Ongoing Ultralite mk 5 issues digital noise on M1max

[Calagan]

Ever since I sent the MOTU audio driver to the trash, my Ultralite Mk 5 has been working beautifully in class compliant mode (Mac Studio Ultra/Monterey).

I'm leaving it alone.

Dan V.

Yep, I know that, but running the unit class compliant means much higher latency and not any DSP and routing possibility...
I mean, I bought a unit with some features/specs, not a class compliant unit that has no routing app and stupidly high latency...

(I need short latency because I use the unit on stage, with vocals and drum triggers : high latency is not an option, or I can use any entry level focusrite at half the price for that)

[mikehalloran]

Ever since I sent the MOTU audio driver to the trash, my Ultralite Mk 5 has been working beautifully in class compliant mode (Mac Studio Ultra/Monterey).

I'm leaving it alone.

Dan V.

Yep, I know that, but running the unit class compliant means much higher latency and not any DSP and routing possibility...
I mean, I bought a unit with some features/specs, not a class compliant unit that has no routing app and stupidly high latency...

(I need short latency because I use the unit on stage, with vocals and drum triggers : high latency is not an option, or I can use any entry level focusrite at half the price for that)

Do you have any numbers for that "stupidly high latency"?

On the M2/4/6, the difference between Core Audio and the drivers works out to moving your ears closer to or further from your monitor speakers 14" or so.

[rplktr]

Sure, 3 milliseconds is 1 meter away from the speakers. But there are plenty of producers working in headphones so this argument is deflecting the issue. The "ultra-low roundtrip latency" is listed on the MOTU M4 page in its second paragraph. It shouldn't be surprising that people depend on this feature.

[Calagan]

Ever since I sent the MOTU audio driver to the trash, my Ultralite Mk 5 has been working beautifully in class compliant mode (Mac Studio Ultra/Monterey).

I'm leaving it alone.

Dan V.

Yep, I know that, but running the unit class compliant means much higher latency and not any DSP and routing possibility...
I mean, I bought a unit with some features/specs, not a class compliant unit that has no routing app and stupidly high latency...

(I need short latency because I use the unit on stage, with vocals and drum triggers : high latency is not an option, or I can use any entry level focusrite at half the price for that)

Do you have any numbers for that "stupidly high latency"?

On the M2/4/6, the difference between Core Audio and the drivers works out to moving your ears closer to or further from your monitor speakers 14" or so.

1- Latency is adding. Device latency, plugin latency, distance from the monitors, etc. After 10ms, latency start to be an issue for a keyboard player and can be felt as a lack of immediacy. You don't feel nothing between 1ms to 7-8ms : there's a bad feeling threshold with latency.
By the way, any latency (even 1ms) is a problem for a singer because of comb filtering that happens in the head. So direct monitoring is an absolute necessity when recording : tell me how to do that without Cue Mix (in class compliant mode) !

2- I don't have the precise numbers, but I remember when I tested it that class compliant mode was more or less the double latency compared to the Motu drivers. I did wrote the numbers for the M2 and I guess it's more or less the same with the Ultralite.
At 44kHz/128 samples, 6.9ms RTL (with Motu drivers) becomes 12ms or so (in class compliant mode) !
Add to this the latency of the plugins used (from few ms to 4-5ms depending on the Live set and plugins used) and you've got a latency of nearly 20ms that is unacceptable, especially when you use the device on stage !
Give someone an instrument with 20ms latency on stage !
20ms is 7 meters : did you ever played on stage without wedge monitors, only listening to the FOH ? Good luck with that.

3- I bought a device with some specific features. Among them, low latency and CueMix mixer (with DSP FX).
How is it acceptable to be obliged to use a device at 50% because some features don't work ?
How can you legitimate that ?

4- I just need low latency and CueMix. Please find a solution to my problem or don't comment.
I don't need judgemental comments on my use of audio interfaces.

By the way, the drivers 93305 fixed my issue. So I guess the issues were coming from MOTU software.
I don't think I will ever update again any MOTU drivers (especially when I read new issues with latest drivers on this forum).

[wo0d643]

It says dont post windows stuff here but I just wanted to let you know that i have this issue sometimes on Windows. I havent found anyone else talking about this. I cannot reproduce the issue. It just happens when it happens. I either restart my daw or restart the motu and it goes away for minutes to days. ok thanks.

[mikehalloran]

…
By the way, the drivers 93305 fixed my issue. So I guess the issues were coming from MOTU software.
I don't think I will ever update again any MOTU drivers (especially when I read new issues with latest drivers on this forum).

