Audiobus related bug, recording long samples crashes Nanostudio

27 replies [Last post]
Alistair
Offline
Joined: 04/11/2015

So I was trying to record some cello samples to the TRG through Audiobus and suddenly Nanostudio crashed.

I ran some tests and found out that the time it takes to make NS crash differs from app to app.

Recording Thumbjam samples makes Nanostudio crash in about 45 seconds. Weirdly, with Animoog it takes exactly 1 minute and 45 seconds.

Since recording samples is apparently the only way to record Audiobus tracks (which frankly is a real pain) I hope that this gets patched.

My device is a 4th gen iPad. iOS version is 8.2.

ElectricWombat
Offline
Joined: 01/16/2015

@Alistair, as you have mentioned above; it is known that NanoStudio DOES NOT record audio tracks (and never has), however there is a 'workaround' (meaning a way to subvert this but not an intended part of its functionality) whereby you can record relatively long audio samples (for a drum machine) up to around a minute (or even longer) directly into the TRG-16.

This means that you can use this technique to create 'pseudo audio tracks'. However as this is a 'workaround' it is not a replacement for audio track recording; and as such it has its limits (around a minute) but if you're clever, this is also easy to subvert simply by breaking the 'pseudo audio track' down into appropriate sections; that is multiple recordings (that break at logical gaps in the arrangement).

You just have to keep in mind that the ability to create drum (or FX) samples around a minute long (or longer subject to platform) is pretty excessive for a drum machine, so this is a useful additional feature to have. However if you approach it like you are, all you are doing is simply filling up the sample buffer until it runs out and breaks; essentially eliciting a crash as its NOT designed to be used like this.

Thus, just simply break your arrangement down into logical performance parts (scenarios or sections) with a target of around a minute long (choosing the most appropriate place to break based upon its melodic structure and progression), record them, assign them to pads in the TRG-16, then trigger them appropriately to recreate your intended performance progression.

NOTE: This is unlikely to be 'patched' as it isn't an error as such - you were just abusing a workaround with the approach you were taking. However I agree it would be more graceful if it simply stopped capturing (instead); but I think you just misunderstood its use and I have provided you with a more suitable workflow above.

Alistair
Offline
Joined: 04/11/2015

Thank you for your reply. Well, you sure made me feel like a chump. Oh well...

After I found out this bug, or as you better said it, a limitation, I did break down my composition into smaller segments and joined them together to make my "audio track". It's more work I guess but it's safer.

It would be nice if Nanostudio supported real audio tracks and had proper Audiobus support.
I also understand that Nanostudio won't probably support these in a while, or maybe never. But one can dream, right?

Well, at least I now know not to record samples longer than 30 seconds.

ElectricWombat
Offline
Joined: 01/16/2015

@Alastair, I'm glad you found my reply useful and that you've managed to get your track together (with a little indirect tweaking and 'Nano' magic). I've discovered that IOS music making in general seems to be full of crazy workarounds and secret workflows etc - lol - but it certainly keeps the music making process interesting and quirky!

You have no reason to feel like a chump for trying to approach 'a regular task' in a manner which you'd expect it to work; only to discover it actually needs to be done in a somewhat unusual way. I frequently have my melon twisted by crazy IOS workflows and workarounds etc; so don't be disheartened!

The veterans on this forum have to help me with quirky IOS workflow phenomena frequently (many thanks guys/girls). Fortunately however this is the best forum/community I've encountered, with excellent creative people from all over the world who are always willing to help and share their knowledge, which you became a part of when you chose NS!

NS was released way back around 2011 - one of the earliest DAW' to appear on IOS; of which the landscape has changed dramatically since then. The code base and hardware architecture it was built on has evolved so dramatically, that all leading DAWS (inc. Cubasis and Auria) have essentially been either partially or fully rewritten consequently (Auria 2 released soon, Cubasis 2.0 on the horizon etc) as is the case with NS2.

NS2 is also under development and will undoubtedly fulfill any omissions imposed upon it subject to the limitations of technology when it was originally released. Features have been added since naturally, but major changes to the engine require a complete rebuild for speed, efficiency and to get the most out of the 'new' platform(s). So in time, these needs will be fulfilled!

