Topic: Latency in Logic Pro X on macOS with the Digiface AES

Hi again, can anyone also tell me the latency in Logic Pro X on macOS at 44.1 kHz with 64 buffers for the Digiface AES?
This would be very helpful.

Example:
https://i.ibb.co/6tVMZGm/Bildschirm-foto-2023-05-18-um-14-32-04.png

Re: Latency in Logic Pro X on macOS with the Digiface AES

With the Digiface USB it says 6.3ms
Wondering what that means "Roundtrip" ?
There was nothing connected to the interface and no signal was transmitted.
If the amount of channels has an influence on that "measure", I don't know.
The Digiface AES has to expect about the same Roundtrip latency.

M1-Sonoma, Madiface Pro, Digiface USB, Babyface silver and blue

Re: Latency in Logic Pro X on macOS with the Digiface AES

Roundtrip means the latency a signal gets from analog input over AD conversion - over USB to computer - DAW - back to interface - DA conversion - to analog out

UCX - FF 400 - Babyface pro - Digiface USB - ADI-2 (original)
Mac mini M1 - Macbook pro - iPad Air2

Re: Latency in Logic Pro X on macOS with the Digiface AES

Thank you very much. As oli77sch explained, round-trip latency (in Logic Pro X) refers to the total time it takes for an audio signal to pass through the software's input, processing, and output stages. It measures the delay caused by the processing required by the computer as it monitors the input signal in real time during recording, which is especially important when modifying the input signal in real time with computer-based effects and/or recording in sync with the computer's playback.

5 (edited by oli77sch 2023-05-25 21:29:17)

Re: Latency in Logic Pro X on macOS with the Digiface AES

With Babyface pro I get 4.8ms at 44.1kHz / 64 samples (Mac mini M1, Ventura 13.4, Logic pro 10.7.7, RME USB driver 4.08, Firmware v203)
https://i.ibb.co/K99mPLx/Babyfacepro-Latency.png

UCX - FF 400 - Babyface pro - Digiface USB - ADI-2 (original)
Mac mini M1 - Macbook pro - iPad Air2

6

Re: Latency in Logic Pro X on macOS with the Digiface AES

This dialog shows the numbers coded into the driver. There is nothing 'measured' and these numbers are often wrong (for non RME devices). The real numbers need real loopback tests and either Logic's special Ping function, or (much easier) RTL Utility from Oblique.

Regards
Matthias Carstens
RME

Re: Latency in Logic Pro X on macOS with the Digiface AES

Taken from a recent Gearspace post:

https://i.ibb.co/P62z1Xt/DFDante.png

Does the Digiface AES use the same driver as shown and has lower latencies than e.g. the Babyface Pro?

8

Re: Latency in Logic Pro X on macOS with the Digiface AES

The Digiface Dante does not have AD/DA, so this is pure digital RTL - and naturally some samples lower/quicker than with additional AD/DA conversion.

Regards
Matthias Carstens
RME

Re: Latency in Logic Pro X on macOS with the Digiface AES

Are the latencies (driver) of the Digiface AES comparable to the figures of the Babyface Pro then?

10 (edited by ramses 2023-05-30 09:44:07)

Re: Latency in Logic Pro X on macOS with the Digiface AES

Any RME driver / solution has figures comparable to PCIe. See my blog for different solution that I own(ed):

https://www.tonstudio-forum.de/attachme … es-v2-jpg/

When playing with VSTi simply stay at around 128 samples ASIO buffersize at single speed (44.1/48 kHz) to stay below (or around) 10ms RTL. In this case (playing with VSTi), you do not have the full RTL, so even a buffersize of 256 samples might work.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub13

Re: Latency in Logic Pro X on macOS with the Digiface AES

Thanks, but I just need the "real world numbers" for the Digiface AES in Logic Pro X. Just the latency numbers as shown in Logic Pro X (buffer size 64 samples at 44.1 kHz). I'm not into Windows based calculations or estimates, although your information is very thorough and helpful. Thank you very much.

12 (edited by ramses 2023-05-30 10:35:52)

Re: Latency in Logic Pro X on macOS with the Digiface AES

The difference between Windows and Logic is not that big.
What I wanted to demonstrate to you with some concrete figures from Windows is, that you can expect excellent values, very close to PCIe, regardless of which RME product you choose.

