sample catcher

borg · Post by **borg** » Sat May 18, 2002 6:57 am

this is an idea i picked up from the internet http://www.padk-rad.com/

would someone with the right kit be able to make a device that records audio continuously into a buffer which can contain x seconds of audio (and which gets overwritten all the time), and if a button is hit, the device should record the buffer and everything behind the x seconds until the button is hit again (or some kind of 'stop' button).
it shouldn't be something fancy (no filters, keyboard spread, envelopes,...), just a recording device you can let run for hours and hours, and just writes audio to your preferred HD folder whenever you hear something you like. you can sit in the couch, watching a program or a movie, listening to the radio,... (which of course is connected to your computer) with your remote keyboard next to you, on which you have two short cuts to trigger the device.

should be simple, i guess, but i've read in another thread, that sampler atoms are not provided by creamware, so???

Neutron · Post by **Neutron** » Sun May 19, 2002 11:13 pm

I thought something like that would be cool to have. press a button and the previous 5 minutes (or whatever you like) get recorded. it would be better if it was a windows program though, because windows has acess to the hard drive, and the program could be controled by keyboards with "internet keys" or a ATI type remote control, and it save it to a networked machine.

I think it would be easy to do in C++ , I will ask my friend who can program.

Spirit · Post by **Spirit** » Sun May 19, 2002 11:26 pm

That all sounds like lots of work reinventing the wheel to me. Why not just hook up a DAT machine to your audio-visual system and hit record whenever you think something likely is happening? Even a VCR on longplay with a 4-hour tape would do the job.... This is what I always do to catch samples and it works fine.

[ This Message was edited by: Spirit on 2002-05-20 00:28 ]

borg · Post by **borg** » Mon May 20, 2002 3:16 am

to neutron: 'windows' thank you very much, i'm on a mac

but i think you must be right. i thought i saw something like it at a friend's. it was a function provided by his soundblaster card, and it recorded whatever that was going from it's outputs, which is cool, but the buffer is really important in this concept.

to spirit: 'hit record when you think something likely is happenin (on dat or whatever)': you will be too late, or do you mean recording non stop? wow, what a job that would be... listening to your recorded material, selecting the good parts, record them into your computer,...
nonono, when you hear something good, you hit record, and the lets say 15 seconds before you hit record (buffer) and all the rest after it (until you hit stop) will be turned immediately into a wav or whatever...by the time you have auditioned your material and selected it (when using dat or the likes) ,... well, lots of things you can do in the meantime. i, for instance, will have turned my caught wavs into well cut and tailored samples and will be

we have the technology so why do it the hard way? for the sake of art?

it would be nice to create a folder which you write the .wavs to. all .wavs recorded in that folder will have the same, but numbered, name (as the folder)
_________________
andy
 the lunatics are in the hall 

[ This Message was edited by: borg on 2002-05-20 04:23 ]

[ This Message was edited by: borg on 2002-05-20 05:33 ]

Spirit · Post by **Spirit** » Mon May 20, 2002 6:37 am

Well, whatever. I find I can usually tell when things are starting to get interesting, but then maybe I'm looking for different stuff...

Neutron · Post by **Neutron** » Mon May 20, 2002 6:28 pm

Well i am sure it could just as easilly be programmed for a mac!
yust not by my friend

borg · Post by **borg** » Tue May 21, 2002 3:54 am

i'll try a work around i was just thinking about:

set up a project with the STS and a few delays. connect the incoming signal directly to the mixer, and connect it to a 'delay cascade' (one probably will not be enough). connect the last delay to the STS. set the delays so that they just 'delay' the incoming signal. like this you have some time to get to the sampler and record the delayed signal.

i haven't tested it though (just thought about it while reading the thread again). soundquality might not be so good...

Spirit · Post by **Spirit** » Tue May 21, 2002 3:54 pm

I use my Organic Predictotron to analyse the incoming signal. When it passes a certain interest threshold it sends a command to my Humanarm which then activates DATswitch. As stated above there is some annoying latency in this system, but depending on how efficiently my Organic Predictotron is operating the results can be quite good.

I note that in your system you still need to use the Organic Predictotron but that the latency issue would be solved

borg · Post by **borg** » Tue May 21, 2002 4:05 pm

spirit, SFP is bad for you!!! stay away from the damn thing. how could we not have seen this coming?

castol · Post by **castol** » Tue May 21, 2002 5:18 pm

On 2002-05-21 04:54, borg wrote:
i'll try a work around i was just thinking about:

great idea borg! i'm definetly going to play with this.