I'm using NS on an iPhone 4S (IOS V8.3), which is amazing to me considering it's the minimum realistic platform now under IOS 8 (as most apps are optimised for running 64 bit hardware). As such, NS provides many of the key features required while being fun to use; and although there are some omissions/limitations (naturally) there are always workarounds etc! ;)

PS: On the iPhone 4S a TRG-16 sample length of 1 minute is the guide time from what I understand, but bare in mind it's likely the code efficiency of external apps and the 'AudioBus' CPU/memory footprint that detrimentally impacts on this which explains why this changes from app to app. Hope this helps! :p

Stiksi
Offline
Joined: 11/21/2010

45 seconds sounds a bit low. Did you already have a lot of big samples loaded into the project? You can check how much available sample memory you have left by going into the MANAGE / Project section. There's a sample memory meter on the right.

One other thing to keep in mind: make sure you kill all other apps while you do big recordings like this, it's easy to get out-of-RAM crashes on iOS and that has nothing to do with any of the apps, it's a "feature" of iOS.

Blip Interactive
Offline
Joined: 04/05/2010

NanoStudio does have a size limit on individual samples, but in general it shouldn't crash if you exceed it. However, NanoStudio stores its samples in RAM (rather than streaming, as an audio track) and iOS's usual response is to just close an app if it decides it's using too much memory, which to a user (me included) still looks like a crash!

I suspect if you close down any background apps you're not using then the problem will be much reduced.

Blip Interactive
Offline
Joined: 04/05/2010

NanoStudio does have a size limit on individual samples, but in general it shouldn't crash if you exceed it. However, NanoStudio stores its samples in RAM (rather than streaming, as an audio track) and iOS's usual response is to just close an app if it decides it's using too much memory, which to a user (me included) still looks like a crash!

I suspect if you close down any background apps you're not using then the problem will be much reduced.

syrupcore
Offline
Joined: 09/26/2010

You can also record into something like audioshare and then paste it into NS. I think the audio file limit in NS is somewhere around 3:30 but don't quote me.

Also, try turning on airplane mode.

ElectricWombat
Offline
Joined: 01/16/2015

@BlipInteractive, thanks for the information Matt, it's always interesting to hear how things work under the NS hood!

@Stiksi, thanks for your comments too! I must admit I've become a total 'app shutt down' obsessive now (ceremoniously force closing everything when I've finished using it). I just figure it's a simple habit to maintain and should help battery life; whilst making sure what ever I'm running is performing as best as it can on my iPhone 4S etc!

It's interesting what you were both saying about memory issues, as it made me think that a good workflow for @Alastair, would be to minimise the session memory footprint by creating a dummy steamlined project for use whilst capturing (with only bare bones reference sounds in it) to maximise performance/sample time.

If he's trying to capture sections but needs audio cues (like a singer using a reference track) he could just select the minimal session components needed to get his track recorded correctly, then just copy the samples over to his master session for actual placement and arrangement etc.

I'm just thinking aloud really as I play acoustic instruments too and I'd like to give it a go at some stage; but you'd think that would make for a good workflow to maximise iPlatform performance and minimise the CPU/memory footprint etc.

syrupcore
Offline
Joined: 09/26/2010

^ Indeed. I haven't been doing acoustic instruments in NS so much lately but when I did, I found it easier to create a mixdown in NS first (possibly just the relevant section) and then do the audio recording in MultiTrackDAW. Once it was good to go (possibly comping takes or whatever), slice it up and paste it back into NS. I prefer no more than 8 bars of audio because there's no random access for samples (the NS play head has to pass the sample's MIDI trigger note to hear it). When you start dropping in 60+ second samples, you have to rewind quite a ways each time to hear it in context. MTD makes it easy to slice on bars if you set the tempo the same as NS' tempo when recording.

ElectricWombat
Offline
Joined: 01/16/2015

@SyrupCore, hey there I'm glad you mentioned AudioShare, as I was contemplating suggesting that previously (considering it's just been on sale for peanuts over Easter) but I have to admit I'm kinda new to AudioShare myself and I haven't used it for sampling as yet (more just as a sample library and manager really) but I love the IAA FX/sampler/wave player it rocks!