Matthias told you already that this pure digital card without analog section will have better values anyway because RTL includes normally the time for A/D and D/A. So careful not to compare chicken with eggs.

Besides this, the RTL is only critical for some use cases, in most cases, the latency compensation in a DAW works fine and fully compensates.
Example1: if I am mixing in Cubase, then I get for external effects like PCM81 and PCM91 full latency compensation because Cubase measured the latency through the device.

There might be other use cases where latency appears to be critical but in fact isn't that much.
Example2: vocalist wants reverb for singing/recording. If you use FX (Reverb) from the DAW, then the little latency doesn't hurt much as Reverb has by itself a little delay. And the rest (EQing, compression) you apply anyway later.

With all due respect, but without more details on your specific request, it seems to me that you are engaging in some sort of "shadow hunting" for numbers without knowing if the small differences are of any relevance to your application. Of course, I may be wrong here, in which case sorry, but currently it seems that way to me.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub13

Re: Latency in Logic Pro X on macOS with the Digiface AES

will.studer wrote:

Thanks, but I just need the "real world numbers" for the Digiface AES in Logic Pro X. Just the latency numbers as shown in Logic Pro X (buffer size 64 samples at 44.1 kHz). I'm not into Windows based calculations or estimates, although your information is very thorough and helpful. Thank you very much.

Since it’s not the same on different computers and even not the same on one computer every time you use it, nobody can give you exact numbers.

UCX - FF 400 - Babyface pro - Digiface USB - ADI-2 (original)
Mac mini M1 - Macbook pro - iPad Air2

Re: Latency in Logic Pro X on macOS with the Digiface AES

ramses wrote:

With all due respect, but without more details on your specific request...

With all due respect, is this Gearspace? How can I make my question more „specific“? I even attached a bloody screenshot for those who have trouble with the terminology. It's precise, simple and clear in my first post:

will.studer wrote:

… can anyone also tell me the latency in Logic Pro X on macOS at 44.1 kHz with 64 buffers for the Digiface AES?

Für die Deutschsprachigen unter uns: Kann mir jemand die Latenz in Logic Pro X unter macOS bei 44,1 kHz mit 64 Samples Puffergröße für das Digiface AES nennen?

Oh dear, I'm not interested in RTL lectures, PCIe cards, mixing in Cubase, delay and latency compensation, recording vocals, Windows antics, mathematics, calculations, Excel sheets, advice on what buffer size to use or any other "estimates".

ramses wrote:

So careful not to compare chicken with eggs.

That's exactly what I don't do, which is why I insist on seeing/knowing these numbers in Logic Pro X and nothing else. See the logic?

However, forgive me for being a snobbish Mac user and of course assuming an Apple M1 based Mac - here all latencies for a given interface model are identical(!) for all currently available models, no place for Windows higgledy-piggledy show.

So allow me to clarify my question:

Can anyone tell me the latency in Logic Pro X on macOS (Apple M1 based Mac) at 44.1 kHz with 64 buffers for the Digiface AES?

15 (edited by ramses 2023-05-30 15:30:21)

Re: Latency in Logic Pro X on macOS with the Digiface AES

Nun mal immer schön mit der Ruhe.

Wenn Du so viel Forensupport machen würdest, wie ich das schon in meinem Leben (nicht nur hier im Forum) getan habe, dann versucht man in vielen Fällen erst mal herauszufinden, was der Hintergrund der Frage ist. Es wäre nicht das erste mal, dass sich dann die Relevanz einer Frage auch mal relativieren kann.

Daher wollte ich Dir einfach nur aufzeigen, dass bei RME die Unterschiede in der Latenz zwischen den Interfaces nur marginal sind. Je nachdem wie alt die Wandler sind, hat man da ggf. noch eine halbe Millisekunde Unterschied bei AD und DA bei single-speed. Die Frage ist halt, ob das für die gewünschte Anwendung entscheidend ist. Beispiele habe ich Dir sogar genannt.

Kurz: ich hätte einfach nur gerne verstanden, warum es Dir da auf jedes "Mü" ankommt, ob das für Deine konkrete Anwendung überhaupt von Relevanz ist.

Wenn Du Dich dadurch in irgendeiner Form belästigst oder brüskiert fühlst, bitte, Deine Entscheidung. Man sollte aber nicht immer gleich vom Schlimmsten ausgehen und denken, dass der andere einem etwas Böses will.

Viel Glück bei der weiteren Suche nach Informationen.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub13