TK.

Virtual MIDIbox SEQ V4 for iPad

84 posts in this topic

But somehow the last time I tried to compile mbSEQ I couldn't (like a month ago) so I don't know if it will work without any further modifications, have you tested this TK?

Have you updated the SVN repository before compiling? If so, then you are probably having an issue with the PATH, as the makefile mechanism has changed recently, see

Share this post


Link to post
Share on other sites

the application has to be adapted to the new interface layer, this hasn't been done yet.

i will try to get a camera connection kit before christmas holidays and do the required changes thereafter.

best regards, thorsten.

Share this post


Link to post
Share on other sites

Last weekend I tested the iPad with iOS 4.2 + iPad Camera Connection Kit+ midi monitor app + gm5x5x5, and...

IT WORKED!!! , ok... just tested moving things on the app, the gm5x5x5 leds shine away like a christmas tree, I'll have to test midi input too later.

Cool, I was hoping to test it this weekend but am still waiting to upgrade to 4.2. At the very least I can hook up a 5x5x5, excellent news.

(tried also with a m-audio midisport 4x4, no go)

Does the 4x4 use bus power (i.e. does it have an external PSU)? If not then I found iPad is quite fussy about powering external USB devices, I tried a few sound cards. The best solution seems to be using a powered USB hub, my iPad was picking up most things I used after that.

Obviously with MBSeqV4 this wouldn't be necessary as the Core32 has it's own power on board. In fact do you have a Core 32 to plug in and test? That would really interest me ;) I'm imagining having a simple Core32 + SDcard board connected to the iPad USB being used as a control surface for MBSeqV4.....

Share this post


Link to post
Share on other sites

Since iPad and the Simulator (running on a MBP) communicate with MBSEQ via WiFi, I know that this is neither a MBSEQ issue, nor a protocol issue - it's either related to several Juce layers, or to the slow iPad CPU, or to the slow network interface of the iPad.

TK, sorry to drag this back up - but I came across a piece of relevant info. I was recommended that if you want to get the best performance from Computer to iPad Wifi connections that you should use an Adhoc Wifi network setup directly between Computer and iPad. This eliminates the wifi router from the chain, routers are well known for adding latency apparently.

Just thought I would mention it as another possible way to solve this.

David

Share this post


Link to post
Share on other sites
you should use an Adhoc Wifi network setup directly between Computer and iPad. This eliminates the wifi router from the chain, routers are well known for adding latency apparently.

Looking at TKs pings (especially comparing ping 1 and ping 2), I don´t think, the router adds 100ms periodically. Its more likely an unresponsive iPad TCP network stack (due to some app or the IOS periodically taking all processing power and not context switching very quickly).It would be interesting to see pings without the Juce app running...

Edited by Hawkeye

Share this post


Link to post
Share on other sites

Does the 4x4 use bus power (i.e. does it have an external PSU)?

No external psu on the 4x4 so yes, it would need a usb hub in between

But the gm5x5x5 worked just fine bus powered by the iPad so no need to use the psu ;)

Obviously with MBSeqV4 this wouldn't be necessary as the Core32 has it's own power on board. In fact do you have a Core 32 to plug in and test? That would really interest me ;) I'm imagining having a simple Core32 + SDcard board connected to the iPad USB being used as a control surface for MBSeqV4.....

I'll have to try this too!

Share this post


Link to post
Share on other sites

I can't wait the day that MIDIbox SEQ v4 is finally released for iPad!!! Keep up the good work!!! :)

Share this post


Link to post
Share on other sites

this fall i will be in the USA and i think i will definitely buy an ipad (in dollars in more convenient than in euro! :P )

there are tons of interesting musical apps.

i still do not understand if the MIDI communication can be achieved with a PC, and how...

will i have to buy additional hardware??

seq+BLM for ipad is such a wonderful thing... :)

Share this post


Link to post
Share on other sites

You will not need to buy additional hardware unless you want to reduce latency. At NAMM I saw a few hardware midi interfaces geared specifically to the iPad. Here are few links to those:

www.iConnectMIDI.com

Akai SynthStation 49

As for connection to PC's: A few apps you can use Tobias's windows rtpMIDI driver that we talked about here because on the iPad side it's basically free to code for that now in iOS 4.2 or >. I think you will get the best results that way anyway. A few other apps use their own methods of sending midi over UDP and in general most apps are good about having a server application available for both Windows and OS X. You will find the details on those applications on the authors webpages.

Share this post


Link to post
Share on other sites

Also every class-compliant MIDI interface works, tested this myself.

