I did not disconnect the FF though (and in the past that lead to freezes anyway) and the computer is still running after more than 10 hours now.

Uninstalling the FF and FW drivers did not help to make the FF reappear (before rebooting). Interestingly the FW driver can be deactivated/uninstalled/changed without Windows asking for a reboot. This means that it is *not* in use even when the FF is plugged in.

Ok, I'm currently using the FF400 successfully again (no buzz, no freeze, no bs for about 15 minutes). I do have several suspicions but like to go through them one by one.

First do set the onboard audio as system default, *not* the FF, so that you can make some settings without a freeze.

Here is the first thing I'd like you to try: *Read* the flash settings from the FF via Fireface Settings. Mine turned out to be quite messed up (no options set, no SPDIFout options set , buffer set to 48 samples). Turn on all options except for "Interleaved", turn on all SPDIFout except for "non-audio", set buffer to 256 samples. You might try different clock settings (auto, master). Then *write* the flash settings back into the FF.

After that do a reboot, *read* the flash settings to check they are correct and report back if it works.

I got my 400 yesterday and have similar problems. I have win 7 (x64). while playback i had lots of cracks. then i deinstallted the nvidia drivers which helped a bit, cracks only all 30 seconds. but i also have a lot of cracks when recording my axe fx ultra (guitar multi effect) via spdif. i played with the samples and i get cracks under 256 samples, which is very annoying because i hat a motu mk iii hybrid before and i had no cracks and a latency of about 3ms at 96 samples. i always read that the rme 400 is the lowest latency device on the market, and now this... :-(

it has obviously something to do with the drivers. so can somebody of rme give an offical word on it? are the aware of the problem and will they fix it?

otherwise will go back to motu soon...

cheers,
dan

I've decided that my quest to solve this mystery meant that I pretty much did exactly what you suggested right as soon as I hit 'post' on my last post.

Initial read:

Set to:

Rebooted and when I read it back the settings persisted.

Now, I've done this as well as set the default playback device to integrated, nothing else. When I loaded Foobar2000 and set the output to the monitoring channels and played back a track, it played fine for several minutes before the buzzing kicked in. It's still buzzing right now (I've turned down the volume) and I've not locked up.

What's more interesting is that Foobar2000's spectrum meter was vibrating/twitching back and forth like it was looping between a few samples. I try shifting the playback bar about and scrubbing and it does change the shape of the spectrum readout even if the buzzing isn't interrupted, but it's completely wedged at that point as if it cannot play back further at all.

Closing the application now, the buzzing stops. I can hear however a slight hiss over my headphones from the Fireface. Opening Foobar2000 again and playing the same track again, it's playing normally but I hear absolute silence. The spectrum meter is bouncing, the timeline is moving and for all intents and purposes it believes everything is fine. It's still set to output over Analog 7+8. Meanwhile, in the Fireface mixer it has zero evidence of anything being played back even though Foobar seems to think everything is fine. The Out 7 + 8 are dead, nothing is moving.

I've not touched anything further than this by the time of my posting, other than obviously opening up my browser and pasting what I wrote in Notepad (paranoid I'd lock up at any moment and lose my notes).

EDIT: With the mixer open, I can adjust the pot on the Fireface and watch it adjust the sliders in the mixer application. So it's obviously responding somewhere.

November, Emphasis should not be checked on the SPDIF output unless you play back material that has been equalized accordingly.

Unfortunately before I saw your reply, I attempted playback for a third time in Foobar2000. This time I got audio playback again, followed by the buzz, then the crash. I've heard audio drivers crash before, but usually I'm used to hearing slight pitch variation or something a bit more reminiscent of what's being fed into the device -- but yeah, you're right that it's being trapped on the sample.

At this point I did walk away for a while so I apologize for the delay in my reply. Anyway, here goes...

I've been able to take control of all the keys that I wanted to via the post I mentioned *including* the enumerated keys and entries created for ControlSet* which is usually handled by removing the Fireface via Device Manager. Though sometimes this does not happen like it should when you remove devices. You gotta struggle a bit because part of the child keys inside these keys actually have no permissions on them at all for you to even view them period. You gotta expand the trees and find these keys and try to push down ownership repeatedly until you hit the bottom of the tree, while assigning yourself full control before you can delete them. Try deleting a key you added yourself as owner/full control to and you'll notice it'll refresh and suddenly be expandable with more formerly hidden keys beneath it. Trust me, with some finesse you can nuke all of this without even having to boot into Safe Mode or otherwise try and force Windows to give up its hold on these entries. It just sucks at recursively assigning yourself ownership *and* then the right to even view the child keys to then go and assign ownership further down.

