02:21
kyak has quit [Ping timeout: 252 seconds]
02:22
kyak has joined #m-labs
03:31
nicksydney has quit [Remote host closed the connection]
03:32
nicksydney has joined #m-labs
07:06
sh4rm4 has quit [Ping timeout: 252 seconds]
07:31
<
playthatbeat >
ysionneau - i wanted to see if it did away with the F1 bug, then i could have 3 video input sources
07:31
<
playthatbeat >
it sisnt
07:32
<
playthatbeat >
in fact the earlier version lost some commands so my patches didnt work
07:32
<
playthatbeat >
but after brick/unbrick it is alot quicker.. ftp, etc. is all snappier
07:33
<
playthatbeat >
i need now to investigate the MIDI implimentation
07:33
<
playthatbeat >
i want to map CC to values in the patches
07:34
<
playthatbeat >
having the 3 bands of audio as modulation sources in a patch is not enough and a bit samey after a while
08:00
lekernel has joined #m-labs
08:26
<
larsc >
meh, all the bistreams that come out of vivado seem to be broken for me
08:47
<
lekernel >
playthatbeat, just copy and paste from wikipedia and other online resources
08:52
<
larsc >
are those the people that publish random wikipedia articales as books?
08:53
<
larsc >
so they took the milkymist wiki and made a book out of it?
08:53
<
larsc >
or your thesis
08:54
<
lekernel >
apparently... I guess a sizable part of their revenue comes from authors who'd shell out $67 to find out, but they don't include me :-)
09:10
<
playthatbeat >
$67 is a bit steep
09:11
<
lekernel >
that's how it works, I don't think they sell a lot of copies. also, it's print-on-demand.
09:11
<
playthatbeat >
we shoud do a book on them and their scam
09:11
<
lekernel >
they basically churn out thousands of titles, most of them never selling, and when one sales happens there's a nice profit margin
09:12
<
lekernel >
bah, they're not even the worst "academic publisher". look at elsevier :-)
09:12
<
playthatbeat >
ironic that they sell via ebay, and anyone that can see ebay and use paypal can surely use a search engine...
09:12
<
playthatbeat >
wait, is 'ironic' the right word
09:13
<
playthatbeat >
maybe i meant 'cheeky' or 'bullshit'
09:13
<
playthatbeat >
never could correctly apply that word
09:14
<
lekernel >
many elsevier books are over $400, and that's just the tip of the iceberg
09:16
<
playthatbeat >
well i suppose it's a legitimate 'transcription service'
09:17
<
playthatbeat >
if you didnt have a printer or hands it would be useful
09:18
<
playthatbeat >
also: is there anything written re: the milkymist 1 and midi?
09:18
<
playthatbeat >
or is it all obvious if i dig about
09:19
<
playthatbeat >
i'm looking to ass CC values into patches
09:19
<
playthatbeat >
*add
09:22
<
playthatbeat >
morning
09:23
<
ysionneau >
08:33 < playthatbeat> i want to map CC to values in the patches < what does "CC" mean?
09:24
<
ysionneau >
I played a bit with the M1 and midi-usb
09:29
<
playthatbeat >
oh CC is 'continuous controller'
09:29
<
playthatbeat >
like the mod wheel
09:29
<
playthatbeat >
or even to some extent pitch bend or aftertouch
09:29
<
playthatbeat >
but there are like 100 CC numbers that can have value 0-127 available in midi
09:30
<
playthatbeat >
so to be able to use a Midi CC source instead of (say) the BASS value in a patch would be awesome
09:30
<
playthatbeat >
then you could build proper animation sequences using a midi sequencer
09:31
<
lekernel >
that's possible, yes
09:31
<
playthatbeat >
orly?
09:31
* playthatbeat
sweats
09:31
<
lekernel >
you have midi0, midi1, ... variables in patches
09:31
<
lekernel >
which you can assign to CC channels in the "MIDI settings"
09:32
<
playthatbeat >
looooooooovely
09:32
<
lekernel >
wpwrak made a more complete (and complex) system later... don't know if you have that one already
09:32
<
playthatbeat >
im on the march 1st firmware from the website
09:34
<
ysionneau >
yep I confirm I used the "midi0 midi1" stuff
09:34
<
ysionneau >
as well as the new wpwrak system
09:35
<
playthatbeat >
the midi settings dialog seems only to allow mapping note value to patch change
09:35
<
ysionneau >
and I could even control a DMX RGB led with the midi controller (via the M1)
09:36
<
ysionneau >
that's the "new" system
09:36
<
lekernel >
fortunately, it's well documented
09:37
<
ysionneau >
what's pretty nice with wpwrak's work it's that it's usually pretty well documented
09:37
<
ysionneau >
at least in an email
09:37
<
lekernel >
there's the pdf doc as well
09:40
<
playthatbeat >
ok this all looks very interesting
09:42
<
playthatbeat >
and this all runs on the last official downloadable firmware for the m1?