Share this post


Link to post
Share on other sites

Hi all,

I checked the source out to have a look, but things aren't compiling/linking. Can somebody explain?

I added ../mois32 to the searchpath, but I'm still left with a couple of link errors.

Cheers!

Share this post


Link to post
Share on other sites

Hi,

I updated the project setup; try it again

Best Regards, Thorsten.

Share this post


Link to post
Share on other sites

Just downloaded the svn files again and compiled the seq with iOS 4.2, all ok.

Any updates on the midi connection kit front TK? tongue.png

Btw, just in case anyone else hates white icons, "proper" Default screen + app icons:

MIDIbox_ipad_default-icons.zip

Laters!

Share this post


Link to post
Share on other sites

Thanks TK.

I can compile again... and again. But there no installing on an iPad, because I don't have an developer account. Any chance somebody could include the installable app here, or is impossible install it that way?

Share this post


Link to post
Share on other sites

Thanks TK.

I can compile again... and again. But there no installing on an iPad, because I don't have an developer account. Any chance somebody could include the installable app here, or is impossible install it that way?

Sorry yoho, it doesn't work like that. You need security keys from both Apple and from a paid developer account to install on an iPad. You should be able to run it in the simulator without those keys though.

Share this post


Link to post
Share on other sites

Would this help?

http://www.buildingi...ete-review.html

Someone would still have to get the keys, but it sounds like TestFlight would make it a lot easier to share the app with other people. It's free, too..

That's a pretty good idea nangu. But someone would need to put together a build machine that gets SVN updates, builds the code with keys that are kept private, then publishes to a group of people with TestFlight. Then there is a limit (100 devices) to the number of devices that that TestFlight and debugging builds can run on. So this is not a good solution to the general public being able to install the app. For that someone will need to actually publish the code as an app on the app store. The other downside is that it really doesn't let someone do quick builds and test, everything checked in would become a live TestFlight automatically.

On the flip side it's only $99/yr to be able do build anything you want on your iPads.

Edited by Narwhal

Share this post


Link to post
Share on other sites

Hi there,

i am new to the midibox world and found this ipad app thread. A friend of mine is a App developer and i might get him to try this out. Where exactly can we find the files that are to be compiled?

Besides any news on the status of the ipad app?

Cheers

Markus

Edited by electrofux

Share this post


Link to post
Share on other sites

I canceled this plan because when you look on existing iPad sequencers today, you will quickly notice that the user interfaces are superior compared to that what I could ever achieve.

It isn't really worth the effort anymore.

Instead I'm now going the Lemur route and design user interfaces with this App, see also:

Since the sequencer kernel is running on a MBHP_CORE_STM32 or MBHP_CORE_LPC17 module, we can expect best MIDI timings - and this is one topic which makes the MBSEQ (and could never be achieved with an iPad)

Best Regards, Thorsten.

Share this post


Link to post
Share on other sites

TK, I have been thinking about the idea of tablet interfaces for MBSeq..... How possible is it to create an entire remote interface? Especially I wonder if it's possible to do remote display (no LCDs on hardware Seq) via MIDI or OSC and still have acceptable performance.

I don't expect you to code these features, just looking for expert advice :)

Share this post


Link to post
Share on other sites

TK, I have been thinking about the idea of tablet interfaces for MBSeq..... How possible is it to create an entire remote interface? Especially I wonder if it's possible to do remote display (no LCDs on hardware Seq) via MIDI or OSC and still have acceptable performance.

I've also been thinking about this. My main motivation is to be able to take iPad + a simple CORE / SDcard for a live set, so that I don't have to take out my whole MBSeq and risk it being damaged etc. I'd hate for all those hours of hard work soldering and building to be lost.

Anyway, I spent an evening looking at the Sysex / CC remote files in the svn/docs directory. My conclusion was that the majority of the settings that we can edit using a front panel could be edited using Lemur / iPad and some clever sysex. The main challenge would be to actually create the Lemur template, something that I won't have time for until summer holidays....

If anyone makes some progress on such a template in the meantime I'd happily be a tester :)

Oh and one other crazy idea I had was for a web interface to MBSeq. This could be interesting from a cross platform perspective. Lemur is iPad only at the moment, and I'm not so sure that they will port to Android. You could have nice features like a Ableton-esque grid view of all patterns showing their names, faders for the MixerMap settings etc.

TK: I'm thinking out loud here, but would such a thing be theoretically possible?

