iPad upgrade ?

Heyez

Ipad Mini2 isn't looking like something I wanna use for BM3..

Hitting cpu limit with glitches/pops etc.

Anyone know what kind of performance improvement could I expect if I switched up to the current 9.7" ipad? The 128gb model seems like a pretty good deal atm..

In the current project 'song' I'd estimate the polyphony is peaking at about 16 voices (approx 8 of these are quite long samples, and played polyphonically with timestrech on). Probably about 15 modulations happening. And about 15 fx.. Getting quite a lot of pops/glitches...

Obvs can't put a definite number on these things but just wanna make sure there's going to be a big increase all round before I fork for a new ipad..

Thanks

philowerx

My old iPad Air 1 has the same A7 processor as your Mini2. Loving my new 2017 9.7". The bump up from 1gb RAM 2gb and the processor increase are quite noticeable especially with some of the more resource hungry synth apps. I wanted a 9.7" Pro for the 4 speakers but couldn't justify the extra several hundred dollars for a meager processor increase over the plain Jane 2017 9.7".

slaq

In 2017 I would recommend 4 gigs of ram and the future is yours!

Tomes

Have you tried using 16bit sound, 44.1k and highest buffer size that still gives you useable latency?

Audiogus

I have the latest 9.7 inch 128gig ipad. I think it is a fantastic deal and has been working great with BM3, lots of stuff going on with all 8 banks and a bunch of fx. No glitches or stutters. It was a MASSIVE jump over the iPad Air and Mini2 I had been running. Hopping between other apps running at the same time is glorious.

Heyez

Thanks guys. Sounds like the upgrade would make a big difference. Probably cop the 2017 9.7 in couple weeks.

Tomes

My air2 cant always handle enough without turning off fxs and stuff to save cpu. And the new basic ipad has similar horsepower. Im hoping to upgrade to new pro

Heyez

@tomes Ugh, that's a bit of a worry then.. Not really down for copping a Pro.

Would be cool if someone did some kind of stress test shoot out across ipads so people knew what they could expect from each model. Like it's cool that people are sharing this info, but without knowing how many tracks/voices/timestrech/fx/modulation/plugins etc it's too abstract and just guesswork to try and estimate how far each ipad model will stretch.

william77

Well, I have a iPad Pro 12.9 with 128gig.

Don't really know if this is much load or not - but the last session I was running:

- two Sampler Banks,
- one instance of SampleTank,
- one instance of Odyssei
- and one instance of sunrizer
- (and couple of effects/filters in the mixer)

• simultaneously in realtime (midi)
  It started to choke a bit - but when changed it to 512 samples buffer - it was all good again.

then I added a reverb on the aux and it started to choke a bit - but was still running ok.

Heyez

@william77 thanks for that detailed info

Any timestrech/polyphony going on with any pads in the sampler banks?

I'm guessing Odyssei is pretty heavy on CPU if anything like ims20 (that used to always flake my Mini2 out if I used it with other stuff running..).

Soon as they add 'freeze track' for plugins/'flaking out' situations etc I guess none of this will really be too much of an issue. Probably grab the 2017 9.7" 128gb and hope for the best

william77

no polyphony on the sampler pads (one at time) and no real timestrech.
(as long as "tuning" samples don't count as such - then yes)

I didn't yet even get to any timestretching features - too much stuff in this DAW
so much stuff to still discover (which is good I guess )

Heyez

@william77 yeah loads of different ways to get creative in BM3 been messing with Timestreching + polyphony + layers + modulation for weird lofi drones and pads from sounds around the house recorded in to ipad mic. Very fun/cool app

Tomes

Quote doesent work, so ill just reply without. How much you can get out of ipad with bm3 depends so much on settings and audio interface used, so this needs to be taken in account when doing some cpu stress comparisons.

Heyez

@tomes, yeah I meant someone using the same project/interface etc on different ipads Deleting elements from the project til each ipad ran smoothly or some other similar method. Doubt many people (anyone...) has a bunch of ipads and the inclination/time tho

winconway

Testing for CPU/RAM with IAA and AUv3 is a false economy, any tests should be done without them, then if you wish to test IAA or AUv3, they needed to be tested individually, one IAA alone could be using 90% of the CPU/RAM.