09:42
<
playthatbeat >
or do i need to compile a fork?
09:42
<
playthatbeat >
or whatnot?
09:44
<
lekernel >
if you don't have the midiX<->channel number mapping in the MIDI settings, then you have this new system
09:47
<
playthatbeat >
all i have in midi settings is 'patch switch channel' and 2 fieds for setting note to filename bindings
09:47
<
playthatbeat >
oh and 'last active controller:'
09:49
<
ysionneau >
playthatbeat: what I do is I open the mini dialog, I move a fader on my controller
09:49
<
ysionneau >
and it will show up in the dialog with the identifier and the value
09:49
<
ysionneau >
I write down the identifier
09:50
<
ysionneau >
and I can then use it in the Patch (with wpwrak syntax) to assign the value to a variable
09:50
<
ysionneau >
-mini dialog+midi dialog
09:50
<
playthatbeat >
kk i get you
09:50
<
playthatbeat >
let me try some midi in and see what i see
09:51
<
playthatbeat >
to turn voltages into midi cc
09:52
<
playthatbeat >
my voltages come from modular gear clocked/synced to audio
09:53
<
ysionneau >
is it midi-usb?
09:53
<
playthatbeat >
standalone
09:53
<
playthatbeat >
no computers
09:53
<
ysionneau >
never tried the midi connectors
09:54
<
ysionneau >
10:53 < playthatbeat> no computers < I meant, is it using a USB connexion?
09:54
<
ysionneau >
for instance my korg nanokontrol2 is using USB (midi over usb) but I can use it standalone with no PC
09:54
<
ysionneau >
and just the M1
09:54
<
playthatbeat >
the midipal is standalone, no usb
09:55
<
playthatbeat >
it has minijack ins (up to 8) for voltages 0-5v and you can translate these to midi CC or events
09:55
<
playthatbeat >
a reverse midi to CV :)
09:55
<
playthatbeat >
so you can feed it things like envelope follower outputs
09:55
<
ysionneau >
sounds nice
09:55
<
ysionneau >
but you kind of lost me
09:55
<
playthatbeat >
or the out of an analogue sequencer
09:56
<
ysionneau >
I don't know much about midi stuff
09:56
<
playthatbeat >
well how about his:
09:56
<
ysionneau >
or electronic music stuff
09:56
<
playthatbeat >
you know the way the M1 can respond to audio level?
09:57
<
playthatbeat >
how about having it respond to voltage levels instead
09:57
<
playthatbeat >
and those voltages can be generated by [insert device here]
09:58
<
playthatbeat >
i just use the Midipal to convert those voltages into midi
09:59
<
playthatbeat >
even think about using a note value in a patch
09:59
<
playthatbeat >
then you could play a midi keyboard and change the patch
10:00
<
playthatbeat >
the nice thing about midi is you get to layer control values and can loop them at different lengths
10:01
<
playthatbeat >
so a 4 bar loop and a 3 bar loop at the same time can appear to the naked eye to be alot more complex then it actually is
10:02
<
playthatbeat >
or 3/4 and 4/4 together is quite pleasing
10:02
<
playthatbeat >
etc.
10:03
<
playthatbeat >
im just tryign to get away from stuff pumping to the beat
10:03
<
playthatbeat >
*trying
10:05
<
ysionneau >
ok I get it now :)
10:06
<
ysionneau >
what kind of devices generate voltages to be used for visual effects?
10:06
<
ysionneau >
or music
10:09
<
playthatbeat >
step sequencers
10:11
<
playthatbeat >
this is a favorite
10:11
<
playthatbeat >
it follows MTC in
10:11
<
playthatbeat >
from this is use it's gate outs to advance [other] sequencers
10:13
<
playthatbeat >
these in turn send out voltages to the midipal which then makes a pattern of midi CC
10:14
<
playthatbeat >
and i use those CC to make stuff happen in various devices and software
10:15
<
playthatbeat >
eg: assigning a CC to zoom level or 3d tilt in Vjamm3 is amusing
10:15
<
playthatbeat >
sure beats 'pumping to the beat'
10:17
<
playthatbeat >
anyway i have the m1 showing CC in now nicely..
