BenIfin
Shredder
- Messages
- 1,505
Since I'm always using the iPad (for the awesome Loopy Pro!), I wanted to measure my latencies in the 'real world' too. So, I shelled out some cash and bought the Latencymeter app to measure the physical latency with a loopback cable.
And then the surprise: With IOS 26 there is a significant improvement!
Here's the table with the comparison values - it's noticeable that my ancient Focusrite (I think it's 2nd Generation) performs way better than the Zoom I bought, because it is really compact for 4in/4out. And to top it all off, I also measured my Helix LT. And what we all know: As an interface, it's really not great...
Code:iPad-OS | Interface | Buffer | 48 kHz | 88,2 kHz | 96 kHz --------+---------------+--------+--------+----------+-------- 18.6.2 | ZOOM AMS-44 | 34 | 7,58 | 6,55 | 6,39 18.6.2 | ZOOM AMS-44 | 66 | 8,96 | 7,30 | 7,05 18.6.2 | ZOOM AMS-44 | 130 | 11,63 | 8,75 | 8,39 26.0 | ZOOM AMS-44 | 34 | 6,48 | 5,33 | 5,17 26.0 | ZOOM AMS-44 | 66 | 7,81 | 6,02 | 5,83 26.0 | ZOOM AMS-44 | 130 | 10,44 | 7,47 | 7,18 26.0 | HELIX Audio | 34 | 8,31 | – | – 26.0 | HELIX Audio | 66 | 9,63 | – | – 26.0 | HELIX Audio | 130 | 12,31 | – | – 26.0 | Scarlett18i8 | 34 | 5,54 | 4,01 | 3,94 26.0 | Scarlett18i8 | 66 | 6,88 | 4,73 | 4,60 26.0 | Scarlett18i8 | 130 | 9,54 | 6,20 | 6,05
Thanks for this - very good info ! :)
It seems to support my "hypothesis" [ <- wanker ] that the CoreAudio Subsystem simply cannot process RTL Audio at anywhere near as well written PC ASIO Drivers or Hardware modelers.
It just seems that no matter how powerful your iPad is and how great your Audio Interface is, it just cant get to the 1.5ms <-> 2.5ms RTL's you get with Top Tier modelers and well written [ie: RME] ASIO PC drivers.
And now Windows ARM is getting its own native ultra-low latency subsystem like CoreAudio, the differences could/will get even greater.