Bizarre audio MIDI fault - it's driving me nuts!

[Digital_Igloo]

It would be great to get some Helix expert opinion on this topic, but it seems it hasn't caught their attention yet.

 

No, we pay attention. We generally just don't jump in until we have a solution. Plus, we avoid posting in tech support threads like the plague, because it would literally add 20-30 hours to our work week. Some of us have new Helix features to design and amps/effects to model.  :D

[soundog]

Get back to work.

[datacommando]

Here's the scoop. Certain brands of keyboards (especially Roland, but some older M-Audio keyboards as well) continuously spit out active sensing messages every 300ms. This basically tells the receiving device "Hey, I'm still here! Hey, I'm still here! Hey, I'm still here! Hey, I'm still here!"

 

IIRC, according to the MIDI spec, the receiving device is supposed to display something like "MIDI offline!" if it suddenly stops getting a "Hey, I'm still here!" message; you know, something to warn you that you've tripped over a MIDI cable in your gigantic room-sized synth rig filled with eighty keyboards.

 

These days, active sensing is all but unnecessary, is almost never used by guitar gear, and Helix wasn't too keen on it. The next rev of the firmware will effectively ignore active sensing so the dropouts should go away. We've tested it with Roland and M-Audio keyboards—all is well.

 

A lot of Roland synths let you disable active sensing in their System menus, but the PC-180 doesn't. I know—I had one back in the day, and hated that my MIDI interface input LED would flash constantly. For those of you experiencing the same issue and can't wait for the next update, manually disable Tx Active Sensing in your keyboard (if it lets you) and all should be well. A few MIDI merge or MIDI thru boxes automatically filter out active sensing as well.

 

This is kinda funny because I suggested active sensing might be the culprit a while ago and no one believed me. Ha!

 

Okay, so it's not very funny.

 

Hi Digital_Igloo,

 

Hmmm, active sensing? Well that's interesting. As I understood it, Roland, implemented Active Sense as some sort of low level self checking "safety" feature and that the "Hey, I'm still here!" thing, along with certain SysEx data dump requests, is what some old greybeards, like me, sometimes used to refer to as MIDI handshake. Although, I think the handshake thing has a sightly different interpretation nowadays. You can check your stuff for active sensing using the old school trick of unplugging the MIDI cable while holding down a note on your master keyboard. If the slave unit stops making a sound, you have active sensing, if it carries on playing/droning – then you don't – simply hit the panic button, and that will kill the stuck notes.

 

As, noted – in one of my previous posts, above – I found that the MIDI Monitor software detected, that as soon as Helix stopped sending the note on/off messages, it looks like some sort of constant handshake thing was happening - until you switch off the MIDI thru and then the "From HELIX Active Sense" messages stop - instantly! Only happening on MIDI Thru?

 

O.K. you have stated that the next Firmware should fix these MIDI problems - yay, that would be fantastic, but it still does not explain why I was able to use my Helix to record MIDI absolutely perfectly from November 2015 up to early March 2016, when it simply just stopped. I had the full thing, aftertouch, pitchbend, portamento, whatever. Guess what, I figured a way to fix how to record MIDI by simply buying a $5 (less than £3:50) Chinese MIDI to USB cable that takes Helix out of the equation.

 

Furthermore, much as I appreciate your comments above, it goes no way to explain why I now no longer have the ability to record audio into a DAW, other than on channels 1 and 2 complete with all effects. I'd like some sort of "active sense" that could show me where everything else went. For 6 weeks I've not been able to use this thing as it has been claimed it can be used - for recording! I now have a very lovely black metal brick! And still no reply from my support ticket!

 

Yep - not very funny!

 

HHHHHHHEEEEEELLLLLLLPPPPPPPP ME!

[Digital_Igloo]

HHHHHHHEEEEEELLLLLLLPPPPPPPP ME!

 

I was at Roland US for five years—know ALLLL about how MIDI works in their stuff—Mr. K was one of the inventors of MIDI (along with Dave Smith), and they follow the spec to a T, to a fault.

 

Can't do tech support here, but make sure you're running the new Mac driver with OS X 10.11.4. Otherwise it won't work. What DAW are you using? We've tested Helix with a ton of 'em, and with the new driver, both multichannel audio and MIDI is working extremely well at 44.1kHz, 48kHz, 88.2kHz, and 96kHz.

 

Provided you're running the new driver with El Cap 10.11.4, the only advice I'm allowed to give is to hit CS up again. Sorry.  :(

[datacommando]

I was at Roland US for five years—know ALLLL about how MIDI works in their stuff—Mr. K was one of the inventors of MIDI (along with Dave Smith), and they follow the spec to a T, to a fault.

 

Can't do tech support here, but make sure you're running the new Mac driver with OS X 10.11.4. Otherwise it won't work. What DAW are you using? We've tested Helix with a ton of 'em, and with the new driver, both multichannel audio and MIDI is working extremely well at 44.1kHz, 48kHz, 88.2kHz, and 96kHz.

 

Provided you're running the new driver with El Cap 10.11.4, the only advice I'm allowed to give is to hit CS up again. Sorry.  :(

DI.

