Yahoo Groups archive

Elektron Musical Instruments

Index last updated: 2026-04-28 23:22 UTC

Thread

Re: Elektron's timing "signature"

Re: Elektron's timing "signature"

2007-05-01 by Leo Cavallo

hi Daniel

Thanks for the prompt reply.

>...it's not that simple to turn something like that on and off, we're
>dealing with a multiprocessor (DSP) system that need to work on audio
>in chunks,

Oh, I see...
Although someone could argue that you turned a limitation of the MD's 
DSP/code into a "cool feature".

Really, I'm just trying to understand... Does that mean that 
improving the MD's timing might be not possible due to some internal 
DSP limits?


>and we've adjusted them according to listening tests and
>going through all available BeatBoxes on the market.


Just curious... Which other beat-boxes did you use for 
reference/comparison while designing the MD?

In an previous email you mentioned the MPC-60. If you did analyze it 
then you've probably noticed that the "magic" timing of the MPC line 
relies in its almost sample-accurate tightness.

In other words, contrary to most people still like to believe, 
there's no magical groove at all in the MPC sequencer, just some 
incredibly accurate MIDI engine at work.

I'm not trying to turn this debate into an "MPC vs MD" kind of thing, 
not really in my intentions, but I still think that providing the 
most stable, reliable, repeatable timing scheme should be the first 
priority for ANY drum-machine.

Then, on top of that, you can have all the "Swing", "Groove", 
"Humanize" functions in the whole world (and I'd be the first user to 
make use of them!!!) but without a solid time base a "magic" groove 
doesn't make very much sense, IMHO.

The MPC is just a good example of that: the most stable MIDI timing 
in the industry still being the n.1 option for any musical genre 
where organic, human-sounding beats are an absolute requirement.

My Nord Modular (not a drum machine at all!) behaves equally well on 
this aspect, showing really low drift values when using its internal 
sequencer/time base.

So why can't a more recent and advanced piece of gear like the MD 
compete - timing-wise - with these two older machines?

>
>..and I know that people will hate me for that. :) But I'm very much
>of the belief that too many options and choices easily stand in the
>way of creativity. When we intruduce a feature we always think hard if
>it would have any negative impact on the workflow.

Not being able to properly sync the MD to other sequencers in 
real-time, due to its built-in "magical" timing irregularities, 
definitely has a negative impact on my work-flow. Many other users 
have expressed the same frustration.

And having to interrupt my jams in the studio because my MD loops 
needs to be edited/tightened up after being recorded in my DAW, 
that's another big annoyance I've always experienced.


>And we're talking about small, almost audible delays here, something
>noone has really heard until they meassured the timing. However, maybe
>they have _felt_ that they prefer a beat from a certain machine before
>another.

Yeah, I agree. The MD is a fantastic instrument but the timing of 
some other machines I have in my studio has always been more 
"inspiring", more convincing to me, without really knowing why...

... until I was suggested to measure the phenomenon to realize what 
was really going on.

>
>Another hint of that the small timing fluctuation is deliberate and
>hosted by a very high precision timer (We're talking MHz that's
>divided down) is that it's never ever drifting, and the different
>delays follow a special cycle. Not exact each turn,

So the fluctuation is there, it's different on every beat and it 
follows a mysterious "special" cycle that doesn't repeat itself every 
single bar....
In other words that makes any MD recording a real nightmare to 
edit/align on any modern DAW's timeline, right?


>What would be very interesting would be if people would do test tracks
>with different drum machines, and see if they enjoy the timing better
>from one or the other.

I've done that. When I have more time I'll make sure to make the 
results public.

>Daniel, Elektron - Pro mysticism and undocumented secrets that users
>dig up sooner or later :)

Well, if I had to defend my own design no doubt that I would adopt 
the same "mystic" attitude...

Best Regards
Leo

[Non-text portions of this message have been removed]

Re: Elektron's timing "signature"

2007-05-01 by Leo Cavallo

hi Daniel

Thanks for the prompt reply.

>...it's not that simple to turn something like that on and off, we're
>dealing with a multiprocessor (DSP) system that need to work on audio
>in chunks,

Oh, I see...
Although someone could argue that you turned a limitation of the MD's 
DSP/code into a "cool feature".

Really, I'm just trying to understand... Does that mean that 
improving MD's timing might be not possible due to some internal DSP 
limits?


>and we've adjusted them according to listening tests and
>going through all available BeatBoxes on the market.


Just curious... Which other beat-boxes did you use for 
reference/comparison while designing the MD?

In an previous email you mentioned the MPC-60. If you did analyze it 
then you've probably noticed that the "magic" timing of the MPC line 
relies in its almost sample-accurate tightness.

In other words, contrary to most people still like to believe, 
there's no magical groove at all in the MPC sequencer, just some 
incredibly accurate MIDI engine at work.

I'm not trying to turn this debate into an "MPC vs MD" kind of thing, 
not really in my intentions, but I still think that providing the 
most stable, reliable, repeatable timing scheme should be the first 
priority for ANY drum-machine.

Then, on top of that, you can have all the "Swing", "Groove", 
"Humanize" functions in the whole world (and I'd be the first user to 
make use of them!!!) but without a solid time base a "magic" groove 
doesn't make very much sense, IMHO.

The MPC is just a good example of that: the most stable MIDI timing 
in the industry still being the n.1 option for any musical genre 
where organic, human-sounding beats are an absolute requirement.

My Nord Modular (not a drum machine at all!) behaves equally well on 
this aspect, showing really low drift values when using its internal 
sequencer/time base.

So why can't a more recent and advanced piece of gear like the MD 
compete - timing-wise - with these two older machines?

>
>..and I know that people will hate me for that. :) But I'm very much
>of the belief that too many options and choices easily stand in the
>way of creativity. When we intruduce a feature we always think hard if
>it would have any negative impact on the workflow.

Not being able to properly sync the MD to other sequencers in 
real-time, due to its built-in "magical" timing irregularities, 
definitely has a negative impact on my work-flow. Many other users 
have expressed the same frustration.

And having to interrupt my jams in the studio because my MD loops 
needs to be edited/tightened up after being recorded in my DAW, 
that's another big annoyance I've always experienced.


>And we're talking about small, almost audible delays here, something
>noone has really heard until they meassured the timing. However, maybe
>they have _felt_ that they prefer a beat from a certain machine before
>another.

Yeah, I agree. The MD is a fantastic instrument but the timing of 
some other machines I have in my studio has always been more 
"inspiring", more convincing to me, without really knowing why...

... until I was suggested to measure the phenomenon to realize what 
was really going on.

>
>Another hint of that the small timing fluctuation is deliberate and
>hosted by a very high precision timer (We're talking MHz that's
>divided down) is that it's never ever drifting, and the different
>delays follow a special cycle. Not exact each turn,

So the fluctuation is there, it's different on every beat and it 
follows a mysterious "special" cycle that doesn't repeat itself every 
single bar....
In other words that makes any MD recording a real nightmare to 
edit/align on any modern DAW's timeline, right?


>What would be very interesting would be if people would do test tracks
>with different drum machines, and see if they enjoy the timing better
>from one or the other.

I've done that. When I have more time I'll make sure to make the 
results public.

>Daniel, Elektron - Pro mysticism and undocumented secrets that users
>dig up sooner or later :)

Well, if I had to defend my own design no doubt that I would adopt 
the same "mystic" attitude...

Best Regards
Leo

[Non-text portions of this message have been removed]

Re: Elektron's "magic" timing signature

2007-05-01 by j3l1yt0t5

> 
> In other words, contrary to most people still like to believe, 
> there's no magical groove at all in the MPC sequencer, just some 
> incredibly accurate MIDI engine at work.
> 

amen

> 
> Then, on top of that, you can have all the "Swing", "Groove", 
> "Humanize" functions in the whole world (and I'd be the first user to 
> make use of them!!!) but without a solid time base a "magic" groove 
> doesn't make very much sense, IMHO.
> 
> 

amen

> Not being able to properly sync the MD to other sequencers in 
> real-time, due to its built-in "magical" timing irregularities, 
> definitely has a negative impact on my work-flow. Many other users 
> have expressed the same frustration.

amen

> 
> And having to interrupt my jams in the studio because my MD loops 
> needs to be edited/tightened up after being recorded in my DAW, 
> that's another big annoyance I've always experienced.
> 
> 

amen





the midi timing should at the very least be as 'magic' as the internal
timing. consistency is a must.

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Niall Munnelly

If you lot would send some examples {kits, patterns, audio}
demonstrating the problem to Elektron, that would probably
help a lot.  I sent some stuff off over the weekend.

-- 
Yours,
Niall.
.. .  .   .    .     .       .           .             .                 .
Aleph Null.                             A Simple Insinuation Around Silence.
http://aleph-null.net
.. .. gpg public key - http://aleph-null.net/niall.gpg .. ..

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Jesse

Do we need to re-post everyones measurements?, It seems to me a lot has already been done in this regard...

-JC
Show quoted textHide quoted text
  ----- Original Message ----- 
  From: Niall Munnelly 
  To: elektron-users@yahoogroups.com 
  Sent: Tuesday, May 01, 2007 2:42 PM
  Subject: Re: [elektron] Re: Elektron's timing "signature"


  If you lot would send some examples {kits, patterns, audio}
  demonstrating the problem to Elektron, that would probably
  help a lot. I sent some stuff off over the weekend.

  -- 
  Yours,
  Niall.
  .. . . . . . . . . .
  Aleph Null. A Simple Insinuation Around Silence.
  http://aleph-null.net
  .. .. gpg public key - http://aleph-null.net/niall.gpg .. ..


   

[Non-text portions of this message have been removed]

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Niall Munnelly

On Tue, May 01, 2007 at 03:14:19PM -0400, Jesse wrote:
> Do we need to re-post everyones measurements?, It seems to me a lot has already been done in this regard...
> 
> -JC

I reckon support@... or somesuch would be a more
appropriate destination than this mailing list.

Elektron have already acknowledged that there's drift, which
is partially the result of buffering and partially a musical
decision.  Sending audio files of impulse sequences and
counting the samples in between won't add to that - we're in
"musically useful or detrimental" territory, now, and it
would behoove anyone who has described this behavior to
illustrate their findings with practical examples of lag,
flamming against other drum machines, and so on.


-- 
Yours,
Niall.
.. .  .   .    .     .       .           .             .                 .
Aleph Null.                             A Simple Insinuation Around Silence.
http://aleph-null.net
.. .. gpg public key - http://aleph-null.net/niall.gpg .. ..

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Jesse

----- Original Message ----- 
Show quoted textHide quoted text
  From: Niall Munnelly 
  To: elektron-users@yahoogroups.com 
  Sent: Tuesday, May 01, 2007 3:26 PM
  Subject: Re: [elektron] Re: Elektron's timing "signature"


  On Tue, May 01, 2007 at 03:14:19PM -0400, Jesse wrote:
  > Do we need to re-post everyones measurements?, It seems to me a lot has already been done in this regard...
  > 
  > -JC

  I reckon support@... or somesuch would be a more
  appropriate destination than this mailing list.

  Elektron have already acknowledged that there's drift, which
  is partially the result of buffering and partially a musical
  decision. Sending audio files of impulse sequences and
  counting the samples in between won't add to that - we're in
  "musically useful or detrimental" territory, now, and it
  would behoove anyone who has described this behavior to
  illustrate their findings with practical examples of lag,
  flamming against other drum machines, and so on.

  -- 
  Yours,
  Niall.
  .. . . . . . . . . .
  Aleph Null. A Simple Insinuation Around Silence.
  http://aleph-null.net
  .. .. gpg public key - http://aleph-null.net/niall.gpg .. ..


   

[Non-text portions of this message have been removed]

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Jesse

>I reckon support@... or somesuch would be a more
>appropriate destination than this mailing list.

Agreed

>Elektron have already acknowledged that there's drift, which
>is partially the result of buffering and partially a musical
>decision.  Sending audio files of impulse sequences and
>counting the samples in between won't add to that 

Also agreed...
Show quoted textHide quoted text
  ----- Original Message ----- 
  From: Niall Munnelly 
  To: elektron-users@yahoogroups.com 
  Sent: Tuesday, May 01, 2007 3:26 PM
  Subject: Re: [elektron] Re: Elektron's timing "signature"


  On Tue, May 01, 2007 at 03:14:19PM -0400, Jesse wrote:
  > Do we need to re-post everyones measurements?, It seems to me a lot has already been done in this regard...
  > 
  > -JC

  I reckon support@... or somesuch would be a more
  appropriate destination than this mailing list.

  Elektron have already acknowledged that there's drift, which
  is partially the result of buffering and partially a musical
  decision. Sending audio files of impulse sequences and
  counting the samples in between won't add to that - we're in
  "musically useful or detrimental" territory, now, and it
  would behoove anyone who has described this behavior to
  illustrate their findings with practical examples of lag,
  flamming against other drum machines, and so on.

  -- 
  Yours,
  Niall.
  .. . . . . . . . . .
  Aleph Null. A Simple Insinuation Around Silence.
  http://aleph-null.net
  .. .. gpg public key - http://aleph-null.net/niall.gpg .. ..


   

[Non-text portions of this message have been removed]

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Andy Tarpinian

On May 1, 2007, at 3:26 PM, Niall Munnelly wrote:

> On Tue, May 01, 2007 at 03:14:19PM -0400, Jesse wrote:
> > Do we need to re-post everyones measurements?, It seems to me a  
> lot has already been done in this regard...
> >
> > -JC
>
> I reckon support@... or somesuch would be a more
> appropriate destination than this mailing list.
>
> Elektron have already acknowledged that there's drift, which
> is partially the result of buffering and partially a musical
> decision. Sending audio files of impulse sequences and
> counting the samples in between won't add to that - we're in
> "musically useful or detrimental" territory, now, and it
> would behoove anyone who has described this behavior to
> illustrate their findings with practical examples of lag,
> flamming against other drum machines, and so on.
>
does the timing issue occur say synced up to the MnM? if so, might be  
the best example to use as elektron could reproduce the scenario

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Robert Krueger

Niall Munnelly wrote:
> If you lot would send some examples {kits, patterns, audio}
> demonstrating the problem to Elektron, that would probably
> help a lot.  I sent some stuff off over the weekend.
> 

Did you miss the long email from Daniel (elektron) regarding the 
"timing"?  :)

It's not a bug.

Re: [elektron] Re: Elektron's timing "signature"

2007-05-01 by Tony Scharf

I think this list should be renamed. "The rehashed, reworded, never
ending discussion of midi timing and arm-chair dsp coding" would be a
better title for it.

Please, everyone, could we move on?  I think now that we have heard
from daniel on this the last word has been given.  The timing issues
you have are by design.  If that makes the MD unusable by you, then
perhaps your resources and energy would be better focused elsewhere.

Right or wrong, elektron has made their choices and they stand by
them.  They have also made it clear this is an issue they are not
willing or unable to address to the satisfaction of all.

Thank you,
Tony
Show quoted textHide quoted text
On 5/1/07, Jesse <jesse@...> wrote:
>
> >I reckon support@... or somesuch would be a more
> >appropriate destination than this mailing list.
>
> Agreed
>
> >Elektron have already acknowledged that there's drift, which
> >is partially the result of buffering and partially a musical
> >decision.  Sending audio files of impulse sequences and
> >counting the samples in between won't add to that
>
> Also agreed...
>
>   ----- Original Message -----
>   From: Niall Munnelly
>   To: elektron-users@yahoogroups.com
>   Sent: Tuesday, May 01, 2007 3:26 PM
>   Subject: Re: [elektron] Re: Elektron's timing "signature"
>
>
>   On Tue, May 01, 2007 at 03:14:19PM -0400, Jesse wrote:
>   > Do we need to re-post everyones measurements?, It seems to me a lot has
> already been done in this regard...
>   >
>   > -JC
>
>   I reckon support@... or somesuch would be a more
>   appropriate destination than this mailing list.
>
>   Elektron have already acknowledged that there's drift, which
>   is partially the result of buffering and partially a musical
>   decision. Sending audio files of impulse sequences and
>   counting the samples in between won't add to that - we're in
>   "musically useful or detrimental" territory, now, and it
>   would behoove anyone who has described this behavior to
>   illustrate their findings with practical examples of lag,
>   flamming against other drum machines, and so on.
>
>   --
>   Yours,
>   Niall.
>   .. . . . . . . . . .
>   Aleph Null. A Simple Insinuation Around Silence.
>   http://aleph-null.net
>   .. .. gpg public key - http://aleph-null.net/niall.gpg .. ..
>
>
>
>
> [Non-text portions of this message have been removed]
>
>

[elektron] Re: Elektron's timing "signature"

2007-05-01 by innerclock2004

> Did you miss the long email from Daniel (elektron) regarding the 
> "timing"?  :)
> 
> It's not a bug.
>

Hi Robert - just to clarify - it is not a 'bug' certainly but by the
clearly random nature of the internal timing stability in the SPS-1 it
can never seriously be considered as a deliberate design concept and
presented as a 'feature' despite Daniel's claims. I refuse to accept
that random timing fluctuations in a rhythmic sequencer would ever be
a design 'feature'.  So, Bug - no. Design limitation - absolutely.
Probability of being rectified - zero. The SPS-1 timing is what it is.
We either love it or leave it and that's cool so long as we all know.
I think the discussions and various debates about timing can only be
of benefit if for no other reason than a company with serious design
skills like Elektron will obviously take timing stability into serious
consideration in future products. Regards - David.

Re: [elektron] Re: Elektron's timing "signature"

2007-05-02 by Niall Munnelly

On Tue, May 01, 2007 at 04:18:33PM -0400, Robert Krueger wrote:
> Niall Munnelly wrote:
> > If you lot would send some examples {kits, patterns, audio}
> > demonstrating the problem to Elektron, that would probably
> > help a lot.  I sent some stuff off over the weekend.
> > 
> 
> Did you miss the long email from Daniel (elektron) regarding the 
> "timing"?  :)
> 
> It's not a bug.

I expect that you missed the email from Elektron Labs asking
for examples.  That's posted in the thread at
elektron-users.com

If nothing gets done about it, so be it.  I'll still make
music, and I've exercised my due diligence.

-- 
Yours,
Niall.
.. .  .   .    .     .       .           .             .                 .
Aleph Null.                             A Simple Insinuation Around Silence.
http://aleph-null.net
.. .. gpg public key - http://aleph-null.net/niall.gpg .. ..

Re: [elektron] Re: Elektron's timing "signature"

2007-05-02 by J D Budris

--- innerclock2004 <david@...>
wrote:

> 
> > Did you miss the long email from Daniel (elektron)
> regarding the 
> > "timing"?  :)
> > 
> > It's not a bug.
> >
> 
> Hi Robert - just to clarify - it is not a 'bug'
> certainly but by the
> clearly random nature of the internal timing
> stability in the SPS-1 it
> can never seriously be considered as a deliberate
> design concept and
> presented as a 'feature' despite Daniel's claims. I
> refuse to accept
> that random timing fluctuations in a rhythmic
> sequencer would ever be
> a design 'feature'.  So, Bug - no. Design limitation
> - absolutely.
> Probability of being rectified - zero. The SPS-1
> timing is what it is.
> We either love it or leave it and that's cool so
> long as we all know.
> I think the discussions and various debates about
> timing can only be
> of benefit if for no other reason than a company
> with serious design
> skills like Elektron will obviously take timing
> stability into serious
> consideration in future products. Regards - David.
> 
> 
> 
If this is such an issue that elektron doesn't rectify
in an o.s. update. Maybe the mystic JJ that has made
the MPC 1000 an absolute wonderfully fun and useful
kit could charge for a new o.s. for the MD. Just a thought.

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

Re: [elektron] Re: Elektron's timing "signature"

2007-05-02 by onosendai@free.fr

>If that makes the MD unusable by you, then
> perhaps your resources and energy would be better focused elsewhere.

Well, it's not like a little problem you can move on : the timing issue make the
Machinedrum incompatible with other machine. You can't play live with a lot of
gear without drifting, you can't record loop in sync in Ableton Live without
rework them... i really don't understand the position of Elektron here. The
Machinedrum force me to record and edit manually each loop ! Why loose all this
time to make simple things.

They want to make the best drummachine ever, but if she can't stay in sync with
other gear it's crazy ! I can't believe it's a "feature". Daniel say yes but for
me it's a kind of "never say bad things about our gear" answer. Like a
politician he have to positive a bad thing. I could understand that but i prefer
"in fact it' really a bug, sorry", or "Maybe it was a mistake, if you are a lot
to think like that we will try to fix that" :)

Daniel, don't blame on me, i love my Machinedrum like everyone else here, but
this problem is a major one for me.

Ono

[elektron] Re: Elektron's timing "signature"

2007-05-02 by Leo Cavallo

hey Tony

yeah, I agree, but if that's really a feature don't you think that
people should know about that timing detail before buying a MD though?

All I see on Elektron's website is:

"The heart of Machinedrum is the Elektron ESP dual DSP hardware
platform, running our custom audio realtime operating system.
Featuring an interrupt driven no latency percussion sequencer which
guarantees >>>super timing<<< and performance"

Maybe they should change "super" into "magic"... :-)

Best Regards
Leo 
Show quoted textHide quoted text
> Please, everyone, could we move on?  I think now that we have heard
> from daniel on this the last word has been given.  The timing issues
> you have are by design.  If that makes the MD unusable by you, then
> perhaps your resources and energy would be better focused elsewhere.
> 
> Right or wrong, elektron has made their choices and they stand by
> them.  They have also made it clear this is an issue they are not
> willing or unable to address to the satisfaction of all.
>

Re: Elektron's timing "signature"

2007-05-02 by fierce fish

I'm wondering do people who are not happy with the timing use swing/shuffle?

I don't have an MD but my Mono has pretty good timing when synced to my other gear, certainly no worse  than my 808, 606 ,303's, Spectralis, Electribes (the exception being the Electribe S MkII which had terrible sync timing and was sold) so also I'm wondering how do people that own the MD and Mono think that they compare, is it something exclusive to the MD?

Also does the problem occur when syncing to other hardware or just computers, is it the same when it is master or just slave?


[Non-text portions of this message have been removed]

Re: [elektron] Re: Elektron's timing "signature"

2007-05-03 by Silvia Tubig

--- onosendai@... schrieb:

! I can't believe it's a
> "feature". Daniel say yes but for
> me it's a kind of "never say bad things about our
> gear" answer. Like a
> politician he have to positive a bad thing. 


out of my own experiance i can say that when i found
the iregularitys in the md\ufffds\ufffdtiming i was looking for
other sources of the problem than the md
itself..because it sounded ok for me..and i am a
person that is extremly sensetiv to bad timings..they
make me feel pain...

So..as said before..the timing derivations of the md
are not just random on a high rate..there is a
pattern..and a a random portion..
the result sound ok..the reality causes edit
problems..

My guess that we have indeed some design here..but
aswell a acting limitation...
 If there is a designed rythmic pattern in the md it
would in ideal dont alter the motiv lenght... so the
absolute lenght of a 2 bar loop stays the same even
when there is some micro timing going on..

That this is not realized might be no bad intention
but the limitations of the dsp inter communikation..

So i dont see a reason to dont belive Daniels
statement that we deal here with a design choice..

but even within this designchoice the random failiure
might be something that can be improoved... 
Its maybe too much talking about the item but in
general the discussion is good..

The MD is a interesting machine with a big sound
potential but in my eyes it still can be optimized...
if timing is a point for optimisation or not is indeed
something that can be discussed... 
I personally vote for the microtiming but would like
to see optimisations on the random factor and tactics
to be implemented to handle situations where higher
accuracy is required..

Avoiding of extra menues is no point here..the global
menu is still very easy to be overseen...much easier
than the soundengine,,,

there could be twice as much global menus to make this
too crowded and abstract... there is easily space for
2 or 5 functional  entrys more..



      __________________________________  Yahoo! Clever: Sie haben Fragen? Yahoo! Nutzer antworten Ihnen. www.yahoo.de/clever

[elektron] Re: Elektron's timing "signature"

2007-05-03 by innerclock2004

For those of you who consider that 2.5 milliseconds is esoteric when 
discussing tempo variation and well outside the range of human 
perception - here is a direct quote from the well respected John 
Klett from a published and detailed article from a few years ago 
called 'Delay in Large Format Digital Music Consoles'

"Feel is a very subjective thing. Let's define feel as the "relative 
placement in time of rhythmic elements". The character of each 
element will make its placement in time more or less a factor in the 
overall feel. The Snare drum has a large contribution to overall feel 
in your average pop mix. The only data we have on this at present is 
empirical. At one time Roger Nichols (Engineer for Steely Dan, Donald 
Fagan) is said to have defined the limit of feel perception at around 
250 microseconds for key elements. My own experience watching how 
certain producers place elements in time on digital audio 
workstations brings me to the conclusion that this perception limit 
is more like 100 microseconds. In any case, people who are very "feel 
conscious" will agree that we are looking at timing shifts well under 
a millisecond as important and affecting feel."

Even if you take Roger Nichols' figure of 250 microseconds - the 
acknowledged SPS-1 random timing variation of 2.18 ms between 
consecutive steps is over 8 times the perceived human limit of feel.

The entire article is here if you like a solid read:-

http://www.technicalaudio.com/reading/digitalconsoledelay.html

Roger Nichols is a legend in his own right and has a great website of 
very informative articles published in EQ magazine over many years.

http://www.rogernichols.com/index.html

Regards as always – David
www.innerclocksystems.com

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.