1* Ipad Pro 10,5 vs. 2* Ipad 2017 9,7

supadom

@ElektrikDiva said:
Yes I did Gus but what does taping that star do? Thank you very much for enlightening me, I never would have figured it out. Cheers, Eletrik Diva

You've made it your favourite,that is why you are getting notified.

mrcanister

i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

AudioGus

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

Right I see. Being a nonmusician with a liberal avant gard glitch fetish it is hard for me to fathom the difference between 6ms and 11ms being relevant but I suppose it could be. I guess some folks like 96000hz because that halves their latency then... ?

mrcanister

yeah, 96000 must be the case, presumably. tho you have exhausted my knowledge on the subject:) i normally have mine set to 256 but i can't say i notice any diff at 512 when i play the keys n pads. probably set it even higher if you're not inputting anything live.

supadom

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tfho.
a buffer size at 512/44100=11ms
256/44100=6ms

I don’t think you can always trust the numbers reported by apps or the system. It happened to me to have unplayable latencies even at 128 setting.

Anyway, watching some of my older videos I’ve noticed that the guitar is strangely behind everything else. I used to use 256 but now I refuse to go above 128 and whould rather use less apps than compromise on latency.

RUST( i )K

@u0421793 said:
For that budget, you could probably accumulate about fifty first-edition version one iPad 1 models.

I actually enjoy your snarky silly comments for some reason............

RUST( i )K

I would not be able to function without 2 iPads. & the iPhone. (Audiobus remote)

RUST( i )K

And here is the view out the window for all you people enjoying winter...........sorry

CracklePot

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

I have always interpreted it that way too. Frames = samples.
So at 48 k sample rate, 256 frames is shorter than at 44.1k.> @AudioGus said:

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

Right I see. Being a nonmusician with a liberal avant gard glitch fetish it is hard for me to fathom the difference between 6ms and 11ms being relevant but I suppose it could be. I guess some folks like 96000hz because that halves their latency then... ?

It may not make a difference to you. Everyone has a different tolerance to latency. I can get by with 512 sometimes since I got used to it years ago, but to someone younger that might be impossible to keep time with while playing live.

AudioGus

@CracklePot said:

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

I have always interpreted it that way too. Frames = samples.
So at 48 k sample rate, 256 frames is shorter than at 44.1k.> @AudioGus said:

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

Right I see. Being a nonmusician with a liberal avant gard glitch fetish it is hard for me to fathom the difference between 6ms and 11ms being relevant but I suppose it could be. I guess some folks like 96000hz because that halves their latency then... ?

It may not make a difference to you. Everyone has a different tolerance to latency. I can get by with 512 sometimes since I got used to it years ago, but to someone younger that might be impossible to keep time with while playing live.

Heh, I plunk, record, chop and edit in almost the same breath.

RUST( i )K

@CracklePot said:

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

I have always interpreted it that way too. Frames = samples.
So at 48 k sample rate, 256 frames is shorter than at 44.1k.> @AudioGus said:

@mrcanister said:
i think the latency related to buffer size is fixed in time, not related to the power of the device.i could be wrong tho.
a buffer size at 512/44100=11ms
256/44100=6ms

Right I see. Being a nonmusician with a liberal avant gard glitch fetish it is hard for me to fathom the difference between 6ms and 11ms being relevant but I suppose it could be. I guess some folks like 96000hz because that halves their latency then... ?

It may not make a difference to you. Everyone has a different tolerance to latency. I can get by with 512 sometimes since I got used to it years ago, but to someone younger that might be impossible to keep time with while playing live.

Thats why I make sure to only perform for senior citizens.

Latency issues don't exist.

Plus I play the same song over and over and they don' t remember.

ElektrikDiva

Ah thanks Supadom!

Bon_Tempi

@AudioGus said:
@Bon_Tempi

Some good news!

On my 2017 iPad using AUM (and Link) I could easily host and record...

-Patterning (main output)
-Elastic Drums
-Model15 with EOS2 (driven by Fugue Machine)
-Syndt (driven by Fugue Machine)
-Shoom with FAC Chorus
-Samplr
-iDensity

Buffer had to be at 512 with this much running (256 was major glitchy) and it would peak at around 80% CPU. I noticed no significant latency and the playback was super solid, no major glitches or sync errors while switching apps etc and it seemed fine for live that could also be recorded to separate tracks on the fly. Recorded for a good five minutes while jamming, no issues, aside from two very very minor audio hiccups that did not even record. Whoot!

