Maybe. I’m reticent about adding too much more stuff to repos without much sense of how shared maintenance is going to pan out in the future, especially given the current focus on trying to ensure that we have as much parity between hosts as possible.
That said, feel free to shine them up and put in a PR.
Something useful on the learn platform – if we can be bothered to collate and flag that its not maintained code but community code – would be to have a “snippets” area where people can peruse community sourced stuff. Keeps it out of the repos but makes it more exposed.
That’s a good idea.
Unless I’m gonna get into the core guts of the stuff, it doesn’t make too much sense to try to edit help files or add snippets if it’s not in the correct style, or going to be implemented at all, etc…
Nor should you feel the need to – having workable but dirty patches on this forum is fine. We can tackle discoverability as a separate issue later 
I guess it’s more the latter.
I only half-remember that there is information on here, and can sometimes find the threads/patches, but I’ve been quite close to the project in that regard. Someone coming across it for the first time isn’t going to have that “luxury”.
I imagine this isn’t the code directly out of the native fluid.ampgate~ as it seems like adjusting the lockout time can break the logic of the ifelse loop.
----------begin_max5_patcher----------
2217.3ocuZszbaiiD9r8uBV7vTNaEqB.78lIoxdatrUsm1KimREDInDlvWKI
jr8jJyu8sA.AeHSJSaSGWwxhMdz8W+BcClue8U16Jef0Xa8Os9cqqt56We0U
JRRBW097U14zGhynMpoYGWlmyJD1eTOlf8fPQ+d998YLqblknzZWMi9MKwAl
UV4ddrYxY7BVb4wB0JHsDSKKDozXljFtkFOQsmk69yawdlUWbLmWjwDMCmIP
r7nvPEMXOa3+kZOInMCIWPyUjs+MV1IlfGSM6eEUDefWreaMKVnUIAtDXwVD
r7Sbfbmrb82fr9C4R9w0WK+3iKTykyZZn6YOQy45hQarm.4nIQNYRjazGZRh
GqXZDXaa8GW.e9Zj0BPTfBsjWI.KX2Cx8SvWUMqhUjXAXnIiGyxXE6EGl.v9
u.K8qDuN9Jf5o+iC48.u4MhxFZdUyeOAFeIdyj4vXCeeAMy9iV1oYkTwEgre
PnDjse91Lv.2.IzffSz5tfIf7NV8D308h3E1cdNriRRHTefZEsF1YAqdKqft
Ki875nzx5bpBx9y5Y.5qczh8KQc43H+zC0qtLRLuvHw8haC8DKYKUHp46NJX
8eqoUY1pMk5rrirxTCYC8gHFXffCVWk3.ZECqmZROU8LZVf4Z+zlnQSKGLCm
aBFMglCk0hErQFkMROjx6x7mWnqVF+DaSSbYEapnHxpjZ7YcFbIp3EecdQGj
qJqA1Yf+PMrGs6mxYA+JCsT3klzTejafF3EQi+11lrRsDeKZCw2w2ygDEh7C
IN3wSTv0FHmMXbj7e99nHW2PWiW5wFAkWncccPA3vHOWhaniu+DG9DMcfaWD
Jb.udmhhbb8cQdDRXfW3Ll.yxpgr+zFVmrh2DE3g7c87QXhqGxr7AAAnMXm.
