Audiobus: Use your music apps together.

What is Audiobus? — Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.

Download on the App Store

Audiobus is the app that makes the rest of your setup better.

Comparing interface latency

dokwok2dokwok2
in Hardware

Is there an easy way to measure interface latency? I’d like to compare the Boss IR-200 with my Motu M2.

«1

Comments

  • filo01filo01
    edited January 14

    It could be easy if you have app for that :) It is just about creating loop between input and output and measuring of time, I did it several times.
    I'm using free RTL Utility on Windows https://oblique-audio.com/rtl-utility.php
    and Onyx Round Trip Latency Meter (paid) https://onyx3.com/LatencyMeter/ for iOS.

    I will post Onyx measurements of any card I have the opportunity to test.

  • dokwok2dokwok2

    @filo01 said:
    Onyx Round Trip Latency Meter (paid) https://onyx3.com/LatencyMeter/ for iOS.

    Thanks for that suggestion. I looked at the version history and reviews: it doesn't seem to have been updated in 5 years, and stopped working several versions of iOS ago.

    AUM has a round trip latency measurement: is that the number I want to compare?

  • mysterycatmysterycat

    I bought Onyx recently, and it worked great 😌

  • filo01filo01
    edited December 2023

    @dokwok2 said:
    AUM has a round trip latency measurement: is that the number I want to compare?

    No, it is not overall latency just inside of AUM.

    Here is my result I did with Onyx several weeks ago - IKM iRig HD2, NuX Mighty Plug Pro and RME Babyface Pro FS.

  • dokwok2dokwok2

    @filo01 said:
    Here is my result I did with Onyx several weeks ago - IKM iRig HD2, NuX Mighty Plug Pro and RME Babyface Pro FS.

    That's a sizable gap!

    Thank you both, @filo01 and @mysterycat, for confirming that the Round Trip Latency app still works.

  • jebnijebni

    I was just about to ask about this very topic to sort out some of my issues, so this is great!

  • dokwok2dokwok2
    edited December 2023

    Since the app requires cabling, I decided to roll my own test using AUM:

    • I ran a short cable between the output of my interface and the input.
    • In settings, I turned off AUM's built-in latency correction.
    • I recorded a snare hit (from Ruismaker).
    • Using an audio editor, I trimmed the resulting file so that the snare hit started immediately on playback.
    • In AUM, I set up one channel playing the trimmed audio file and one channel to record the input from my interface. I armed the recording channel. Since the output of the interface was currently plugged into the input, I would be recording the playback of that snare. With the input channel volume at 0 dB, I got some feedback, so I backed off the volume until the feedback disappeared.
    • In AUM, I backed up the transport to the beginning, so that as soon as I hit record, the snare file would instantly start playing and the input channel would instantly start recording.
    • Using an audio editor (in this case, Twisted Wave), I zoomed into the file I had just recorded to see how many milliseconds elapsed between the beginning of the file and the beginning of the snare.
    • I repeated the last two steps with different settings.

    With my MOTU M2 interface running at 48k, 24 bits, and 64 buffer, I clocked a little over 8 ms, about a millisecond slower than that RME Babyface. With 32 buffer, I got it down another ms. For comparison, AUM estimated roundtrip latency at around 6 ms. With latency correction on, I got just under 4 ms.

    Curious to see the numbers for the new Boss IR-2...

  • filo01filo01
    edited January 14

    Thank you @dokwok2 for your findings. I don't have AUM so I will continue to show results from Onyx app here in 32/64/128 format at 48kHz.

    Audient EVO 8

    RME Babyface Pro FS
    These values are big disapointing for me. Probably due to the lack of low-latency drivers (ASIO) in iOS.
    EDIT: The question is whether it makes sense to buy an expensive sound card for iOS just because of low latency. Considering the other results, I would say it doesn't. Latency difference between RME and e.g. Focusrite Scarlett or Audient Evo is negligible.

    NuX Mighty Plug Pro
    It is a little bit tricky to use this NuX as audio interface for monitoring/tracking. You need to pair the NuX with control app via Bluetooth at first and then set USB audio slider to 100% wet there.

  • mojozartmojozart

  • dokwok2dokwok2

    @mojozart said:

    I saw that too. I think that's the latency from the cab simulation. Will the latency from the audio interface be higher? Maybe not, or maybe not a lot, since the A/D conversion has already happened. But you never know until you test...

    Speaking of testing: thank you, @filo01, for reporting those test values. I think you're right about the drivers: I expect my MOTU would perform better on a MacBook or Windows machine for the same reason. Recent versions of iOS do allow for custom drivers, but I'm guessing the market's too small for companies to invest in writing and maintaining them.

  • mojozartmojozart
    edited December 2023

    The IR-2 results are strictly for the pedal features.

