Elektron-Users

Hi all - I love my MD SPS-1 to bits - no question - so this is in no way a rant/slag off of the MD or Elektron who I have a great respect for on all counts - I've done some fairly detailed testing and found the step/tempo clock jitter on my MD/SPS-1 is not that crash hot. I wondering if anyone else might verify this by doing a few test just in case it's my machine that has the issue? The numbers and testing details/instructions are on this page of my site: http://www.innerclocksystems.com/index.asp?action=page&name=34
I'd love to know if any SPS-1 owners out there get similar results. All the very best - David.

The testing method if flawed. You cannot use you DAW software as a reference as (ALL) computer's timing is absolutely pathetic.

There is also error introduced as the graphical wave form drawn by your DAW will be an average of sample at every plot point. And another error for the markers position. Although the computer timing error will be by far the largest.

So, I advice you buy a 100MHz+ DSO (Digital Oscilloscope) and repeat your tests.

I disagree totally. Even if the errors are out by even a few samples the DAW/Windows/SoundForge Method is more than accurate enough to show timing deviations and drift over relatively long periods (as in tempo testing) - You don't need a CRO to show timing flaws that you can hear. If Sound Forge was that inaccurate as a sample editor to begin with - no one would use it. Any many professionals do. Regards - David

Also - yes - computer [Serial] midi I/O timing is flawed and pathetic - I agree but that is not being used here for these test. This is 100% audio from MD outputs to Firewire Soundcard direct into Soundforge. If Digital Audio was that poor as to misrepresent such corse variations in signal it would be unusable to record music in the first place - which we all do and have done for quite some time. Regards again - David.

And - Elektron have looked at these numbers and acknowledged the data and made no comment as to the nature of my test methods either - regards again David.

Sorry to go on about this but it is worth clarifying that these results are indeed valid considering Jedi's quick dismissal of my initial post. Take a close look at the SPS-1 test numbers: you will see that the different period jitter/offsets between steps is far too regular for a flawed analysis method.

Similarly all the other devices I have looked at so far. Sound Forge accurately represents the timing characteristics - certainly well enough to show tempo timing stability over a given period.

To prove this even further - I use the same method to test Sync between two machines. If I lock two machines via Midi Clock and pan one machine hard left and the other hard right to a Stereo recording in Sound Forge and record a few bars I can zoom in a see the sample offset between them.

If I use the Sync Shift or use another method of sync/offset [say Midi Clock - FSK - Midi Clock] so I can delay the FSK using a standard DDL (Digital Delay Line) with enough precision (say a TC-1280) - I can adjust the slave device start time to compensate for the initial sync lag reported in Sound Forge by setting the delay time to the exact number of samples the two devices are out by.

If I then record the two machines again into Sound Forge [L/R] and zoom in on the waveforms I see an almost perfect waveform alignment - give or take a few samples error possible introduces by Markers/DAW or Windows.

And more importantly than the numbers themselves - it sounds better when the numbers are spot on.

Hope that clears things up.

Regards David.
www.innerclocksystems.com

Maybe they had nothing nice to say about your method of "testing" so they chose to say nothing?

"If Digital Audio was that poor as to misrepresent such corse variations in signal it would be unusable to record music in the first place"

People recorded on tape for years... Tape was/is so famous for its speed up/down during record and playback that people have gone to great lengths to creat plug-ins simulating it.


Based on my professional experience it is my opinion that whilst the machines in questions quite possible have timing issues the method outline to quantify these timing issues is flawed.

Again - I don't want to start a nasty dispute here but in this case you are incorrect in your assumption. Elekton has indeed acknowledged the numbers in question - and to quote 'David - that was thorough testing!' in my last reply from Elektron Support.

I'm not sure what recording to tape or plug in simulation of Tape Wow/Flutter has to do with this thread....?

The best analogy I have for these tests is like a Dyno for car engine performance/power output. The Dyno will not provide accurate kilowatt/Horse Power engine output but it will show if an engine is providing more or less power output after modification and this is still a valid test.

Again - have a look at the numbers - they are too accurate and reliable/repeatable to be ignored.

And again - even if they are out by the odd sample given Windows/DAW/Graphic errors - they still show reliable deviation that is worth taking into consideration.

I'm not sure why you seem determined to negate my finding so strongly. I have spent along time discussing timing issues in sequencers with many individuals with far more scientific background than my own and all feel these results are indeed valid along with the method of testing to achieve them.

In the end, as I said on my test page - I am only interested in getting timing better across the board for all users concerned and this can only be a good thing can't it?

Again - I love Elektron and my SPS-1 without question.

If you have a Dual Trace Storage CRO - I'd love if you could verify/improve on my test results?

Regards David.

Hey David,
thanks for testing/posting
great idea for a website, I'll have to give it a good read later.

do you know if there's a quality stand-alone midi clock out there ? if I slaved all my 'grooveboxes' to a single clock (using a splitter on the midi/in for each unit-- one cable from the clock, the other cable from a sequencer) , in conjunction perhaps with some kind of midi-delay for the clock signal where necessary, is this the way towards synchronisation utopia ?

OR....is it just a handful of classic drum machines, basically the mpc60/mpc3000 , that spit real tight midi clock ? I'm hoping my googling skills are terrible and quality stand-alone midi clock (with multiple outs maybe) exists...

also, do you think its fair to assume the ASQ10 (just the midi sequencer part of the mpc60) would have as tight timing as the mpc 60 ? I assume the mpc60 is tight as hell going by anecdotal evidence I've seen on the interweb.


any info/links/ideas are most appreciated

Hi David,

the method sounds fine to me as long as the jitter of the sound card's a/d process is much less then the jitter you are measuring. given the short length of a single sampling period (1 second / 44100) that does seem likely, doesn't it.

so, you have found that the max deviation between quarter notes is ~2ms... am i correct in thinking that implies that the max deviation of a given quarter note from the 'perfect' time is smaller than that? i seem to remember elektron giving a figure for the latter that was around 1ms. and i remember thinking 'that's good enough for me, and better than most'. would you agree? of course, it would be great if the mpc3000 set a new benchmark that would be followed by all modern machines

also, do the errors accumulate in such a way to cause a variable drift over longer periods... or do the errors 'self adjust' over the course of each couple of bars keeping the total drift smaller?


on your site you talk about inter-ear timing; i believe that is computed in the brain by a different mechanism from what we call 'rhythm'... I have seen claims from psychoacoustic testing that highly skilled musicians typically can't distinguish timing errors of less than 4ms or so. (Of course, smaller jitters still do matter, because relative changes between different voices will cause different phasing etc, similar to the inter-aural phenomena. i suppose that's what you were thinking.)


pls correct me if im misinterpreting any of your findings; its interesting stuff.