 

Damn, I hope those 5 years weren't during the period when a certain firmware upgrade for the Roland GR33 Git Synth totally fritzed a whole bunch of 'em. I shall just have to hope that you were the "guardian angel techie" who slipped out the engineers "factory default" disk images containing the MIDI files that enabled some of us end users, all around the world, to fix it. Of course Roland denied such a software fault had ever existed. They insisted the main board needed a replacement at 3 times the price I had paid for my pre-owned unit, even though there was no actual hardware fault. (I've still got the email audit trail on that one) - good fun eh? The usual run around - got to be expected really.

 

So, regarding the new Mac driver with 10.11.4 - for my part - sorry, but it didn't work for me. Fact is, it made everything worse, so I removed it - you can find comments in my various rants regarding this, especially on the thread which you started:-

 

http://line6.com/support/topic/19809-helix-mac-driver-100-public-beta-now-available/

 

Not only was I minus MIDI but the Helix audio is now no longer able to create and re-amp a dry signal - I can only get fully processed audio on chl 1 and 2. Various DAWs were used to test eg: Logic Pro X, Pro Tools First, Reaper and of course the real no brainer - GarageBand. I was going to check out Ableton Live and Reason but was so depressed I couldn't face even more disappointment. Ye Gods - this thing worked right out of the box, recording all required MIDI info and multi-channel audio, for 4 months and now - nothing!

 

With the month of May fast approaching, I'm seriously considering sending this unit back to the retailer for a replacement, and let them chase Line 6 for a fix.

 

Does no-one listen?

 

The new driver does not work in my instance!

 

Now hear my screams!

 

This lemon needs to be fixed.

 

Tell me I have a rogue unit!

[Digital_Igloo]

Damn, I hope those 5 years weren't during the period when a certain firmware upgrade for the Roland GR33 Git Synth totally fritzed a whole bunch of 'em. I shall just have to hope that you were the "guardian angel techie" who slipped out the engineers "factory default" disk images containing the MIDI files that enabled some of us end users, all around the world, to fix it. Of course Roland denied such a fault ever existed (I've still got the email audit trail on that one) - good eh? The usual run around - to be expected really.

 

Hmm... Don't remember that particular GR33 thing—that was before my stint there and I dealt primarily with synths and groove products.

 

I'm using Helix with Logic and GarageBand all day every day with no issues (actually, Logic is open right now). Reason and Live have been tested by QA extensively as well. Something very well may be up with your particular unit; CS will have to confirm.

[datacommando]

Something very well may be up with your particular unit; CS will have to confirm.

Let's hope that the confirmation will be sooner, rather than later - then I can get back to recording instead of typing this stuff every day!

[datacommando]

I was at Roland US for five years—know ALLLL about how MIDI works in their stuff—Mr. K was one of the inventors of MIDI (along with Dave Smith), and they follow the spec to a T, to a fault.

Sorry about dragging out this thread, DI, but I guess I had to just throw in the extra comment because you mentioned these real heroes.

 

Do you actually think that Mr "Sequential Circuits" Dave S and "Sensi", Ikutaro Kakehashi San (梯 éƒå¤ªéƒŽ) would tolerate with this unsolved MIDI stuff. I don't think so!

ã‚ã‚ŠãŒã¨ã†ã”ã–ã„ã¾ã—ãŸ

[Digital_Igloo]

Oh, there was plenty of MIDI craziness at Roland too, I can assure you.

[soundog]

I finally found time to test this out again. Earlier in this thread I had noted that "Helix receiving MIDI "note on/noteoff" data via DIN inputs only sends out "note on" (and not "note off") through the USB port or MIDI out port."

 

Based on Igloo's suggestion to try filtering out Active Sensing, I looked into this. I have an old iConnectMIDI box that will filter out selected MIDI commands. Sure enough, I had a newer Roland TM-2 in my MIDI setup, and it was sending out Active Sensing pulses. I filtered it out, and lo and behold — my MIDI note ons and offs are now being passed through (DIN in, USB out) intact and unpulverized! Its interesting to me that Roland is still using Active Sensing, even in its newest products!

[datacommando]

I finally found time to test this out again. Earlier in this thread I had noted that "Helix receiving MIDI "note on/noteoff" data via DIN inputs only sends out "note on" (and not "note off") through the USB port or MIDI out port."

 

Based on Igloo's suggestion to try filtering out Active Sensing, I looked into this. I have an old iConnectMIDI box that will filter out selected MIDI commands. Sure enough, I had a newer Roland TM-2 in my MIDI setup, and it was sending out Active Sensing pulses. I filtered it out, and lo and behold — my MIDI note ons and offs are now being passed through (DIN in, USB out) intact and unpulverized! Its interesting to me that Roland is still using Active Sensing, even in its newest products!

 soundog,

 

The active sense thing definitely seems to be doing something in there:

as I had previously detailed from my Midi Monitor record way back in this thread, at the beginning of April, I got this –

 

16:09:01.242 From HELIX   Note Off 16 D3 0

16:09:01.243 From HELIX   Note Off 16 G♯2 0

16:09:01.244 From HELIX   Note Off 16 A♯7 0

16:09:01.324 From HELIX   Active Sense

16:09:01.624 From HELIX   Active Sense

16:09:01.925 From HELIX   Active Sense

 

Yep! that really is an active sense message approx every 300ms - except I could only pick that out of the stream coming from MIDI Thru?

