[sdiy] Fwd: Odp: Re: Analysis of the TB-303 CPU timing

Tony Kalomiris weplar at gmail.com
Fri Mar 17 19:07:26 CET 2017


---------- Forwarded message ----------
From: Tony Kalomiris <weplar at gmail.com>
Date: Fri, Mar 17, 2017 at 2:06 PM
Subject: Re: [sdiy] Odp: Re: Analysis of the TB-303 CPU timing
To: chromatest at azburners.org


Having owned and made music with these Roland machines there is definitely
a 'groove, 'feel' or what have you, that is part science part perception,
but it's there if you listen to enough music, especially DIN synced studio.

 I always was suspicious of the clones that used a generic Roland style
pattern sequencer, and even the so-called 'replacement upgrade' cpus for
old monos (respect to the programmers) - a modern processor running faster
with MIDI does not mean it's tight, well written optimized code. One of the
reasons I didn't jump on the Europa bandwagon -and partly because I had
been writing my own.

I am no engineer (I dropped out) , but how can people say there is no
difference between a micro with an LC interrupt oscillator, and an external
Tempo oscillator, plus gate delays, is beyond me.

I am in agreement with Colin Fraser here, that in isolation, it is hard to
detect. Add an 808 synced in there and some more variation in patterns and
it's a whole other ball game. Sadly I don't own these instruments or I
would do the tests as well. If all this sounds like more rubbish, then
promptly ignore the subjective opinion.

p.s. I had a look at the ROM area mask of that link, and counted 128 x 128
(2K x 8). Question, can some semiconductor physicist here explain please
the short and longer 'bars', I am trying to make sense of the 1's and 0's,
to map this out. Besides being a work of art, it's highly meditative,
especially while listening to Aphex' SAW Vol.II ;) Yes I am a masochist, I
read out the SH-101 bit by bit before I wired something up to do it
automatically.

Tony K

On Fri, Mar 17, 2017 at 1:09 PM, Chromatest J. Pantsmaker <
chromatest at azburners.org> wrote:

> Agreed, except that many many people claim that a an original 303 acid
> line sounds better than any clone... In isolation...
> On Mar 17, 2017 1:07 PM, "Colin f" <colin at colinfraser.com> wrote:
>
>>
>> > That's exactly the point of this test.
>> > People claim that nothing sounds like a 303 because nothing has the same
>> > processor of the 303.  If people can't hear the difference in 4 CPUs
>> designed
>> > to be listened to, then that's another step closer to calling it
>> nonsense.
>>
>> Demonstrating that people can't hear the timing variations of a 303 in
>> isolation proves nothing, other than that people can't hear the timing
>> variations in isolation.
>> It would be like saying that multiple DCOs with a common clock sound
>> exactly like multiple free-running VCOs after listening to static saw waves
>> from each in isolation, and not considering the behaviour when more than
>> one oscillator is playing.
>> I'm not saying the 303's timing definitely does make a difference.
>> Just that the scope of this test is narrow.
>>
>> Cheers,
>> Colin f
>>
>>
>>
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy
>>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20170317/f207ce1c/attachment.htm>


More information about the Synth-diy mailing list