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.

Many synths out of tune at 48khz sample rate

We’ve been discussing this in the stagelight thread here: https://forum.audiob.us/discussion/29594/stagelight-is-live-on-the-app-store/p32

In Stagelight and other au hosts, a number of synth apps play out of tune by just over a step when the host is at 48khz. This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

Here’s a list of some synths that show this behaviour:

  • Zeeon
  • Laplace (and the other Ice Gear synths)
  • Poison 202
  • Phasemaker
  • Unique
  • Volt

Some that don’t have this problem are Model D and Mood.

I gather this has been discussed before but not resolved. Can any devs shed some light on the problem? It’d be great to get it fixed across the board.

@giku_beepstreet @brambos @ka010 @MatthewAtStagelight

«1

Comments

  • edited December 2018

    I’m fairly certain that Phasemaker adapts to samplerate changes on the fly. Which leads me to believe that Stagelight doesn’t call AllocateRenderResourcesAndReturnError() correctly when changing samplerates.

    Edit: or the host is running at a different samplerate than was reported to the plugin...

  • in NS2 phasemaker without issues on any sample rate, same zeeon, ice gear synths, poison 202..

  • @brambos said:
    I’m fairly certain that Phasemaker adapts to samplerate changes on the fly. Which leads me to believe that Stagelight doesn’t call AllocateRenderResourcesAndReturnError() correctly when changing samplerates.

    Edit: or the host is running at a different samplerate than was reported to the plugin...

    This sounds like very useful information :smile:

    @MatthewAtStagelight

  • I adding to the list these two apps:

    -Syntronik
    -Synthmaster Player

    Both apps tested in several hosts, but the best test is to just load them in AUM in an iPhone without any audio interface or headphones attached (it will force 48KHz no matter what you select in your host).

    I reported the issue with Syntronik about six months ago to IK Multimedia and they acknowledged the situation, but unfortunately it hasn’t been solved yet. @ikmultimedia @Ryan_IK

    Synthmaster Player has been updated to AU just a few days ago and I have comfirmed the issue but haven’t reported it yet. If someone knows the devs or they have an account here please tag them. I’ll send them a bug report later.

  • @gregsmith said:
    We’ve been discussing this in the stagelight thread here: https://forum.audiob.us/discussion/29594/stagelight-is-live-on-the-app-store/p32

    In Stagelight and other au hosts, a number of synth apps play out of tune by just over a step when the host is at 48khz. This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

    Here’s a list of some synths that show this behaviour:

    • Zeeon
    • Laplace (and the other Ice Gear synths)
    • Poison 202
    • Phasemaker
    • Unique
    • Volt

    Some that don’t have this problem are Model D and Mood.

    I gather this has been discussed before but not resolved. Can any devs shed some light on the problem? It’d be great to get it fixed across the board.

    @giku_beepstreet @brambos @ka010 @MatthewAtStagelight

    I have tested Zeeon, Phasemaker, Laplace and Poison 202 in my iPhone 6S (without any audio interface nor headphones attached, therefore forced 48KHz) and they work fine, so I guess the problem must be related to the particular host as @brambos mentioned, in this case StageLight.

  • @Rodolfo said:
    Synthmaster Player has been updated to AU just a few days ago and I have comfirmed the issue but haven’t reported it yet. If someone knows the devs or they have an account here please tag them. I’ll send them a bug report later.

    @kv331audio_bulent @kv331audio

    I will re-check when I get home but really want to say I experienced the Stagelight sound shifting issue without any external synths involved--just using included drums and loops--but by plugging/unplugging my lightning connector earphones.

  • @gregsmith said:
    This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

    Disturbing! This means less than ideal sample rate conversions for anyone making
    music that will be played on a CD player or MP3 player that can’t accept non 44.1khz files. 48khz was originally a format for video, because it divided evenly into 24 frames per second. 44.1kHz has been the standard for music since the late 80’s.

  • Not even Apple is 100% 'sane' when it comes to sample-rates...

    I mean the iPhone 8 has 48K as the 'hardware' sample-rate (can't even select others in AudioShare or AUM) so why on earth does Apple insist on using 44.1K when recording Video when 48K is de-facto standard for Video work?!
    (The 44.1K audio that ends up in the video-file is most likely down-sampled from 48K).

    Is this yet another 'iOS Bug' that Apple has not solved yet?!

  • edited December 2018

    @Processaurus said:

    @gregsmith said:
    This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

    Disturbing! This means less than ideal sample rate conversions for anyone making
    music that will be played on a CD player or MP3 player that can’t accept non 44.1khz files. 48khz was originally a format for video, because it divided evenly into 24 frames per second. 44.1kHz has been the standard for music since the late 80’s.

    ? Nothing is keeping apps from using any sample rate they desire, even if it's 47.738628287 x Pi kHz, for mixdown, as that's entirely decoupled from the audio hardware.