hmaDBGAVhPuKEe+r+1Y3R4El8Gg.kpWjaP.HC3v1IjvhoONV665i8bBa+L3B
9M3fM9QC9ILBbehhTYSh7.k+fevX4ARC8ozb1nhhPa7c.aTjmia.IJvmbltr
SDQaBPdgfigGIxECl+06LsN2yIiFwS6qLcFamIpQSFR87G10+sKF05DpBW87
UgqND0Y8DmWYfI6uRnwSgZb3KHGDZdwE6oJ9x2U8GWOymqYII+iIAPvZjDs0
nbIChAgHc4VdduohPt7ICX+eJmLz5iICl8jERF1a0dONXXd357SAt3PxOI3N
W6kDuoPO4Uza3TsEDosl5p9HZrhVYDflB.3UE.NssyE8l.vk83B+o3v4EnSR
n8zHDsOk7oY83Hu1bl6T7uu6kSb18m3M7c7Lt3wQ3HMsg0Jg2pNc41gmamUF
+MVRRMceSbcYV1PiVLz462DGpKOt+vP55B3O7zEnG3zSGPyEyImFp61mWlLl
RYcB.pATn06aupG6ulxg1FjMDUer4.qYCkmJe7qYkkUvWvxuepTdXaO9F3EL
8QGFF00pPBKuro7XQxFPiCpY60+rEclIGW88zn77CiV2SO2yJ9aquBkqbnlA
sBkk.mnAOml1S.dd70aXIuUmKduWCKfZcuTfkq1Hs2uEQWpzvD5lfhyZfU52
bhU2vKKF10oMspZ.4w81lS+yR0FE10yH.SMImNR0LYP23sU5yBht.j6i0ZWp
G7c6a8T5yWW.sLIWiooSiHorqRWwlp1qWzNooZCXMMafcuBA5rRkhwU6Ag8z
kPh50HxXLcr2.WcvNTwJ3EUfi.jzmJZAP2vIrT5wLw1QU6R1L43laAcxA6ho
9W0xFiLSYeMOorPJDiLGRxF1IK.SeqWCAiZFEzpIVL3RB5kYFD5gTbrYGsVk
hTe2CDyfhxxrwC0stLVpnc3JdQwYZQQY07CVy2e3BqcWILX9k1a0HMaOVnGc
K3XH1Juzlwyilk0lAX71+.sfmSELSGYDT2fOI8Md7HmlXjDvQOlcOOQbXz0H
IGAlNux3DY2YkS36YMhwzDTc17dJMhG0J8AjNZNcaqfkWkAnX7DfPDdi.Rhc
eS6DMNZCU.8uhfgA1CSmNh9kRqNN0JjEyhLbjItN3YyLNeMRyVmj9.B2vt7c
CaDYvcVYN030.2XHojblCDkohtm4hgeInkbFKlHAy72TR+8Jd9aF4W+2PbuJ
oYxWrurB0UU5kupJM+H8UmFMnlHch31JSr+MvUqr9QKYND1MnO7Iq6JLzj5r
iEwmSFRrd5+JuzzV52U7ejW2n0fSiuAqGpcfAGKei2fAFe97Mxyma2Qd5MJI
x5yeFp14W9EKdg0WFxAIsNAw5WOeHinqV+Gfs762UXYAZcr0msvexxxR9b+r
NST.gPHmPqLHWwcE+.9kk0vz61cBPF61fuH4xfsq6q21t1wPBqgDVJ3CTNev
r0F9hZWaOR+rbYRp.VHvSpY9IK6ULX4B4F.q.dlTC3YCVPyFrfuXDwHG1Ycz
w5xLCeWxa7LoImSW3t9oI0f0MxaAYI6qnPVk6LmSn3gb7oUCPKB0wFSQaxPq
w4PfiFE7ht5q98NrasTivqQFvKPF7eGkA7B0CdyJCsDeUuClYZGJM6HOYCMu
JkQkkkC8Fc.pPqBVVZM6+IKOBY80TnSxZqa8fNiRgxsfLvUGqrb5eJo79BKm
PfRSV48siCc560SPMEIok1HEdcajZIsP09tjcz+mCv47KgPsCiCMzp9ycEZM
.OwEfLzMbFyOYjehcei+SZzWSNi8mly32DmGCmYv76AiIKfwxqb0Zt6WXoLx
e4H7MwGukXCQilzJ48rDN+t373t.Uq2JnZCVhux6Qb4RhNBeOX7hP73IkySp
J4Eh1je3Hr5EH3peS59ntmV8THKRXCeWxdsXVu5Acx2Z+ymWAqY8amSK5bIx
ZvIxR3jyZvImkvI20fStKgSdqAm7VBm7WCN4uDNErFbJXIbJbM3T3R3TzZvo
nEE49VO3hfWZdb7pWe0hprC87GdoKq9rW5fjqm8xFN6EM7zWxv7ufgye4BpW
rvTuTAYSTW+iq++fyt12.
-----------end_max5_patcher-----------
Judge for yourself. The difference isn’t leaping out at me, but I can see the gen~ version getting stuck
Hmm, looks similar enough to me.
Maybe something to do with the fact that this is taking in a differential vs having an absolute threshold as the input in fluid.ampgate~ so maybe there’s a signal that can jump out of the logic somewhere?
