Elektron-Users

ZiggY wrote:

rgmccaig wrote:

Black-Man wrote:
http://www.notable.com/index.php?page=about

Hmmm... 808 tight? ROFLMAO.


Here's what your page says:

Roland TR808 NA ? 0.1 ms ? 0.1 ms but slower tempo
(rimshot)



So that's damn tight, yes ... ?!?

0.1ms, indeed what we wish the sps had; or am i missing something?

No, it says NA where the timing measurement would be. .1ms is the variation in measurement accuracy.

Look how the other measurements are shown:

Emu SP12 Turbo 376.1 ms ? 0.1 ms ? 2 ms
(rimshot)

I didn't read the whole thing, but just below it was this:

"The TR808, an analog drum machine, could not be measured for tempo accuracy because it has no digital readout. Strangely, it was not less consistent when parts were added. But its tempo did slow down."


I disagree:: I believe the missing number is the startup time, and the plusminus numbers are the jitter. That's my reading of the page.

(After all, 376.1 ms couldn't possibly be a jitter figure.)

And here's a pretty clear quote from the page:
"The quarter note test showed fairly **stable tenth of a millisecond resolutions** among the SP12, TR808, and RX5 sequencers."

Boy, I hate to argue yet another point ...

*shrugs* I don't know, I've only glanced at it.

"at a slower tempo", does that still qualify as being in time?

ZiggY wrote:

"at a slower tempo", does that still qualify as being in time?

The TR-808 has no way of setting an exact decimal tempo value (no keypad entry or display) when running on its own internal sync. It only has coarse and fine analogue/VCO circuit based rotary tempo setting knobs. So, running under self sync you can only approximate tempo by ear.

Most TR-808's I have worked with (being over 20 years old) have potentiometers that get a bit scratchy and temperamental - mix volume as well as the Tempo rotary controls.

Under self-sync and when adding step/voices or other real-time play pattern changes, unless the Tempo Potentiometers are clean and calibrated with no corrosion 'drop out spots' then fairly large tempo fluctuations can happen.

A quick rapid spin of the Tempo knob usually settles this down as it clears the corrosive build up on the potentiometer and the CV tempo value derived from the VCO becomes stable once again.

I have a feeling this is may be what the Tempo reduction is due to.

Once this is sorted what matters most as far as stability goes is not the actual tempo but step deviation at whatever tempo the unit happens to be running at and a well calibrated TR-808 is indeed tight.

If you drive it as a slave Din Sync off a tight sync clock source - same precision is evident which is the important thing.

Regards - David.

Thunder wrote:
Shut up and die bounce.

This kind of talk is not approriate to this forum. Please refrain from such outbursts in the future.

Glaive
Site Admin

glaive wrote:

Thunder wrote:
Shut up and die bounce.

This kind of talk is not approriate to this forum. Please refrain from such outbursts in the future.

Glaive
Site Admin

Yes, that's correct

Geir Helgi
Awesome Admin

Copied from an email I sent to the Elektron mailing-list a few days ago.

-----
Hi all

To make the timing thread even more interesting I just found out that
when my MD-UW is triggered by my MPC4000 (playing a simple 120 bpm
sequence containing a single MIDI note triggered every quarter note)
it shows a very peculiar timing pattern.

(length in samples between each quarter notes @ 120bpm)
...
22144 x
22016
22017
22016
22144 x
22016
22017
22017
22144 x
22017
22016
22015
22144 x
22016
22016
22017
22144 x
....

Can you see the pattern?
Some other tests confirmed that the 128 samples lag is not coming from
the MIDI out of the MPC.

Considering that in this specific case the MD sequencer is not even
running it looks like the lag comes directly from the MD sound engine.

Interestingly enough the lag stays the same independently of the type
of sound triggered (synthesized sound or UW sample).

I don't wanna draw any conclusion but this test seems to confirm that
the "magic" factor in the sequencer is only partially responsible for
this whole timing issue.

The sound engine clearly plays its part as well.

----------------

To me the fact that the MD shows a regular timing lag even when triggered by a pretty tight sequencer confirms the fact that there must be some problem in the DSP architecture.

Anyone with other possible explanations?

lcvl wrote:

The sound engine clearly plays its part as well.

----------------

To me the fact that the MD shows a regular timing lag even when triggered by a pretty tight sequencer confirms the fact that there must be some problem in the DSP architecture.

Anyone with other possible explanations?

Hiya - matches my tests also using the trigger input method driving the SPS-1 from the Roland MC-4B which under it's own tempo stability tests clocked in at a maximum of 8 samples variation (0.18ms) recording the quarter note output trig/gate pulses.

These are the numbers:

MC-4B

[22381/22380/22382/22387/22382/22380/22372/22379/22381/22380/22382/22378/22381/22371/22387/22382]

Maximum variance between consecutive quarter note intervals: 8 Samples (0.18ms)

SPS-1 (Track 1/MD-Imp - driven from same MC-4B Gate Output)

[22402/22402/22402/22530/22402/22402/22530/22402/22402/22530/22402/22402/22402/22530/22402/22402]

(Note: more precise overall than running under the SPS-1 internal sequencer but with a significant and very regular error of exactly 128 samples every few intervals which equates to 2.9ms)

I have to agree with lcvl here - the 'magic' groove template does indeed seem to have as much to do with DSP/Voice Triggering than it does just in terms of the sequencer stability alone.

I was as equally surprised by the regular instance of one trigger every few having a precise 128 (2.9ms) sample timing error.

In terms of accurate external drum triggering - 2.9 ms drift off the mark is a tad on the wide side I feel.....

If your replacing Kick drums with this method - anything above 1.5 ms you really start hearing tonal/phase changes and once you're up around 3 ms off the actual trigger point the transient punch really starts to suffer especially if you have any bass notes happening at the same time.

David

I prefer the posts about obsessively improving timing compared to the posts obsessively attacking other people's posts.

Nobodies improving anything and I've made 4/5 posts on the subject. Obsessive? yes someone who makes 40/50

Don't crucify people who obviously know enough and care enough to stick their neck out regardless of where they work. That was a low blow Dreg. Shame on you.

Stating facts is a "low blow"? get real he should have declared it and I can go "lower" if you like, ask neonleg but it will be a fact. yawn!

Ziggy and Dreg, your mediocrity is really beyond words.
taking the whole debate to a personal level,

Read some of his replies to see who got personal, and now who's getting personal no nothing dullard.

This site is mainly for Elecktron users wanting further their experience with their equipment, not for some employee of another company to make users feel that their equipment is sub standard. Which it is not!
WITH NO ANSWERS ON HOW HIS PERCEIVED PROBLEM CAN BE FIXED.
I left it alone for along time but at some point I will stand my ground and say CLOSE THIS THREAD!


And guys defending the Timing BS please read posts carefully it shows when you haven't fully read replies or understood them. Slag me all you want I've got broad shoulders and tough skin.
And I'll get back to making some killer tracks and wild sounds ..!., :-x

'And I'll get back to making some killer tracks and wild sounds ..!.,'

ok you go back to doing that and we'll have some intelligent discussion on the forum .

:-o

dreg wrote:
And I'll get back to making some killer tracks and wild sounds ..!., :-x

I've never stopped doing that. It's the way I earn my living.

And that's also the reason (having to use these tools every day for my professional activity) why I'd like to see the MD timing fixed.