Active Sense messages would stop as soon as I turned off the MIDI Thru option - very strange!

 

Let's hope that Digi_gloo is correct about the next Firmware Upgrade putting an end to this very odd problem.

 

Now I just need to fix the screw up with the audio only happening as fully processed on Channels 1 & 2, no re-amp options whatever!

[Digital_Igloo]

Let's hope that Digi_gloo is correct about the next Firmware Upgrade putting an end to this very odd problem.

 

It does. Igor has an M-Audio controller and Dom has a Roland SPD-S. Both were funky before; both work fine now. Update's coming this week (hopefully).

[datacommando]

It does. Igor has an M-Audio controller and Dom has a Roland SPD-S. Both were funky before; both work fine now. Update's coming this week (hopefully).

 

 

Yay!

 

Coolio!

 

You've made a happy man very old!

 

...(or something like that)

[jpdennis]

It does. Igor has an M-Audio controller and Dom has a Roland SPD-S. Both were funky before; both work fine now. Update's coming this week (hopefully).

Updates: including more cool factor stuff?

 

Thanks again DI

 

Dennis

[datacommando]

It does. Igor has an M-Audio controller and Dom has a Roland SPD-S. Both were funky before; both work fine now. Update's coming this week (hopefully).

 

Really got to hope that this one is gonna work!

Finally - I got a reply from the Support Ticket and, if this is correct, I will soon be able to get outta here and back to my real life.

Here's the message:

Hi,sorry, took some time :QA in the US checked it and reproduced the issue. It will be fixed. I don't have a definitive timetable but it will be fixed relatively soon.

Well, waddayano - QA "reproduced the issue".

Seems I didn't imagine it - Vindicated, at last!

"It will be fixed" - Yay! Nirvana!

Digital_Igloo, if you were involved in progressing this and getting it resolved, then thank you very much, but that's also going out to the front line troops who really did the job!

 

Point me at the download!

When this thing gets installed, and I have checked it does what I expect, only then will I be able to flag this thread "ANSWERED"

Yippee ki yay! Mr Falcon.

[datacommando]

O.K. Helix MIDI users.

 

Heads up for the Final Status Report on my Helix MIDI Issues

 

This “mystery†was taken onboard and has been investigated by Digital Igloo (Eric) who, along with Marcus, did some extensive testing with dozens of various synths to figure out if that active sense messaging issue was only happening with Roland (and some M-Audio) keyboards. The general consensus is that it seems to be the case – something odd with the Roland and Helix MIDI combination (unless anyone else knows a different answer).

 

Anyhow, just to bring everyone up to speed, the current situation for me is thus:

 

As I had previously noted, time and again - I had no issues if I used the work around of connecting a mucho cheapo Chinese MIDI DIN plug to USB to hook up my old PC-180A to the Mac. Also, I have a whole studio rack full of vintage MIDI sound modules with DIN sockets that would always play nice with that particular controller, e.g.: Roland D110, U110, Yamaha TX81Z, SPX90, Kawai K3m, K1m, Akai XE8 drum expander and the ubiquitous Akai S950 Sampler. That's a lot of MIDI connections, not including the Roland GR33 Git Synth, Boss GT6, the PODXT Live and HD500, but they are all gathering dust over in a corner, so that's O.K. The thing is – I only had access to that one controller keyboard which I've used for years, so I had nothing else to use in a comparison test with the Helix. Trouble is, most new controller keyboards no longer have old school DIN sockets and use a single cable to supply power and USB communication. Damn! Screwed again!

 

Except, I did happen to find a pretty good one - the Arturia Keylab 61 Hybrid Synth. Now this mutha is the mutts nuts and it has DIN MIDI in/out along with the more modern USB. Plus it's made of aluminium and wood, semi weighted keys with velocity and after touch. Way cool.

 

Plug it into the Helix MIDI ports and away we go – perfect response instantly. Excellent! Note: Hooking up via the Helix for the MIDI connection frees up a USB port on the computer, which is always useful to have. Right – my life is worth living again, especially as the last Firmware update seemed to fix the equally mysterious fault I had with re-amping the Helix. It all works again. Yay!

 

Now, because I don’t have another Roland keyboard to test the Helix with, I guess we will never know if it was spurious Active Sensing messages being transmitted to Helix that caused stuff to screw up. If anyone else has an problem like this, and they use a Roland (or M-Audio) keyboard, maybe they will have an idea of where to start on a fix it.

 

As for the old PC-180A, I no longer care if Brian Eno still uses his! I think I will just drag it down the yard, stick a couple of rounds of 12 gauge through it, then give it a Viking funeral. 

 

I know this is not exactly a proper solution, but for my part – I think we can put this puppy to bed now.

 

Thanks to the Helix Staff and Forum Users for all your assistance and patience in this matter.

 

Happy trails

 

Now, I’m off to get a studio tan!