Firmware Update Failure consistently at 82% complete

[Cagey]

I'm having trouble updating my MFC firmware to 2.15 using Windows 7 on a laptop via MIDI-OX and USB->MIDI interface. I got "ERROR 10" (Flash Write Failed) several times and noticed it was happening at about the same place. I thought perhaps the Axe Fx was chattering at the thing and the comm interrupts were messing with the timing, so I undid the CAT5 connection and switched to the stand-alone power supply so there'd be no external interference. Same problem. So, I started reducing the size of the buffers and slowing down the packet delay, to no avail. At 82% (consistently), I get "ERROR 10". I'm down to 64 byte buffers with 192ms delay between and it still tilts.

I could try using the Axe FX as the interface, but everything seems to be working fine as it is until it gets to 82% complete, so I don't think the interface is the problem.

Anybody else encounter this, or have any advice?

[MrGuitarabuse]

Since it always stops at 82% perhaps file got corrupted when you downloaded it. Try download it again and see if it works

[Cagey]

Since it always stops at 82% perhaps file got corrupted when you downloaded it. Try download it again and see if it works

Thanks for the reply. I tried that, and no joy. Same result.

I still have the 2.02 file so I'll try to restore that. Otherwise, this thing is bricked.

[MrGuitarabuse]

I checked my MIDI-OX and my settings are 128 bytes and 60milliseconds. Im also on Win 7 and those settings worked for me. could be worth a try

[hussamd]

I use 128 and 16 on the buffer page. Try that. Lowering the buffer was key for a stable update.

You can load my MidiOx profile from here. I have used it successfully many times. Import it from the File menu.

[Cagey]

Update: I was able to upload firmware 2.13 to the MFC successfully, which sorta says Windows 7, MIDI-OX and its settings, the laptop and USB/MIDI interface are all ok. I have a fresh copy of 2.15, which also exhibits the problem. So... now I'm wondering if there are bad memory locations in the upper reaches of the MFC. The 2.13 firmware file size is 167K, while v2.15 is 170K. I wonder if there's any way to test the memory on the MFC? Maybe try to upload some other file that's as large or larger than 2.15 and see if it tilts?

[axme]

I haven't done the latest update but I had the same problem last time. After over 20 tries getting to about 70% I learned that the buffer delay parameter in Midi Ox was was the problem not size. Try adjusting it up or down.

[Cagey]

I haven't done the latest update but I had the same problem last time. After over 20 tries getting to about 70% I learned that the buffer delay parameter in Midi Ox was was the problem not size. Try adjusting it up or down.

I suspected packet size or timing was an issue early on, and have adjusted those parameters up and down without effect. I'm not sure where they are now, but they were good enough to upload 2.13, so I can't imagine 2.15 would require something different. It almost has to be the file size or its contents, and I've replaced the file itself so I'm down to size, or there's something else I haven't considered. Or, the MFC is wonky. I can easily imagine a bad memory location that doesn't get used until I try to upload something as large as 2.15, at which point the problem appears.