Report on sequencer problems due to RAM

Anashwaran · Post by **Anashwaran** » Thu Oct 18, 2007 2:42 pm

I have made 2 passes with the Memtest without any errror.
So ?

Anashwaran

Mag66 · Post by **Mag66** » Thu Oct 18, 2007 3:32 pm

Anashwaran wrote:I have made 2 passes with the Memtest without any errror.
So ?
Anashwaran

Yeah.. the errors I got were quite intermittant.. I think I got a couple of errors every 25-30 passes.. (Corrected number of passes)

Anashwaran · Post by **Anashwaran** » Thu Oct 18, 2007 4:22 pm

Mag66 wrote: Yeah.. the errors I got were quite intermittant.. I think I got a couple of errors every 25-30 passes.. (Corrected number of passes)

Do you mean you had your test running for 15-20 hours?
Because it took almost 2 hours for 2 passes.

Anashwaran · Post by **Anashwaran** » Thu Oct 18, 2007 4:29 pm

I must sincerely doubt that it is due to ( now suddenly discovered) RAM-problems that this sequencer is completly getting crazy. I believe it has something to do with the update!

They changed some things also due to the new playfunction, and suddenly many people have sequencer-problems never seen before.
If it will continue I will shurely go back to the previous OS until this matter is cleared. It is not possible to work anymore.
Anashwaran Unless someone can convince me about the issue

loneguitar · Post by **loneguitar** » Thu Oct 18, 2007 5:27 pm

Reading this post and a thought came to me does this mean the memory that Korg uses in the Oasys is the cause of this problem, if swapping the Ram cures it

Could it be a ram quality control issue

sebbytriton · Post by **sebbytriton** » Thu Oct 18, 2007 5:53 pm

loneguitar wrote:Reading this post and a thought came to me does this mean the memory that Korg uses in the Oasys is the cause of this problem, if swapping the Ram cures it
Could it be a ram quality control issue

loneguitar, it's perhaps a good idea.

For myself, got a problem with 1.1 --> 1.2 migration and 1.2 --> 1.3 with a large kind of message "Error Code 2" !!!!!!!!!

My OASYS is working after about 10 updates/restore procedures, it is hard to be "ZEN" for the future !

IF THERE IS A DDR SRAM PROBLEM, PLEASE TELL US !!

danatkorg · Post by **danatkorg** » Thu Oct 18, 2007 6:38 pm

loneguitar wrote:Reading this post and a thought came to me does this mean the memory that Korg uses in the Oasys is the cause of this problem, if swapping the Ram cures it
Could it be a ram quality control issue

I believe that the few cases we know about are issues with user-installed RAM.

Note also that timing issues can cause compatibility problems between two DIMMs which work separately.

These are issues with any system that uses RAM.

Best regards,

Dan

EJ2 · Post by **EJ2** » Thu Oct 18, 2007 8:23 pm

I took it upon myself to replace the factory stick with another stick I had lying around here that matched my additional stick exactly, and since then I have noticed a marked improvement in stability. Maybe using matched sticks had something to do with it too?

Funny thing, Nigel, that thought occurred to me just after making the switch. For now, I'll keep the factory stick where it is, but I'm tempted to go out and get another Kingston.
Cheers,
Eric

elvisjohndowson · Post by **elvisjohndowson** » Thu Oct 18, 2007 8:54 pm

My system has identically matched DIMMs, but I can still freeze the OASYS if I rapidly press switches in the program mode. I think this has something to do with overflowing the input buffer for the O/S. This has happened before.

Elvis Dowson

danatkorg · Post by **danatkorg** » Thu Oct 18, 2007 9:11 pm

elvisjohndowson wrote:My system has identically matched DIMMs, but I can still freeze the OASYS if I rapidly press switches in the program mode.

Which switches? Is this a repeatable crash?

- Dan

elvisjohndowson · Post by **elvisjohndowson** » Thu Oct 18, 2007 9:37 pm

Hi Dan,
It's the row of switches just above the sliders, on the left hand side.

With an older version of the O/S it used to exit back to a linux command line prompt and show an error message that referenced something related to stdin (cant remember the error message precisely cos it happened a few months ago, I think I post that bug on the KLF). Since the error occurred as I was rapidly pressing the switches and it reported an stdin error, I inferred that it has something to do with a routine that samples the state of the switches and control surfaces at regular intervals.

Elvis Dowson

elvisjohndowson · Post by **elvisjohndowson** » Thu Oct 18, 2007 10:02 pm

Hi Dan,
Yes, this is reproduceable. Here are the steps:

1. Select Program User-D 005:Peanut Lead, click on the EXi to bring the Main GUI edit screen (Program P4: Polysix Main).

2. Click on the Tone Adjust switch. Play something on the keyboard and simultaneously start pressing the first five switches on the left side, both rows (i.e. Mix Play/Mute, Karma Scenes 1 to 5 and Mix Select, Karma Switches 1 to 5) and rapidly keep pressing these two rows of switches at random. While doing this, stop playing the keyboard and start playing with the touchscreen and select some of the knobs on the touchscreen. For me, it freezes eventually when I select the Resonance knob on the touch screen.

At the end of these two steps, the entire OASYS's control surface has frozen. However, you can still play the keyboard and the program will still sound. And, after this, you'll have to reset the keyboard to bring it back to normal.

Dan, can you please additionally include a function to warm boot the OASYS instead of cycling it each time with a cold boot?

Elvis Dowson

danatkorg · Post by **danatkorg** » Fri Oct 19, 2007 12:53 am

elvisjohndowson wrote:Hi Dan,
Yes, this is reproduceable. Here are the steps:

Thank you! Your clear and careful instructions made this one easy to reproduce. We'll be working on a fix.

- Dan

elvisjohndowson · Post by **elvisjohndowson** » Fri Oct 19, 2007 5:13 am

Heheh, I'm glad it helped and you were able to reproduce it!

There is a good chance that overflowing the input buffers will probably cause the OASYS to freeze in other modes, such as the sequencer, if the user is rapily working with the control surfaces and the touchscreen.

danatkorg · Post by **danatkorg** » Fri Oct 19, 2007 5:28 am

elvisjohndowson wrote:Heheh, I'm glad it helped and you were able to reproduce it!

There is a good chance that overflowing the input buffers will probably cause the OASYS to freeze in other modes, such as the sequencer, if the user is rapily working with the control surfaces and the touchscreen.

The bug seems restricted to the graphics screens of the PolysixEX and MS-20EX, so far. If you can reproduce this elsewhere, please let me know! Pressing buttons quickly is one of the standard QC tests, so I don't think this is as simple as a general problem with buffer overflow...we'll find out.

Thanks again for your help!