dam updating/epoms!

T

Tesseract

Member
Think i've just gone through my 5th eprom now?
I managed to update to v9.03 finally a few weeks ago using Axe Edit and it was working for a day...then the next day when i switched it on all the lights are on and it wouldn't load so i swapped in a new one and it was sorted. So i can't see it being a bad checksum or whatever if it managed to load it fine then die the following day.
Ive tried uploading using different computers, tried midi ox and Axe edit and still get problems.
It tends to always get past the good check sum point, finishes verifying the flash and dies when it restarts itself.

Gah time to order some more eproms so i can try to update to v10!
/rant
 
Tesseract said:
Think i've just gone through my 5th eprom now?
I managed to update to v9.03 finally a few weeks ago using Axe Edit and it was working for a day...then the next day when i switched it on all the lights are on and it wouldn't load so i swapped in a new one and it was sorted. So i can't see it being a bad checksum or whatever if it managed to load it fine then die the following day.
Ive tried uploading using different computers, tried midi ox and Axe edit and still get problems.
It tends to always get past the good check sum point, finishes verifying the flash and dies when it restarts itself.

Gah time to order some more eproms so i can try to update to v10!
/rant
Click to expand...

dude 5 eproms = a bigger problem. i'm going on 3 years and have never had an issue with the eprom. Doesnt mean it wont happen, but i'm sure it's supposed to be updatable being an eprom and all.

i would scrap my method of updating and try something else.
 
Hi,

Pretty much the same story here. I have already exchanged 3 EPROM's and the next is on the way. I was hopping to update my Ultra using other person set up and to see if there is problem with my unit or set up. However, it may be just the Ultra. I will try one more time if it fails again I will have to send my unit for service to Fractal. I have been waiting way to long thinking that is maybe my computer or RME Fireface 800 malfunctioning.

I try using different midi device and see what will happen.
 
Rezonans said:
Hi,

Pretty much the same story here. I have already exchanged 3 EPROM's and the next is on the way. I was hopping to update my Ultra using other person set up and to see if there is problem with my unit or set up. However, it may be just the Ultra. I will try one more time if it fails again I will have to send my unit for service to Fractal. I have been waiting way to long thinking that is maybe my computer or RME Fireface 800 malfunctioning.

I try using different midi device and see what will happen.
Click to expand...

Ahhh you're using the RME Fireface 800 too? This must be where the problem is. I still find it suprising and a bit of a joke that the Axe has issues with certain interfaces.
I don't have any other interface to try. Does anyone know a good, cheap alternative for MIDI interface? Does the Axe FX get along with USB>MIDI?

Whats more annoying is i'm now being asked to pay for the eproms. I'm totally fine with that as I understand i've been through alot of them. However they're asking for 60euros for a chip worth a fiver? And 10euros for shipping when the previous parcels they have sent clearly say 3.40 euros for the postage? :shock:
Looks like im stuck on V9.03 for a while.
 
We have three reports of this, all using the same interface. I believe the interface is causing multiple bit errors which makes the checksum ok but the data itself actually corrupted.
 
FractalAudio said:
We have three reports of this, all using the same interface. I believe the interface is causing multiple bit errors which makes the checksum ok but the data itself actually corrupted.
Click to expand...

Maybe time for an MD5 checksum?
:D
 
FractalAudio said:
We have three reports of this, all using the same interface. I believe the interface is causing multiple bit errors which makes the checksum ok but the data itself actually corrupted.
Click to expand...

I'll delete my post. To avoid any confusion on this. I have no idea what kind of checksum you use.
 
AlbertA said:
FractalAudio said:
We have three reports of this, all using the same interface. I believe the interface is causing multiple bit errors which makes the checksum ok but the data itself actually corrupted.
Click to expand...

Maybe time for an MD5 checksum?
:D
Click to expand...

Too late obviously.
 
Inasmuch as I'm the third party who experienced/reported the same problem with the Fireface 800, I'll chime in here.

After going through the same cycle of corrupted EPROMs I switched interfaces and my updates have thus far been flawless. YMMV

David
 
Just for the record, I have to say that I've never had a problem with a Fireface 400.
There must be something wrong with the 800...
 
I'm with Albert and Sean here -- I have no idea what kind of checksum is being used, but it is almost statistically impossible to have just the right bit errors occur randomly to pass a checksum with bad data.

With that said, I would definitely try a different interface and/or program for updating your Axe-Fx. I've had various problems with my setup when I was transitioning from Vista x64 -> Windows 7 x64 and started using Axe-Edit. However, after reinstalling the drivers for my MIDI devices (I have a E-MU X-Midi 1x1 and E-MU 1212M) the only major change I've made is to use Bome's SendSX to send the firmware updates rather than relying on Axe-Edit. For some reason, it just seems to work more reliably for me; I haven't had a single checksum error using that approach.
 
FF8 here (never used to update firmware, as far as I can remember), and preliminary tests do show a possible flaw in updating (either that or a flaw in my testing) -- info already sent to Fractal.
 
steverosburg said:
I'm with Albert and Sean here -- I have no idea what kind of checksum is being used, but it is almost statistically impossible to have just the right bit errors occur randomly to pass a checksum with bad data.

With that said, I would definitely try a different interface and/or program for updating your Axe-Fx. I've had various problems with my setup when I was transitioning from Vista x64 -> Windows 7 x64 and started using Axe-Edit. However, after reinstalling the drivers for my MIDI devices (I have a E-MU X-Midi 1x1 and E-MU 1212M) the only major change I've made is to use Bome's SendSX to send the firmware updates rather than relying on Axe-Edit. For some reason, it just seems to work more reliably for me; I haven't had a single checksum error using that approach.
Click to expand...

XOR8 checksum is used. Which can't do much for 2-bit errors.
 
AlbertA said:
steverosburg said:
I'm with Albert and Sean here -- I have no idea what kind of checksum is being used, but it is almost statistically impossible to have just the right bit errors occur randomly to pass a checksum with bad data.

With that said, I would definitely try a different interface and/or program for updating your Axe-Fx. I've had various problems with my setup when I was transitioning from Vista x64 -> Windows 7 x64 and started using Axe-Edit. However, after reinstalling the drivers for my MIDI devices (I have a E-MU X-Midi 1x1 and E-MU 1212M) the only major change I've made is to use Bome's SendSX to send the firmware updates rather than relying on Axe-Edit. For some reason, it just seems to work more reliably for me; I haven't had a single checksum error using that approach.
Click to expand...

XOR8 checksum is used. Which can't do much for 2-bit errors.
Click to expand...
True; XOR is too simple. I had assumed CRC or similar checksum, and thus my statement. MD5 is probably overkill for detecting simple bit errors. :)
 
Did some more tests and have every reason to believe the Fireface 800 is at fault.

Haven't tested different buffer settings yet.
Using MIDI-Ox (yay, found the Logging feature) and some fun Notepad++ acrobatics, testing between a known-good config (my UM-880, on the ports tested against, has already been used to update my Ultra to 10.01) and my Fireface 800 (latest drivers/firmware) -- the Fireface seems to, in a large SysEx stream, pull 16 (sometimes 14) bytes from later on in the stream and insert them ahead of time. This happens every 256 bytes -- I suspect different buffer settings may alter that (I have MIDI-Ox set for buffer sizes of 256 bytes), but it takes about an hour or so to really do a full test and I should be elsewhere two hours ago.

I love my Fireface, but in this, it seems to be utterly failing.
Given the known-good comparison versus the Fireface and the results I got, I can't think of any other point of failure that's very likely.
Tested on both oldschool 4.05 firmware and the latest 10.01.

Dug up some 2008-era posts on the RME forums that also admit to bugs in lengthy Fireface SysEx transfers -- didn't see anything more recent though.
 
You must log in or register to reply here.
Back
Top Bottom