As of September 23rd elektron-users has been replaced by elektronauts.com. Find out what this means here.
Elektron-Users Elektron Forum Elektron Gear MD SPS-1 Timing Performance Issues (1 viewing)
Go to bottom Post Reply
TOPIC: MD SPS-1 Timing Performance Issues
#17284
Boo
Posts: 173
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago
the problem with the electribes is not that they are not tight..they are late...

with ableton live you can have independend clock delays on each midiinterface.. so when you have an extra one for the electribes and make it 12 ms earlier than the others you will see that it fits much better
  The topic has been locked.
#17285
Posts: 0
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago

rudebop wrote:
and one thing..never use mp3?s to measure timings..the compression is altering the timing and therefore all measurements are irrelevant..


I must contest this - like jitter in ASIO/CoreAudio recording, there's no way it could be an accepted behavior, and would have been redressed years ago, if it ever happened.

Please find the linked Audacity project. Mp3 is on channel one and wave on channel two. You'll see no timing deviations.

http://syncretism.net/snd/MD-Audacity.zip
  The topic has been locked.
#17287
King Koopa
Posts: 242
0
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago
Ignoring outgoing Clock Sync which has zero bearing on how the SPS-1 behaves internally - the majority of these findings do seem to suggest that the Voice generation/Step tempo accuracy/stability of the SPS-1 is a tad on the rubbery side of things - particularly if you compare it to an MPC-3000/4000 and even an old MC-4 Sequencer. Even if you personally feel 2.18 ms is not that big a deal - considering the MPC-3000 is over 10 years old in technology terms and the MC-4 is nearly 30 years old - both these sequencing devices have a voice/step tempo stability that outclasses the SPS-1 by at least a factor of 24 times! These numbers start to mean something when you run devices side by side and 'feel' the difference. The SPS-1 was the first drum sequencer I felt happy buying in a sea of mostly average 'groove boxes'. It was a lot tighter than my PC and much more of a joy to compose with. Then I was talked into buying an MPC-3K for sampling duties. It's like buying a new pair of shoes - suddenly the rest of your outfit looks a bit shabby. That's not a bad thing - and sometimes - with things like higher sample resolution, colour screens, more patches, USB etc - these things are just fluff mostly and hanging on to your old 12bit/32KHz 4 MB Casio Sampler is far more rewarding than upgrading to get all the trinkets. However - timing stability IS NOT FLUFF. It is the foundation of sequencing. It's not OK that we went backwards by a factor of 24 times in the timing stakes between MPC3K (mid 90's) and SPS-1(2004). My musical partner had a Future Retro Mobius. It always bothered us - it never sat right and it didn't do Kraftwerk very well. We never tested it but we never let it run for very long. Six weeks ago I convinced him to purchase an MC-4B for analogue sequencing. Three days later he sold the Mobius. Without doing the Mobius testing I can almost bet the timing improvement between it and the MC-4B was by a factor even greater than 24! And you can still move notes around if you wish - it doesn't have to be grid quantised. But - BUT.....if you do lock a 16th arpeggio down it makes your eyes water. And that is down to the tempo/step precision. The Mobius was never in the race.

