Audio Interface for Mainstage - need answer fast!

[ElmerJFudd]

In theory a proprietary driver has the potential to offer better performance and features than a class compliant driver (which needs to support a wide range of devices from many hardware developers.  That said, there are always unforeseen circumstances and bugs to fix regardless of which route one goes.  If an interface sells well there is money available and interest in continuing to support it. 

 

SSL2+ uses USB 2 Type C connection

On Mac OS it is core audio class compliant

For Windows SSL provides an ASIO/WDM driver

 

They say the unit is capable of 2.5ms round trip latency performance.

However, as a user you have to provide a high performance computer, you may need to make tweaks and adjustments to OS settings, and you may need to run your session at 192khz to actually see that claim realized on your setup. 

 

[octa]

Motu and especially RME are known for their drivers, with extra low latency mentioned often as an advantage. I like to hang out at 32 buffer size (96k) and it's a rare occasion to have to bump it to 64 or 128. In home studio, I use a UFX II by RME and since it's almost 2x the price of the MacBook Air M1 it interfaces with, the UFX II had better sound better than the 1/8" headphone out! (and it does)

 

Going to hook up this M4 + MacBook tonight to see how things sound.

[octa]

But, I must add: In my tests, using an RME audio interface with MacBook offers more sound improvement than using the same audio interface with iOS device. I've compared 1/8" out on my iPad Pro 2nd gen to original RME Babyface interfaced with iPhone/iPad and the sound is quite similar. When I occasionally hook up the MacBook to same setup, there is a noticeable improvement. (crispness, detail, more solid sounding). So, iPad users, maybe headphone jack is the way. Audiophiles, maybe MacBook + interface is the way. 

[ElmerJFudd]

Note: there is a difference between real time midi in to audio out (as you would use for feeling good when playing live) to mixing large projects with many plugins.  One requires low buffer settings, the other does not.  The difference is less obvious these days, but anyone still using a previous generation CPU is still very much aware that their computer chokes on a 32 buffer setting when demands are too high.  It is amazing to me that in just a few generations of CPU development that laptops have reduced audio in to audio out (and by association midi in to audio out) significantly enough to use as an FX rack in live mixing.  

[ElmerJFudd]

24 minutes ago, octa said:

But, I must add: In my tests, using an RME audio interface with MacBook offers more sound improvement than using the same audio interface with iOS device. I've compared 1/8" out on my iPad Pro 2nd gen to original RME Babyface interfaced with iPhone/iPad and the sound is quite similar. When I occasionally hook up the MacBook to same setup, there is a noticeable improvement. (crispness, detail, more solid sounding). So, iPad users, maybe headphone jack is the way. Audiophiles, maybe MacBook + interface is the way. 

A few possibilities there on why sound and perhaps feel are different when running the same software on different devices…

 

how the sound is generated (synthesis, modeling,  or sample playback), what bit depth and sample rate the engine is running at on the device, the quality of the DA and headphone amp on the device.  Does the CPU throttle for battery life or is real time audio performance prioritized. 
 

Also has the software developer actually offered exactly the same software on both iOS and macOS or is the iOS version a lite variation - smaller sample library for example, less velocity layers, more sample stretching, shorter length samples, things like that.  Or in order to function well on many iPads (older and newer) does the engine default to a lower sample rate, lower polyphony, higher buffer?  Has the developer offered preferences to the user to change those settings on iOS?  

[Reezekeys]

I guess I'm lucky - I'm mostly doing gigs with my laptop where, until I got my new MacBook Air, a 128 buffer was the best I could do - and it felt perfectly OK to me. With my new computer I can easily go to a lower buffer, but I tried it - and thought I could tell a slight difference, but it was 1) very slight and 2) might have been a kind of confirmation bias, i.e., I was expecting to hear a difference (and wanted to hear it) so my brain told me it was there. In reality, of course there was some improved latency - there had to be, by virtue of the smaller buffer - but it wasn't noticeable enough for me to feel like I should go to a lower number. At this point in time, I'm comfortable at 128.

[BluMunk]

Not to disrupt the driver/buffer size discussion (which is fascinating), but I thought I'd just update a little more on what I'm doing in the moment:

 

My sound guy's Focusrite would not maintain a USB connection with the mac- not totally sure why and didn't have time to do too much troubleshooting. So, fortunately the keyboard has a USB midi out, so we're using that to get MIDI into the mac and taking the headphone out back to the board.  Working great.

[ElmerJFudd]

1 hour ago, Reezekeys said:

I guess I'm lucky - I'm mostly doing gigs with my laptop where, until I got my new MacBook Air, a 128 buffer was the best I could do - and it felt perfectly OK to me. With my new computer I can easily go to a lower buffer, but I tried it - and thought I could tell a slight difference, but it was 1) very slight and 2) might have been a kind of confirmation bias, i.e., I was expecting to hear a difference (and wanted to hear it) so my brain told me it was there. In reality, of course there was some improved latency - there had to be, by virtue of the smaller buffer - but it wasn't noticeable enough for me to feel like I should go to a lower number. At this point in time, I'm comfortable at 128.