    In practice, the effective audio interface latency is dominated by the buffer size, which depends on how much processing you want to add. The buffer size is the latency.

  • dokwok2dokwok2

    @mojozart said:
    In practice, the effective audio interface latency is dominated by the buffer size, which depends on how much processing you want to add. The buffer size is the latency.

    As @filo01 's tests indicate, some interfaces are never going to get below 10 ms, even with a small buffer: the iRig HD2, for example. I expect the Boss will do better than that; question is: how much?

  • filo01filo01
    edited December 2023

    Arturia MiniFuse 1
    Measured with Onyx app. https://onyx3.com/LatencyMeter/

  • filo01filo01

    Presonus Studio 26C

  • filo01filo01
    edited December 2023

    Focusrite Scarlett Solo 3rd gen
    Almost the same values as Audient EVO8

  • xmortenxxmortenx

    @filo01
    Awesome that you post this :)
    Any ideas of how Motu M2/4/6 fares with RTL ?

  • filo01filo01
    edited December 2023

    @xmortenx said:
    @filo01
    Awesome that you post this :)
    Any ideas of how Motu M2/4/6 fares with RTL ?

    No idea if you mean RTL under iOS. I don't know anyone who has both Motu and Onyx app and can measure it. As I stated in another thread, the measurement results from PC/Mac cannot be applied to iOS due to lack of low latency drivers.

  • dokwok2dokwok2

    @xmortenx said:
    Any ideas of how Motu M2/4/6 fares with RTL ?

    I posted some numbers for the M2 above, using the method I described there.

  • filo01filo01

    @dokwok2 said:

    @xmortenx said:
    Any ideas of how Motu M2/4/6 fares with RTL ?

    I posted some numbers for the M2 above, using the method I described there.

    Oh, I see now. Short memory, I'm sorry not to mention your result.

    With my MOTU M2 interface running at 48k, 24 bits, and 64 buffer, I clocked a little over 8 ms, about a millisecond slower than that RME Babyface.

  • dokwok2dokwok2

    Yesterday I did the same test with a Boss IR-200. The numbers aren't exactly comparable, since the IR-200 only operates at 96kHz. Here's what I found:

    6.7 milliseconds (ms) with a buffer of 64
    6.1 ms with a buffer of 32

    Not too shabby, but see my notes on monitoring here: https://forum.audiob.us/discussion/comment/1288583/#Comment_1288583.

    For comparison, I tested my MOTU M2 interface and got round trip latency at 48kHz/24 bits of 8.8 ms with a buffer of 64 and 7.5 ms with a buffer of 32.

    This morning I did some further testing with the M2, to measure the difference between running the M2 directly into my iPad Pro M1 (1 USB-C cable) and using a powered hub (2 USB cables, the hub, plus a USB-A to USB-C adapter). I'm sure there is a difference, but I couldn't measure it (which is nice, because I use that hub to add MIDI ports).

  • dokwok2dokwok2

    Some more numbers, this time for a TC Helicon Go Twin (which appears very similar to an iRig Pro Duo). Same test as described above, playing a snare hit in AUM:

    48 kHz, 24 bits, 32 buffer: 11.7 ms
    48 kHz, 24 bits, 64 buffer: 13 ms

  • filo01filo01
    edited January 3

    Focusrite Scarlett 2i2 3rd gen

  • dokwok2dokwok2
    edited January 14

    Focusrite Scarlett 2i4 1st gen:
    48 kHz, 24 bits, 32 buffer: 11.1 ms
    48 kHz, 24 bits, 64 buffer: 12.4 ms

    Notes: (1) filo01 is using different tools to measure; and while they should produce the same results, I haven't tested that. Caveat emptor. (2) I'm not rebooting between tests or activities. When I have retested things (occasionally), the results have been consistent. But this isn't Consumer Reports.

  • filo01filo01
    edited January 14

    Zoom U-24

  • filo01filo01
    edited January 14

    Yamaha AG03 MK2

  • NoiseHorseNoiseHorse

    @dokwok2 said:
    Is there an easy way to measure interface latency? I’d like to compare the Boss IR-200 with my Motu M2.

    My money’s on the Motu.

  • dokwok2dokwok2

    @NoiseHorse said:
    My money’s on the Motu.

    See above: https://forum.audiob.us/discussion/comment/1288587/#Comment_1288587

  • filo01filo01

    Swissonic Audio 1 (Thomann)

  • gkillmastergkillmaster

    I just bought an Audient ID14 MkII and wonder how it would fare!

  • filo01filo01

    @gkillmaster said:
    I just bought an Audient ID14 MkII and wonder how it would fare!

    If you're willing to spend $9.99, there's an easy way :) https://apps.apple.com/us/app/round-trip-latency-meter/id1427507645

Sign In or Register to comment.