So - the million dollar question. Are we going to put pressure on Elektron to put the SPS-1 back in the ballpark?
  The topic has been locked.
#17288
King Koopa
Posts: 242
0
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago
Also - it's worth mentioning I think - we all have emotional and financial investment in the equipment we buy and use and with that investment we put faith in the tools and really wish and hope and believe that they are worth their weight and cost. However - sometimes we can be blinded by our investment - a close friend who spent over $5K on a Mac/Logic rig a few years back spent a good year refusing to acknowledge what he could plainly hear - that his external Midi stuff sounded sloppy as all hell. He did everything - groove templates, micro event adjustments, moved his room around, changed his mix position - he even stopped listening to tight production because it reminded him of how rough his own shit sounded! His emotional and financial investment kept him from seeing the simple truth - the rig didn't do Midi justice - period. After a few beers one day I convinced him to let me bring over my Atari/Notator/Unitor system and hook it up to his Midi gear. Next we copied some of his midi files to Notator Format. I'll never forget the look on his face when he pressed the space bar. Priceless - the illusion falls away and you can go back to what you know is right even if you have to buck the trend and use a computer that is 20 years old to get there. But from there you can move forward. My purpose of this topic/thread is not to slag or negate the value of different gear and certainly not Elektrons'. It's about listening truthfully and saying 'yes - that 20 year old sequencer makes the hairs stand up on the back of my neck' and 'no - this new one that I have just spent a months salary on doesn't cut it'. Only then can you begin to ask 'why?' and once you start to find out why you are then in a position to do something about it. Regards as always - David.
  The topic has been locked.
#17289
Boo
Posts: 150
0
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago
I don?t mean this to be a diversion away from the technical issues but I thought I might give some observations from a drummers perspective. Interesting that Innerclock?s reading of the MD show a regular irregularity if you might call it and Rudebop?s TR808 observations show solid groove with occasional wide fluctuations. IMO Elektron boxes have already confirmed their reputation as funky boxes, capable of creating sick grooves. I can appreciate that the MPC3000 is also a funky box capable of robotic or tight electro beats or funkiness can be created by offsetting notes manually against a very stable midi clock. My point is that perhaps an irregularity is a contributing factor to the funkiness when heard against a theoretical perfect pulse.

I don?t want to get caught up in a theoretical vs. practical aspects of timing debate as science and art will always be coming from different perspectives but as a point of departure two things that have interested me are:

1) Tactus in early Rennaisance vocal music. In a historical period where no barlines existed in music manuscript the importance of a pulse that each singer recognized universally, albeit via a conductor/choirmaster (master clock), was essential for the proper interpretation of a piece of music. If you ever see original manuscript from the period there are individual pages given to each performer with nothing but a series of different note values and pitches, each singer having to accurately place complex syncopated melodies against the other singer's parts.

2) The influence of African rhythm in American music. When listening to BigBand Jazz of the 1920?s onward, Funk, Rock and RnB, the relationship to dancing is unquestionable. Why is it that bands who never even listened to click tracks grooved so hard people couldn?t help but get down on the dance floor? There is a constant in all these styles whether you call it The Funk, Swing or Groove. Take any number of great bandleaders, Count Basie, James Brown, George Clinton etc. Miles Davis was very particular about feel, the list goes on?but what is it raises them above the rabble?

From a drummers perspective, but this could go for any musician, the most important thing to develop is your internal clock. All the other stuff will come but if you can?t keep good time internally it doesn?t matter how good your chops are it?s not going to groove. Sure you can push and pull the feel but against what? There has to be a constant that all band members follow. Take for example a simple 4/4 pattern playing 8th notes on the Hihat with your right hand and accenting the 2 and 4 on the snare with your left. If you switch to a pattern playing 16th notes on the hihat leading with your right hand suddenly your right hand is accenting the 2 and 4 on the snare and you have to calculate, or delay compensate, for the time it takes your right hand to get from the hihat to the snare and back for the next hihat 8th note without interrupting the flow of the groove. This is by no means an easy feat and impossible if you don?t have a good internal clock.

So my point is that if a box does have reliable clock with the occasional stutter then this is funky but if it?s all over the place and you don?t have a reliable reference point to begin with then the overall feel is going to suffer. Computer sequencers present a whole range of synchronization problems but maybe there is something simple at the core of it? One machine needs to be master but if they are all dancing to the beat of a different drum they?re not going to sound like a tight band are they? I?ll leave it to the more technically minded to discuss these finer points of computer sync and I hope this wasn?t too off track from the issues. I just thought a drummers perspective might not be out of place, since most often a drummers job is to make sure people tap their toes and boogie down.
  The topic has been locked.
#17290
Boo
Posts: 173
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago

niall wrote:

rudebop wrote:
and one thing..never use mp3?s to measure timings..the compression is altering the timing and therefore all measurements are irrelevant..


I must contest this - like jitter in ASIO/CoreAudio recording, there's no way it could be an accepted behavior, and would have been redressed years ago, if it ever happened.

Please find the linked Audacity project. Mp3 is on channel one and wave on channel two. You'll see no timing deviations.

http://syncretism.net/snd/MD-Audacity.zip



I tested mp3s in the past and found them jittering the timing..maybe this dont happens allways... i just dont have the time to make a scientivic thing from that... just.. i prooved it to myself that it can happen..thats enough to warn about timing measurements based on mp3?s...

Jitter in a normal digital recording is way below the range we talking here about its measured in us not ms...

However... i found the MD sounding tight for me..i was shocked the first time i realized how wobbely it is in relation to other drum machines... but it dont hurts..it wobbels nicely.. its only a problem when you want to sett the refference looplenght.. with tghe MD i ve to cut 10 loops and take the common denominator..so i rather record a click of one of my nord modulars and take that as refference..

The important issue for me was that i feared that the clock output of the MD is wobbely aswell..but as i proofed for myself today..its not...thats the biggest issue. Its a capabel clock master..when it swings its own sounds around the clock a bit its ok... I would perfer this kind of "human factor" centered on the one..but it dont sounds bad ...that the main thing...
i started timing measurements mid 90?s because i got pain from that jitter...
Aqnd i had to proove to c-lab (later emagic) that i had a point..
so i started measure my sytem..and bought an mpc 3000...
So Jitter is a bad thing when it fucks up a groove by its nervous and chaotic behaviour...
A slightly wobble in the timing is a different thing..Our brain interpretates that as natural...

So to answer this thread..yes..the MD wobbels around the sounds...but it has a good timing never the less.. a drum machine with inbuild human factor...pretty unique
  The topic has been locked.
#17291
Boo
Posts: 173
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago

mvmono wrote:
I hope this wasn?t too off track from the issues. I just thought a drummers perspective might not be out of place, since most often a drummers job is to make sure people tap their toes and boogie down.


no..not at all... the problem with the devil timing problems is that they keep you from the main goal..to do music...

So its important to dont dig deeper into that than necessary... and get reminded to the main goal again... this machines are tools or instruments... not the main theme...


i however run into that trap in the past and its good to isolate problems...

its not an easy goal to mix systems... either DAW+ plug ins..that works...
Or pure hardware.. (with atari seen as hardware)...that works...
All mixtures can have real big problems... and you just dont feel not good..and dont know why..than you start to look around..and look and look..

When a machine hurts it needs to be replaced..point..

i today replaced my MD... with another MD...its much better now )

I
  The topic has been locked.
#17292
Kong
Posts: 1312
mr. applehead
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago
thanks for doing all these tests guys, its a very interesting discussion though a little hard to follow @ times.
has anyone considered that perhaps elektron have taken into account the leap year phenomenon?
time is rubbery at best.
btw can someone please test an emu xl7 my friend needs to know how his clock rocks.
greetz :-D
  The topic has been locked.
#17293
Boo
Posts: 173
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago

innerclock wrote:

So - the million dollar question. Are we going to put pressure on Elektron to put the SPS-1 back in the ballpark?


I wouldnt want the MD to be as sterile as the mpc 3000..but i would like to have a clean 1...

Would be nice if this "human factor" would be a switchable option, or if at least the one of a bar would be stable
  The topic has been locked.
#17294
King Koopa
Posts: 242
0
Re: MD SPS-1 Timing Performance Issues 17 Years, 2 Months ago
Hi Rudebop - please don't ever think tighter event precision means sterile. The MPC-3K funks and grooves like a motherfucker. This type of thinking is what lets manufacturers off the hook! With good swing and quantize/tick placement options - you can create any feel you like - rubber or robo! The trick is locking the clock drive chain down so that it never budges between steps. Trust me when I say - if you like the MD now - it will blow your head off if they can straighten up the event precision. Best regards - David.
  The topic has been locked.
Go to top Post Reply
Powered by FireBoardget the latest posts directly to your desktop

Login Form

start Player