Audio interface,yes or no.

[Bikerden]

New to electronic drums and just bought SD3 which i've not got setup properly yet as still learning about it.
Anyhow is it better to use a separate audio interface or just stick with the USB cable.
Whats the pro's and con's?.

[Nick74]

Latency resp. performance (especially on heavy load) + audio quality will be good with a decent audio interface. ...But don't mistake cheap audio interface headphones jacks (which propably won't be load enough nor provides good audio quality) for a decent phone amp.

[haluksarisel]

Look at the link below about the latency:

https://www.vdrums.com/forum/advance...90#post1262690

[perceval]

Will need a lot more info about your setup.

In general, if you have a Mac, you can start with the headphones out of your Mac, you will need to spend good money to better that.
Windows usually will benefit from an audio interface. Again, depending on your setup and needs, they vary from $150 to over $2,000 ...

[jpsquared482]

Will need a lot more info about your setup.

In general, if you have a Mac, you can start with the headphones out of your Mac, you will need to spend good money to better that.
Windows usually will benefit from an audio interface. Again, depending on your setup and needs, they vary from $150 to over $2,000 ...

Does anyone know what the RTL is for a new Macbook Pro with an M1 running SD3? I heard there's no need for an interface b/c the latency is non-existent (as in less than 5 msec?) with the new Apple Silicon chips.

[perceval]

Does anyone know what the RTL is for a new Macbook Pro with an M1 running SD3? I heard there's no need for an interface b/c the latency is non-existent (as in less than 5 msec?) with the new Apple Silicon chips.

Check out Ribot's thread on the new M1. Looks very nice. Short launch for SD3, minimal latency.
I also have the M1 MBA, 16G, 512HD. Didn't get the chance to try it out yet with SD3 yet.

[Nick74]

Does anyone know what the RTL is for a new Macbook Pro with an M1 running SD3? I heard there's no need for an interface b/c the latency is non-existent (as in less than 5 msec?) with the new Apple Silicon chips.

Audio RTL is not relevant for e-drummers because of not having any audio input latency. e-Drum RTL: (from hit to sound) will also depend on the trigger device latency.

Just for comparison: you can get down to 1.5ms audio output using a Babyface Pro FS and this will be stable even at super heavy load. You'll propably have 3-5ms using the onboard sound device. The M1 chip hasn't changed anything in terms of core audio latency.

[jpsquared482]

The M1 chip hasn't changed anything in terms of core audio latency.

Thanks, this is precisely the bit I was looking for - whether the M1 chip did improve core audio latency.

Further, do we know if the D/A converters of the new M1 machines have improved, imparting lower latency to the new M1 platform (if that even makes sense; I'm a lay person in this area.)

As perceval said, Ribot 's experience is that latency was greatly improved with the M1 machines. I'm wondering whether that's real and measured by someone, and why it seems to be the case.

Thanks!
JP

PS EDIT: For the lay person like myself, this may be useful for others' understanding of Core Audio.
https://developer.apple.com/library/...CoreAudio.html

[perceval]

Ribot is usually serious about his tests, so I am inclined to trust him. He was previously using a RME PCI version audio card I believe and it was very low latency. So, if he feels the latency of the M1 is quite good, we actually have something to weight in against.

[Nick74]

As perceval said, Ribot 's experience is that latency was greatly improved with the M1 machines. I'm wondering whether that's real and measured by someone, and why it seems to be the case.

Of course it was not latency but performance (vs his old outdated and pretty weak Macbook) - so due to the new chip you'll be free of glitches when using lowest buffer settings. I've seen his MBAir M1 onboard latency at lowest buffersize when I visited him: it was not so sexy (more than 5ms audio out at 48kHz). Then with his Digiface we got 1.5 ms.

[jpsquared482]

Ribot is usually serious about his tests, so I am inclined to trust him. He was previously using a RME PCI version audio card I believe and it was very low latency. So, if he feels the latency of the M1 is quite good, we actually have something to weight in against.

Thanks, and at the same time Nick74 's point, "The M1 chip hasn't changed anything in terms of core audio latency," may also be true, in that Core Audio itself (the OS workings) haven't improved latency, but other elements of the overall new Mac architecture and hardware are what's at work in making the improvement, perhaps. I just bought an M1 and will be setting it up with SD3 after I make the upgrade. I'll be happy to report my user findings.

[perceval]

I just bought an M1 and will be setting it up with SD3 after I make the upgrade. I'll be happy to report my user findings.

Thanks!

[Anyware]

More on the M1 performance. SDXs load much faster:https://www.vdrums.com/forum/advance...oad-time-tests

[perceval]

Thanks too!

[jpsquared482]

Hi,
I thought I would revive this thread.

Firstly, I never did get around to testing my M1's RTL performance with my MOTU M4, but I intend to do so as i look at newer Audio Interface options to get my RTL to the lowest possible without breaking the bank. After some research, these seem to be the current best bang for the buck AI's for under $1000.

• RME Babyface Pro. It seems possible to get RTL in the range of 2-3 msec. It seems to set the standard and yet is still pricey at around $900 street price.
• MOTU UltraLite-mk5 is a very close second. MOTU claims 2.4 msec. for $600.

MOTU's claim on their web site is: "Optimized drivers for both Mac and PC deliver ultra-low latency performance. With a high-performance DAW such as Digital Performer, the UltraLite-mk5 delivers an astonishing round trip latency (RTL) of 2.4 ms (milliseconds) over USB (at 96 kHz with a 32-sample host buffer)."

A key question I have is: Does the standalone version of EZD3 and/or SD3 allow for this same performance level? With EZD3 I can select 96kHz and a 32-sample host buffer, so this seems to be the case.

And: Does anyone have experience with this MOTU?It seems very competitive vs the RME for less $ and more I/O ports.

Thanks,
JP