So I used to own a M-audio Audiophile Firewire sound card. Everything worked great with both my desktop PC and my Dell m65 laptop with ableton.
The only thing the m-audio card lacked was the additional inputs/outputs I needed to route to an external mixer such as the pioneer djm-800.
So I decided to purchase a Motu 828 Mk3, which was the best sound card available for the money, offering 10-in/10-out, which is exactly what I needed to route 4 channels from ableton into an external DJ mixer.
The new Motu worked great with my PC. However when I went to use it with my laptop I ended up getting a blue screen of death when installing the drivers. The BSOD reported a DRIVER_IRQL_NOT_LESS_OR_EQUAL error.
So I contacted Motu's technical support and they told me that they only support Firewire chipsets manufactured by Texas Instruments. Well thanks for NOT stating that on the box f**ckers! You clearly claimed that you were IEEE 1394 compliant. Obviously f**cking not! Maybe you should have stated you were IEEE 1394 compliant with TI chipsets only!
So I took the liberty in ordering a Syba Expresscard (model# which contained a Texas Instruments XIO2200 chipset and re-attempted to get the Motu working. After connecting the Syba expresscard it showed in the device manager that the firewire chipset card was indeed Texas Instruments.
So my question to Motu (which they never answered) was "How come my existing M-Audio firewire worked with BOTH my PC and Dell m65?, where the Motu does not". Apparently Motu blames it on the non-TI manufacturers of the firewire chipsets. But really can they claim this?
I have a degree in electrical engineering and IEEE 1394 is a specification. If M-Audio can get the IEEE 1394 spec right why can't Motu? To my understanding Motu manufactures much higher quality products than M-Audio.
So it appears to me that the only real solution is to purchase a laptop with an onboard TI firewire chipset. Reason I know this? Because my friend owns a Macbook which I tested the Motu on and everything worked fine.
Anyways,
Word of the wise -> Don't buy a Motu unless you have an onboard TI chipset
Posts
17 comments:
I understand your pain but things aren't always quite straight forward. As an example, I work in the ASIC industry and one of the components of our ASIC is to support the I2C protocol. We follow the protocol to a T, but, unfortunately, other manufacturers do not. As it turns out, those who did the right thing ended up not working properly with many of the vendors who did not implement things properly. However, those who didn't implement the protocol properly could often work with each other. Of course, the end consumer suffers and is largely oblivious to these issues---especially if vendors end up pointing the finger at each other.
It's possible that TI is one of the few manufacturers that actually got the implementation of 1394 down right. MOTU isn't the only one that's fallen victim to 1394 implementation incompatability---Presonus has it and even RME, though perhaps to a lesser extent, are suffering from it.
Anyway, the situation is certainly not ideal but at least there's a fairly cheap workaround (i.e. get another firewire card that DOES have the TI chip). Does it suck? Of course! But you'll move on and for the next machine that you purchase, you'll almost certainly be wiser. :)
One last point...yes, 1394 is a protocol that is presumably well-defined (by the IEEE). But, as a fellow engineer, you probably know that implementing something to a spec---especially if the spec has areas that are open to interpretation (and, believe me, this is *often* the case)---is much harder to do than it seems. Many EDA vendors provide verification IPs to ensure that your design conforms to a spec but there's nothing to say that their implementation is really correct either. Often times, it's just a matter of who gets entrenched first. They may not have implemented things correctly but if their implementation gets adopted in products early on, their implementation almost implicitly becomes the "standard."
I do understand that everyone can implement the spec differently which can lead to incompatibility issues. I see it all the time as im now a software engineer.
The funny thing is that I did purchase a firewire card with a TI chipset. The Syba Expresscard http://www.syba.com/Product/Info/Id/438
And I cannot replace my laptop as it is over a year old.
Motu should really not specify they are IEEE 1394 compliant. Or at least have proof that someone has validated and tested their IEEE 1394 implementation, not just their own engineers and testing team.
I am now convinced that it is an IRQ issue, most likely related to sharing devices on the same IRQ. Gonna try and see if thats the problem.
What chipset is in your laptop?
Compliance is one of those really nebulous things. Normally, the way it works is that there are these conferences where all implementers get together and test their products against some sort of reference design...and if it passes, you get the "compliance" stamp. There is a level of interoperability testing but one can never be fully sure.... I think TI's implementation seems pretty much compliant with everyone. It'd be interesting to see which implementations are the bad ones (hence my question about your laptop above).