  • @Processaurus said:

    @gregsmith said:
    This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

    Disturbing! This means less than ideal sample rate conversions for anyone making
    music that will be played on a CD player or MP3 player that can’t accept non 44.1khz files. 48khz was originally a format for video, because it divided evenly into 24 frames per second. 44.1kHz has been the standard for music since the late 80’s.

    well, the PC 'standard' soundcard for decades (Creative Labs Soundblaster) did all internal processing at 48khz, while most applications used 44.1 as their samplerate.
    The samplerate conversion was pretty slick (based on EMU's sampling technology iirc) and I assume Apple's current routines are quite solid, too.
    At least when converting in Audioshare I never noticed any (obvious) decrease in quality and I was much surprised by the speed of the process.
    (never did picky A/B comparisons, though)

  • @SevenSystems said:

    @Processaurus said:

    @gregsmith said:
    This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

    Disturbing! This means less than ideal sample rate conversions for anyone making
    music that will be played on a CD player or MP3 player that can’t accept non 44.1khz files. 48khz was originally a format for video, because it divided evenly into 24 frames per second. 44.1kHz has been the standard for music since the late 80’s.

    ? Nothing is keeping apps from using any sample rate they desire, even if it's 47.738628287 x Pi kHz, for mixdown, as that's entirely decoupled from the audio hardware.

    Yes, but the issue is bouncing down at a different sample rate than the session was recorded at, that is the moment where the less than ideal re-sampling would happen, because you weren't able to record at 44.1khz, when you ultimately want to be bouncing down to 44.1khz.

  • @Processaurus said:

    @SevenSystems said:

    @Processaurus said:

    @gregsmith said:
    This is a problem in later Apple products such as the new iPad Pro and iPhone X, as the sample rate is locked to 48khz.

    Disturbing! This means less than ideal sample rate conversions for anyone making
    music that will be played on a CD player or MP3 player that can’t accept non 44.1khz files. 48khz was originally a format for video, because it divided evenly into 24 frames per second. 44.1kHz has been the standard for music since the late 80’s.

    ? Nothing is keeping apps from using any sample rate they desire, even if it's 47.738628287 x Pi kHz, for mixdown, as that's entirely decoupled from the audio hardware.

    Yes, but the issue is bouncing down at a different sample rate than the session was recorded at, that is the moment where the less than ideal re-sampling would happen, because you weren't able to record at 44.1khz, when you ultimately want to be bouncing down to 44.1khz.

    Ah got it, I almost exclusively make EDM "in the box" nowadays so recording sampling rates are not an issue ;) When I need vocals, I just grab one of the 72 bazillion hours of archived recordings from the better days ;) and chop them up (oh, and those ARE 44.1 kHz :D)

  • @Processaurus said:
    … 44.1kHz has been the standard for music since the late 80’s.

    A standard but not the standard. In the late 80s or a couple of years + I bought my first DAT recorder and ran it mostly at 48kHz, and further into 90s when I moved to minidisc recorders (because my two DATs were skewing all over the place so I lost trust in them) they were recording at 48 all the time, too.

  • Bit late to the party here, but @gregsmith, do they work okay in AB, just out of curiosity?

  • Just did the 'crude conversion' in Audioshare (a 44.1khz 24bit guitar DI track peaking at -8dB) upward to 48k and then back down to 44.1.
    The put the original source and the twice converted version in Auria Pro, inverted one of them and saved the mixdown.
    This mixdown file is supposed to only contain the difference between the original and the processed version - it's some high frequency stuff beyond 16khz at -54 dB.
    Pretty good for 2 instantaneous conversions ;)

  • @Michael said:
    Bit late to the party here, but @gregsmith, do they work okay in AB, just out of curiosity?

    Michael, I loaded them in AB and tested Zeeon, Phasemaker, Laplace and Poison 202 in my iPhone 6S and they all work fine.
    There’s a clear issue in StageLight (which makes several apps to be out of tune) and there are some specific apps that are out of tune in ANY host.
    So far, the missbehaving AU apps I can confirm are:

    -Syntronik
    -SM Player
    -FM Essential

    I am checking all my AU synths to see if I find another one.

  • edited December 2018

    I can confirm on my side on iPhone 7 Plus the issue for AB/Aum/ApeM/GB hosts on internal speakers 48 kHz and headphones 48 kHz for the following synths:

    • Syntronik
    • SM Player

    They play well on headphones 44.1, which is setup I use most of the time, as 7 Plus allow for that. But that means I can’t use them reliably on speakers.

    Stagelight has additional 44.1/48 issues with Zeeon and certainly others.

    Even worst for X and new Pro users.

  • Well, I tested all my AU synths (34 apps in total, not counting Drums apps) and I confirm the following apps go out of tune when sample rate is set to 48KHz:

    -Syntronik
    -SM Player
    -FM Essential
    -Addictive Pro

    I have already reported Syntronik and SM Player to their corresponding devs, and will do it soon for the other two.

