RME Fireface 800

P

PaleRider

New Member
Hi all. New user here, and looking to flash my Standard up to version 11. However, I am using a Fireface 800 unit, which Fractal seem to have singled out as the Devil incarnate, with warnings on the update page etc.

I have contacted RME directly about this and they have told me that they cannot reliably comment about this problem, as it comes as news to them, and they have never been formally contacted about this by Fractal or anyone associated with them.

I have always found RME's products and support to be of the highest possible calibre, so I have no reason to doubt them on this.

So, my question is this: has anybody managed to update their Axe Fx using the RME Fireface 800's midi interface, and if so, what driver/RME firmware level were you using? OS X,if it makes any difference.

Is there any hard evidence that the problem lies with RME's device, and if so, has it ever been passed directly on to them in Germany, as opposed to their less than sharp US agents?

Thanks in advance.

Tony
 
Here's the thread with the info I was able to gather...
http://forum.fractalaudio.com/ultra-zone/26882-dam-updating-epoms.html

I haven't *completely* ruled out other aspects of my setup, my experiments were not perfect, and I haven't experimented with other operating systems or the buffer settings I was thinking of testing, but everything else that I tried seemed to point to a fault with the FF800. I didn't follow-up as I don't rely on the FF800 for my firmware updates. I'd hoped to do more tests and then send something on to RME when I had more exact circumstances to pick out, but I never got around to it.

I still have the analysis files here if you want to take a look through them and my notes (which are a little weird... I can't remember why I used "zombie" as a sort of variable name... :p). Between all the iterations of the log files from MIDI-Ox and the various stages of tests I did, it's just shy of 120 megs.

Here's the notes from TEST004, which is probably the most detailed, but the regex strings won't make a lot of sense unless you know what a MIDI-Ox log file looks like. The Search and Replace and general text editing was done in Notepad++.

Code: 
-= TEST 004 =-
buffers/delays same as before, ultra4_05.syx
Testing with MIDI-Ox logging -- only works on MIDI-Ox OUTPUT channels, so the routing is...

MIDI-Ox uses firmware file to generate firmware SysEx message, sends through software to
the Fireface 800 -- Fireface 800 hardware output sends to Edirol UM-880 midi interface
hardware input 2 -- MIDI-Ox, monitoring UM-880 input 2 passes data to UM-880 hardware output 1
This last hand-off should render the post-Fireface SysEx into the log.

...and in summary...

Ox (in) >> FF8 (out) >> 880(2) (in) >> 880(1) (out)

Since both the FF8 and 880(1) are outputs, a filter is applied to all the FF8 Output messages.
The "filter" is taken care of by way of stripping the FF8 data from the log file afterward.
The reason the FF8 outputs messages are not used directly is because I'm not certain that what is SENT
to the outputs from MIDI-Ox is what is sent FROM the outputs on the FF8 (which is what's being tested).

To error-check later, need to test the Inputs and Outputs of the 880 WITHOUT the FF8 in the loop.

Then (paying attention to leading/trailing spaces (might be able to ignore by unwrapping first))...

Search:  \s[0-9A-F\s]+MOX  2     F0  Buffer:   [0-9]+ Bytes   System Exclusive      
Replace: \r\nzombie

Search:  \s[0-9A-F\s]+6   6     [0-9A-F-]+  Buffer:   [0-9]+ Bytes   System Exclusive      
Replace: \r\ntarget

Search:  \s[0-9A-F\s]+6   6     [0-9A-F-]+  Buffer:   [0-9]+ Bytes   SysEx Continue      
Replace: \r\ntarget

Search:  \s[0-9A-F\s]+6   6     [0-9A-F-]+  Buffer:   [0-9]+ Bytes   SysEx End      
Replace: \r\ntarget

TextFX > Text Edit > Unwrap Text

Search:  ^zombie.*$
Replace: 

TextFX > Text Edit > Delete Blank Lines

Search:   SYSX: 
Replace: \r\n

Search:  target
Replace: 

TextFX > Text Edit > Delete Blank Lines
TextFX > Text Edit > Unwrap Text

manually delete some extra spaces and blank lines...
...should now be equivilant to original firmware file.    :)

...but I'm showing a 12 character discrepancy?!...
12 more chars in the original firmware file versus the streamed file... somewhere

Copy "12" (unrelated to above char discrepancy) to the clipboard...
TextFX > Text Edit > ReWrap Text to (Clipboard or 72) width
(wait)

Compare files...

Same pattern of errors as TEST003.
If firmware file...
	ABCDEFGHIJKLMNOPQRSTUVWXYZ
...is sent, the likely resulting output will be...
	ABDCEFHGIJLKMNPOQRTSUVXWYZ
...etcetera.
There is some variation in this.
Usually each misplaced string is 16 bytes and occurs regularly every 256 bytes.
Sometimes, the misplaced string is only 14 bytes (looks like more in ExamDiff, but I'm pretty sure the actual difference is just the 14 bytes being misplaced in the same way as the 16-byte strings).

And there's still that 12-char discrepancy somewhere... I think from the 16-byte line comparison, that it happened about 3/4ths of the way into the transfer, but I haven't pinned down exactly where the extra characters are coming from or whether it's just an error on my part (which seems likely through all the RegEx manipulation).

