Analogue-sequencer

> I was given one of these 
> http://www3.sympatico.ca/sales_emblab/about.htm dyn
> sync to midi converters & ive been trying to drive the p3 
> from my tr808.
> The timing is off though (p3 drifts and resets occassionally) 
> & it only
> works on tbase 32 setting on the p3 when driven by the 4 bar 
> pre-scale on
> the 808.
> Is most likely just a ropey converter, 808 probs or am i 
> missing something?

P3 will sync to midi clock no problem. Midi clock runs at 24 pulses per
quarter note (really clock bytes rather than pulses), and Roland DIN sync
runs at 24 ppqn, so a good DIN sync to midi clock convertor is fairly simple
to put together - you just watch for a pulse, and send a clock byte when one
arrives. Send a start byte when the RUN line goes high, and a stop when it
goes low.
I make a PIC-based DIN to midi to DIN sync convertor myself, and test them
with my 808 driving my P3, and I know for sure that works.
The need to use a higher than normal tbase setting on P3 suggests you may
have the patterns on your 808 scaled at a different timebase than you think.

Have you tried syncing the 808 from the P3 DIN output ?
The 808 can be quite fussy about the timing of DIN sync signals.
I've actually been working on this a lot recently, moreso on the 303's
internal sync signals, but the clock circuitry on the 303 is identical to
the 808. I've been perfecting the DIN sync to midi clock conversion feature
of my 303 midi interface board so that it can precisely drive my P3, and
indirectly my 808. I've also made some improvements to the P3 DIN sync
output - sometimes the 808 needs a little time to realise the RUN line has
gone high before it sees the first clock pulse. The 303 is the same, though
it seems that sometimes the 303 will jump one tick ahead when it shouldn't -
this is causing me occasional problems generating midi clock from the 303...

Cheers,
Colin f

--- In analogue-sequencer@yahoogroups.com, "Colin f" <colin@c...> wrote:
> > I was given one of these 
> > http://www3.sympatico.ca/sales_emblab/about.htm dyn
> > sync to midi converters & ive been trying to drive the p3 
> > from my tr808.
> > The timing is off though (p3 drifts and resets occassionally) 
> > & it only
> > works on tbase 32 setting on the p3 when driven by the 4 bar 
> > pre-scale on
> > the 808.
> > Is most likely just a ropey converter, 808 probs or am i 
> > missing something?
> 
> P3 will sync to midi clock no problem. Midi clock runs at 24 pulses per
> quarter note (really clock bytes rather than pulses), and Roland DIN
sync
> runs at 24 ppqn, so a good DIN sync to midi clock convertor is
fairly simple
> to put together - you just watch for a pulse, and send a clock byte
when one
> arrives. Send a start byte when the RUN line goes high, and a stop
when it
> goes low.
> I make a PIC-based DIN to midi to DIN sync convertor myself, and
test them
> with my 808 driving my P3, and I know for sure that works.
> The need to use a higher than normal tbase setting on P3 suggests
you may
> have the patterns on your 808 scaled at a different timebase than
you think.
> 
> Have you tried syncing the 808 from the P3 DIN output ?
> The 808 can be quite fussy about the timing of DIN sync signals.
> I've actually been working on this a lot recently, moreso on the 303's
> internal sync signals, but the clock circuitry on the 303 is
identical to
> the 808. I've been perfecting the DIN sync to midi clock conversion
feature
> of my 303 midi interface board so that it can precisely drive my P3, and
> indirectly my 808. I've also made some improvements to the P3 DIN sync
> output - sometimes the 808 needs a little time to realise the RUN
line has
> gone high before it sees the first clock pulse. The 303 is the same,
though
> it seems that sometimes the 303 will jump one tick ahead when it
shouldn't -
> this is causing me occasional problems generating midi clock from
the 303...
> 
> Cheers,
> Colin f

Sussed it - dodgy power supply on the converter. 
Swapped it out & its running lovely.

So should the p3 drift slightly along with the 808 in a musical
fashion? ;)

/m

> So should the p3 drift slightly along with the 808 in a musical
> fashion? ;)

There shouldn't be much drift in the 808 DIN clock.
The tempo may gradually change with temperature, since the clock is built
from some logic gates, resistors and caps. The thresholds will move slightly
as things warm up, but not so it should really be noticeable as any kind of
'feel'.
The way the 808 CPU resonds to the DIN clock causes a timing variation that
you may notice...
This effect is more noticeable on the 303, which uses the same CPU (with
different code, obviously) and the same method of following the clock.
There is an extra timer circuit interrupting the 808, 303 and 606 CPUs every
2ms or so.
The DIN clock line is checked to see if it has gone high during the
interrupt routine triggered by this extra clock.
So in effect, the 808 CPU quantises the incoming DIN sync to 2ms intervals.
This introduces a very subtle timing shift depending on the underlying tempo
of the clock signal. You very likely wont notice it consciously, although
you can notice the effect on the 303 if you set up a pattern to repeat the
same note every 16th, set up a sound with a very snappy envelope, then
gradually increase the tempo. There is a cyclic shift in the sound as the
relationship between the tempo clock and fixed interrupt clock changes.

Cheers,
Colin f

> Dude; sometimes you scare me.

Sometimes I scare myself.
I discovered tonight if you spend too long listening to variable high-tempo
resonant clicks, then take your headphones off, you can find the sound of
reality has a slight flange. Nice ;-)
This was discovered while repeatedly starting a 303 pattern at maximum
tempo, then bringing the tempo right down to check sync accuracy.
On the plus side, I found a cap in the right place fixes the sync bug on the
303.

Cheers,
Colin f

> Dude; sometimes you scare me.
> 
> Paul

..you beat me to it. get a freak talking about a passion..;-)

On Thu, 30 Dec 2004 21:09:45 -0000, "Colin f" <colin@colinfraser.com>
wrote:

>So in effect, the 808 CPU quantises the incoming DIN sync to 2ms intervals.
>This introduces a very subtle timing shift depending on the underlying tempo
>of the clock signal. You very likely wont notice it consciously, although
>you can notice the effect on the 303 if you set up a pattern to repeat the
>same note every 16th, set up a sound with a very snappy envelope, then
>gradually increase the tempo. There is a cyclic shift in the sound as the
>relationship between the tempo clock and fixed interrupt clock changes.

Dude; sometimes you scare me.

Paul

---
Paul Nagle / Soft Room Music
Email: paul@softroom.co.uk www.softroom.co.uk
                           www.BogusFocus.com

I was given one of these http://www3.sympatico.ca/sales_emblab/about.htm dyn
sync to midi converters & ive been trying to drive the p3 from my tr808.
The timing is off though (p3 drifts and resets occassionally) & it only
works on tbase 32 setting on the p3 when driven by the 4 bar pre-scale on
the 808.
Is most likely just a ropey converter, 808 probs or am i missing something?

Thx.

/m

Colin,

> Sometimes I scare myself.
> I discovered tonight if you spend too long listening to variable high-tempo
> resonant clicks, then take your headphones off, you can find the sound of
> reality has a slight flange. Nice ;-)

hehe, never mind mind altering substances, buy colins 'reality adjustment audio
CD!'

> This was discovered while repeatedly starting a 303 pattern at maximum
> tempo, then bringing the tempo right down to check sync accuracy.
> On the plus side, I found a cap in the right place fixes the sync bug on the
> 303.

hehe, you can spot the people who are obsessive about there gear and how it
works :-)

Paul