I assume it's possible to get AudioShare to produce a metronome click that isn't recorded? The only thing is though that from my own experience with recording acoustic material, it really helps if you have a reference track that you can play over (to get the feel/groove/intonation right); but I wouldn't think you could set something like that up in AudioShare, as its asking a bit much lol!?

I'm sorry I'm being totally lazy here (trying to) type and play on Xbox at the same time - yes I'm a geek! I will go and have a peep at AudioShare and see what it says about this kinda stuff, as its interesting and it would be great to sample say some acoustic guitar plucks/strums and things sometime! ;)

ElectricWombat
Offline
Joined: 01/16/2015

@SyrupCore, ha ha :) I was trying to keep up with the narrative progression of your posts but you can tell from above that I was falling behind lol! Yeah the workflow you mentioned above sounds really good (bounce down and reference etc), all the audio cues and control you need to get the 'perfect take' with multing/comping etc. It's interesting what you said about the long sample RAM as well because that totally fried my head on one occasion (i.e. trying to experiment on a part of the arrangement between sample triggers - no scrub audio - doh (noob me)! Ha ha :)

PS: Just remembered - AudioShare has the ability to cut/slice on the tempo beat/bar too (if I recall correctly) which is useful once you've got the timing aligned, but again; trying to know what's right (intonation wise) is impossible with only a metronome for reference as the finest ms shift is equivelent to being swung or unswung etc.

PPS: Consequently - is there a simple way to apply 'non permenant' micro timing shifts to samples in the NS arrangement window (like a a sort of latency compensation nudge type thingy)? I'm sure there's a work around if there isn't a direct approach! The comments above just made me think about that in relation to the nature of this discussion etc. I know I just love the way you can zoom into the samples in the NS sample editor! Woa caffeine rush - right I'm going to shut up now. :p

syrupcore
Offline
Joined: 09/26/2010

You can just push the start of the trigger note, no? If you need finer resolution you might be able to send it through a delay, 0 feedback and 100% wet. Or you could put the sample in Eden, presuming mono is ok and play with the note triggee time along with the sample start offset (XY pad->patchbay->sample start).

ElectricWombat
Offline
Joined: 01/16/2015

@SyrupCore, thanks for your comments and suggestions, brilliant as ever - much appreciated - You'll have to excuse me I was pretty wired from lack of sleep and caffeine abuse yesterday (speaking of which) and I was just kinda thinking out loud without the thinking part - lol. :0

However with regard to latency issues when capturing audio it is really useful to be able to make tiny micro adjustments etc (in the ms domain). At the time I totally forgot that you can take quantise off and just slide it into position (yes I'm a donkey - hee haw) though my digits (or hooves) are are pretty clumsy on my tiny iPhone 4S screen too. I guess it's much easier on an iPad with the finer digitiser and mega resolution etc.

Anyway I was inspired by the thread dialogue about acoustic recordings in NanoStudio, that's really nice to see (and definately more obscure/rare)! It's an interesting topic regardless and if you're still following @Alistair - it would be great to hear the track you composed using a Cello, the sophistication of it :p musical diversity is awesome! :)

Slam-Cut
Offline
Joined: 09/07/2011

OK, I'm trying to catch up on this thread. Didn't Alistair say in his original post that he was recording cello "samples" into TRG? And since he mentioned Thumbjam, I assume the samples were from that app. You guys went off in the 'recording acoustic audio' direction when I don't think that was the problem, unless I've missed something. He may have tried to record long passages, but it seems like he just wants a TRG bank of cello single note samples, which can be a great way to get around the lack of audio tracks. Granted a lead melody line will likely need more expressive control and be better handled as longer audio phrases, but I am guessing he was using cello as a voice in the harmony where some fairly plain long samples would work just fine. He didn't really clarify what he was doing with the samples, so maybe I've misread his intentions.

Thinking about it a bit more, if one had access to samples with various expressions (legato through staccato let's say), each on different TRG banks, one could play notes on a TRG track (with an expression/duration right in the middle), and then later adjust the individual notes in the piano roll editor By moving them to other TRG tracks to change the note duration & expression over the course of each phrase. Move notes to different TRG tracks until one is happy with the feel of it, then bounce that phrase down to a single TRG pad, repeating as neccessary.

