MOTUnation.com

I am at the end of my gosh darn rope with this piece of •••• Ultralite. Every time I start Live it starts this speaker shredding popping and screetching and locks the machine up hard half the time. I'm very very close to taking it outside with a sledgehammer and then making sure none of my friends ever buy a MOTU product again.

For the record, I've already tried:

1. rolling back to sp1 firewire drivers
2. factory reset
3. running with dc adapter power
4. installing a "supported" belkin ti chipset card

Anything else I should try before I smash my first and last MOTU product into a million tiny little pieces?

same happens here
i own a 828 Mk1
and never as any problems like with this ultralite

One weird thing about this - it only happens in Ableton Live. Other ASIO hosts work ok. Reason, Cubase, EnergyXT all fine.

Set the latency higher!
Im using live with 256 samples 6ms latency, overall lagtency 12.8ms...

tm21 wrote:Set the latency higher!
Im using live with 256 samples 6ms latency, overall lagtency 12.8ms...

I get this at any buffer size.

I feel for you. I have exactly the same disguisting whine with my new travelller. I'm using both windows and OS X and the whine seems to happen in both. My previous audio interface had no such problems. At first I thought the traveller was being very fussy, and it seemed to help to have it switched on when I turned on my computer, but it still occasionally happens, in both operating systems, at any buffer size. I've downloaded and installed both latest drivers for mac and win, and installated them with the traveller disconnected, just to be sure. What the hell is the problem with this thing?
Also, just been using the traveller in live and had endless clicking and scratching, apparantly coming froms ins 1 and 2, despite nothing being connected to either and their input levels set low.
GRRRRRRRRRRRRRRRRRRRRR...MOTU!

On a hunch I completely disabled my network interface on my XP machine and it *seems* to have solved the problem. I'm not sure why the network would conflict with the firewire bus but so far it seems to be working.

kuniklo,

I've seen the same thing with the USB circuitry on something here. It's like there's some sort of polling going on behind the scenes that interrupts the audio device. If either was a PCI card, I'd say try to change slots, but there still may be a way to change the IRQs of the built-in stuff and/or lock one to an IRQ of it's own in the BIOS or Windows somehow. The "virtual IRQ" thing that ACPI does isn't supposed to be a problem with modern hardware, but as usual, YMMV.

Worst case, some recommend doing the "standard PC" install of 2K/XP and running with ACPI off. (you'll need to hit a function key when you launch the installer)

Hope it stays clean,

George

If I get this happening again I'll try a non-ACPI install. The firewire port is a pci card at least.

I'd like to avoid a total XP reinstall if possible though just to avoid the Ableton C/R tango one more time.

If it's a card, you might try the slot swapping. I've heard that there's still some connection between the shared slots and devices on motherboards, even with ACPI, but I'm not sure if that's right. The "sharing map" should be in the MB manual, but I've had to go looking for the info online for boards before. There's probably a slot or two which isn't linked to anything.

Be warned on standard PC. It's sort of a PITA after dealing with XP/ACPI, but some people swear by it. You'll eat all your IRQs up really quick, but I think some of your "regular" stuff can still be set to the same numbers with no problems. I've run it before on my studio machine and it seemed to do OK.

If you can drag another drive or partition into the picture, you may want to Ghost what you've got now, just so you can go back if needed.

Sorry if you know any of this already, but hope it helps,

George

PS- The standard PC install is F5 at boot. At the point where it says "hit F6 if you have a third party SCSI controller" or something, you'll hit F5. You may also need to change some ACPI/APIC plug&play stuff in the BIOS first too.

Thanks for the suggestions. I'd already tried moving the card around with no luck but I've made sure it's not sharing any IRQs at least.

I'm crossing my fingers that leaving the network off is a real fix. I don't need this machine to be online anyway.

I found that the only way to get my Ultralite to stop popping (in Reason, DP,Live) is to set it to 9600...WASTE of CPU!!! Tech supp just said "That's weird"...

I had this same problem. As a firmware programmer I have some extra insight into this problem.

I own a Belkin 802.11b PCI card. At first my ultralite was working great, but then I turned on the wireless network card: any time the wireless card sent an appreciable amount of data the sound would go nuts.

All the meters on the Ultralight would all begin spiking randomly even though I was sending audio only out through one channel. In addition to that, there was a lowercase "m" between the analog meters and the L/R meter where there is normally no character at all.

After disabling the network card OR not accessing the net with any programs AND halting any sound output to the unit the nasty noises would eventually stop. Once everything was quiet, I could restart my audio and the unit would behave normally again right up until I tried to access the internet.