Share this post


Link to post
Share on other sites

Interestingly Reaper has recently added a web interface plugin - it seems highly configurable. My suspicion is that other protocols for using MIDI / OSC will offer less latency though.

http://charlie-roberts.com/Control/ Seems to be a promising solution for crossplatform use, but it's very much less polished than Lemur. I asked the Lemur dev (Liine) if they were porting to Android and they suggested they would consider it but can't promise anything.

Share this post


Link to post
Share on other sites

TK, I have been thinking about the idea of tablet interfaces for MBSeq..... How possible is it to create an entire remote interface? Especially I wonder if it's possible to do remote display (no LCDs on hardware Seq) via MIDI or OSC and still have acceptable performance.

I don't expect you to code these features, just looking for expert advice :)

Mainly there are two possibilities:

1) a 1:1 emulation of the MBSEQ frontpanel. Button/Encoder events have to be sent to MBSEQ, and MBSEQ sends LED/LCD changes to Lemur.

In fact I already implemented this into the MBSEQ V4 firmware two years ago to allow remote-control of a second "core-only" MBSEQ, but it's undocumented and well hidden in a configuration file ;)

This mechanism could also be used to control the MBSEQ remotely from an iPad once Lemur supports SysEx messages with more than 128 bytes (this limitation should be fixed in a future update)

2) creating a completely new control surface which is optimized for the iPad, because the usage of virtual encoders is too cumbersome, and at many places you would like to control certain functions with dedicated buttons (on dedicated pages) instead of clicking through the menu pages.

This is feasible as well, it could be implemented on the same way like I did for the and the implementation at the MBSEQ firmware side is easy - but creating a perfect Lemur control surface for controlling all the parameters is a challenge.

If somebody with Lemur skills would like to do this, I could take over the MBSEQ firmware part.

The main challenge would be to actually create the Lemur template

Exactly!

Oh and one other crazy idea I had was for a web interface to MBSeq. This could be interesting from a cross platform perspective. Lemur is iPad only at the moment, and I'm not so sure that they will port to Android. You could have nice features like a Ableton-esque grid view of all patterns showing their names, faders for the MixerMap settings etc.

TK: I'm thinking out loud here, but would such a thing be theoretically possible?

No, this isn't possible, the HTML protocol isn't made for this, and there are no RAM resources free for a webserver anyhow.

My suspicion is that other protocols for using MIDI / OSC will offer less latency though.

Haha, totally wrong! ;)

MIDI (over USB) has the best performance, OSC is a bit slower since it requires additional string based parsing at the receiver side, and all other protocols are not compatible to MIDI synthesizers or DAWs anyhow, and would require some kind of proxy/bridge which will add latency.

Best Regards, Thorsten.

Share this post


Link to post
Share on other sites

Mainly there are two possibilities:

1) a 1:1 emulation of the MBSEQ frontpanel. Button/Encoder events have to be sent to MBSEQ, and MBSEQ sends LED/LCD changes to Lemur.

In fact I already implemented this into the MBSEQ V4 firmware two years ago to allow remote-control of a second "core-only" MBSEQ, but it's undocumented and well hidden in a configuration file ;)

This mechanism could also be used to control the MBSEQ remotely from an iPad once Lemur supports SysEx messages with more than 128 bytes (this limitation should be fixed in a future update)

2) creating a completely new control surface which is optimized for the iPad, because the usage of virtual encoders is too cumbersome, and at many places you would like to control certain functions with dedicated buttons (on dedicated pages) instead of clicking through the menu pages.

This is feasible as well, it could be implemented on the same way like I did for the and the implementation at the MBSEQ firmware side is easy - but creating a perfect Lemur control surface for controlling all the parameters is a challenge.

If somebody with Lemur skills would like to do this, I could take over the MBSEQ firmware part.

Hadn't seen the MBCV V2 development.... looks great!

My suggestion is a combination of 1 and 2. Have a main interface similar to MBSEQ frontpanel, but keep a space for a 'pop up menu'. Press and hold on the virtual encoder (or other button) and all available options pop up in the menu area. Then you could have direct selection of notes etc.

I will investigate the purchase of a second hand iPad. If I get one I'll help with the Lemur stuff.

Share this post


Link to post
Share on other sites

No iPad yet (I will wait till release of iPad 3 in early March), but here are some early tinkerings:

post-6793-0-28917900-1329206077_thumb.jp

The encoders work with vertical touch movements - I think this would be comfortable to use but I haven't actually tried it myself. For some actions quick selection options could come up in the container at bottom right.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now