To get closer to the original goal I then set up Audiobus running in parallel to AUM for sending DRC into LoopyHD. I then sent LoopyHD to AUM. Doing this caused more audio glitches that were a bit worrisome (not good for live), but again did not show up at all in the track recordings.

So really if you wanted to spread the love out across two 2017 iPads, both running AUM, I say it would clearly handle the load, and then some. One could be your 512 high buffer power lifter sequencer and the other the low buffer midi ninja.

Dear AudiGus -

wow - thanks a lot for this good news and also for the effort and time which you have put into this - I deeply appreciate this ...

and I think it is really good news that the Ipad 2017 is that capable in real life performance ...

and about your advice:

So really if you wanted to spread the love out across two 2017 iPads, both running AUM, I say it would clearly handle the load, and then some. One could be your 512 high buffer power lifter sequencer and the other the low buffer midi ninja.

are there some relevant informations to take care of in such a setup - whill the syncrnoization with LINK be a stable and reliable way to do it - is there any alternative to Link? is there a way to sync apps which are not Link-enabled?

but again - thanks a lot for your very helpfull test ...!!!

Bon_Tempi

@RUST( i )K said:
I would not be able to function without 2 iPads. & the iPhone. (Audiobus remote)

Dear Rustik -

may I ask how you syn you sync the two ipads and how your exoerience is there - is the syncing reliable/stable/easy to handle?

Bon_Tempi

@Zen210507 said:

@Bon_Tempi said:
can you describe your way of using the 10,5 Pro? how many apps at the same time? which apps? recording at the same time? and at which buffer size?

>

How long is a piece of string.

We, as in IOS studio band RTM, work with iPad Pros. We use apps such as Blocs Wave, Auria Pro (with filters and FX on most tracks), Sample Tank, BeatHawk, TwistedWave, AudioBus and AUM both with whatever apps are required for making a particular track. On individual iPads there might be one, or half a dozen apps open at a time, with additional in-app AU and/or IAA FX. Files then fly around via Dropbox. So far, we have not had any issues doing this stuff. You can hear the results via our BandCamp page, and regardless of whether it’s your kind of music, you will hopefully be able to hear that polished results are possible.

Many others here can provide other examples, which may help. Among those others, of course, there will be entirely different workflows. Many will be using a lot more actual tracks than we do, and working much more in midi. Obviously, such people are infinitely better qualified to tell you what limits they have found than I.

You could also look up last year’s Doug Woods benefit album, link below, featuring lots of music makers from these parts. Just find who is doing stuff in the same area you like, and ask them how they do this work.

thanks for these nice examples - and the explanation ...

Bon_Tempi

@supadom said:
I'd be quite interested in finding out a practical way of using 2 iPads and how they can be tempo synced. Link is great but it only bloody works via network.

Dear Supadom - can you share some of your expperiences with LINK and such a kind of set-up - you don´t sound very happy and/or content with it up to now?

and have you been reading the result of AudioGus´stresstest ...

@AudioGus said:
On my 2017 iPad using AUM (and Link) I could easily host and record...

-Patterning (main output)
-Elastic Drums
-Model15 with EOS2 (driven by Fugue Machine)
-Syndt (driven by Fugue Machine)
-Shoom with FAC Chorus
-Samplr
-iDensity

Buffer had to be at 512 with this much running (256 was major glitchy) and it would peak at around 80% CPU. I noticed no significant latency and the playback was super solid, no major glitches or sync errors while switching apps etc and it seemed fine for live that could also be recorded to separate tracks on the fly. Recorded for a good five minutes while jamming, no issues, aside from two very very minor audio hiccups that did not even record. Whoot!

To get closer to the original goal I then set up Audiobus running in parallel to AUM for sending DRC into LoopyHD. I then sent LoopyHD to AUM. Doing this caused more audio glitches that were a bit worrisome (not good for live), but again did not show up at all in the track recordings.

So really if you wanted to spread the love out across two 2017 iPads, both running AUM, I say it would clearly handle the load, and then some. One could be your 512 high buffer power lifter sequencer and the other the low buffer midi ninja.

by your experience with the IpadPro 10,5 - do you think it would be able to deal with such a set-up on the lower buffer sizes (256/128)? and would it be possible to run it all on just a single 10,5 Pro - also the additional Loopy-action and the low latency midi playing ... without needing to source this out to a 2nd device?

supadom

@Bon_Tempi my problem with Link is that it doesn't support stop /restart for all connected apps which is something I really need. It seems this feature is coming in the next version apparently.