I tried adjusting the PCI latency of my firewire interface very high and dropped the settings of everything else, but this had no effect.

Finally I just removed the card altogether and replaced it with a USB 2.0 802.11g card and now everything works flawlessly.

Here's what I think is happening:

I believe there is a bug in the driver that causes a mis-sync in the raw data structure being sent from the computer to the UltraLite. When the mis-sync happens, all the data becomes misaligned. The design of the ultralight software seems to be to recognize that there is a problem with the data stream (either a checksum error, length error, boundary error, or token error), and display an 'm' as a result. I think this 'm' display is related to MIDI output and is to to the buffer overrun.

The good news is that if I'm right the unit knows the data stream is bad and it's probably not the fault of the unit. That means the problem is most likely in the windows driver code and can be fixed with enough diligence.

I'm pretty sure the problem has to do with frequency switching at this point.

The bad news is that this kind of bug can be very hard to track down. Any time something that seems to have nothing to do with what you are working on effects your outcome, you know something very obscure is going on. This kind of bug stinks of an uninitialized data/race condition/order of execution problem. That means sometimes it will work -- and sometimes it won't. You could step though the code a hundred times and never see the bug because the circumstances to cause the failure aren't present. You have to catch the fault in action and have enough debugging output in place that you can figure out why the fault occurred. The upshot of all this is that it might be VERY hard for them to find and squash this bug.

I'm a little pissed that the unit isn't more careful about pops and clicks. Personally I'd love to see a firmware mod with a setting I can turn on and off that either a) detects and prevents random clicks and pops from passing through or b) has some sort of hearing-safety threshold compressor. I fear for my hearing every time a full-volume pop goes off. The frequency response of the unit's headphone amps and my headphones are good enough I fear my hearing will be damaged by the sudden impulse.

To try to find the offending component try first removing any network cards in the system or disabling them in BIOS. If that doesn't work, try removing other cards and devices. Pull everything not needed to boot -- basically leave the video card and the hard drive in place. Pull USB devices, firewire devices, flash cards, EVERYTHING else -- except perhaps for the mouse -- and if things start working then add the hardware back in one unit at a time until the problem reappears.

Of course, before you do this, make sure you're using a TI chipset and have tried all the "standard" fixes.

If any MOTU techs are listening, go to eBay and buy a belkin 802.11b PCI card: the old old old version. Install it and set up a wireless router. Visit a few websites while playing some music through the Ultralite and the problem should present itself.

-Chiem

I had the same issues of screeching, popping, playback pauses and other not so good sounds coming from my UltraLite. Found a solution that was very by accident and not something I would have ever guessed. Little story...

Have a MB with a TI chipset 1394 on it so I tried that first. No go. Then tried 3 other cards all with TI chipsets based on good reviews I found here. No go either. Tried two non TI chipsets with still the same problems. At this point it's 4am and I••™m too tired to do anything productive anymore. Fast-forward to today. Come home from work expecting to spend another evening of battle and defeat. Boot things up and to my amazement everything works perfectly! Somewhere between confused and delighted, I just take the victory and move on. In the process I pop in a CD into my external 1394 CD burner and find out that my cat had crawled behind my computer again and pulled the cable out the back of the burner. Shutdown, replug, reboot. UltaLite freaking out again! Shutdown, unplug burner, reboot, success! The cause of all my troubles ended up being my external CD burner, on a different 1394 card completely, yet was causing the UltaLite to loose its mind. No shared resources or IRQs between the UltaLite and the burner. Figures, my cat fixed my MOTU••¦

Hi
I'm new to this forum, 'cause I'm new to MOTU products.
Bought an UltraLite 2 weeks ago and tested with most popular DAW software. I have to say that all is working correctly! I bought also a PCMCIA FireWire card with TI chipset but... i didn't needed. My built in firewire (4 pins ) with RICOH chipset works great. No pops, no clicks. Only few problems, setting latency at 256 using guitar rig for processed monitoring. But at 512 every thing is ok. Wavelab, Nuendo and Live are working great.
Only 1 cons: an annoyng noise in my monitors (not dependent from sound signal) that disappear as i unplug my Notebook AC adaptor . I've read that a solution for that is to cut the ground wire (green/yellow wire) from che AC connector / socket.

Perhaps someone will find useful these informations on such a lucky combination (once in a while.. it's right! ).
ASUS z92F
Motu UltraLite
Windows XP SP2
Wireless (always on)