  • edited December 2018
    The user and all related content has been deleted.
  • edited December 2018

    @Michael said:
    Bit late to the party here, but @gregsmith, do they work okay in AB, just out of curiosity?

    Good point. Just tried Zeeon and Phasemaker as au in AB and they are in tune. So it is a stagelight prob, and AEM which also has the problem (but does let you work at 44.1 as a workaround)

  • @gregsmith said:

    @Michael said:
    Bit late to the party here, but @gregsmith, do they work okay in AB, just out of curiosity?

    Good point. Just tried Zeeon and Phasemaker as au in AB and they are in tune. So it is a stagelight prob, and AEM which also has the problem (but does let you work at 44.1 as a workaround)

    Right! Okay good to know

  • @Telefunky said:
    Just did the 'crude conversion' in Audioshare (a 44.1khz 24bit guitar DI track peaking at -8dB) upward to 48k and then back down to 44.1.
    The put the original source and the twice converted version in Auria Pro, inverted one of them and saved the mixdown.
    This mixdown file is supposed to only contain the difference between the original and the processed version - it's some high frequency stuff beyond 16khz at -54 dB.
    Pretty good for 2 instantaneous conversions ;)

    Good experiment, thanks!

  • In fact I was quite surprised and it should be considered that the converted (altered) signal was converted (altered) again to be able to phase out the original. So a single pass might even yield less than -60dB of difference, but only in the final half of the top octave.
    No idea what was altered because I can't hear in that range of the spectrum o:)

  • edited December 2018

    You can add Quanta in the list. I was so shocked when it first happened. This has to be fixed cause we don't want to mix at 44.1 rate in 2018!!!

    Edit: It happened in NS2 for me.

  • edited December 2018

    @Telefunky said:
    Just did the 'crude conversion' in Audioshare (a 44.1khz 24bit guitar DI track peaking at -8dB) upward to 48k and then back down to 44.1.
    The put the original source and the twice converted version in Auria Pro, inverted one of them and saved the mixdown.
    This mixdown file is supposed to only contain the difference between the original and the processed version - it's some high frequency stuff beyond 16khz at -54 dB.
    Pretty good for 2 instantaneous conversions ;)

    This sounds really interesting.

    So the inverted waveform is cancelling the other, with any sound remaining being the 'difference'?

    Cool. Any easy steps on how to give this a try? (beyond what you already stated). I think I get it, but a step guide would help my brain! :) Cheers.

  • @Norbert wrong thread - it's not about restriction to one particular sample rate, but about messing 44.1 and 48khz by overlooking some conditions during developement.

    If you prefer 96khz (and above), set your device to that rate or use an interface that supports it.
    Which btw would be rather pointless since practically all (qualitity) IOS sound generators/processors do internal oversampling when appropriate.
    Not to forget that careless use of today's so called 'mastering strategies' can easily wash away any tiny benefit of a pure 96khz source.
    Use your ears and don't rely on spec sheets. ;)

  • edited December 2018

    @Telefunky said:
    @Norbert wrong thread - it's not about restriction to one particular sample rate, but about messing 44.1 and 48khz by overlooking some conditions during developement.

    If you prefer 96khz (and above), set your device to that rate or use an interface that supports it.
    Which btw would be rather pointless since practically all (qualitity) IOS sound generators/processors do internal oversampling when appropriate.
    Not to forget that careless use of today's so called 'mastering strategies' can easily wash away any tiny benefit of a pure 96khz source.
    Use your ears and don't rely on spec sheets. ;)

    Why is it the wrong thread? The fact Quanta cannot be mixed down in 48khz because the pitch is wrong is not the right thread? But it seems to me the title shows it's exactly that: 'Many synths out of tune at 48khz sample rate'.

  • @Norbert said:

    @Telefunky said:
    @Norbert wrong thread - it's not about restriction to one particular sample rate, but about messing 44.1 and 48khz by overlooking some conditions during developement.

    If you prefer 96khz (and above), set your device to that rate or use an interface that supports it.
    Which btw would be rather pointless since practically all (qualitity) IOS sound generators/processors do internal oversampling when appropriate.
    Not to forget that careless use of today's so called 'mastering strategies' can easily wash away any tiny benefit of a pure 96khz source.
    Use your ears and don't rely on spec sheets. ;)

    Why is it the wrong thread? The fact Quanta cannot be mixed down in 48khz because the pitch is wrong is not the right thread? But it seems to me the title shows it's exactly that: 'Many synths out of tune at 48khz sample rate'.

    Not the wrong thread as far as I can tell 👍

  • @Norbert sorry, I misread your comment as '44.1 isn't good enough today'

  • edited December 2018

    @Telefunky said:
    @Norbert sorry, I misread your comment as '44.1 isn't good enough today'

    If "today's youth" ™️ continues to listen to music at these volume levels, soon 32 kHz will be good enough :D

    (don't take it too seriously -- I also love to listen at Tinnitus-inducing levels)

Sign In or Register to comment.