I'm I making up a solution to a different problem from what the OP addressed?

syrupcore
Offline
Joined: 09/26/2010

Sounds like an overbuilt solution (my speciality, normally). As long as it's not a solo cello part, you can get quite a bit of sample articulation within Eden. With 5 controls (XY, XY, and knob) you can set up attack, release, vibrato, tremolo and whatever else your track calls for. The knob can set the rate for the tremolo/vibrato or filter frequency or reverb send...

Slam-Cut
Offline
Joined: 09/07/2011

My overbuilt solution is probably born out of my inability to manipulate samples. I've never really been able to get much expressive variation out of a single sample Eden preset. I've dabbled a bit, but I never really know what to do with the X/Y pads and knob. Experimentations have created some interesting synth presets, but not so much when it comes to acoustic samples. I'd love to get some of the expensive sample libraries, one day...
Your tips sound like a good place to start, so I'll have to try experimenting with that some day. ...but realistically I'll probably wait until NS2 before I put a lot of time into acoustic presets. NS2 should be worth building a huge preset library for. Multi-sampled Eden presets of acoustic instruments could be amazing!

ElectricWombat
Offline
Joined: 01/16/2015

@Slam-Cut, I have absolutely no idea what 'Alistair' was referring to as he made an extremely vague post about 'Cello' samples and being frustrated about recording long sample lengths? This made me think he was talking about an acoustic setup - but who knows!?

I love it when someone asks a question then dissapears without so much as a thanks to all the people who took the time to try and help lol. If he was talking about samples of notes - then I can't understand why he didn't just import them into the TRG and compose using the score as you would imagine!?

Bizarre - regardless it was interesting to engage in a dialogue about acoustic samples anyway, as this is an area that interests me, and the prospect of multi sampling in NS2 is exciting! I fell in love with multi sampling when I purchased the Acoustic Samples 'Kawai Ex Pro' grand piano for an insane bargain price off some U.S. dude on Piano World!

That wet my appetite, so I purchased Steven Slate Drums (Ex then Platinum) as I'm a total drum fiend and currently own about 6 ancient drum machines (you can never have enough drums right ha ha)! I have to say Slate drums are mesmerising (as is the Kawai) - just track down some live triggered MIDI sessions and days melt away ha ha!

syrupcore
Offline
Joined: 09/26/2010

> as I'm a total drum fiend and currently own about 6 ancient drum machines (you can never have enough drums right ha ha)! I have to say Slate drums are mesmerising (as is the Kawai)

Do spill all six. Are you talking about the Kawai XR-5? There's one available locally for about $100 that I keep eyeing. Don't know if I'm ready for that much menu dividing though!

ElectricWombat
Offline
Joined: 01/16/2015

Ha ha, hey SyrupCore - sorry man, that statement was a little incoherent (I meant the Kawai Ex Pro sample library from previously)! Gonna check that Kawai XR-5 out though he he! The drum machines I own are a Roland R8M [rack version] (All ROM cards), a Yamaha RM -50 (all ROM cards) [basically a rack version of the Yamaha RY-30], 2 x desktop Yamaha RY-30's (yeh I guess you could say I like the RY-30 ha ha), a BOSS Dr-5 (guitar focused drum machine) and a Roland MC-303 (the last 2 are toys really, but interestingly the DR-5 has a 4 track sequencer, lots of sounds and runs on batteries - so I've taken it hiking/camping etc - hilarious)!

I always loved the RY-30 (lush sounding resonant filters, amazing pattern/song editing with 96th note quantise) and it accepts SY/TG-xx ROM cards! I went backpacking round the world and took it with me in my ruck sack - I can't believe it survived! The sound quality of any of these items is destroyed by any modern pristine drum machine, but they have a certain Lo-Fi charm; especially in the way you program them which I love! :)

Slam-Cut
Offline
Joined: 09/07/2011

@ElectricWombat: Disappearing Original Posters made me laugh. Some clarification would be useful. Yes, I agree, the resulting discussion was good nonetheless. Sooner or later most of us want to use acoustic instruments or vocals, so work-arounds with NS are a necessary discussion. Contemplating the various workflows we come up with can be a useful source of insights for all of us

My hope is that if I bring up Multi-Sampling in NS2 often enough, the development team will feel compelled to include that feature. We'll see how that pans out...

syrupcore
Offline
Joined: 09/26/2010

@ElectricWombat I'm jealous of that collection! Suddenly looking for ry-30s on craigslist...