Anyway... for the consistency of the error I was getting, all the steps I took to isolate the FF8 as much as I could in my tests, the prior reports of problems with that interface (mentioned in the linked thread), and the consistency of non-error results I got when pulling the FF8 out of the chain, the FF8 seemed the most likely cause of the errors.
 
Hi and thanks for your reply. Interestingly, one of the driver updates since June/July last year, when your investigations seem to have been taking place, claims:

New in Version 2.66

- MIDI Out: Sending SysEx did not work reliably

This was in August 2010, so I wonder whether anyone has tried the update via the RME hardware since then. I guess not, due to the stark warning.

Ah well. Too small a user base on both sides for there to be much volume of information on this sort of thing.

Thanks for your own very useful contribution. Much appreciated.

Tony
 
Looks like that was a February, 2010 release? -- RME User Forum / Latest driver updates

I wouldn't have picked up on it anyway though, as I'm on Windows here. XP. There *is* a new driver than the one I'd been using though (2.9992, firmware 2.77) -- I might have to re-test things with 3.034 and see if the issue's still there. If it's not, that would be very cool!
 
Ah- yes. I think you're right. It doesn't look like the OS X and Win drivers follow the same branching though, so maybe this little feature of the Win driver (I am assuming it is a host-side software issue) which never made it into the OS X one.

Not thrilled about being the tester though. :-/

Thanks again.
 
I have a Fireface 800 and I've had issues trying to update the firmware with it, though I've never made the Ax unusable. I have never used any method that was entirely reliable. On a Macbook Pro I've used SysEx Librarian and the AxeFx editor with M Audio Uno and a generic from Amazon midi to usb cable. The generic cable at $7 is definitely the value winner. It seems to work as well as anything else (meaning it works within 3 tries) and it's super cheap. It also works fine as a general midi interface. With the M Audio Uno and the editor I managed to update to firmware 11 on the Axe on the first try.

A bonus for both the Uno and the generic midi adaptor is that they both work fine with my Ipad 2 and the camera connection kit for a really cheap midi solution for Ipad. I haven't done it yet, but I'm going to set up a template on my Ipad to edit parameters on the Axe. The Ipad likely to be as good as anything for doing that. It strikes me that one x/y pad for amp gain/master and one for drive pedal gain/master would be a really great thing and a super easy way to control gain structure.

Mike
 
Last edited:
Presumably, transmitting the Bank A/B presets to the unit via the FF800 would also fail, for the same reason? If they transfer across without corruption, maybe the indication is that my software/hardware combination is OK?
 
Hi, new member and first time poster here. Just thought I'd share my experience with you. I have a FF400 running on Win XP so this might not add to your confidence, but...

Because of the Fractal warning about the FF800 (and assuming the FF400 would have the same problem), I tried for two days to use an alternate midi interface (Mackie Universal Control with 3 built-in midi interfaces) to update my AxeFX firmware. No matter what I tried, I kept getting the dreaded "Bad Checksum" I tried using both MIDI-OX and AxeEdit to make the transfer and I also tried numerous buffer and delay settings to no avail.

I was so frustrated after hearing all the rave reviews of FW11 and not being able to update that I decided to try the FF400 -- it worked perfectly on the first try (Driver 3.034, Firmware 1.7)
 
Been performing firmware updates with my FF800 since V8.03, never had a problem (so far); current FF driver /firmware versions: 3.034 / 2.77
 
As an update, I have just successfully updated from v10.03 to v11.00 using the Fireface 800 in the following configuration:

OS: OS X 10.6.7
Application: Axe Edit 0.9.191
FF800 firmware: 2.77
FF800 driver: 2.90

Direct connection between FF and Axe FX.

Worked perfectly. First attempt too.

Hope this helps those in a similar position (if there are any!).
 
Just remember, that IF things should go wrong, you've been warned. Expect generous use of the handpalm icon in that situation.

Warnings like on the Support page typically only get posted if there have been multiple issues.
 
Yes, but without people willing to test these things as software/driver revisions advance, then there would be no progress.
It seems that under my current configuration, SysEx dumps are problem free, as I have been copying banks back and forth all afternoon with not a single error.
Maybe the risk conditions as well as the extent of the risk need to be better defined on the support page.
 
I have always followed the below guideline:

1. Make sure you have the latest firmware from RME installed and working
2. Fireface OUT only Axe Fx IN only midi connections
3. Follow Fractal Audio recommendation in regards to firmware update if using third party MIDI program software

After upgrading to W7 last year I have had never any problems with firmware update.

Good Luck!
 
Rezonans said:
I have always followed the below guideline:

1. Make sure you have the latest firmware from RME installed and working
2. Fireface OUT only Axe Fx IN only midi connections
3. Follow Fractal Audio recommendation in regards to firmware update if using third party MIDI program software

After upgrading to W7 last year I have had never any problems with firmware update.

Good Luck!
Click to expand...
Well, it's like 4 years after your post, but anyway :)
How do you manage to have the axe edit working if you only use axe fx in only ??
 
You must log in or register to reply here.
Back
Top Bottom