[ This Message was edited by: castol on 2002-05-21 18:22 ]

spacef · Post by **spacef** » Tue May 21, 2002 11:51 pm

That's a feature that existed in notator (atari, for midi).

Anyway, you better use musys for all your improvisation recall needs. I think it also record audio manipulations. For the rest, I think that's called ""work"" (uuuh,

http://www.muzys.com/product.html

[ This Message was edited by: spacef on 2002-05-22 01:00 ]

spacef · Post by **spacef** » Wed May 22, 2002 12:04 am

There's also "Live" by Ableton which specifically adress these issues of recalling anything you do

http://www.ableton.com/

at least that's from what i understand on their websites I neither tried Live or Musys, but those 2 soft look like really killer apps for impro and recalling what you do while not recording).

Everything you do during the session is recorded and can be post-edited in detail, including all the automation of the mixer and effects.

yummy

[ This Message was edited by: spacef on 2002-05-22 01:07 ]

algorhythm · Post by **algorhythm** » Thu May 23, 2002 7:51 pm

AFAIK - Live doesn't record audio - you drop in wav's and mess with FX, all of which can be automated, but I don't think that it retroactively samples wav or aiff data. I am messing around with the demo, I will look for this "feature" - I am seriously considering buying Live for, um, playing live.

(edit - the following is supposed to be funny - think POW! BANG! BOOM! - not "Spirit I hate you!")
To Spirit - "the King of Ease" - you and your constant ranting about export audio, and yer belittling borg for this wonderful lil "retro-active idea capturer"? Can your Organic Predictotron (tm) sense that my HumanFoot is gonna kick you in your HumanAss in defense of my Minimal-Techno-Wielding "Circuits of Fury" Cy-Borg?
Don't worry too much though, there is still the latency issue of flying to the land down under from USA, so at least you are forewarned, and forewarned is Forearmed. Human(fore)Armed.
please forgive my sarcasm. it is getting late. goodnight

[ This Message was edited by: algorhythm on 2002-05-24 07:00 ]

Spirit · Post by **Spirit** » Fri May 24, 2002 12:03 am

I am not belittling anyone. Above you will find a suggestion followed by humour.

.... Edited for the sake of international harmony .....

[ This Message was edited by: Spirit on 2002-05-24 09:36 ]

spacef · Post by **spacef** » Fri May 24, 2002 12:08 am

i don't know, algo, that was just an idea.

[ This Message was edited by: spacef on 2002-05-24 03:50 ]

borg · Post by **borg** » Fri May 24, 2002 4:26 am

i'm gonna have a shower...

please guys. cool down! i must say that my first reaction to spirit's reply was like 'well, go ahead and start listening to your hours of captured audio if you want and if you can't see the hours gained (and on a carreer it will be days, weeks,...) but i also thought he expressed his doubts in a rather funny fashion, with maybe,yes, a slightly disonant sounding touch. but that didn't bother me that much. this forum is a medium with limited possibilities regarding expression. emoticons can give a phrase a whole different meaning. but the eyes and the intonation of a person speaking cannot be replaced. so i try to look at the words from as many different angels as possible and take the most positive approach... naive maybe, but that's me.

algo, this somewhat heated expression of sympathy towards my person is well taken, but if i felt really attacked or spoken to in an unfriendly way, i'd have reacted myself somehow (have done it only once before, you know who...).
now back to work. i'm trying to master the STS so i can start on the algorhythm remix.

what was this all about? ah yes, the sample catcher. i like that name

_________________
andy
 the lunatics are in the hall 

[ This Message was edited by: borg on 2002-05-24 05:30 ]

spacef · Post by **spacef** » Fri May 24, 2002 4:55 am

....

[ This Message was edited by: spacef on 2002-05-24 07:09 ]

algorhythm · Post by **algorhythm** » Fri May 24, 2002 5:54 am

um, it was a joke guys! joke? joke! as in haha funny

). I like the fact that he is always after ease of use in sequencing, and I have badgered him about it in the past and he took it in good spirit (no pun intended).

I was following up on his rediculously funny "Organic Predictotron theme" to the extreme. It was not serious(!). Jeez. the word "sarcasm" was in my post at the end, but it must mean something different here than elsewhere? Maybe it wasn't obvious enough, sorry. I guess you don't get intentions in forum text, only the words. And my intentions were not violent or evil, more like spoof comic book stuff.

And believe me, violence? Jeez, I must have been taken differently than I intended. I quite literally could not hurt a fly, and definitely not a human being. And God forbid hurting a Spirit!

That's a pretty low blow about being a "silly college boy." Thankfully, it misses, because I am done with school.

However, I am definitely silly sometimes (although it appears not always obviously so!).