Ok, I don’t see how this is the case from looking at the code, but I think if the minslicetime is adjusted while state = 1, then it breaks the logic.
I tested this by setting the values to extremes when playback is paused (or specifically timing it between hits).
The native fluid.ampgate~ version never seems to break in the same way (from my tests) so I guess there’s code above what you posted that takes care of that?
Lol, found it. It’s because you’re passing a float into as minslicelength and the code assumes it’s an integer. When it’s a float, debounce will go below zero in the decrement loop, and the condition to change state (debounce == 0) will never be true.
So you could either change that condition to debounce <= 0, or make sure that debounce is an integer by debounce = int(minslicelength);
I love that you thought I was holding out on you though 
Hah, that makes sense.
I know you guys are super busy over there so didn’t want to presume anything.
Here’s an updated version of the example patch from above for posterity (I’ll also edit the meme-post version of it too).
----------begin_max5_patcher----------
1894.3ocuYszbaaCD9r7uBN7PGmN1Z..e2zjI8VuzY5odotiGJRHIzPAxRB4
GMS5u8t.ffDTlTg1lNdhkL2E.6qu8AX9xEqb2T9.sw04mb9SmUq9xEqVoHII
rp84UtGReHqHsQsL2B1cz0MYkUz+y8JMeVthS4l+9ZBxPje7.iWPEpcQ5IVd
TXnhaopIIdrhpUD2Mo7ctN+UK6pTQ1dFe2s0zLgdE9jn0nqbBQpu7P9xuvXu
0ntcUCmQ64gTbk7jr95EWH+3pmi8ll2TejYLsTgHM6y21TTp03qQqIgdgAdj
jXTXLwCObgB1A057ViwIx+EFhR78i8aWVywFQJiKWBZsGJBGmD3S7i8BCepG
FmLpGFiLdKZ5m0mTRhmenOJfPhiBhmHDX1VMsfl1P6zU75jn.TnePHBS7CPl
sy3LAKsPKArWDIvOAgSfHQbvTgS2qlwucAtsLt47QHvoFj3GEA5.NtcA4zrz
GG588CwAdwseFcFbCNZcXh0OwI.7IIQBPHIAfy25GLFPMA1XJsjMtnDz5POH
FkD34GQRhBIm3K6TQz5HTPL.LBHI9XH7+Bvgb58P32D6EzGDCgmilMhGGqfG
EK3YB.kbQC6e0v.v7mJr1v1IiTWY+WmMq0KVktFDpRW8HgJ2t2KLwj9u4oYi
Y033mQMHzzpKN.qJx3q9xOv74KRamH78iiZ.QKQQz1fx4BHFKDQTQh.kIRHu
lJkS1Y.G9coyPKFSlLG.oudw8Qs2hFCSatdeWLWbL46j4lUd3.kKdBBlDLl0
SFuvyyMELJQGM87LHS3SzBaAnwL.7hZ.dgpDMbxqx.NOhK96BfKHRWjPizHD
MlR9zjHNxKsl4Fk7o0FC6NF896XMrMrBl3wA1w1sMzVM7ZU2kqs6aWTl8YZd
dc5tlr5xhB6fVVAK6yh80kG2s2lNkmtoft+oaPy3tmxPKESmSC0M6NTlOjRY
cNXTVTRq20NCt6m1xJnxNoapO1rm1rNksU93mJJKqf+.K+66JkMa6sOKTv3s
NLBhmpGcAlj4PYS4Qd9ZviCtY2ku2htxjmJZPzH+3jks64NJ++b9DLtx9ZZy
9xhbniF771s8DfmA+PCDjggx36D6c78wcY7ELNMCbCB6TE6AndFUAHydNo4T
P2W+EQOpjcAcSRg1a05tjiMCfRZcCqjaoi.1ppxh7JqsH8w+co5fZGsVRhw0
j75HUSkIcCOVIlETcAn2Gq0PpGB8c6OF.yWygqLI2ilHDsaUIUbUBEapRyZw
iMUqgno4.b6cHvMqTkX70HHbfdDRTuGQlioy8rf5PbnhxY7J.H.E8SEsFPG6
b51ziEhaGLsKY8n721plixrKm5WpkWLxrjc0r7RtTIFDNjjMhSN.lx1BrMF0
J3oUirY.RB9kIXB2gTbrYSZspDYA0BSBoLkkECY0suB5VQK6JFmehWTTVMMy
Z1t8mYuaJAlGN2Yq3zb6Qtl6s.vPbaS5cC81hzhh1J.CO9GR4rCoBp4FYDTG
ymT9FOjycivIG.5Yz6Y4h8s2+rmCrbVkAD41EkyY6nMhgzDo5p48TZDOpc5V
