Apple announces new Mac Mini, Air + 13" MBP featuring their own M1 chip.

[zvenx]

https://steinberg.help/cubase_pro_artis ... ard_c.html

ASIO-Guard

The ASIO-Guard allows you to shift as much processing as possible from the ASIO realtime path to the ASIO-Guard processing path. This results in a more stable system.

The ASIO-Guard allows you to preprocess all channels as well as VST instruments that do not need to be calculated in realtime. This leads to less dropouts, the ability to process more tracks or plug-ins, and the ability to use smaller buffer sizes.
ASIO-Guard Latency

High ASIO-Guard levels lead to an increased ASIO-Guard latency. When you adjust a volume fader, for example, you will hear the parameter changes with a slight delay. The ASIO-Guard latency, in contrast to the latency of the audio hardware, is independant from live input.
Restrictions

The ASIO-Guard cannot be used for:

Realtime-dependent signals

External effects and instruments

I know Logic has something similar, maybe others do too.
rsp

[zvenx]

Unless the DAW can 'look ahead'.......which they can do once it isn't a live track.
rsp

Logic has an option to do multicore on live-input tracks as well tho.

I just wish apple finally disabled auto-arm for instrument tracks, it can really murder computers on track stacks when you just navigate around the project.

Above my paygrade to try to explain that .


About to read this:
https://support.apple.com/en-us/HT205975

Or we can just take Markus's word (which I have already done)
rsp

edit:

From Logic:

The Playback & Live Tracks setting distributes the load to different threads in cases where a Track Stack that contains multiple software instrument channel strips is selected, or when more than one audio track is recording. If the computer has enough cores and there are no other existing DSP demands on the computer, each live channel strip can be processed by a different core.

so what I understand it all to mean is for prerecorded things it can multicore even if it is one track (my screenshots show that).
and if it is live track in cubase and logic if it is only one track, it will probably use only one core.

[Ploki]

Unless the DAW can 'look ahead'.......which they can do once it isn't a live track.
rsp

Logic has an option to do multicore on live-input tracks as well tho.

I just wish apple finally disabled auto-arm for instrument tracks, it can really murder computers on track stacks when you just navigate around the project.

Above my paygrade to try to explain that .


About to read this:
https://support.apple.com/en-us/HT205975

Or we can just take Markus's word (which I have already done)
rsp

damn, i thought i knew everything.
that setting actually improves things here, because in the end-game mix i always make submixes with stacks, and just clicking on one murders it.
I'll try disabling it again.

Freezing a single track inside the track stack prevents stack from getting armed. it's just a nuissance to do so tho.
A simple "mix mode" where there's no live input and related cpu load would be very welcome.

[spacepluk]

It seems that beeing a full-time plugin-developer for nearly 20 years is 'not experienced' enough to know how plugins work.

Hmm, when I said I would love to hear from someone with more experience that included you.

[spacepluk]

From Logic:

The Playback & Live Tracks setting distributes the load to different threads in cases where a Track Stack that contains multiple software instrument channel strips is selected, or when more than one audio track is recording. If the computer has enough cores and there are no other existing DSP demands on the computer, each live channel strip can be processed by a different core.

so what I understand it all to mean is for prerecorded things it can multicore even if it is one track (my screenshots show that).
and if it is live track in cubase and logic if it is only one track, it will probably use only one core.

The way I understand this it doesn’t explain why it’s using multiple cores in your experiment. It says: “multiple software instrument channel strips”. Now I’m really curious to know how it works.

[Sascha Franck]

so what I understand it all to mean is for prerecorded things it can multicore even if it is one track

Of course it can as there's barely any buffer constraints anymore. Playback tracks will be set to 1024 samples buffersize. That allows for quite some core/thread jumping. 32 samples however don't.
And besides, all that still doesn't mean serial signal paths can be processed parallely.

[spacepluk]

I would argue that for a single track with a serial chain jumping between threads actually makes things slower...

[Sascha Franck]

Btw, first singlethread results for Logic are out on GS (I created a litle test project and someone with an M1 MB Air has been so kind to run it). I used one EXS instance and ran it through an array of Space Designers (all serially on auxes running into each other). Space Designer is Logics IR loader also offering a socalled syntesized IR (which I think is some white noise that just gets some tweaking). CPU hit is in the medium range so it's perfect for comparisons.

Ok, on my 2x2.66GHz cheesegrater Mac Pro (still a pretty decent computer), at 44.1 and 64 samples, I can run 24 instances of Space Designer (quite good for a 10 year old computer). Someone with a 16 core new Mac Pro managed 52 at those settings.

And now fasten your seatbelts: The baseline Macbook Air (yeah, with just 8GB of RAM, even if that shouldn't matter much for this test) pulls of 80 (in words: EIGHTY!) instances! A fanless Starbucks gadget is running circles around a 12k machine in terms of single core performance!

[Sascha Franck]

I would argue that for a single track with a serial chain jumping between threads actually makes things slower...

Exactly (if it would happen at all, which it isn't on "live" tracks - at least I don't know of any host doing so). And that's not exactly desirable, more to the opposite.

[Ploki]

Btw, first singlethread results for Logic are out on GS (I created a litle test project and someone with an M1 MB Air has been so kind to run it). I used one EXS instance and ran it through an array of Space Designers (all serially on auxes running into each other). Space Designer is Logics IR loader also offering a socalled syntesized IR (which I think is some white noise that just gets some tweaking). CPU hit is in the medium range so it's perfect for comparisons.

Ok, on my 2x2.66GHz cheesegrater Mac Pro (still a pretty decent computer), at 44.1 and 64 samples, I can run 24 instances of Space Designer (quite good for a 10 year old computer). Someone with a 16 core new Mac Pro managed 52 at those settings.

And now fasten your seatbelts: The baseline Macbook Air (yeah, with just 8GB of RAM, even if that shouldn't matter much for this test) pulls of 80 (in words: EIGHTY!) instances! A fanless Starbucks gadget is running circles around a 12k machine in terms of single core performance!

Holy f**k

[zvenx]

From Logic:

The Playback & Live Tracks setting distributes the load to different threads in cases where a Track Stack that contains multiple software instrument channel strips is selected, or when more than one audio track is recording. If the computer has enough cores and there are no other existing DSP demands on the computer, each live channel strip can be processed by a different core.

so what I understand it all to mean is for prerecorded things it can multicore even if it is one track (my screenshots show that).
and if it is live track in cubase and logic if it is only one track, it will probably use only one core.

The way I understand this it doesn’t explain why it’s using multiple cores in your experiment. It says: “multiple software instrument channel strips”. Now I’m really curious to know how it works.

I use Cubase not logic and I explained in the previous page.
rsp

[Sascha Franck]

And as this hasn't been posted here yet:
https://www.pro-tools-expert.com/produc ... -logic-pro

No matter how you put it, no matter your gripes with Apple (check, for me at least...), these new machines are something to lust for in case you're into audio work on computers.

[zvenx]

I would argue that for a single track with a serial chain jumping between threads actually makes things slower...

You can argue as much as you want.
Clearly Cubendo is using multi cores on single track.


Same Test in Studio One Pro 4.6

Repro1 then All

rsp

[david.beholder]

Still, serial signal flow is serial signal flow.

Serial signal flow doesn't imply single threaded or single core processing.
Btw it's serial only on summing say if you have master. Lot of people have multichannel analog summing so it's parallel.

[spacepluk]

I don’t have a Cubase license anymore to do this test. But if I’m wrong, disabling ASIO Guard should at least saturate one core. If I’m right and the meters are misleading you should see aprox 1/4 cpu usage.