Anyway, I've removed every trace of the Fireface 400 drivers from both the system and the registry, in addition to nuking the cached drivers (what Windows tries to automatically reinstall with) using pnputil and then rolled a brand new fresh installation after a reboot.

On a lark, I went to Fireface Settings and it did not have any of the previous settings set. I hit Read Flash Memory and OK'd out of it.

(By the way, MC, I'm just syncing to what Timur had set, Emphasis was not set prior when I was having these issues. Though I am curious why you honed in on that screenshot and not the 17 other posts I've made in this thread? Anyway, I've unchecked it now regardless.)

EDIT: -- Ignore the following crap below because I am an idiot and it's 6:45 AM and I need to sleep soon. I'm keeping the following two paragraphs because it might amuse someone.

Here's where we get to an interesting problem. There isn't an Analog 1+2 output anywhere. Not in Control Panel/Sound, not in Foobar2000. I've got Analog 3 through 8, SPDIF, ADAT 1 through 8, etc. Everything that should be there other than AN 1+2 for Playback. It actually DOES exist for recording/input. On OS X, this is certainly not the case. If this functions different on Windows, I'm totally unaware (and mind you, I've not had a chance to find out given this issue). I honestly haven't noticed if it was like this before during my other ordeals and trials.

I've tried routing the other channels (3+4, etc) and playing with Foobar, but I hit the buzzing issue every time. Anyway, if you can confirm if my missing AN 1+2 is odd or not, that'd probably help a good bit (Again, I've not used this really EVER under Windows, so I don't know if this is normal or not, nor do I have the Fireface manual on hand at the moment).

EDIT: It was staring me in the face. It's Speakers, isn't it? I'll mess with that soon and see if it changes, but given everything else looks bugged, I'm not optimistic.

I've tried the Speaker output, same buzzing result.

Good to hear the registry is irrelevant, now if we could only figure out where stuff is going wrong.

I'll try giving the chipset update a go, though usually those filter drivers aren't too too terribly important beyond killing Windows installs dead after a motherboard swap.

If that doesn't work you will have to try my oddest suggestion: If you are using W7 Ultimate then download the german language pack and set Windows' "Display Language" to german. Do a reboot after that (not just a relogin).

This actually made me laugh, but I know where you're going with this. I often (though not currently) usually have my default fallback language set to Japanese to fix certain programs. I'll actually give it a go just because my German is terrible (I actually did study it!) and it'll be fun to test.

Unfortunately all of this will have to wait now. The sun is up and I really need to get some sleep. Thank you for your help so far, Timur. I'll try to get back to this thread when I've had more time to try things out.

Okay, some good tips to play with! (Yes, I'll be going to bed shortly, but I suddenly wanted to test something out first!)

Both Jeff Petersen and several others here have mentioned DPC Latency Checker a few times. I've decided to go pretty hog wild with this,
despite the fact I have some reservations about the idea of latency causing a bluescreen:

First I've disabled pretty much *everything* I could even remotely think of safely disabling. I even rolled back to the VGA driver after that screenshot, rebooted *and* killed BootCamp.exe after booting into glorious 800x600. I've made sure to include every device I've disabled in that screenshot just for fun.

This ran for a while. There's no spikes whatsoever past green. Now I closed the app, made sure the Fireface 400 was hooked up, Speaker was the default output device and I fired up an mp3....

So it looks like whatever this is isn't latency related or related to any of the devices I disabled, which is damn near everything. I'll be trying your suggestions next, but I figured I might as well get this big one out of the way first.

Today I had the Blue Screen problem popping up again. Which is a good thing, because it gives me another opportunity to try some things out.

The only thing that should have changed on my setup between my last successful use (last monday) and today was the installation of Office 2010 Beta (which I had installed and uninstalled on that machine before).

One thing since monday is that I repeatedly send the MBP/Windows 7 to sleep via closing the lid. Interestingly Windows 7 by default send the computer into Hybernation (suspend-to-disc) even when the settings were set to *Standby* (Safe Power). I reset the settings back and forth and now Windows properly uses Standby instead.