jNZ5tcqfdnp.rhgK.RQXMBnH18MsKz.zrc.8u6F6Da6xoCnetxpCKsBUwbH1
brpJFZSerJiSOizjyIoaP3G2Uuy9hHqLcLZqi7BM2LnnjbkVpxXY2mXsAu.q
kbhHFo.yzuojtRAlc2Ug4m+MHuWUzL+itm2g5qF8JTMkVXh5KuDqYhzEhamL
w8WAnVY8iNxZHzKQu68N2vMzj9ri7rSICEVu6ORKNZneC+2SqSO3X0M9RrlU
KCq1xWFXwXX+4Kk8maOQ11KUZjyG9.LsyO7CNLtyGskfjVmh37ymxxn5p8+N
3H+xMbGGvqic9fC98NNNxm6WEb9hKGpNRMQHWUqhH21M7uB+RKZn5i7FAnnc
mxGkhx5L69yqa26P6BqsKrT6s7PuybzF4hZ2au4JUW0IBFDAdRsx263tfYLm
o.ADJvSTe.OYFCZxLF7YSKFfZmDsi0yZF+lT73aTqbJeg+xWqTar9IAynTY+
XExQcmnYgRFR9i6Ff6ITmYBEswWmgERf9iBFuaHq+rqxoybCBOWcPd13YnCg
uw5vb7C9SpCsD0TVl6Dss3HKec5gpszT4r4vEj1CioUAaaaM8ejyHgb9zV35
j0NWG.WOZKLyETFt5XkiW+S4k2yc7hAJMEk22xGttePOA0Rjjl6sovK6soly
8nvApWxrWDp+dTV9Z0ILL0P65OEJzF.dBDHZv8vGO7OHcws+x+iFymqfwyPv
wuEBlLGK1qGw+lZgjwsP7q1BweKKL7sPxyBMMbQGX4UkvXKs3WbBV8hf80+O
hBWS17zRqq3YorwuEto4KZ7hm1kLGTIVK5WujvyQRjkPRj4HIukPRdyQR9Kg
j7mijBVBIELGIEtDRJbNRJZIjTzbjT7RHo34HojkPRIyJy8014hfmacb7h26
ZVcMQuUCFLq5UeCIqmI6jWasTnm75pO4UU+zWS8zuh5Se8zpWM8XuVZ4D3W7
0K9e.3LCj1C
-----------end_max5_patcher-----------
In my second (of likely more than a few of) reasking the same or similar questions about pd implementations, is it possible to do this kind of per-sample logic/debouncing in pd without gen~?
History state(0);
History debounce(0);
History prevValue(0);
Param onthreshold(10);
Param offthreshold(5);
Param minslicelength(4410);
if(state == 0 && in > onthreshold && prevValue < onthreshold && debounce == 0)
{
out1 = 1;
debounce = int(minslicelength);
state = 1;
}
else
{
if(debounce > 0) debounce = debounce - 1;
}
if(state == 1 && in1 < offthreshold)
{
state = 0;
}
prevValue = in1;
out2 = state;
I can make sense of doing some of this logic with vanilla objects but setting/incrementing conditional counters is not something I’ve done with msp~-esque objects.
yes. I am (unsuccessfully) offline but I can help next week. for now, check the uber-powerful expr~ - I cannot remember which of my many helpfiles have something doing exactly that or even if that survived the latest edit, but hey, you can do a counter in expr~ and a comparison.
The JIT-NMF example has the circular buffer recording with (very confusingly syntax’d!) fexpr~, so a sample counter makes sense, it just breaks down in how to fire/start/stop logic at audio rate (in pd). I’ll have a further poke today.
Also, turns out that threshold~ would be perfect on its own, as it has built-in time hysteresis, but it outputs message rate, which makes it useless for this purpose.
For the time being I’m doing this, and it’s working pretty well:
Not ideal, but the ADSR~ I have further downstream is triggered via message rate anyways, so just driving it with a parallel fluid.ampfeature~.
