i’m trying to record a longer part played with the keyboard … but the keyboard i like to play stands a few meter from the HAPAX
so i thought i can enable punch in … so it only starts to record when actually some notes arrive. That works when the Looper mode is off.
But as soon as i set it to AutoLength the record starts after a bar or quicker , no matter if i play something or not.
Is this a bug or am i thinking wrong?
I have Looper Auto Length enabled, and with a pattern selected, pressing Rec to start recording a pattern and then again to end the pattern works like I’d expect. I also use a foot pedal set to Rec, and that works as well. What exactly are you doing that isn’t working, and what do you expect it to do?
i thought i explained that briefly. I want the recording to start when i start to play/hit a key on the keyboard and not in the moment i hit record.
The option punch in does exactly this. But it doesn’t do it when enabled the looper.
Seems to be working fine here !
punch in = on
len = auto len (aka looper)
transport off
arm track
leave it for a while… nothing happens…
play some notes , transport/ recording starts…
so obviously there is some other aspect to the issue you are having…
I dont have time right now to try other aspects which perhaps are tripping you up.
e.g Im using internal clock - are using internal or external? does it work for you with internal?
apart from that, I guess, check no other midi data is arriving…
im not sure punch in will start with other data… eg. cc?
I guess, what Id do is have a look at the midi monitor when it starts and see if anything else appears
of course my sequencer is running already … seem pretty strange that i have to stop it , because than i don’t hear the other track i’m playing to anymore
and also i can’t stop it when it’s synced to ext clock
“of course” would be helpful, so I we can identify limitations…
and I means I don’t waste time testing, as if it was “obvious” then you could have simply sent this as a feature request to Squarp - “FR: allow Looper Punch In to work when transport is running”
as it is currently, this is I would say working as intended… but is just not functioning as you had hoped/expected.
i was writing this thread to figure out why it’s not working … now i found out.
Thank you for helping identifying the problem. I will - of course - send now a feature request and and not a bug report.
Btw it’s is nowhere written that this combination doesn’t work when the sequencer plays.
Also writing all circumstances that could possibly have an influence on a certain function is close to impossible… so i don’t understand you being annoyed wasting your time because i didn’t write this in first place.
I’m a bit confused as to how this would work. Let’s say you have things set up like you’re describing. Punch in is enabled, and the clock is running. What happens when you play a note on your synth that doesn’t line up exactly with the downbeat, maybe due to human error. Let’s say slightly after the beat. Where does the pattern start? Does it start when Hapax detects the note, let’s say 20ms after the downbeat? Does it retroactively set the beginning of the pattern to whichever beat was just before? Or the first beat of whichever measure you’re in? What if you play slightly ahead of the beat so that your first note is slightly before the first beat of a measure?
Or what if you play exactly on the beat, but you play your first note on beat 2. Where does the pattern start and end then? Does it start on beat 2? If so, how would that appear in step mode? Because AFAIKT, all patterns in pattern mode start on beat one.
I just don’t see how punch in makes sense when the music is already playing. That said, as I said in my previous post, Hapax supports using a foot pedal, which could be all the way across the studio, and works very well to record loops while you’re not next to Hapax!
Im not annoyed… and I’ve no issue with helping to track down problems.
rather your “of course my sequencer is running already”, struck me , as claiming it was ‘obvious’.
and that I think is something, thats not true at this stage of development.
reality with v1 software/hardware is there, are a huge number of combinations… so not everything will be figured out for release.
but that also does not mean ‘basic things’ have not been tested…
so my usual approach is…
if something doesn’t work, go back to basics…
95% of the time the basic function will have been seen to work during testing.
so we test, this, and then start adding ‘complications’.
and yes, things like external clock are a “complication” , since whilst external clock is tested as a ‘feature’ in itself, if its works… most other features will be assumed to work the same with external clock as internal… (*)
BUT as is the case, here there can be some exceptions… and Im sure this wont be the last !
the same is true of the manual, for sure somethings may need clarification… but at the same time, it cannot (without being 500 pages long, and impenetrable) cover absolutely every scenario.
so during this phase, its useful to point out to Squarp, something things that might need to be clarified in the manual… because they are not ‘obvious’.
don’t get me wrong, for some they just want to use it, and don’t feel they should be part of this ‘testing cycle’.
I totally get that… but honestly, this is pretty normal these days, so Id say if thats the case then waiting a few months/firmware releases for things to mature is perhaps a better option for those users.
but for those helping to find get to that maturity, it’s useful if we do things systematically.