There may be some connection between Sleep mode and the problem, especially when knowing that the NVidia chipset came with an updated "Sleep Mode" driver. Seeing how there are issues in connection with sleep on OS X as well it's at least something to take a look at. DeadHorse

Hi guys
Did anybody made Windows 7 and FF400 work right?
I have tried everything you guys wrote on this forum but still have the same high latency on DPC.
Please anybody knows something?
We could also use Teamviewer so you check my system your own.
Thanks,
George

HI,
this is common with older laptops
this has to do with ACHI/IDE modes.
if the laptop has the option for ACHI it might work but it will break the XP install.
conversely newer laptops cant install XP generally.

Scott
ADK

Installed my FF400 to a Win7 x64 system this past weekend after getting tired of all the blue screens I was getting on shutdown using a tascam fw1884.  Had much better success after borrowing the ff400 from my mobile rig.  My system is an Intel 975bx2 mobo with a q6700 processor and 8 gigs ram oc'd to 3.2 ghz using mostly Sonar x64.  It's actually been running many years under xp64 using the tascam interface but I just upgraded it to Win7 x64 (fresh install to a new c: ssd boot  drive) last week.

Installed the ff400 29992 drivers and flashed to 1.70.  Had previously set the Windows sound scheme to "No Sounds" and shut off the start-up sound.  Also previously reflashed the Intel mobo with the latest bios, installed all of Intel's newest infs, and did all my Windows updates.

I have always used WDM/KS in Sonar but the ff400 spidif interface kept getting reported by the rme driver as a 16-bit device. Fired up Reaper x64 and it got the same info from the ff400 driver.  I use spidif for my hardware reverb unit and 16-bit truncation before mixdown wouldn't be good for me.  Switched to asio and that seems to be good. Only issue was shelling out of the daw to edit audio clips with SoundForge... asio doesn't share well.  Got around that by giving SoundForge Analog 3/4 (daw uses Analog 1/2) and routing it in the Matrix to the same outs as 1/2.  And all is well.

I did experience a bit of a problem doing a spidif-send to my reverb unit and recording its spidif-out.  Got allot of crackling and popping.  Turned out I had to shut off write-caching in Sonar which I never did with my tascam unit.  No big deal once I guessed the solution.  I was able to get 2 solid days of work done this past (us holiday) weekend on an 88.2k/24-bit project without any major incidents. So far, so good.

I have the buffer-size set to 128, nothing checked on spidif-out, everything checked on options, and dds-disabled.

Howard

My pleasure! :-D

If you are on a laptop turn off the "ACPI Comliant Battery" thing via device-manager.

If your problems are happening with DAW software then either try deactivating Aero or manually setting the priority of the software to "Realtime" (start Task-Manager "as Administrator" to do that or turn off UAC).

The priority problem is not a problem of Ableton, but of nearly all current DAW software. They all use the same *non*-realtime priority that Aero uses. The only exceptions I know about are Sonar (MMCSS) and Samplitude (highest *real*time priority as an option). With Live it's easy to fixvia Task-Manager

But if you also got the problems with Windows Media Player then the problem lies somewhere else.

Something interesting happened: for the second time now my Macbook Pro frooze without having any external audio interface attached or even playing back any audio/video. But the behvaior was *very* similiar to how video-playback crashed/frooze with the FF400 before.

In these situations the display was put to standby (black screen) while Windows kept working on some simple downloads in the background. I moved the mouse to wake up the screen. Windows stopped responding to any input (including CTRL-ALT-DEL), but the mouse was still moving. Then, after a couple of seconds, the whole system froze (no BSOD).

Even alot worse than what we experienced with the FF400: Windows could *not* be started anymore except for Safety-Mode, not even from the Installation CD! Removing the graphic-driver via Safety-Mode did not help. The startup process would either halt after the Windows 7 boot-loader animation (where the 4 colored dots form the Windows logo) or the screen would remain black.

Now comes the interesting part:

1) Starting into OS X and then rebooting back to Windows (Bootcamp) made Windows work again. After that I switched the NVidia driver back to the one shipping with Bootcamp, but the freeze happened a second time sometime later.

2) Since the last crash (BSOD) with video-playback via the FF400 I did not have time to check it again, but after the second non-playback freeze I not only started into OS X, but also reset the PRAM/NVRAM. Now I checked video playback via Media Player Classic over the FF400 and no crash happened.

Since November is using a bootcamped Mac, too, I am very much inclined to suspect a Mac specific problem!

@November: Please do a PRAM/NVRAM reset (http://support.apple.com/kb/HT1379) and try again. If you didn't start into OS X for some time then also try that.