10:18
<
playthatbeat >
thanks guys..
10:18
<
playthatbeat >
new lease of life for the m1 here
10:19
* playthatbeat
buys everyone drink
10:23
<
ysionneau >
I'm glad you won't put the M1 in the bin !
10:37
<
playthatbeat >
now to convert the stock patches to midi cc modulation
10:38
<
playthatbeat >
subtle changes in all of them will make for a more syncronised display
10:39
<
playthatbeat >
btw - the key to this is using an audio-mtc engine
10:40
<
playthatbeat >
hardware that listens to the music and on the fly generates mtc in sync
10:41
<
playthatbeat >
this means all you need is an audio feed and everything will sync as long as you kick it off on a downbeat
10:41
<
playthatbeat >
then you can just work on pulling faces and tweaking knobs and faders
10:42
<
playthatbeat >
even if the ausio source changes tempo you can rely on your timing source to chase it
10:42
<
playthatbeat >
*audio
10:43
<
playthatbeat >
you can set scene changes every 4 bars or whatever
10:44
<
playthatbeat >
some basic patching of sequencers can make for fractal-like evolution
10:52
<
ysionneau >
what's a "bar" ?
11:05
<
playthatbeat >
a bar is 4 kick drums
11:05
<
playthatbeat >
you count 1,2,3,4 to the music
11:05
<
playthatbeat >
thats 1 bar
11:05
<
playthatbeat >
there are 4 beats in a bar
11:05
<
playthatbeat >
this gives you 4/4 timing
11:06
<
playthatbeat >
and you if you use the number 4 as a modulus in patches or sequences
11:06
<
playthatbeat >
then you get sympathetic changes
11:06
<
playthatbeat >
4 is the magic number lol
11:06
<
ysionneau >
if the music timing is 4/4 then indeed it's natural to use this timing
11:06
<
playthatbeat >
listen to modern music - it is all made up of patterns of 4
11:07
<
ysionneau >
if it's 6/8 then I guess it's more 3
11:07
<
playthatbeat >
6/8 or 3/4 is fun
11:07
<
playthatbeat >
running 3/4 alongside 4/4 makes for interesting interactions
11:07
<
playthatbeat >
like every 6 bars the patterns seem to sync
11:08
<
ysionneau >
interesting :)
11:08
<
playthatbeat >
triplet timing is big these days in some music
11:08
<
playthatbeat >
like psy trance
11:08
<
playthatbeat >
so you get sections that drop into 8t timing
11:09
<
playthatbeat >
3 beats per bar
11:09
<
playthatbeat >
'swing'
11:09
<
playthatbeat >
this can confuse some BPM engines
11:09
<
playthatbeat >
which is always lulz
11:20
<
playthatbeat >
unrelated q: do you get the patch name on screen if you hit F1 like i do?
11:20
<
ysionneau >
I never tried
11:20
<
ysionneau >
+ I don't have any working keyboard for the M1
11:20
<
ysionneau >
I only have USB mouse ^^
11:20
<
playthatbeat >
orite
11:21
<
playthatbeat >
well F1 F2 F3 should switch between the 3 video inputs
11:21
<
playthatbeat >
this works fine, but F1 always splashes up the patch name :(
11:21
<
playthatbeat >
weird
11:22
<
ysionneau >
what do you mean by "3 video inputs" ?
11:23
sh4rm4 has joined #m-labs
11:23
<
ysionneau >
I thought there was only 1 video input (the camera)
11:44
<
playthatbeat >
no the phono sockets on the mm can be 3 composite ins
11:45
<
playthatbeat >
or i think a single s-video or single componant
11:45
<
playthatbeat >
while it is running you can flip between the 3 ins using the keyboard f1 f2 f3 keys
11:45
<
playthatbeat >
very useful
11:45
<
playthatbeat >
so you can flip from camera to feedback or whatever
11:46
<
playthatbeat >
ok i have this responding to midi like a champion now
11:58
<
lekernel >
playthatbeat, :) cool. please send videos/pictures :)
11:59
<
playthatbeat >
it's coping well with 5 CCs at once
11:59
<
playthatbeat >
1 is an envelope follower, 4 are from sequencers
12:01
<
playthatbeat >
ok this is POC for me
12:01
<
playthatbeat >
now ill edit up all the stock patches to follow midi CC and build a bank of midi-synced fun
12:02
<
playthatbeat >
THEN WE SHALL SEE
12:02
* playthatbeat
adds reverb for a long cartoon vbad-guy laugh
12:29
mwalle has quit [Quit: WeeChat 0.3.8]
13:10
<
ysionneau >
yes please send pictures/videos :)
13:16
<
playthatbeat >
im getting too sucked in to tweaking
13:16
<
playthatbeat >
er..