93305 is the latest driver for the UltraLite mk5. Unlike most of MOTU''s drivers, these are published as Sonoma compatible and don't work on other interfaces. All these other "new issues with latest drivers" affect other devices, especially MIDI, not yours.

The rest of your post twisting things I never wrote into conclusions I never meant isn't even good fiction. Many things can cause the latency numbers you describe, chief among them is buffer size but tracking with certain plug-ins will, also, even in bypass mode. MOTU specs RTL with a buffer of 32 @ 96kHz 24 bit — any comparisons need to be measured there or they are meaningless. My math uses the average speed of sound at sea level and 40% atmospheric humidity.

Since there is no such thing as zero latency, any I/O mix that is 50/50 will result in phase distortion and comb filtering. A 60/40 or 40/60 monitor mix along with a low buffer will make this negligible but nothing eliminates this. Physics is not just a good idea: it's the Law. CueMix is what MOTU recommends.

[Calagan]

…
By the way, the drivers 93305 fixed my issue. So I guess the issues were coming from MOTU software.
I don't think I will ever update again any MOTU drivers (especially when I read new issues with latest drivers on this forum).

93305 is the latest driver for the UltraLite mk5. Unlike most of MOTU''s drivers, these are published as Sonoma compatible and don't work on other interfaces. All these other "new issues with latest drivers" affect other devices, especially MIDI, not yours.

The rest of your post twisting things I never wrote into conclusions I never meant isn't even good fiction. Many things can cause the latency numbers you describe, chief among them is buffer size but tracking with certain plug-ins will, also, even in bypass mode. MOTU specs RTL with a buffer of 32 @ 96kHz 24 bit — any comparisons need to be measured there or they are meaningless. My math uses the average speed of sound at sea level and 40% atmospheric humidity.

Since there is no such thing as zero latency, any I/O mix that is 50/50 will result in phase distortion and comb filtering. A 60/40 or 40/60 monitor mix along with a low buffer will make this negligible but nothing eliminates this. Physics is not just a good idea: it's the Law. CueMix is what MOTU recommends.

I'm sorry Mike but you don't read my posts (and actually a lot of other posts), and this causes a lot of misunderstandings on this forum. It truly doesn't help anybody.

1- The point of the discussion is using the device in class compliant mode and conclusion is that it's a huge limitation. I don't know about what you are speaking.
2- 93305 are not the last drivers. They are dated from June 2023
3- Latency numbers when used in class compliant mode are the double of latency numbers when the MOTU drivers are used. There are no "many things" that cause high latency : using the device in class compliant mode is the only one that is significant to my point.
I'm not stupid and I know the difference between 64 samples and 1024 samples, I know that tracking with plugins can cause latency... I'm speaking about the HUGE difference between using the Ultralite as it was intended (with Cue mix and Motu drivers) and with these drivers in the same conditions (same buffer size, same sample rate, etc. etc.).
3- There is such thing as zero latency : direct monitoring is zero latency ! It doesn't produce comb filtering during monitoring when recording a singer. And it's not possible to use direct monitoring on the Ultralite mk5 without the drivers and CueMix. CQFD !

The whole point is : it's not acceptable to use this device in class compliant mode.
We payed for Cue Mix and the latency numbers advertised.

[mikehalloran]

2- 93305 are not the last drivers. They are dated from June 2023

93305 was the latest driver when I wrote that post a few days ago. As of yesterday, MOTU posted new ones. If MOTU gave you beta in the meantime, good for you.

Screenshot 2024-02-07 at 11.26.55 AM.png (144.7 KiB) Viewed 1478 times

.

[HCMarkus]

Of course, no digital mixer is "Zero Latency." Near Zero is the best we can do with AD and DA involved.

For true Zero Latency, use an analog mixer for cue mixes.

PS: we could argue there is no such thing as Zero Latency, because even electrons take time to move thru a wire, but THAT is splitting one very fine hair. And I'm just fine with a digital mixer for my cue mixes.

[wo0d643]

It says dont post windows stuff here but I just wanted to let you know that i have this issue sometimes on Windows. I havent found anyone else talking about this. I cannot reproduce the issue. It just happens when it happens. I either restart my daw or restart the motu and it goes away for minutes to days. ok thanks.

Thought id update this just in case someone found it. This hasnt happened again. Its been about two weeks. I guess it was an anomoly.

Thanks