I had more bootcamped Windows 7 freezes on my Macbook Pro 17" 2009 and it turned out that reseting the PRAM/NVRAM does not help to make Windows boot again (not even the recovery enviroment can be loaded), but only starting into OS X helps.

Unfortunately I did not have any other FF400 related crashes yet, so I could not test if there is any relation between the two.

Nubie here.

+1 for bootable USB with O\S
In my experience (using win7 since the first rc's), hidden partition that is installed with Win7 is a bad idea. If the files here are damaged, it will never boot.
I never install with hidden partition (google) How can Acronis (or any other) make a full image to restore if it is not included? Maybe someone knows that answer?
I have had the same problem as Matthias, many times, long time ago. It was due to bad ram corrupting the files on the hidden partition (Memtest86+) and believed my HDD's were failing.
No problems at all 24\7
Dedicated PC's for audio. (Fireface 800 Win7x64)

I hope this can help someone with similar problems. Best regards.

Hi all,

Thought I'd share some info with all of you re: my FF 800 bluescreen problems (sorry if this is a double post from my other thread, but it seemed relevant - maybe the other post should be deleted).

I run Win 7 64bit (not via bootcamp - I have a PC) and the RME drive has been causing bluescreens on a regular basis. I'm running the latest firmware and 2.9992 driver (problem was still occurring with the older driver too). Reinstalling Win7 hasn't helped.

The error that is displayed is IRQL_NOT_LESS_OR_EQUAL or IRQL_GT_ZERO_AT_SYSTEM_SERVICE (kicking myself that I didn't write down exactly which one....). But I do have my crash dump files, and when I analyse them, the error is occurring in audiodg.exe every time.

The reason I'm fairly certain it's the RME drivers is because when I go into Sounds in control panel and disable all the RME ins and outs in Windows, the crashes stop occurring. I'm still able to use the Fireface in applications that use the ASIO driver, but not in those that use that go through the normal windows sound drivers.

For the others who are experiencing these bluescreen errors, have you tried downloading the window debugging tool and analysing the crashdump .dmp files in your c:\windows\minidump directory? Posting the results may help RME get to the pottom of this? If everyone's error has the same cause, it means it's a non-system-specifc driver problem that needs to be fixed.

Here is the analysis of my latest crash dump (they all report the same error). It's a raised IRQ fault - possibly related to switching from ASIO usage to normal windows driver usage (eg. using Cubase, then trying to run something like Media Player). But I'm just guessing.

--------------------------------------------------------------------------------------
Microsoft (R) Windows Debugger Version 6.11.0001.404 AMD64
Copyright (c) Microsoft Corporation. All rights reserved.

Loading Dump File [C:\Windows\Minidump\022510-51870-01.dmp]
Mini Kernel Dump File: Only registers and stack trace are available

Symbol search path is: SRV*c:\Symbols*http://msdl.microsoft.com/download/symbols
Executable search path is:
Windows 7 Kernel Version 7600 MP (8 procs) Free x64
Product: WinNt, suite: TerminalServer SingleUserTS
Built by: 7600.16385.amd64fre.win7_rtm.090713-1255
Machine Name:
Kernel base = 0xfffff800`02a68000 PsLoadedModuleList = 0xfffff800`02ca5e50
Debug session time: Thu Feb 25 03:35:55.284 2010 (GMT+11)
System Uptime: 0 days 16:24:08.110
Loading Kernel Symbols
...............................................................
................................................................
...........................................
Loading User Symbols
Loading unloaded module list
............
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 4A, {7759ff2a, 2, 0, fffff88013483ca0}

Probably caused by : ntkrnlmp.exe ( nt!KiSystemServiceExit+245 )

Followup: MachineOwner
---------

3: kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

IRQL_GT_ZERO_AT_SYSTEM_SERVICE (4a)
Returning to usermode from a system call at an IRQL > PASSIVE_LEVEL.
Arguments:
Arg1: 000000007759ff2a, Address of system function (system call routine)
Arg2: 0000000000000002, Current IRQL
Arg3: 0000000000000000, 0
Arg4: fffff88013483ca0, 0

Debugging Details:
------------------

PROCESS_NAME:  audiodg.exe

BUGCHECK_STR:  RAISED_IRQL_FAULT

FAULTING_IP:
+0
00000000`7759ff2a ??              ???