I can't say for sure but I've heard somewhere that link is piggy backing on IAA sync which in some apps is still not the tightest.

studs1966

@RUST( i )K said:

And here is the view out the window for all you people enjoying winter...........sorry

No probs matey. I see that everyday here in Cyprus. At the mo it's a Bonechilling 17C here, &'there is snow on the Troodoos Mountains.

Bon_Tempi

@supadom said:
@Bon_Tempi my problem with Link is that it doesn't support stop /restart for all connected apps which is something I really need. It seems this feature is coming in the next version apparently.

I can't say for sure but I've heard somewhere that link is piggy backing on IAA sync which in some apps is still not the tightest.

"piggy backing" means? - I am not a native speaker - so I cant understand this expression - sorry ...

and what about your thoughts about the ipad-stresstest in comparison to your experiences with the Ipad Pro 10,5?

BlueGreenSpiral

Link is flawless for me. I'm 99% sure the audiobus team helped Ableton out with development?

There are at least 2 apps out there that transform midi clock to Link and vice-versa.

Yesterday I had AB3 running triqtraq and troublemaker into aum with cpu at around 60%. This was on an 8gb iPhone 5c with less than 100mb free space. Didn't check the latency setting but it was very responsive and no crackles.

Bon_Tempi

@BlueGreenSpiral said:
Link is flawless for me. I'm 99% sure the audiobus team helped Ableton out with development?

There are at least 2 apps out there that transform midi clock to Link and vice-versa.

Yesterday I had AB3 running triqtraq and troublemaker into aum with cpu at around 60%. This was on an 8gb iPhone 5c with less than 100mb free space. Didn't check the latency setting but it was very responsive and no crackles.

thanks for the tip - this helps ---

Bon_Tempi

@soundshaper said:
I’m on a new 2017 iPad Pro 10.5 and while I can agree that it’s fast and runs Model 15 beautifully, I can’t seem to get the latency down to 128 or lower without experiencing crackles and CPU spikes when the UI/graphics move. It will handle a lot of instruments and effects in AB or AUM at 256 though. Not a huge deal for synths, but not usable for live drumming.

I’ve heard others say the iPad Air 2 is performing just as well as mine or slightly better at low latencies. I was expecting to be able to go as low as 64, which is why I put down the dough for the latest Pro.

Dear Soundshaper - since you own the 10,5 Pro .. - how would you compare it´s performance to a set-up which got tested by "AudioGus" with his 2017 9,7 model ...

@AudioGus said:

@Bon_Tempi

Some good news!

On my 2017 iPad using AUM (and Link) I could easily host and record...

-Patterning (main output)
-Elastic Drums
-Model15 with EOS2 (driven by Fugue Machine)
-Syndt (driven by Fugue Machine)
-Shoom with FAC Chorus
-Samplr
-iDensity

Buffer had to be at 512 with this much running (256 was major glitchy) and it would peak at around 80% CPU. I noticed no significant latency and the playback was super solid, no major glitches or sync errors while switching apps etc and it seemed fine for live that could also be recorded to separate tracks on the fly. Recorded for a good five minutes while jamming, no issues, aside from two very very minor audio hiccups that did not even record. Whoot!

To get closer to the original goal I then set up Audiobus running in parallel to AUM for sending DRC into LoopyHD. I then sent LoopyHD to AUM. Doing this caused more audio glitches that were a bit worrisome (not good for live), but again did not show up at all in the track recordings.

So really if you wanted to spread the love out across two 2017 iPads, both running AUM, I say it would clearly handle the load, and then some. One could be your 512 high buffer power lifter sequencer and the other the low buffer midi ninja.

RUST( i )K

@Bon_Tempi said:

@RUST( i )K said:
I would not be able to function without 2 iPads. & the iPhone. (Audiobus remote)

Dear Rustik -

may I ask how you syn you sync the two ipads and how your exoerience is there - is the syncing reliable/stable/easy to handle?

Here is how I have had best luck with LINK and 2 iPads / 1 iPhone.

I FIRST OPEN Coding Cod's MIDI LINK sync (free app) on secondary iPAD device.

I find that this independent first start LINK source is super dependable for keeping things that way.

Yes reliable.

I am actually syncing the Circuit with this Coding Cod app.

