Oasys Glitch/Problem [Video]

ryan42 · Post by **ryan42** » Mon Dec 13, 2010 8:44 pm

Anyone else experience this? This is on Combi INT A-78 but I'm fairly sure I've had it happen elsewhere.

http://www.youtube.com/watch?v=e04uar0130M

cello · Post by **cello** » Mon Dec 13, 2010 9:20 pm

I know this is a useless post, but had to say the pain I felt when I heard what your O did...

positive waves from here to wherever you are. Hope you get her back to her old, better, ways soon...

Edited to say - to answer your question; no, never had anything happen like that on IA-78 or anything else...

ryan42 · Post by **ryan42** » Mon Dec 13, 2010 10:05 pm

Thanks. Send your happy vibes to Texas, USA.

I'm trying to track down if it's related to the sound generation or effects processing. If anyone knows of a Combi patch on the Oasys that's mega-killer complex or has made one that is mega-killer complex, let me know so I can try with it.

I've been messing with the Oasys for about three hours now related to this problem, and I have a few other troubleshooting steps that I'm going to consider.

1) Pipe a MIDI in and let it spend a few hours bouncing that track to disk. Looking at the waveform afterwards from afar will show if there was a glitch on that. Maybe it's in the output/preamp/whatever circuitry.

2) (I've started on this one) Look at a TFX (which shows graphical audio level meter) while playing and see if the cutoff also happens in the VU meter onscreen. I did this, and indeed the glitch happened, but it kept making sound through the glitch, so I couldn't make a determination.

Other advice is appreciated.

Edit: Update. That was fast. I went back on the TFX screen and it did it.

The VU meters for the input signal slam to the right, and slowly come down, accompanied by what I refer to as the "8-bit noise." It looks like something in the Oasys is protecting against blowing your ears out.

This doesn't look good.

Edit 2: After the glitch, the VU meters on screen stay stuck off of their baselane, even though nothing is coming out of the speakers.

ryan42 · Post by **ryan42** » Mon Dec 13, 2010 11:27 pm

A video of test 2.

http://www.youtube.com/watch?v=Rv8oK_EbrrQ

danatkorg · Post by **danatkorg** » Mon Dec 13, 2010 11:41 pm

Sorry to hear that you're having difficulties. This sounds like a hardware problem. If I understand correctly you're in Texas, so call Korg USA and arrange for service!

- Dan

danatkorg · Post by **danatkorg** » Mon Dec 13, 2010 11:46 pm

It's interesting that it shows up on the meters. Are your audio inputs routed to the outputs in any way (including through effects)? If so, make sure that they're completely un-routed (Bus Select, FX Control Bus, Rec Bus etc. all set to Off, sends set to 0, no audio inputs used for EXi Audio Input etc.) and see if the problem recurs.

SCHWEATS · Post by **SCHWEATS** » Tue Dec 14, 2010 6:28 am

I have experienced similar behavior. I 've experienced the 8-bit noise
or similar to the sound when a guitar amp tube goes 'microphonic' ,
if that was the appropriate term I remember from working with various
guitar players from decades ago.

I will watch this thread with great interest

Schweats

danatkorg · Post by **danatkorg** » Tue Dec 14, 2010 6:35 am

One thought that comes to mind is a DC offset in the analog inputs. Check to make sure that the inputs aren't routed through in any way.

Charlie · Post by **Charlie** » Tue Dec 14, 2010 6:44 am

I don't understand yet what you're doing to produce this glitch.

Just play for some time and keep the hold-pedal pressed until ... Oasys runs out of polyphony? Or do you just play for some time with a "complex" combi until you get the glitch? I've never experienced such a problem ...

danatkorg · Post by **danatkorg** » Tue Dec 14, 2010 7:01 am

Charlie wrote:

I don't understand yet what you're doing to produce this glitch. Just play for some time and keep the hold-pedal pressed until ... Oasys runs out of polyphony? Or do you just play for some time with a "complex" combi until you get the glitch? I've never experienced such a problem ...

This is not related to running out of polyphony. See above.

