it is an issue specifically with SanDisk drives? But then again, maybe not. Kawai need to either fix the firmware or do some extensive testing across all the brands and issue a bulletin on which brands are compatible... preferably both. In the meantime, try the workaround I suggested above, creating a 2 GB FAT partition.


32/64-bit refers to the bits used for addressing RAM, and should not be a factor in mass storage. We've had multi-GB hard disks for a long time on 32-bit machines.

Are you saying that the DP uses FAT to format the USB drive? Because that would limit the storage to 2 GB, so it's impossible to format the whole of a drive larger than 2 GB in FAT. But there are no problems creating a 2 GB partition.

I am going to add this to the list of things that aren't helpful.^

Okay, so I have had my better half sleuthing on this. Between us both, we have amassed quite a bit of computer knowledge over the years. Here's what I have tried

-Fresh 16GB Patriot Stick, zeroed out and then formatted on the MP11
-Formatted on the MP11, powered down, and then removed

I first tried it on Windows 8.1 x64, where I was promptly notified that it needed to be scanned. Sounds like a corrupt filesystem. Hmm...


So we put it on a Linux system running Gentoo and analyzed it and this is what we found:



1) Dirty Bit - This means the filesystem signified that it had not been unmounted properly prior to removal. I couldn't find anything in the MP11 menu to do this, so I powered down. Seeing as it says "Goodbye." when you shut down, and it's a soft switch, I felt it was a sure bet that the system would safely cap off writing to the USB stick while the piano powered down. Turns out apparently not. Maybe I'm missing something here?

2) No Partition Table -- While not entirely unheard of, this is unusual for most removable volumes formatted on a device, ie digital cameras. This is certainly a WTF moment and more than a little weird.

3) Boot Sector Missmatch -- Dunno what the implications of this are, but apparently the MP11 makes a copy of the bootsector but this backup doesn't match the current one. Huh.

4) Bad Magic Number -- In short, there is a certain field that is incorrect in the formatted volume that should tell the system that it is, indeed, a FAT32 filesystem.

Right now, we are imaging the volume so we can send it to James, but right now it seems like the two main problems are that 1) the MP11 has no way to safely unmount the drive when in use and 2) the MP11 does not write the filesytem header completely or conventionally when formatting a volume.

We will be doing more testing soon, hopefully, and maybe can actually figure out why the MP11 creaks and groans on Windows-formatted volumes, but I would be willing to bet it has something to do with how the filesystem drivers are written for that part of the firmware. Hope this helps.

Another thing I'm wondering is what cluster size the MP11 uses when formatting or what it expects. The default cluster size for any FAT32 volume formatted in Windows varies, depending on the volume's capacity. If the MP11 only knows how to play nice with a certain cluster size, it may explain the slowdown on volumes it didn't format itself.

Update: MP11 is using 32KB Clusters (again, not unheard of, but unusual) and it is missing bootcode in the header. It's something that most systems don't NEED, but again, is unusual. We will try some other tools and see if the results agree and it's not the application misreading the volume.

This is great work. And yes -- I forgot to mention that the USB drives I tried that do at least show up as formatted in Windows, always prompt the "needs to be scanned" message from Windows.

I really think the MP11 firmware is doing an abhorrent job with reading and writing to USB drives (both small and larger capacity), and it's creating big headaches and compatibility issues.

Considering how many pages in the manual are dedicated to the recording system, it's rather surprising how poorly this must have been tested. I seriously get the feeling that whoever coded this part of the MP11 used whatever USB stick was lying around, and just made sure everything worked with that stick. I don't mean to be flippant, but really -- this is ridiculous.

Can you point me to the thread where you first raised your issue?

Anyway, I can look into partitioning, although I know Windows won't do that natively to flash drives, so it requires using a third party tool. Certainly, I'd prefer that Kawai hop to it and issue a firmware fix for all these USB issues.

My guess is they DID test it...with a Mac

What, you're a musician who uses WINDOWS???

Just tried the stick it in a late model Macbook Pro running Yosemite. Works perfectly.

It's a common (though no longer reasonable) assumption to think that most serious musicians are still using Macintosh--especially those with the dosh to drop on an MP11.

Apple's BSD-based OSX may not care about the MP11's--shall we say--cavalier fashion of formatting volumes, but Windows is apparently very sensitive to this. James, I would urge you to take sticks of various sizes, ranging from 4-32GB and do a deep dive with you engineers on formatting cross compatibility and performance between Windows 7, 8, 8.1 and the MP11. I am not a programmer, but I am sure this is code that can be fixed in a firmware update. I'm not worried--I know you guys will do right by us.

Also may want to make users aware that in the most recent versions of Windows, FAT32 is only an option for formatting volumes 32GB and under. This is due to reasons that could only be described as mean-spiritied.

Just formatted the same stick in OSX and it is readable by the MP11, but it says "Checking" for almost ten seconds. Looks like the MP11 comes from a parallel universe where FAT32's specification was just a little...off.

With regard to the CA95 and CA65 and our tireless Lolatu, I can only assume they share most of the same firmware code, since I thought the MP11 was just a CA95 with no cabinet or speakers.