CUSTOMER_CRASH_COUNT:  1

DEFAULT_BUCKET_ID:  VISTA_DRIVER_FAULT

CURRENT_IRQL:  2

LAST_CONTROL_TRANSFER:  from fffff80002ad9469 to fffff80002ad9f00

STACK_TEXT: 
fffff880`13483a68 fffff800`02ad9469 : 00000000`0000004a 00000000`7759ff2a 00000000`00000002 00000000`00000000 : nt!KeBugCheckEx
fffff880`13483a70 fffff800`02ad93a0 : fffffa80`0a56b3e0 00000000`01e1fc58 fffff880`13483bc8 00000000`00000001 : nt!KiBugCheckDispatch+0x69
fffff880`13483bb0 00000000`7759ff2a : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!KiSystemServiceExit+0x245
00000000`01e1fc28 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : 0x7759ff2a

STACK_COMMAND:  kb

FOLLOWUP_IP:
nt!KiSystemServiceExit+245
fffff800`02ad93a0 4883ec50        sub     rsp,50h

SYMBOL_STACK_INDEX:  2

SYMBOL_NAME:  nt!KiSystemServiceExit+245

FOLLOWUP_NAME:  MachineOwner

MODULE_NAME: nt

IMAGE_NAME:  ntkrnlmp.exe

DEBUG_FLR_IMAGE_TIMESTAMP:  4a5bc600

FAILURE_BUCKET_ID:  X64_RAISED_IRQL_FAULT_audiodg.exe_nt!KiSystemServiceExit+245

BUCKET_ID:  X64_RAISED_IRQL_FAULT_audiodg.exe_nt!KiSystemServiceExit+245

Followup: MachineOwner
---------

Here is one interesting Windows system log entry concerning Windows 7 Sleep/Standy on a Macbook Pro 17" 2009:

That was at 3:22 in the night when I send the system to sleep/standby. At around 10-11 in the morning I restarted the computer and used it for about 2 hours without problems (no FF connected).

Then I connected the FF400 and my external projector to the computer and it crashed (connecting an external display before did not crash it, so I suspect the FF400 connection to be the culprit). After a restart I could use the FF400 for normal playback via Windows Media Player and Media Player Classic without problems.

Some additional information about the latest BSOD crash:

After restarting Windows 7 the FF400 was *not* detected eventhough the HOST light on the FF was not lit (at least I didn't notice). I turned the FF off and on again (bus-powered via backside switch), heard the Windows Hardware detection sound and everything worked again.

@panu -  This is EXACTLY the same issue I'm having (happens in the same kind of situation too). Could you try a couple of things and let us know the results? Firstly, try going into Control panel - Sound and disabling all FF inputs and outputs from "Playback" and "Recording" tabs (they will disappear from the list). Basically, the idea is to make sure Windows isn't using the FF for anything. To work around getting sound from your windows applications, you could run sound out of the digital or analogue out of your motherboard's onboard soundcard into the inputs of the FF (this is what I'm currently doing). Also, Sonar will still work since disabling the FF ins and outs in Windows Sound will not disable the ASIO driver.

If this works for you as a fix (it did for me) - can you report back and let us know?

Also, it would be great if you could download the Windows 7 Debugging Tool (http://www.microsoft.com/whdc/Devtools/ … fault.mspx) and analyse your crashdump file (it will be the latest .dmp file inside the c:\windows\minidump directory). This will allow you to check whether the problem was with audiodg.exe, as it was with me.

@MC - it seems I'm not the only one having this bug - could you let us know if it is something that is being looked into for the next driver update?

Best bet would be to install the drivers for your onboard soundcard - the Asus P6T has an onboard Realtek ALC1200 chip (if it is disabled in bios, re-enable it). Once this is enabled, set Windows -> Sounds to the onboard sound input and output. Then, you'll need to run an optical cable from the output of the Realtek (i.e. on your motherboard's back panel) to the ADAT2 input on the FF, and switch "SPDIF IN" in the FF Settings window to "ADAT2". Your other alternative is to get a 3.5" stereo to 2 x 6.5" mono cable and run the analog out of our Realtek into 2 of the analog inputs of the FF.

Volume is then controlled via the respective inputs in Total Mix. Eg. need to route the SPDIF input to analog 1 and 2 (or whatever outputs your speakers are connected to).

It's a pain in the behind workaround, but most importantly - it works. Since I've done this, my computer has gone from crashing every 1 to 2 days to staying on 24/7 and not having crashed once in 2 weeks. Look forward to hearing your results.

P.S. Are we going to get any kind of acknowledgement from RME that there's an issue here? No replies from anyone yet - to my forum post or to my email to support??

P.P.S. I wonder if this is just an issue with Win 7 64bit and not with 32bit?