Hi there, I just discovered this thread recently, and it seems that besides the different opinions concerning additional processes and wifi and so on, there‘s a consensus that MBP2018 owners using their machines for low latency audio work are struggling with issues mainly caused by the Smartbatterymanager, right?

There‘s been the talk about disabling this driver with the .xkext method (not really an option as I want to use a mobile device in a mobile way, which isn’t really nice without knowing my battery stats) and about guys knowing apple officials and so on...

Does anyone have an estimate about when this could be fixed, if even? Did anyone already get a hotfix or something similar?

Indeed there is.

But, why am I not seeing it in App Store/Updates..?

EDIT: Now I know. The update requires APFS-formatted drives. My System drive is a bootable "clone" created with ChronoSync and is formatted as HFS+ (Journaled).

WTF.

I'm not really sure what's going on. I think "clones" (which they really aren't) created with CCC are also HFS+ formatted as they were seeable by HFS+ partitions when I tried it, which shouldn't be possible.

It seems all backup solutions currently create HFS+ formatted "clones" which makes no sense to me. Apple's own Disk Utility is useless, as discussed in the Mojave thread.

Yeah, man.

Apple, pull your head out!

EDIT: It occurred to me that in order for the "clone" to be APFS-formatted, the destination might have to be pre-formatted as APFS as well. If it is HFS+ formatted, it will stay that way. I will have to try this out.

damn. i thought i was the only one with these issues. looking forward to apple solving the problem.

In case anybody is still interested, I posted a summary of my findings on Mojave here:

MacOS 10.14 "Mojave": share your experiences here

Regardless if this is the way. i appreciate the post on this forum and any way to get it resolved. I too am running RME and thought it was their driver.

I mean how are you going to report a bug when you are 4 levels deep into hardware and software. its ****ing difficult. So i'm running OSX High Sierra, Fireface UC, Ableton, 30 different plugins. It glitches intermittently. What do i say? If you have any advice about how to write a bug that would be useful to apple I would love to hear it.

OK, here are some ideas on how to structure a bug report:



-- give it a descriptive title

This could be something like 'intermittent CPU spikes in audio applications'



-- list all hardware and software in use when the issue occurs

That includes hardware specs like Mac model, CPU speed, RAM specs - how much is installed, and whether it's original or 3rd party RAM, OS version, exact version numbers of all software in use, driver versions, etc



-- give an exact, step by step description on how to reproduce the issue

Word this like you're speaking to a child, meaning, include things that you wouldn't think are necessary, because they are so obvious. Every step matters. Also, be clear on whether the issue is intermittent.



-- state the 'expected behaviour'

..like, for example, 'continuous audio playback without interruption'



-- state the 'observed behaviour'

Again, be precise, like 'audio drop outs during monitoring input while recording, not audible on the recording itself when listening back to it'



-- list all the troubleshooting steps you have taken, and your observations

This includes all the basics, like testing without any external hardware connected, testing with just the audio interface and nothing else connected, testing with a fresh user account, testing with all plug-ins disabled, etc.
Example: 'the issue occurs when the RME interface is connected, it does not occur when only internal audio is used'



-- optional: if you think you've found relevant log entries, copy and paste them into the bug report

..again, be precise, state which logs they come from. Including logs is not necessary in most cases - if engineers reproduce the issue, they can look at their own logs, and sift through them in a much more systematic manner, and with a much better understanding of what is important, and what is normal behaviour.

What's the latest on the T2/OSX audio bug? We are still getting glitches on recording and playback through various RME interfaces (Core Audio) and their ADI-2 Pro (i.e. Class Compliant). Are you guys having problems with both third party and Apple's drivers too? Does it also affect Logic users? We're using a 2018 13" MBP running Reaper with genuine Apple USB-A adapters.

It seems the first of the two OSX 10.13.6 'supplemental updates' improved the problem to the extent that we recorded 2 or 3 shows with 20 or so channels at 96k with no glitches.

Then the glitches began again with the second update, and have continued with Mojave. That said, it's hard to test a sporadic problem exhaustively, so was it ever fixed? Neither Apple's Activity Monitor nor Reaper's Performance Meter are showing overloads when I hear the glitches. I haven't yet checked the Console but would guess this is the next step...

What's the best formal route to raise the issue with Apple?

At first I thought it was just the RME thing... but I realized that I get the glitches even without the RME hooked up on a 2017 macbookpro 15". It is not good. I haven't written music in weeks because of it.

Macbook Pro 2018 Audio Glitch - Apple Community

I'm curious what the update is with zenwolf

Wow, that's bad. I am sorry to hear.

Could you explain more about the "coreaudio crackling bug"? Does it affect primarily/only USB interfaces?

I thought it was limited to the latest generation MBPs?

Right.