(*) also bare in mind, for many users… the primary sequencer is very often the master clock!.
i unterstand your point … and i’m not upset because this is not working. I can live with it.
But more generally speaking, having to stop a sequencer to do certain thing is very old school and it just didn’t came to my mind that this could be necessary. Especially since really everything else works without the need to stop it. It was quite a surprise to me … so my reaction “of course” .
To some degree i’m also fine doing some testing and i expect that when i buy the newest device.Thats not my problem with this issue.
Btw i’m thankful for your always helpful input here!
For me we can close the topic please and maybe we can forget it and take it as a miscommunication / misunderstanding.
Remember live recording notes are micro-timed so the recording should start and the notes should be exactly where they are. If that isn’t further along to be in step 2, then hapax should highlight step1 and place the note there with its micro-timing.
I think @verstaerker 's expections are reasonable. You should be able to enable punch-in, autolength on, set transport playing and take your time meandering across the studioscape to your midi-enabled CS80 and be able to capture your opus with confidence in your interpretation of the manual. 
Feature request I think.
i don’t fully understand what you’re saying.
A pattern is a fixed length. By default of 16 steps. It starts at step 1 and ends at step 16 - wich makes one bar. Then the next bar…
if the notes i’m playing are within this 16 steps my recording is only 1 Bar. If one note goes even only a few milliseconds above that line my recording is 2 bars long.
As long as i didn’t start playing any note the pattern loops at it’s default length of 1 bar synced to all other patterns.
And basically it is working as expected. I only have to start it the Hapax manually. Wich i can’t when it’s synced to midi-clock and i don’t want to stop all the other gear, as i want to play along to my beat and i need to “hear into the groove”. Also i really don’t like to stop everything just to use one feature. I could stick to roland sequencers from the 90s if i don’t mind start/stopping my sequencer all the time.
So my current solution could be either pre-setting the length of the pattern - most likely to it’s maximum and then just start record and let it loop.
Or i use a pedal to start/stop record. Can you let me know what kind of pedal i’d need for that?
I understand the confusion and expectation…
PUNCH IN’s current functionality is to control the transport.
so its not going to currently work if the transport is already running, or controlled by an external source.
however, I think the underlying issue here though is @verstaerker wants the looper to wait, rather than just start (at sync boundary) - totally reasonable.
the question here is really … is this punch in?
the issue is, unlike the pyramid, the hapax (currently) auto syncs looper start to a bar.
(pyramid used offsets, but we all remember the confusion that caused users, and how many were very vocal that they did not want this)
so the issue with punch in is… if the transport is running, and you press a note half way thru the bar… what does it do? it cannot just start recording !
and its not reasonable, to use this as a request to record at next bar boundary…
perhaps it should create half an empty bar?
but the point is, this is very different to what punch-in is currently doing…
(starting transport, and by definition its at start of bar)
Im sure with a bit of thought and discussion with Squarp it can be made to do something reasonable, but it’s not trivial/obvious what the behaviour should be.
solution/workaround currently?
well the issue is not that looper requires the transport stopped… thats not true.
its that punch-in requires a stopped transport.
for now the solution, is indeed to use a pedal… so that we don’t need punch-in
( I will say, Hapax currently has limited pedal functionality compared to Pyramid, which Im hoping we get back in a future firmware e…g. undo was very useful !)
My question is, if punch in shouldn’t control the transport, what exactly should it do? Since a foot switch can be set to toggle Rec on and off, and if Looper is set to Auto Length and no pattern has been recorded, Rec will also set the start and end point of patterns, what else is there to do?
The term punch in is most often used in an audio recording context to talk about “punching in” to start recording over a running track (replacing the existing contents) and then “punching out” to stop recording, both at very specific spots, allowing precise control over replacing mistakes with new audio. Which makes sense, because you can’t overdub audio in a single track, and you need some way to fix mistakes.
But when Rec is enabled, MIDI patterns are in a constant state of overdubbing. You can always add to them. I actually really like the idea of punch-in for MIDI, because I’d love foot pedal control to be able to actually punch in on the currently playing pattern, replacing any existing notes with new notes. Basically, a combination of temporarily enabling “Hard rec” mode with enabling / disabling Rec. I’m going to send in a feature request for this, because it would be useful to me.