14:05
lekernel has quit [Ping timeout: 252 seconds]
14:18
lekernel has joined #m-labs
14:23
<
lekernel >
quizz: what does Verilog print?
14:23
<
lekernel >
$display(2'd2 + 2'd2);
14:23
<
lekernel >
$display(2'd2 + 2'd2 >= 2'd1);
14:24
<
lekernel >
(hint: it's as fucked up as always)
14:27
<
lekernel >
and for $display(3'd4 >= 2'd1) too
15:02
<
GitHub91 >
misoc/master e464935 Sebastien Bourdeauducq: downscaler: add chopper module
16:08
playthatbeat has quit [Ping timeout: 248 seconds]
16:10
playthatbeat has joined #m-labs
16:16
<
ysionneau >
$display(2'd2 + 2'd2); <= 1 ?
16:17
<
ysionneau >
or no, 0
16:31
<
ysionneau >
so the 2 bits value gets transformed into 3 bits value
16:32
<
larsc >
well that's expected, isn't it
16:32
<
larsc >
addition adds one bit
16:36
<
lekernel >
yes, this one is actually sane
16:36
<
lekernel >
but then
16:37
<
lekernel >
3'd4 >= 2'd1 is 1, which is also what you would expect
16:37
<
lekernel >
and 2'd2 + 2'd2 >= 2'd1 is 0 :)
16:39
<
larsc >
and is that expected behaviour?
16:46
<
lekernel >
I guess this is another pedantic imbecility of the verilog standard, but I can't pinpoint it yet...
16:56
<
lekernel >
"This description is left over from Verilog-1995, before there were signed operands. It is simply wrong now. [...] This description wasn't even correct in Verilog-1995."
17:07
Alarm has joined #m-labs
17:09
Scopeuk has joined #m-labs
17:09
juliusb_ has quit [Ping timeout: 260 seconds]
17:09
ysionneau has quit [Ping timeout: 260 seconds]
17:09
juliusb has joined #m-labs
17:14
ysionneau has joined #m-labs
17:15
<
lekernel >
hmm, the standard says the smaller operand should be extended
17:15
<
lekernel >
so it should work...
17:15
<
lekernel >
maybe it's just a iverilog bug
17:18
<
lekernel >
xilinx isim produces 2'd2 + 2'd2 = 0, 3'd4 >= 2'd1 = 1, 2'd2 + 2'd2 >= 2'd1 = 0
17:29
playthatbeat has quit [Ping timeout: 264 seconds]
17:33
<
lekernel >
I think I understand - the width of the right handside of the comparison determines the width of the addition
17:35
<
lekernel >
"Verilog's rules for operand expansion reflect how hardware works" - WTF
17:38
sh4rm4 has quit [Remote host closed the connection]
17:45
sh4rm4 has joined #m-labs
18:35
mumptai has joined #m-labs
18:36
mami_ is now known as mami
18:44
playthatbeat has joined #m-labs
19:11
playthatbeat has joined #m-labs
20:17
<
ysionneau >
ah xilinx just did the tape out of their 20 nm Virtex ultrascale
20:25
<
ysionneau >
playthatbeat: the result looks good, well done :)
20:25
<
ysionneau >
so you're not at all using the audio input ?
20:25
<
ysionneau >
only input is midi then?
21:22
<
playthatbeat >
all midi
21:27
<
playthatbeat >
im working thru the patches now converting them to midi cc
21:27
<
playthatbeat >
replaceing the time,bass,mid and treb variables with midi CC
21:28
<
playthatbeat >
because i can
21:28
<
playthatbeat >
just opened up the raindance.fnp and i see it has already got extensive midi support
21:30
Alarm has quit [Quit: ChatZilla 0.9.90.1 [Firefox 26.0/20131205075310]]
22:40
lekernel has quit [Quit: Leaving]
22:50
mumptai has quit [Quit: Verlassend]
23:57
nicksydney has quit [Remote host closed the connection]
23:58
nicksydney has joined #m-labs