I doubt your problem is with IRQ sharing. IRQ sharing has not been an issue for a very long time with PCI enumeration. It's somewhat unlikely that another device would occupy the same interrupt in the PCI address space. The error message is kind of bogus in my opinion...but then, I can't be 100% certain either. :o
yeah you are correct.
i have ruled out the IRQ issue as the syba express cards TI chipset shows it is using IRQ 19 and no other devices are shared
ah well. time to ditch the laptop and build a rackmount solution
So no success even with the Syba card?
nope no success.
so its not just a chipset compatibility issue.
i get a feeling when motu designed their audio interfaces they only tested it with TI chipsets conected via the PCI bus.
no PCI-e, or express bus or even pcmcia cards.
.
Hm...so you can't even get the driver installed on the laptop?
most of the time the the driver installs. other times i get a BSOD when installing the drivers.
when the drivers successfully install and i attempt to use the motu ASIO drivers via ableton things are incredibly slow to the point where ableton cant even load a clip.
.
Not to insult your intelligence (since I'm sure you've already tried it) but are you running with the latest driver and firmware?
Does the entire system come to a crawl or just Ableton? Does the Task Manager show what process is hogging the CPU? It would be interesting to know whether it's the MOTU driver or some other firewire driver that's hogging the CPU.
At least there is hope for you...if this problem is really just a software problem rather than a fundamental hardware incompatibility, then maybe things will be fixed at some point. Instability is the price we pay for living on the bleeding edge. :-o
I am running 1.04 of the motu firmware. Haven't upgraded to 1.06 yet because I read other people having issues, but I may just take the risk.
It is a combo card however the chipset clearly states in windows that it is a TI chipset.
Also nothing is utilizing the CPU because my mouse actions are still working, and process explorer shows nothing significant (ie: nothing above 5%). The application that attempts to interface with the ASIO drivers (eg, ableton) is incredibly slow and comes to a halt.
I am also on the belief now that is is likely a software driver issue.
I've done exactly what Motu tech support suggests and that is buy a TI chipset.
I also should add that I am running the latest drivers that came packaged with firmware 1.06. I am wondering if these drivers only work with firmware 1.06?
I am going to try upgrading to 1.06 today and attempt this again.
Crossing fingers.
I am not sure which version the drivers are but the timestamp on the motu website for the file fw32.zip is Date: 2008/06/30
Hm...I guess in your situation, it probably doesn't hurt to go to 1.06. The worst thing that could happen is...it won't work which is where you're at right now anyway.
Why can't you work for Motu?! They haven't responded to me in over a week!
I agree with charlton_wang. Especially when implementing drivers for various devices. Not only do some manufacturers not implement the specs right, but sometimes not at all. Just enough to work with "most" cases.
In any aspect, there are bugs, and the poor software writers will often depend on those bugs to achieve a goal. Or come up with creative ways to get around it.
I haven't looked at the 1394 spec. But more often than not, the specs can't even be implemented. I don't do a lot of hardware work. But in software, ones we use every day, like the RFC (or specs) for E-Mail or SQL, are so complicated, that nobody has implemented the full RFC. They are such complicated problems, or insecure, that it just isn't worth it.
MOTU also is a company that is mostly Mac. So they probably tested there. They probably use something that other chips don't implement. Even other companies typically specify Via or TI. I have a feeling it has something to do with the sustained transfer, which most other applications don't even need.
So really, it is probably all the other chip manufacturers that shouldn't advertise support 1394.
I realize this is an old post. I found this while googling for "windows vista motu drivers suck". I got a MOTU 828 mark I (yep, the first one).
Just wanted to point out that I never had any bluescreens on Windows XP, which luckily also happens to be my beloved music installation.
For everything else I've used Vista x64 and now Windows 7 x64. Blue screens happen when turning on the interface, but not every time, or when I start Vista/Seven while having the interface turned on. BSODs are IRQ_NOT_LESS_OR_EQUAL and sometimes BAD_POOL_HEADER.
It happened quite a few times but I can't see a pattern. It may happen or it may not. Since the official Windows 7 drivers came out turning on the 828 does not cause BSODs.
By the way I'm sure MOTU didn't test the interface on a dual-boot configuration, because I'm not the only one that has to power-cycle the 828 prior to starting up the other partition. Rebooting to XP results (sometimes!) in Windows not recognizing the interface. Rebooting to Vista/Seven results in a BSOD or sudden reboot during startup.
Just wanted to share my experience.
I swear my next MOTU will be a 2408 with a 424 card.
Anyway, hope you managed to solve your BSOD problems.
Cheers.
---
Specs:
AMD Athlon X2 6000+
Foxconn Destroyer (nVidia 780a SLI)
4GB RAM
Texas Instruments 1394 Chipset Onboard
Texas Instruments 1394 PCI Card
(same problems on both cards, onboard and PCI)
check here
http://createdigitalmusic.com/2006/12/12/motu-releases-audio-drivers-for-vistaxp-vista-driver-changes-in-store/comment-page-1/#comment-1037405
i think all solutions are found
your concern really helped
thanks
Post a Comment