Mannequins W/2 Beta Testing

Haven’t dug into this firmware yet, but based off the docs, I’m wondering, with i2c, if W/Tape affords the creation of something like a tape sequencer.

I’m thinking of a recording process where the tape head skips around in a programmatic and/or stochastic way (modulate timestamp?); tape speed (freq) and erase head (rec_level) could also be sequenced.

Imagine keys, drums, or even a human voice cut up, splayed, stretched, layered, and re- and/or de-constructed along some portion of the W/ tape… Dada and/or Cubism in 2 HP :slight_smile: .

Now that I think about it, this is also an idea for a norns script… :thinking:

3 Likes

What you’re describing reminds me of a lot of the Norns scripts like compass, cheat codes, and some others.

2 Likes

That’s right! I knew I was thinking of something… so then it sounds like a lot of those functionalities are possible in W/Tape, no?

Also — sorry for the double post, but what makes W/Tape different for me over the norns scripts is the persistence of the tape on the module.

One of the most compelling and evocative aspects of W/1 for me was finding old sections of tape — moments of shows I’ve played, loops I’ve captured, and so on — in short, pieces of my life. In those moments, W/ became, for me, a nostalgia machine :sob:

I’m just trying to think of how W/2 can afford for, and expand on that theme

6 Likes

and this is something I like on the orig fw - I can create a long loop, splice within it, CV around those splices, while CV’ing overdub/write values. as I’m not using i2c, if i wanted control of playback speed, I’d have to swap with overdub/write but it’s all there.

2 Likes

how reasonable is a request to split user-based development for crow/i2c/W/ to another thread?

5 Likes

super reasonable, apologies for flooding. i’ll round 'em up.

1 Like

Oddly, today when I turned my synth on, W/ won’t let me seem to record at all. My recordings from yesterday are still there but I can’t Overdub or Overwrite… Everything else seems to be as normal.

1 Like

Thanks for the info, much appreciated :slight_smile:

i tried w/ tape today and it worked fine, but imo - it is ridiculously overcomplicated for something so simple that could be achieved with 2 taps of the same button.

So v2 to make a loop on the fly (in overdub mode which autoplays) you have 5 steps.

1 - press record
2 - press play
3 - press loop to start loop
4 - press loop again to close loop
5 - press record again to stop recording ( if you are lucky you just might catch it right without overwriting what you just recorded? - a serious live performance no go)
loop plays

  • The issue with this mode is having to do Step 5. When you close the loop in step 4 here, record should autostop and it should autoplay your loop. There should be an autoplay loop setting in the menu somewhere.

V1 to make a loop on the fly (in overdub mode) which autoplays you have 4 steps

  1. down toggle + record (together)
  2. press play
  3. down toggle (loop start)
  4. up toggle (loop end)
    loop autoplays

A looper that does the same thing is 2 steps - pressing the same button twice.

  1. Press Record.
  2. Press Record.
    Loop autoplays.

How I would like to see it done is similar to v1 but with less steps and still using toggles.

  1. Press Record
  2. Toggle Down (loop start)
  3. Toggle Up (loop end)
    Loop autoplays

I am not understanding how making a loop from scratch, that autoplays, has become so complicated.
More importantly there’s a higher chance of screwing up when this being done in live performance by using button presses instead of the toggles, plus the additional step 5 that could wipe your loop.
I just thought i’d bring it up from the practical side of using it.
Its not a major issue for me personally, because i have 2 x w/.
So 1 will remain on the old firmware specifically for live looping.
I do love the synth mode and delay mode - really amazing additions, but the tape mode i won’t be using on v2.

6 Likes

Could you post the crow firmware you used? The link at the top doesn’t work - 404 error. I tried v2 RC on the releases page, but i think that one didn’t have w/ support.

Click “Artifacts” on the right hand side here https://github.com/monome/crow/pull/342/checks

Grabbing new builds like that was a nice tip dropped in one of Trent’s streams for checking out new features before releases are made.