ElectricWombat
Offline
Joined: 01/16/2015

@Slam-Cut, I'm glad that you made you laugh :) it was weird there were a couple of new users after the NS sale that briefly appeared then dissapeared without trace (like ghosts); perhaps it was an illusion; we've just been generating topics based on the echoes of our own subconscious?

Man I should sleep more lol he he. Yeh acoustic stuff is really nice, I love the contrast between digital and acoustic, even if it's just being able to utilise bespoke samples like acoustic guitar phrases or brief vocal expressions etc. Musics amazing we're so lucky to be surrounded by such wonderful technologies!

Yeh a multi sampling feature would rock and open up many possibilities for incorporating external libraries into NS2 (such as grabbing portions of my Slate Drums collection etc) or just subverting it to create weird velocity reacting patches etc! Poor Matt - just when the NS2 wishlist had been quiet for a while ha ha ha ;)

ElectricWombat
Offline
Joined: 01/16/2015

@SyrupCore - thanks man, yeh I'm very lucky indeed to have these things - the RY-30 obsession actually all started when it was originally released (in 91), my mate bought it straight away for occasional use in his obscure projects; and that is basically what I used exclusively for experimentation until 1999 when I finally acquired my own!

In fact, I almost forgot - in addition to the aforementioned; I also have (wait for it) yet another RY-30 which is actually the original one above (totally forgot as its buried somewhere)! However it's been stripped for parts and in a sorry state! It got hammered from all the action since 91, and naturally I'd primarily contributed to its wear so I bought my mate a fresh one off eBay!

Man the stupidity of some people, the douch who sent it just wrapped it up in brown paper - a vintage drum machine via standard post in brown paper - no bubble wrap; I hate flea-bay! Miraculously it somehow survived unscathed - how I have no idea but he was stoked and it's still going strong! :)

The old one mentioned was screwed as the modulation wheel just rests on the edge of the PCB, and heavy use broke the motherboard - but I took it for parts and slightly later managed to track down a 'broken' RY-30 via SOS magazine which I managed to completely fix using parts from the cannibalised one - and that's how I scored another!

If you think I'm a freak, one of my old school friends was also an RY-30 obsessive and used them primarily (but bolstered with an original 303, 909 and SH-101) he got some tracks out via Wasp records which paid for them - nice choice (though far easier to obtain in 1996)! Anyway I opened up a drawer in his house one day and there was an RY-30 graveyard (around 5+ demolished machines). He was actually lucky enough to talk Yamaha into burning a custom EPROM with his own sounds on it - though what happened to it who knows.

That's the reason why I acquired an RM-50 (rack mount version) as it has additional features such as a tiny amount of RAM which you can load your own samples into (and I mean tiny) but it's useful for adding additional 'playable' periodic waveforms (instead of drums) for lots of sonic debauchery! Also as mentioned previously you can use SY/TG-xx ROM cards with it (such as Synth Strings) which really expand its sonic repertoire! I've tracked ROM cards down from all over the U.S. (Venice to Detroit and beyond) they can be pretty tough to acquire - but they're out there!

Yup there is something very special about the RY-30! Autechre loved it so much they created a special EP named after it (promo: We R are why) featuring tracks made exclusively in it! Well if you do manage to track one down SyrupCore you'll certainly have one kick ass acid machine, and I'd be happy to share everything I've learned about it since its incarnation - man I feel old ha ha ha ;)

syrupcore
Offline
Joined: 09/26/2010

Thanks for breakdown and the offer of help.

ElectricWombat
Offline
Joined: 01/16/2015

@SyrupCore, no worries man - you'll have to excuse the rambling - just reminiscing, it can be crazy how attached you get to particular instruments ;)

Slam-Cut
Offline
Joined: 09/07/2011

I think a lot of us are old gear junkies and can relate, even if we had different gear. I was never into drum machines (although I have one somewhere...) but I still have too many keyboards that I'm not using. I feel guilty. I need to get a set up with NS2 and my old gear all working in one little mini studio and then I'll feel like I didn't waste all that money. :o/

syrupcore
Offline
Joined: 09/26/2010

@Slam-Cut Happy to help you with the keyboard surplus.