FWIW, I just received a mail by Static Z Software since I use their Sound Control app on my current Mini, and I'm also on their beta list. They sent me a link to a new version and the release notes say (amongst other things):

"Initial compatibility with T2 Macs. (iMac Pro, 2018 MacBook Pro, 2018 Mac mini)."

So, it appears the T2 chips are very much relevant to audio and that older software and drivers may not be compatible.

Do we know for sure whether this is related to the 'crackling' speakers posts on other sites?

To me it sounds like a buffering issue either with a silent dropout on playback, or a repeated glitch noise during recording (neither sounds like the crackling people have posted on YT).

What's the definitive text to search for in Console to diagnose?

And why doesn't disabling the battery manager fix it?

I'm one of the posters in the other thread(s). I have a MBPro i7 2.2 and I have the same USB audio dropout issue, though I've managed to mitigate it to about once every couple of hours.

I don't have a direct correlation, but I have bluetooth, siri, location services and iCloud turned OFF.

When the glitch happens on my setup, it is *always* preceded by a "timed" daemon event which relates to clock setting. My theory is that the internal clock controlled by the T2 is sometimes reset as a result of internal clock drift from the time servers and resync via the "timed" process. I have not tested whether turning off wifi (and thus blocking access to the time servers) will prevent this.

The actual error is reported by the kernel for HALS_IOA1Engine.cpp, but there's a bunch of IOAudioEngine related messages in the logs that indicate that the audio engine gets paused and restarted (several times) immediately following the "timed" messages. That's the glitch in the audio stream.

I filed a bug report and sent in the requested sysdiagnose. Apple's helpful response was to upgrade to Mojave because "This is running a pretty old version of our driver, and we have certainly made improvements since then." Ironically, I'm running 10.13.6 with all of the updates except the problematic security update 2018-002. So basically "take a hike". Its also clear from users running the latest Mojave, that the issue is *not* fixed there either.

Dorian

Thanks Dorian. Which update/problems are you referring to?

I'm interested in your comments because there were two 'supplemental updates' to 10.13.6 and I found the first of these delivered the best audio performance (we recorded several 2hr concerts with 12-22 channels at 96k and no glitches).

However, the glitches returned with the second update, and continued on all subsequent versions of 10.14

Sorry I wasn't a bit clearer. I have both *supplemental* updates installed, but not the latest *security* update (2018-002 for HS). See MacRumors for example. Mojave 10.14.1 includes those security patches FYI.

Normally I would install a security update, but I'm a bit leery of destabilizing a "sort-of" stable system. Also why I'm still running LPX 10.4.1 instead of 10.4.2 - because its currently stable!

I'm guessing this problem is only related to USB audio interfaces.

I just ran a successful 4 hr recording session @ 44.1 with my 2018 2.2 i7 MBP w/32G RAM through a
USB-3 -> Thunderbolt 2 dongle attached straight to my Antelope Audio Zen Tour Interface without any glitches, stutters, dropouts or breaks.

...And that's with another dongle going from USB C -> a USB2 hub supporting a Push 2 and other things inside of Ableton.

It's unfortunate this issue is still affecting people using USB interfaces with 2018 Mac Minis and MBPs but I'm quite happy and satisfied with thunderbolt and this computer's performance for audio work.

This was on Mojave 10.14.1

Thanks for all the reports, guys, keep it up. This is the only way to get a true sense of what issues there might be before making buying decisions.

BTW, I run a RME card inside a Sonnet Thunderbolt chassis with a 2012 Mini i7 and Mojave without any issues. I'm also looking at getting a new Mini.

10.14.2 same issue

stockholm syndrome is the only rationale for staying on OS X for audio now.

A user with similar symptoms to me and others reported that turning off the automatic clock sync in sys prefs seems to have helped with the glitching since they also saw evidence of the glitch in tandem with the timed clock sync process. I'm gonna try that this weekend.

Audio interface issues with 2018 Macbook Pro | MacRumors Forums

Maxwelldub for what its worth i totally agree with you. I used to use macs but they are worse than microsoft now. And way more greedy.

I had to recently buy a cheap MBP 8.1 whilst Metric Halo forfill their promise to support windows. I have noticed massive glitching to do with multi core usage of plugins. If they are able to (eg kontakt, sd3), switch them to single core and only allow the daw to be multicore.

But yeah guys smell the coffee, dont waiste money on new mac hardware. Hackinosh or cheap second hand is all thats left. Now go ahead and flame me

Good to hear that someone above has had a good experience (xZenx). Myself and 2 of my other friends with 2018 I9 MacBook Pros with 32 GB of RAM are all having the same issues, but with different interfaces and different DAWS. Same problems though, namely low latency audio is completely impossible: glitches that sound like the devil's ring mod, stuttering, popping and dropouts. I am using the latest High Sierra on my old Mac Pro and its flawless, 16 channels at 96k with an apogee symphony at 64 buffer no problem. The single core speed is nowhere near what this new machine is capable of, but I can't do that on the 2018 I9 MacBook Pro without glitches. My 2011 MacBook Pro also just hums along and smokes the new computer for recording glitch free.