It is a great tool for this.

It is the most reliable of "Link sync" apps I have tried.

Hope that helps.

Bon_Tempi

@RUST( i )K said:

@Bon_Tempi said:

@RUST( i )K said:
I would not be able to function without 2 iPads. & the iPhone. (Audiobus remote)

Dear Rustik -

may I ask how you syn you sync the two ipads and how your exoerience is there - is the syncing reliable/stable/easy to handle?

Here is how I have had best luck with LINK and 2 iPads / 1 iPhone.

I FIRST OPEN Coding Cod's MIDI LINK sync (free app) on secondary iPAD device.

I find that this independent first start LINK source is super dependable for keeping things that way.

Yes reliable.

I am actually syncing the Circuit with this Coding Cod app.

It is a great tool for this.

It is the most reliable of "Link sync" apps I have tried.

Hope that helps.

yes - thanks, that helps a lot - and may I ask whick models of the Ipad you are using?

RUST( i )K

@Bon_Tempi said:

@RUST( i )K said:

@Bon_Tempi said:

@RUST( i )K said:
I would not be able to function without 2 iPads. & the iPhone. (Audiobus remote)

Dear Rustik -

may I ask how you syn you sync the two ipads and how your exoerience is there - is the syncing reliable/stable/easy to handle?

Here is how I have had best luck with LINK and 2 iPads / 1 iPhone.

I FIRST OPEN Coding Cod's MIDI LINK sync (free app) on secondary iPAD device.

I find that this independent first start LINK source is super dependable for keeping things that way.

Yes reliable.

I am actually syncing the Circuit with this Coding Cod app.

It is a great tool for this.

It is the most reliable of "Link sync" apps I have tried.

Hope that helps.

yes - thanks, that helps a lot - and may I ask whick models of the Ipad you are using?

Air2 on the left
Air1 on the right
iPhone7+ not seen

BlueGreenSpiral

Just to add that I checked and Midi Link Sync is also the one I use and it does exactly what it's supposed to..

Bon_Tempi

@RUST( i )K said:

@Bon_Tempi said:

yes - thanks, that helps a lot - and may I ask whick models of the Ipad you are using?

Air2 on the left
Air1 on the right
iPhone7+ not seen

and may I ask - since you own an ipad air 2 - in relation to the stresstest AudioGus was setting up for his Ipad 2017 9,7 - would something like this be possible with an Air 2 as well performance-wise?

Bon_Tempi

@BlueGreenSpiral said:
Just to add that I checked and Midi Link Sync is also the one I use and it does exactly what it's supposed to..

thanks )

RUST( i )K

@Bon_Tempi said:

@RUST( i )K said:

@Bon_Tempi said:

yes - thanks, that helps a lot - and may I ask whick models of the Ipad you are using?

Air2 on the left
Air1 on the right
iPhone7+ not seen

and may I ask - since you own an ipad air 2 - in relation to the stresstest AudioGus was setting up for his Ipad 2017 9,7 - would something like this be possible with an Air 2 as well performance-wise?

I guess I need to know what you are talking about so I better go read....LOL

Bon_Tempi

@RUST( i )K said:
I guess I need to know what you are talking about so I better go read....LOL

ah - ok - sorry ) -

it´s this stresstest ...

@AudioGus said:
@Bon_Tempi

Some good news!

On my 2017 iPad using AUM (and Link) I could easily host and record...

-Patterning (main output)
-Elastic Drums
-Model15 with EOS2 (driven by Fugue Machine)
-Syndt (driven by Fugue Machine)
-Shoom with FAC Chorus
-Samplr
-iDensity

Buffer had to be at 512 with this much running (256 was major glitchy) and it would peak at around 80% CPU. I noticed no significant latency and the playback was super solid, no major glitches or sync errors while switching apps etc and it seemed fine for live that could also be recorded to separate tracks on the fly. Recorded for a good five minutes while jamming, no issues, aside from two very very minor audio hiccups that did not even record. Whoot!

To get closer to the original goal I then set up Audiobus running in parallel to AUM for sending DRC into LoopyHD. I then sent LoopyHD to AUM. Doing this caused more audio glitches that were a bit worrisome (not good for live), but again did not show up at all in the track recordings.

So really if you wanted to spread the love out across two 2017 iPads, both running AUM, I say it would clearly handle the load, and then some. One could be your 512 high buffer power lifter sequencer and the other the low buffer midi ninja.