1 Like

this may be early in the process but I have to ask if, once 2.0 is out of beta, will v1.2.1 be supported by Mannequins?

edit: Q not directed at @mlogger

1 Like

Well if the issue is only the last step of pressing record again to stop recording maybe that’s something that can be reverted back to the way it was on the previous version? I can see how having record automatically turn off might be more universal and a bit more conducive to live performance because it’s more “careful”, less prone to accidents, etc…

I definitely don’t like the idea of having just a two button press looper, I like the idea of it being more of a “tape” then simply a digital looper… but I could probably be convinced. I think what I miss the most is being able to arbitrarily change the loop by adding cues within… cutting the loop in half or doubling it might be more “musical” in some contexts (especially if you’re looping with a clock signal over i2c) but it also seems to take it away from the “tape” metaphor make it more like a live looper… My gut feeling is that this foot in both worlds is limiting and makes it not quite doing one thing or the other as well as it could. These metaphors are only useful where they’re useful, so “this is not enough like real tape!” is not by itself a legitimate criticism, I realize, but tapping into that tradition of music was one of the things I liked about v.1

Okay so, solutions. To me the thing that seems most obvious (although maybe way too much work) is that tape mode should either be fully “tape” mode (more like v.1) or it should commit to being “loop mode” — focused on time synced live looping, clear buffer on start up (don’t need a 6hr buffer), maybe your two button press loop option. Much simpler interface (more similar to delay and synth modes in their simplicity—I’ve pretty much already learned them, whereas tape mode i have to refer back here every time). Is there a world where this mode doesn’t even require an SD card? People would be sad about losing the tape, and the ability to play back recorded audio from previous sessions, but maybe that’s a sacrifice that would be worth it for a newly imagined instrument geared towards use as a looper (which the current version seems to only halfway do). Is that the kind of sacrifice you and @baleen are asking for with this version?

I’m not advocating for this personally, I’m just trying to distill how you think this can be improved. For me, improvement would mean moving away from the looper pedal metaphor and more towards the tape machine metaphor—play means playing till you press it to stop and can continuously be sped and slowed and reversed, record means recording till you press it to stop with control over the overdub, cues persists and can continue to be added to manipulate loop points like chalk on a tape (okay the metaphors are falling apart lol, please check me if I’m out of pocket with this). This combination of functions is not easy to imagine on this interface: for example, if you set cues via the switch, you can’t also use it to continuously speed or slow or bump the “tape”.

In the beta’s current state, tape mode seems like the weak link to me as well (and it’s got the most glitchy/weird behavior where it’s difficult for me to even report whats going on because I don’t know if it’s reproducible or what’s happening exactly). I tried doing some loop tests to emulate @mlogger 's workflow as described, and it seems like if you press record off right before setting the loop end then nothing gets recorded? And sometimes it seems like the loop points don’t get put where I expect them—sort of like when you set a cue point to loop on v.1 and then you pass a cue point so when you set the final cue to begin looping it loops from the middle cue point… except there’s no cue points here so I must be doing something else wrong? Or there’s a bug? Sometimes things don’t seem to record at all if I’m toggling record on and off when looping (trying to punch things in). I’m not sure what’s going on there because it works sometimes and I can’t figure out what the difference is…

I bring up this idea of a “loop mode” because it seems like that would be able to have simple enough controls where it might be able to fit into the “language” of “delay” and “synth” modes… whereas the current tape mode has an almost entirely different language both from the other modes and from the previous version, making it difficult to learn… or it’s just buggy in it’s current state and I’m a nincompoop which is entirely possible and likely. I suspect it’s a combination of the two :stuck_out_tongue: I hope this long post is helpful somehow.

TLDR: I would be sad to loose some of the tapeyness of w/ but maybe the halfstep towards a livelooper/livesampler with this firmware indicates that that’s the proper space for the “tape mode” to live in? And that it should fully embrace that?

PS if this “tape mode” became a more simplified “loop mode” would that make v.1 support more likely? I would hope so, but v.1 support is probably completely untenable in either case :frowning: (unless it becomes open sourced at a later date, which I got the sense was implied in the livestream, but I may be wrong about that…)

PPS this post was so long it had a TLDR and a PS, smdh.

7 Likes

I don’t care about the SD card as the material on it was inaccessible. 8 hours and different tapes was a ceiling I didn’t reach, and I never used a prior sessions recordings. This reads stubborn but what I concretely want from W/2.0 is the original W/ design and functionality to not contain many critical failures. A “live looper” which I am imagining based on your post, is a very different module. is that okay? maybe (if I can erase the tape). is that why, a little over a year ago, I bought W/? no.

my real concern rn is not about what sacrifices I would accept in W/2.0, it’s that I don’t want W/2.0, as the features I use are the features that break the module.

ikr i mean it happens to all of us tbh and fwiw its what the internet is about amirite

2 Likes

However, if you recorde/loop live played acoustic instruments, it is more smooth when the release of the sound naturally fades to the beginning of the loop. Chase bliss blooper has this option and it works very well

1 Like

Reformatted SD card. Updated firmware again. Same issue… Except I can now record if I’m running a loop. And the recording will persist on the “tape” if I cancel the loop. But still can’t just record in normally.