- Dan

Akos Janca · Post by **Akos Janca** » Tue Dec 14, 2010 8:00 am

Just to wish you good luck and quick recovery for the OASYS. That "Bell Bell & Bell" patch is a beautiful Combi! Works well over here.

ryan42 · Post by **ryan42** » Tue Dec 14, 2010 6:34 pm

Thanks Dan for all your ideas and suggestions here. Also, indeed it does sound sort of like a microphonic tube in a way! Interesting, no?

Not to jump to conclusions, but I think I've determined what may be the problem.

As an independent IT consultant and hobbyist Hammond Organ fixer by a few nights a month,

I've found the issue I've been having with the Oasys most intriguing since the keyboard combines beautifully written software with awesome hardware. From all I could look at and determine, it looks like all the sound generation and effects occurs in the computer, in software, and the handling of inputs and outputs and controls is handled through proprietary hardware.

I began to look at this more like a computer software problem, and as I often do with computers that have random crashes in the field, I popped a copy of MemTest86+ in the CD drive, plugged a USB keyboard in, and started it up. When I woke up seven hours later, this is what greeted me:

For those of you who may be unfamiliar with MemTest86+, yeah, all that red stuff, that's bad.

So, I'm going to put some of this on order http://www.newegg.com/Product/Product.a ... -_-Product and as a bonus I'll finally be getting that 2GB that you need to use ESx1 and ESx2 simultaneously.

Also, my god this thing has a fan in it. As soon as MemTest started it kicked the fan to high. I woke up last night and instinctively went to go put my PowerMac G4 back to sleep before I realized I sold it two years ago and the Korg was still making all that noise.

Also, at the completion of the test, the Boot CD sent the ACPI off event, turning the computer off even though the power switch was on. Since there's no soft power button on the Oasys, this was a fun one to figure out.

Anyway, I'm pretty happy that I think it's nailed down to being a simple fix. I'll let everyone know what happens.

Edit: Indeed, Akos, the "Bell Bell & Bell" patch is a beautiful one.

Akos Janca · Post by **Akos Janca** » Tue Dec 14, 2010 7:14 pm

Thank God, it seems to be solved! Interesting that you don't have the RAM expanded. Although it's better now: only one stick broke down, not two.

What can be the reason? Hopefully the OASYS doesn't have an additional hardware problem that damages the memory.

Have you tried the Test mode?

Also, thank you for the memory testing idea (I truly hope I will never need it). Dan, please advise: do you discourage it for any reason?

ryan42 · Post by **ryan42** » Tue Dec 14, 2010 7:42 pm

Success! I robbed 1 GB out of a Pentium 4 desktop and it isn't screwing up anymore!

Heh, I always meant to get around to upgrading it but I just never went and did it.

As for a hardware problem damaging memory, I don't think that is the case. My Oasys is over three years old, and there's a three year old PC living inside. It is very reasonable to me that a three year old RAM stick would fail. I've seen newer RAM die just the same or come dead on arrival. It's fairly common.

I wouldn't readily encourage others to follow my footsteps. My Oasys is out of warranty by over a year, though, and I have quite a bit of experience dealing with PCs in embedded applications, and a ton of experience dealing with PCs by themselves.

Just like I had a problem with the Oasys getting soft-powered-off (which was admittedly not MemTest's doing

), so others may have problems as well. I would hate for my thread to be the genesis of someone else damaging their Oasys.

That being said, though, if you want to test the memory in your Oasys or any computer you have a feeling may be ill, http://www.memtest.org/ is the place to go.

On the Oasys if you get the MemTest ISO from that link (mine was off of a testing boot CD I made) you don't even need a USB keyboard. You just insert the CD-R with MemTest on it into the drive and power the keyboard up. It will autoboot and autostart the test. Errors show up in red, as you can see. Once you've had your fill, eject the CD and turn the keyboard off and back on.

The test needs to run for a long time. Off the top of the screen you'll see the progress bar for one "pass" of the tests. Because memory failures often produce random results, you may want to let it run overnight, as I did.

Thanks again everyone for your insight and ideas. I'm going to play around some more, but I'm reasonably sure that this problem is solved!

Edit: Ah, I didn't know about Test Mode, but I'll take a look at it!

cello · Post by **cello** » Tue Dec 14, 2010 8:11 pm

Well done sir! Great that she's sorted.

Lesson for all here - memory is volatile and probably only a matter of time before we all have something like this to deal with...