Similarly, I am having problems with the Slate VRS-8s. I wouldn't blame it on them since its the same problem I have with the USB Apogee, so its clearly an OS/hardware issue. So far I've just gotten a runaround from Apple with a lot of conflicting info. Here is a list:

1st call: "never heard of this before. Try reinstalling your system software."
2nd call with a Logic specialist: "never heard of this before, wait the 1st guy told you to reinstall your system software? Thats not right! I don't know why this is happening, let me get you to an OS specialist."
3rd Call (OS specialist): "never heard of this before. Lets find the drivers that are older than 2017, they are probably the problem. Even though that Apogee driver works flawlessly on your 2 other machines (2011 MacBook Pro and Mac Pro) it won't necessarily work on this computer and is probably causing all of your audio problems even if you aren't using it." But why are all of my friends having the same problem in different DAWs and with different interfaces? "Well, probably no developers have tested with these new computers."

So basically I got nowhere in 3 hrs of talking with Apple.

Anyone else have success with low latency audio with the 2018s?

If three different support agents at Apple tell you that they never heard of this before, it's a very good indication that this is not an issue that has been reported with any frequency before. This may come as a shock, but internal communication at Apple about new and emerging issues is not by word of mouth.


If you look at this thread, you'll see various different reported issues, on various different hardware.

I'm not saying it's impossible that there is actually a real OS bug at the core of it, but I haven't seen any kind of conclusive indication that this is actually the case.

Is this only happening with USB audio interfaces? How are thunderbolt interfaces doing? Anyone have both types and can confirm it is isolated to just USB?

No my thunderbolt interfaces weren’t working either.

Actually, it’s mostly people talking about the 2018 MacBook pros (which I and 2 of my friends have) and low latency audio. You have a 2010 (?) MacBook pro that has no issues. I have a 2011 MacBook Pro that has no issues.

Looking around the internet, there seems to be a consensus that there is some kind of common problem with these machines. You can keep arguing if you want, but if you had one of these machines those arguments would probably be more resonant.

Also, I’d like to hear from anyone else that has a 2018 MacBook Pro that’s working great for recording/monitoring at low latency. Anyone else?

Wow! that is bad. I was testing out a 2018 Mac Mini and had this problem with my RME UFX connected USB.

Sounds like an Apple issue either with the T2 chip, new hardware somewhere or with MacOS.

I'm just trying to argue for a rational, systematic approach to problem solving.

If someone claims it's an issue with the AppleSmartBatteryManager.kext, and your and my older MBPs use the exact same kext (same version number) without issue, then the issue is not caused by that kext - to name just one example.


If it's an issue with the OS's USB driver, then the issue would not occur on Thunderbolt, or with any audio interface that uses it's own USB driver (meaning, an UAD or RME USB driver, rather than running driverless/class compliant with the OS's driver).


There's so much contradictory information in this thread ('it's a USB issue - no, it also affects Thunderbolt; it's an issue with 2018 MBPs - no, I see the same issue on an older Mac; it's a driver issue - no, it's a hardware issue; it's an issue with T2 only on MBP's - but not with the same T2 chip on iMacs; etc) that it's fairly impossible to draw any meaningful conclusions.

I wouldn't doubt anybody who says they've got problems - except for a very small number of obvious trolls who make things up.

So that's a nonissue.

If I was in this position - and I might be soon because I'm planning on getting a new Mac - I'd first wipe the internal drive and install OS X again, then update. Lastly, install audio drivers and see what happens. If the problems persist, I'd take the machine and interface to an Apple Center and let them deal with it.

Try turning off time syncing in system prefs. Both the sync itself and the setting for time zones. There’s some evidence that the time sync process “timed” is causing the internal clock to resync, causing a pause and restart of the audio engine. I’ve seen this error in my logs and there’s a user on macrumors who claims that turning off the time sync appears to fix the issue. I *think* that turning WiFi and any other network interfaces off would have the same effect since timed will only run with a network interface attached.

anyone interested in solving this issue in this thread should ignore this user ('stratology')

also its quite obvious that apple's failure to properly identify & track this bug is at least as much of a customer service/internal process failure as the bug itself is an engineering failure because reporting this to apple and going through the motions has resulted in zero progress on this in 6 months. you can bring your machine to the Genius Bar, call apple support, even get a replacement computer but you won't be given a machine that's usable for audio. so its clear we have to come up with a workaround ourselves.