I think you do piano type gigs, solo and in small combos, right?  This is where you’d notice it most of all.  Not as much on the big stage where latency varies due to distance, where your own monitor is, or using in ears from FOH, etc.  

 

Lower latency, of course feels better for acoustic instruments, like the piano, where we are used to, and expect to get sound back to our ears quite quickly because we are right on top of the instrument.  If you have a computer that can do a 32 buffer, even with high polyphony and peddling that is going to be extremely tight timing wise (especially if you use headphones or have the speaker right under your proverbial ass).  Even compared to a real piano.  In fact, a real piano might feel more like a 64 or 128 buffer.  

18 minutes ago, BluMunk said:

Not to disrupt the driver/buffer size discussion (which is fascinating), but I thought I'd just update a little more on what I'm doing in the moment:

 

My sound guy's Focusrite would not maintain a USB connection with the mac- not totally sure why and didn't have time to do too much troubleshooting. So, fortunately the keyboard has a USB midi out, so we're using that to get MIDI into the mac and taking the headphone out back to the board.  Working great.

Check the Focusrite website to see which version of the firmware they recommend for your Mac’s operating system.  
 

Yes, your Mac’s audio chipset and the headphone out just work with Logic/MainStage.  It’s a no muss no fuss solution for live playing. 

[Reezekeys]

45 minutes ago, ElmerJFudd said:

I think you do piano type gigs, solo and in small combos, right?  This is where you’d notice it most of all.  Not as much on the big stage where latency varies due to distance, where your own monitor is, or using in ears from FOH, etc.  

 

Lower latency, of course feels better for acoustic instruments, like the piano, where we are used to, and expect to get sound back to our ears quite quickly because we are right on top of the instrument.  If you have a computer that can do a 32 buffer, even with high polyphony and peddling that is going to be extremely tight timing wise (especially if you use headphones or have the speaker right under your proverbial ass).  Even compared to a real piano.  In fact, a real piano might feel more like a 64 or 128 buffer.

 

I look at it like this - there's no perceived latency, minor or barely perceptable latency, and uncomfortable latency. For me it's not about the numbers per se, although I know that for my particular setup, going from 128 to 256 is something I can definitely feel. I still put 256 in my "minor" camp because I've done gigs with that setting and could tell it wasn't my usual 128, but I wasn't uncomfortable.

 

Yes, piano is what I play most of on a gig. My preferred piano is the NI New York as it has very dry and close-miked samples that, with my two speakers directly behind me, gives me the kind of connection I feel when I'm sitting at the real thing (except the sound is coming from in back of me, of course!). I.e., the sound is right there. Maybe that explains why 128 is acceptable to me? I've stopped thinking about it actually.

[Reezekeys]

1 hour ago, BluMunk said:

My sound guy's Focusrite would not maintain a USB connection with the mac

 

That's the exact same issue I had with my MOTU MicroBook IIc. It kinda sucks because this MOTU is a very powerful and full-featured interface - tiny, with built-in DSP (just EQ and dynamics, no verb though), and a decent mic pre. It crapped out once in the middle of an AWB show and that's when I said "no mas." The headphone output has been trouble free and sounds fine, as you found. Good luck figuring out what's up with your interface - I never did with mine, even with a lot of back & forth with MOTU. What I got was pretty much the boilerplate response to reinstall drivers, my Mac's OS, etc.

[ElmerJFudd]

33 minutes ago, Reezekeys said:

 

That's the exact same issue I had with my MOTU MicroBook IIc. It kinda sucks because this MOTU is a very powerful and full-featured interface - tiny, with built-in DSP (just EQ and dynamics, no verb though), and a decent mic pre. It crapped out once in the middle of an AWB show and that's when I said "no mas." The headphone output has been trouble free and sounds fine, as you found. Good luck figuring out what's up with your interface - I never did with mine, even with a lot of back & forth with MOTU. What I got was pretty much the boilerplate response to reinstall drivers, my Mac's OS, etc.

I’m always suspicious of the “printer port” on the back of the usb interfaces.  How many plug/unplug cycles are they good for?  A lot of newer interfaces have ditched it for USB-C which isn’t a pro quality connector either - not like 1/4”, XLR, BNC, speakON, or ethernet for that matter.   🤷‍♂️ 

[Reezekeys]

As I've mentioned a few times here, the USB-B jack on my first A800 fell apart from the cable flexing as I played on my little x-stand - the innards of the jack came loose. I replaced the jack a few times. My newer A800's jack holds the cable more tightly and it hasn't been an issue (yet). I agree, USB-B connections aren't built for repeated connects/disconnects.