Topic: FF400 Disconnects from Host

Hi,

I have been building an interactive application for a customer.

Setup:
Macmini 2.4GHz - Current Model - New
FF400 - New
FW800 to FW400 Cable
MaxMSP/Jitter (Latest Version)
Reason5
MIDI input device.

This interactive installation is interacted with via MIDI. It needs to run permanently without being shutdown (day and night).  It creates sound (via Reason patches) and vision from Max Jitter.  Max also cleans up the MIDI and triggers Reason.

Problem is that the FF400 loses connection to host for no reason typically in under 12 hours.

I have tried:
-Replacing FW800 to FW400 cable with different higher priced cable.
-Using Bus powering for FF400
-Using External powering for FF400

It still drops offline.

When the FF400 disconnects it can not be reconnected without the Mac being restarted AND the FF400 being restarted (external power to the FF400 must be cycled).

I have had a FF800 for around 5 years.  It is a great box.  I substituted the FF400 for my FF800 to try and resolve whether the issue was with the Mac software combination or with the FF400.  The FF800 has been up now for a week without disconnecting.

I would really like to get my FF800 back.

Any ideas?  Any incompatibilities with this setup that I should be aware of?

Thanks,
James.

Re: FF400 Disconnects from Host

Can you test the FF400 with another computer?


Regards
Daniel Fuchs
RME

Regards
Daniel Fuchs
RME

Re: FF400 Disconnects from Host

Hi Daniel,

I have had the FF400 plugged into an old Dell X1 Laptop with a 4pin FW400 interface.  It is running Windows XP Pro.  It has been up for 6 days now running Reaper with a session permanently open routing audio in to out (I have basically being using Reaper as an FX unit and the FF400 as the IO).  I have had no issues.

I would also note though that I had no issues with the FF800 which, up until the dropout problems with the FF400, was connected to the Reaper/Dell X1 setup for 4 weeks without issues permanently operating as an FX unit (the exact same setup).

I don't have a second new Macmini to do the same tests on as the curent one (maybe) being a problem (and don't really want to buy one).  I didn't update the drivers on the Macmini for the FF800 when I plugged it in, so as I see it the Macmini FF400 and FF800 test only has one thing that is changing (being the FF400 to FF800).  They are all latest drivers and firmware on the Macmini/FF400/FF800 setup.  The X1 drivers wouldn't be far off if not the latest (not that I am having issues with the X1).

Thoughts?

Thanks,
James.

Re: FF400 Disconnects from Host

mecanome wrote:

Hi Daniel,

I have had the FF400 plugged into an old Dell X1 Laptop with a 4pin FW400 interface.  It is running Windows XP Pro.  It has been up for 6 days now running Reaper with a session permanently open routing audio in to out (I have basically being using Reaper as an FX unit and the FF400 as the IO).  I have had no issues.

I would also note though that I had no issues with the FF800 which, up until the dropout problems with the FF400, was connected to the Reaper/Dell X1 setup for 4 weeks without issues permanently operating as an FX unit (the exact same setup).

I don't have a second new Macmini to do the same tests on as the curent one (maybe) being a problem (and don't really want to buy one).  I didn't update the drivers on the Macmini for the FF800 when I plugged it in, so as I see it the Macmini FF400 and FF800 test only has one thing that is changing (being the FF400 to FF800).  They are all latest drivers and firmware on the Macmini/FF400/FF800 setup.  The X1 drivers wouldn't be far off if not the latest (not that I am having issues with the X1).

Thoughts?

Thanks,
James.

Are you runnning OSX or windows on the mini? It might be a firewire bus related power saving thing. Also did you use the same firewire cable for the 800 and the 400?

Isaac P

Re: FF400 Disconnects from Host

Hi Isaac,

Macmini 2.4GHz is running the as supplied OS (OSX 10.6.*).
Replaced the firewire 800 to firewire 400 cable as a first measure to confirm if it was suspect.  The cable made no difference.  We didn't buy a cheap cable either (not that cost necessarily is an indication of performance, sometimes it is marketing more so...).

The only change is the FF400 was unplugged and the FF800 was plugged in.  Software drivers, cable and all other software aspects were left unchanged on the Macmini.  The only change made in software was to reconfigure the audio routing in Reason and MIDI routing in Max.

Any thoughts into why the FF400 falls over to the point that it needs to be power cycled to recover where the problem doesn't exist when an FF800 is physically substituted with no other changes?

Cheers,
James.

Re: FF400 Disconnects from Host

You did not list it specifically, but did you also both FW ports on the FF400? Did you try to slightly shake the cable/connection while transferring audio to make sure that there is no loose contact?

Would it be possible to run the FF400 connected to Reaper on the Mac (OS X version) for some hours like you did with the Dell?

Re: FF400 Disconnects from Host

mecanome wrote:

Hi Isaac,

Macmini 2.4GHz is running the as supplied OS (OSX 10.6.*).
Replaced the firewire 800 to firewire 400 cable as a first measure to confirm if it was suspect.  The cable made no difference.  We didn't buy a cheap cable either (not that cost necessarily is an indication of performance, sometimes it is marketing more so...).

The only change is the FF400 was unplugged and the FF800 was plugged in.  Software drivers, cable and all other software aspects were left unchanged on the Macmini.  The only change made in software was to reconfigure the audio routing in Reason and MIDI routing in Max.