But I would call the existing punch-in something totally different, because it’s not really punching in. It’s just starting the transport in Rec mode when the first note is played. Maybe it should be called “Start Rec on note played.” instead of “Punch in.”
Also, yes, undo! Hapax needs more foot switch functionality. And external MIDI control of Hapax, since I know I’m going to want more than 2 foot switch buttons! 
yeah, often these terms are ‘re-used’ differently when we moved to a slightly different model/context.
the idea of punch-in here on the hapax, is “kind of” consistent with how you would use punch in/out in audio realm.
the main use case here is , set a loop point on hapax, then punch-in to record that new bit. so behaves differently, but can be used in a similar way.
I do agree for that use, ideally we’d also be able to hear a few bars lead-in, which gets us to…
how do daws deal with this, what do they use punch in/out for?
well, if we look at (e.g.) Ableton Live’s use of the term its a bit different… more common to audio realm.
the way it works is… you set punch in/out and a loop point, the you start the transport and it will only record with that loop point (aka punched in)
BUT, thats arguably only really necessary because its AUDIO, which by definition is destructive.
if you are using midi, you can simply user overwrite behaviour, and just not play any notes 
(… so it would really only come into play if you wanted to use HARD record for midi)
however… I think this is all a bit of a moot point for hapax
a) we cannot play outside loop points currently, so this kind of functionality is not currently going to be able to work this way, I suspect its not a minor dev to change this… but again FR if you think its what you want.
b) it doesn’t solve @verstaerker ‘issue’ at all, which is a way to control when the auto-len starts without a pedal
its an interesting area… and just goes to show, how there are many different ways to approach these things, all have pros/cons… and complexity. so, this is why its useful for Squarp to hear the use-case you have in mind, to see how it can be applied to their model/workflow.
can i ask again, what kind of pedal i’d need to control this? Sounds like a good solution to use a pedal
any switch pedal will do…
ideally, you’ll buy one with two pedals, that has a stereo output - as this is supported by hapax, as two different “actions”
I guess, you could also buy two single pedals, and then use a 2 x TS-> TRS adapter…
not done this, but perhaps this is useful if you have two master keyboards… and so want to control from two different pedals.
note: pedals have polarity (i.e. open or close on down), but I think all pedals allow this to be configured.
I use a Boss FS 6 - which I like a lot
(you can look at it for specs)
I’ve not tested with Hapax, as Ive got it hooked up elsewhere… and currently the Hapax is sitting on my main keyboard, so I don’t really need a pedal.
but Im sure it works with Hapax (I’ll test if you need), as its worked flawlessly with the Pyramid, and synths.
also, as stated above, we currently have only a few actions we can assign to the pedal (unlike pyramid)… though in your case, all you need is record, so it’ll be fine!
why FS6?
dual switch, which can be momentary or on/off.
has stereo output which can directly go into Hapax/Pyramid
(also has mono, which is useful for other synths etc)
battery powered (*)
its a quality product… had it for years!
(*) theres a pro/con here…
the FS6 is quite ‘clever’… when it does not have a jack plugged in, it powers off… so you can leave a battery in it, and it wont drain… so it doesn’t have/need a power on/off switch.
but, sometimes I forget to unplug it… and so drain the battery.
that said… in fairness, I guess if it had a switch, Id be just as likely to forget to switch it off
I think the other thing perhaps Squarp could consider is supporting MIDI pedals for these kind of operations.
whilst generally, something like the FS6 is idea for this… we are of course limited to 2 foot switches, which for looping purposes can be a bit limiting.
e.g. I’ve an old Keith McMillen Softstep… with this you could assign many more functions - play, stop, track select, undo , snapshot etc.
the other option I guess, is for Squarp to get more sophisticated with the pedal input…
so alot of hardware loopers, also only one or two switches.
however, they have quite a lot of gestures… short press/long press/double (or even triple) tap, which are also often ‘context sensitive’ (so whats is currently happening)
though in fairness, this may be expecting a bit much (at least in early days) …
this pedals are 100% dedicated to looping functionality, so its not surprising there excel in this area!
thanks. When i checked the FS-6 that looked familiar to me, and i remembered that i have a FCB1010 somewhere lying around … and i found it
it has two pedal outs… i’ll see if they work for that purpose
I don’t know the FCB1010…
but to be clear, these need to be switch outputs (as used on sustain pedals) NOT expression pedal outputs which are quite different.
(simplistically - sustain pedals are a switch using 2 wires, whereas expression pedals are potentiometers using 3 wires)