Doesn’t read as stubborn to me, but here’s what’s unclear: describing all the functionality you didn’t use and then saying you want it to do the same thing but better. The point of not using the SD card is to avoid critical failures, which are associated with wearing out an SD card. It seemed you and mudlogger were using it as a live looper (not saving buffer info between power cycles, being able to erase buffer, quick looping functionality with ability to overdub, etc these are the functions I associate with this use-case but please correct me if I’m way off base). If this is not the use case you bought w/ for then I’m misunderstanding something.

It’s clear that your ideal would be improvements of reliability without any change to the functionality of the module, again I’m not trying to convince you to make compromises—that’s your own decision. It seems like the improvement in reliability requires a change in the use of the SD card which is part of the purpose of this update. The tape mode is only slightly changed, things like clearing the tape on startup, bringing back something like cues (your requests), or toggling recording OFF when looping is engaged (@mlogger 's request) seem very much like open questions which could be reconciled… this is why we’re here giving feedback, right? (As I got into at length in my post above tape mode has some other problems right now other than these issues, so I wouldn’t say it’s working as well as or similarly to v.1 at this particular moment, but theoretically). I keep saying I’m not trying to convince you and maybe I should stop saying that lol because it seems like that’s what I’m doing! I think my point is that support for v.1 is unlikely given the size of whimsical raps/ the cost of the module/ etc. for all intents and purposes I think we should assume that v.2 is not intended to be a new or alternative module but is intended to improve the function of the module as it existed and add features. Concrete feature requests might get some traction—that’s what I’m trying to get us talking about: what do we want specifically and how do we want them implemented and accessed? There’s no obligation to do any of this work of course, but that’s the purpose of this thread, I think…

Anyways, regardless of whether or not I’m trying to convince you of anything, I’m certainly not trying to “win” an argument with you (or even argue at all!). I’m going to get off this thread until I have better suggestions for how to reconcile the relative benefits of v.1 and the v.2 tape mode, it seems like there must be a way to do it… (sorry again for taking up so much space with my ramblings)

2 Likes

Yea in the livestream. Delay and Synth modes don’t use the SD card at all, and Tape mode I guess uses it better(?) then v.1. Part of the cause of failures is continuous looping/reuse of the same “part” of the SD card for way more hours than it was spec’d for. The new version allows use of a “lower quality” SD card because the way it’s used is less wearing I guess. I don’t know the technical details but this is what I understood from what Trent said.

1 Like

we are probably arguing around the same thing despite probably general agreement bc we are different people with different perspectives and language. I didn’t watch the livestream so you may have additional context I am missing.

I want W/ to do the same thing it already does (happy to pay as a cost the features we are already tossing out) but as a stable module.

my understanding is that these conditions are mutually exclusive:

•the module as originally designed requires the SD to do the functions I want it to do.
•using the SD in this manner eventually creates critical failures on the card.
•2.0 creates new modes to maintain the viability of W/ as a hardware platform but doesn’t resolve that the original design (again, which is what I use) is essentially flawed and unreliable.
•“delay” and “synth” bypass the card completely, and my guess is that there is not enough available realtime memory on W/ to create a live looper of a length beyond the max of the “delay” mode.

I also do not want to noise up the thread as by now I think my perspective has been articulated, tho I am happy to clarify, but it’s probably time to just wait for more official perspective on this thing.

4 Likes

Seems like you’d have to get a crow to accomplish the v1 stuff you liked on v2. Not my concern to sell you on crow, but it’s probably worth looking into a bunch of other cool stuff you might like to do with a crow simultaneously. It’s super useful and affordable for a multitude of tasks and flexibility. Good luck! I do appreciate your predicament.

I will say, as somebody who struggled a ton with v1, I have taken to v2 immediately and having a total blast! No cheat sheet required this time around. Amazing work and thank you Trent!!!

Definitely hearing some funny artifacts when modulating delay times. Sounds cool though. I’ll probably miss it when it’s “fixed“ like I missed the crunchy low speeds in v1 lol

1 Like