Any thoughts into why the FF400 falls over to the point that it needs to be power cycled to recover where the problem doesn't exist when an FF800 is physically substituted with no other changes?

Cheers,
James.

I don't use OSX so I don't know what power saving is going on in the background. I would guess that the way the FF800 and the FF400 interact with the firewire port is slightly different. This could mean that the minimac is putting the port to sleep to save power. Timur does this make sense to you at all? I would also definitely check the things mentioned by Timur above.

Isaac P

Re: FF400 Disconnects from Host

Hi,

both FW ports appear have exhibited the same behavior, but this question leads me to ask if there is a reason or a know issue where one port on the FF400 behaves better than the other.

Port seems reasonable tight in terms of connection.  Handles wiggling, although I didn't feel it necessary to apply a magnitude 9.1 style tremor to the cable.  I would also add that in the installation the cable and equipment is never touched and is out of reach.

Unfortunately I can't play around with the Macmini install easily.  It is installed in a location where the client only wants me to work on it outside of public access hours.  I also don't want to start messing with the nice cleanly installed setup (which is still working fine with the FF800).  I really don't have the option to start conducting a full gamut of tests on that system.

I do have to say, in my experience of fault finding, substitution of one component at a time usually deduces the problem item.  Currently the substitution indicates it is likely the FF400.

I am keen to hear if anyone has had similar issues with the FF400 (particularly with Mac's and more particularly with Macmini's).  I am also keen to hear that users have long uptimes with FF400 on Macmini's (particularly the current model) without issue.

Ideally the test I would like to try is to get another FF400 and substitute that into the system and see if it behaves the same.  Unfortunately I doubt that is easy to arrange.  RME - can you send me one to trial? smile

The main problem I currently have is that I have loaned my personal FF800 to a project to make it work (which it does...) and my business is going to get left holding a FF400 without a use (when it is not trusted for these types of installs) and will have to buy the more expensive FF800 to solve the issue...  As such I would obviously like to get the FF400 to work or be given a reason that allows me to return it.

Thanks,
James.

Re: FF400 Disconnects from Host

Thanks Isaac.

Are power saving issues with FW ports on Macs an acknowledged problem or acknowledged potential problem by RME?

Subtle differences between different RME firewire interfaces and the way they interact with the host is not something that, as a user of the product, I could ever factually determine (only speculate about) - I would likely need to be a design engineer at RME to access the sort of knowledge required to provide those types of answers.

I would note that the FF800 has FW800 and FW400 ports as opposed to the FF400 which only has FW400 ports. That likely means different FW port related silicon for the two models.  It may be the case that the silicon and firmware driver aspects of the FF400 when combined with the current model Macmini software/hardware are causing the issues and that the issues are not seen when using (speculated) different silicon and related silicon firmware subsys of the FF800.  Also, my FF800 is 5 years old.  RME may have changed the FW port electronics since then.  Like I was saying earlier, I would have to be an RME design engineer to get a clear full picture of that stuff.  All I can do as a user is try swapping, playing around and asking questions of others and support.

Thanks for your help,
James.

10 (edited by Timur 2011-05-13 20:14:10)

Re: FF400 Disconnects from Host

My own FF400 works on a multitude of Macs, both OS X and Windows:

MBP 2008 via faulty LSI/Agere revision 6 chipset via repeater cable. MBP 2009 via fixed LSI/Agere revision 7 chipset without the need of a repeater, but works with and without. MBP 2010 via LSI/Agere revision 8 chipset, with and without repeater. Imac 2010 21" via LSI/Agere chipset (if I remember right) and 27" via TI chipset (if I remember right).

Unfortunately I did not get hand on a Mac Mini yet and cannot give feedback on that. You should check which Firewire chipset is used within (likely LSI/Agere). I'd not expect some power-saving feature of OS X to be the culprit, because that should affect other OS X Macs as well. At least unless the Mini uses something exotic (in Mac context) like JMicron or Via.

What happens when you daisy-chain the FF400 to the working FF800? Does the FF400 still drop out? If so then like you I'd suspect the FF400 to be faulty and try to have it exchanged with another one at your dealer (you wrote "new" unit).

PS: The reason why I suggested trying the other port on the FF400 and checking the connection was not because of a known (to me) problem, but simply because it could be that your specific single device may suffer from a fault.

Re: FF400 Disconnects from Host

Thanks Timur,

I will daisy chain the FF400 off the working FF800.  Initially, I will still leave the audio coming out of the FF800 etc (so the FF400 won't really be doing anything).

Hopefully if the FF400 falls over again it won't take out the FF800 (so that the installation stops working sad ).

It is very strange that the FF400 was able to "lock up" to the point that it needed to be power cycled.  It was like its firmware had crashed.  It did not cause the Macmini to lockup when it happened.  Operation was able to be restored by power cycling the FF400.  Operation could not be restored if the FF400 was not power cycled, including just restarting the Macmini. The FF400 had to be power cycled for it to work again after it locked up.  Has anyone experienced this sort of "firmware crash" type of event before?

Regards,
James.

Re: FF400 Disconnects from Host

So how is it going?

Re: FF400 Disconnects from Host

Haven't had time yet to chain it off the FF800...  I'll let you know when I get it installed.

Have you ever heard of "firmware" style crashes of FF's before (where the only way to make them work again is to power cycle)?

Thanks,
James.