Re: DJRND2

[Kim Flint <kflint@annihilist.com>]

At 7:48 AM -0700 7/9/99, PERILLE wrote:
>> In fact, your device still requires more preparation than any of these
>> earlier devices, as the user must know the BPM ahead of time and type it
>> in! For Jamman, echoplex, etc, you don't need to know anything about the
>> source before you start looping it. You just tap the record in time and
>> it's immediately sampled and looping.
>>
>> I don't mean to put you down, you are doing a very cool thing. Just keep
>> the history straight and respect what has come before you! :-)
>>
>
>For live musicians, A tap record method is certainly sufficient since
>they can still manage to sync on their own.

Sounds like a bit of dis of live musicians, eh? :-)  Believe it or not,
many live musicians are very demanding about accuracy. Tapped recording can
give you excellent results for this, assuming the user has a sense of
rhythm. Most dj's I've encountered seem to have an excellent sense of
rhythm, so it seems fine to me.

>But a more accurate approach should be usefull for DJs, and this is
>maybe why existing tiny loopers integrated in DJ mixing tables have not
>been a big success.
>
>A DJ can still make one loop at a time. If he wishes loop polyphony, he
>still has got a loop a editing phase to do before triggering back his
>loops. But a BPM based cyclic method lets him work just only with one
>parameter to keep sync in real time, which is also a more similar
>approach to his mixing art.

Sure, entering a BPM is useful too, assuming you actually know the BPM
ahead of time, and you haven't changed the speed knob on the turntable.
(and the turntable is actually calibrated.)

My point though, is you are claiming your method as being superior because
it does not require preparation. But it does require preparation! You have
to know the BPM and enter it ahead of time. This is certainly superior to
the idiotic phrase samplers that are built into some dj mixers, but it is
not as real-time friendly as a tapped record method. For tapped-record, all
you need ahead of time is a decent sense of rhythm. If you don't even have
that, OB echoplex and Lexicon Jamman both have sync possibilities to help
you get it perfectly accurate.


>
>> >I have reached a certain amount of experience in the use of BPM 
>looping,
>> >and this is one of the reasons I could have my looping method patented.
>> >As I told you just before, I feel I come out to soon in comparison with
>> >the other existing products, but I still hold on.
>>
>> Well, we know the feeling of coming out too soon. ;-)
>>
>> I'm not sure about your claim for patent. You have some interesting new
>> ideas, certainly! but the basic method of sampling the loop to the BPM 
>is
>> very much like what we invented for the echoplex sync to midi clock in 
>'94.
>> We have the same notion of the loop cycling around to the tempo, so that
>> would likely be prior art. The primary difference is we get the BPM from
>> the midi clock or an analog sync pulse, and you get it from the user
>> entering in on the front panel. And you have some interesting variations
>> with the gear ratio ideas, as opposed to our approach to that with 
>multiply
>> and insert and 8ths/cycle. Otherwise it is quite similar....
>>
>
>In fact, the methods are lightly different.
>
>Using midi clock or an analog sync pulse to determine the loop lenght
>deals with terms of physical durations, even if it comes out from a BPM
>based metronome. It is a kind of time quantized method where is counted
>a number of sync pulses for a loop recording. This also means that the
>loop to record is still determined with A start and B stop points, the
>starting point corresponding to the next incoming sync pulse immediatly
>after the rec keypressed. And then the recorded loop runs with a given
>sampling rate and a finite number of x(n) samples going from A to B.

not necessarily!

>But a permanent cyclic loop straightly defined in terms of tempo and
>bars set you free from any need of external sync, and rec start can
>appear wherever around the wheel, the loop being perfectly made once a
>whole revolution made. The method works in terms of Omega/phi (Omega =
>pulsation = tempo) and (Phi = phase = differential bar synchronization),
>Omega commun to every loop, Phi given for each loop. It is a BPM
>gearwheels recording machine. No start/end points. Free running in your
>own beat where you can record at any time during your playback. An
>eternal and fiendish grooving method which allows BPM changes.

Yes, this is what I'm trying to tell you. This is almost exactly how the
echoplex works, and we invented this years ago. Of course, we don't force
our users to remember angular geometry calculations from their high-school
mathematics courses! :-) So our terminology is different, but the method is
very similar.

(although, I'm ashamed to admit, the current shipping soft has a slightly
disabled sync feature, in that we force the record to start at the
startpoint of the sync. The underlying software is actually capable of
starting anywhere in the cycle and maintaining sync. we kept it like the
previous version for, er legal reasons. And it is capable of starting
immediately after the sync starts. Since then, we've done a lot more with
this....the sync capability in the next version is fucking amazing, if I
may be blunt. :-)

kim

______________________________________________________________________
Kim Flint                   | Looper's Delight
kflint@annihilist.com       | http://www.annihilist.com/loop/loop.html
http://www.annihilist.com/  | Loopers-Delight-request@annihilist.com