If you guys don't turn down the sensitivity knobs on your "Organic-Forum-Vibe-Predictotrons then maybe I'll have to kick ALL of your arses with the "Algorhythm Sonic Bazooka!" (TM)

BE VERY AFRAID.

There is latency with it too, though -> sub 20hZ waves are nice and SLOOOOWWW

I hope that you all take this apology in good spirit ->

_________________
algorhythm</algorhythm - "I enjoy occasionally wandering around randomly, and often find that when I do so, I get to where I wanted to be."

[ This Message was edited by: algorhythm on 2002-05-24 08:32 ]

algorhythm · Post by **algorhythm** » Fri May 24, 2002 6:03 am

On 2002-05-24 05:26, borg wrote:
this forum is a medium with limited possibilities regarding expression. emoticons can give a phrase a whole different meaning. but the eyes and the intonation of a person speaking cannot be replaced. so i try to look at the words from as many different angels as possible and take the most positive approach... naive maybe, but that's me.

Oh, so you'll read Spirit's post from a positive angle but not mine? I'm hurt!

Playing favorites are we?
I am on my way to Europe with my Algorhythmic Sonic Rupturer (its the improved model)!

_________________
algorhythm</algorhythm - "I enjoy occasionally wandering around randomly, and often find that when I do so, I get to where I wanted to be."

[ This Message was edited by: algorhythm on 2002-05-24 07:04 ]

Spirit · Post by **Spirit** » Fri May 24, 2002 8:17 am

Thanks for your gracious reply. I obviously have to study harder about how to spot humour, but this is the problem with filtering human content through machines...

No harm done.

Now, what I think SFP really needs is some way to export audio ...

But really, on the topic of this thread, I was never trying to be flippant, for years I have had devices in the house and on me (currently an Ipaq 3850) ready to instantly record anything interesting. It's a problem I've dealt with for years. My original comment was more of a warning about over-engineering.

I always used to have a VHS tape in the recorder waiting for the interesting bits and would continually be hitting "pause-record" to be ready to grab immediately instead of suffering the agony of the usual 4-second VHS start-up. And yes, there's nothing worse than sitting through hours and hours of audio waiting for that good sample - I know, I've done it many times.

But VHS being what it is there is that horrible take-up lag and roll-back.

So I tried normal tape and then DAT. Both are OK, but if you're recording off TV then you miss the visual clues which enable you to fast-forward through the material. It's all a compromise.

In the end I have become pretty good at predicting and not ending up with hours of garbage....

I think half of what you want is already in Sound Forge 6: one continual take with stop / start pauses being recorded as a series. You can also "record with regions" which will (depending on your chosen settings) divide each word, phrase, beat etc into marked regions - like most beat slicers I suppose. But the beauty of this is that you can then "export regions" and have SF6 automatically name and save each region as a wav file. Nice

After all, you don't want the time you've saved with your precise Sample Catcher to be wasted again when it comes time to chop up the audio and export individual files....

And if you combine this with Kontakt (my love at the moment), then you can just define all these wavs and dump them into a new instrument in one click & drag with everything tuned and spaced on the keyboard. There you can also do "group edits" to every sample....

I'm getting a bit sidetracked here, but SF6 and Kontakt are a fantastic team... Of course now we're talking serious $.

but if you already have this software then the only thing the Sample Catcher really needs to do in this scenario is to delay the original audio feed into SF6. This could be a simple Scope-Fusion device that merely sits between the audio input and wave destination module.

But imagine this: if you're listening to the original source and the delay unit (whatever it happens to actually be) is operating say 10 seconds behind, then you've got your happy 10-second buffer. But what happens when you press stop ? You of course hit "stop" when listening to the original material, not the delayed version. So does the unit know to really stop in 10 seconds time ? If so then what happens if you immediately hear something else interesting and hit "record" again ? Your buffer is still unloading into the recorder, but you now want to start it again... In that case I suppose a second "record" command while the buffer is unloading just cancels the previous "stop" command.

Ideally I think the Sample Catcher would make a very good plug-in for Sound Forge - maybe as some sort of DXi effect ? In fact there may be some real commercial applications of this idea. ( Speaking of which don't many radio stations operate on an 8-second delay in case someone says something libelous or obscene ? They must already have soemthing like this. Maybe even Creamware's radio divison has something like this ?)

But if you wanted to get really simple then the Sample Catcher could simply be a 100% wet delay with 0 feedback and variable time and you could just deal with the messy "stop" situation manually in SF. In my experience the problem with this sort of thing is not so much when to stop recording, it's when to start - usually too late

regards,
King of Ease