MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

It's official - Behringer 808...
MUFF WIGGLER Forum Index -> General Gear Goto page Previous  1, 2, 3 ... 8, 9, 10 ... 13, 14, 15  Next [all]
Author It's official - Behringer 808...
toxoplasma_gondii
lilakmonoke wrote:
Quote:
AFAIK Mr Roger Linn himself has refuted that there was some special mojo in vintage drum machine clocks.


roger linn is an engineer, not a musician i doubt he can hear the difference :-) i love this topic, every time i bring this up all hell breaks lose.
.


Roger Linn is a musician. He has worked as a touring guitarist and co-written several hit songs.
chvad
toxoplasma_gondii wrote:
lilakmonoke wrote:
Quote:
AFAIK Mr Roger Linn himself has refuted that there was some special mojo in vintage drum machine clocks.


roger linn is an engineer, not a musician i doubt he can hear the difference :-) i love this topic, every time i bring this up all hell breaks lose.
.


Roger Linn is a musician. He has worked as a touring guitarist and co-written several hit songs.


all true!
Panason
It's all academic but the house swing is still being done (eg check recent DJ sets by Green Velvet on YT).. it's mostly sampled 909 loops but you can get the same groove with an MPC or a DAW with the right groove template ... occasionally you will hear a track that is clearly not a 909 but still has that swing.

I think people ascribe more sophistication to Roland's engineers than was actually there... let's not forget that the 303 and 808 were not sold as awesome future techno machines and they were commercial flops.

Roland's drum machine releases after the 909 confirm this - they didn't really have a clue and likely didn't care much for what a bunch of hip hop guys in Detroit were doing...
lilakmonoke
Quote:
Roger Linn is a musician. He has worked as a touring guitarist and co-written several hit songs.


i dont know the man but if he claims there is nothing special about the timing of a 909 or like i once heard him say the old mpcs then i suspect he doesnt hear it. these timing "errors" are around 0.5 - 5 ms and can easily be measured ie. there is a scientific software called "sonic visualizer" that will show you even the tiniest tempo drift and jitter.

anyways its a fascinating subject just like the difference between bernard purdie and steve gadd is only a few milliseconds but purdie can do 20 minutes of talking without drift and not miss a single beat :-)

umma gumma
Bernard Purdie is the man!!
anselmi
lilakmonoke wrote:
Quote:
AFAIK Mr Roger Linn himself has refuted that there was some special mojo in vintage drum machine clocks.


roger linn is an engineer, not a musician i doubt he can hear the difference :-) i love this topic, every time i bring this up all hell breaks lose.

its not a myth at all, here is a simple test that you can do: if you have a 606/808/909 record 3 minutes of a simple pattern without tempo synch and then take it in your daw. you can immediately see that the tempo drifts around an imaginary BPM in a breathing way, not random - which is exactly what real drummers do, they synch to the "pocket". thats why analog drum machines sound so organic.

why the drift? the best description of the timing mechanics behind those machines is in this analysis of the 303 cpu by the guy behind sonic potions:

http://sonic-potions.com/Documentation/Analysis_of_the_D650C-133_CPU_t iming.pdf

what is organic timing? its all about synching between independent clocks. there is quite a bit of research around this but the best summary comes from james holden an electronic musician who has been experimenting with timing for years and written a max patch for synching a sequencer to real musicians:

https://www.ableton.com/en/blog/james-holden-human-timing/

here he is live with a band, listen to the timing and the crowd reaction:

http://vimeo.com/119763896

finally here is an amazing mix by mark fell. this is new york house music from around 88-92 so no daw but synched machines recorded to tape. lots of roland drum machines, high end analog audio production and a totally different groove than all the contemporary "sample precise" ableton house garbage. now, if you cant hear the difference you are an engineer ;-)

https://www.mixcloud.com/secretthirteen/secret-thirteen-mix-073-mark-f ell/

anyways all this goes to prove that the analog roland drum machines were not only designed by engineers. im sure they tweaked them until the timing sounded just perfect - and that to me is the magic of these machines.

im not sure that can be easily reproduced with a microprocesssor just like VA still doesnt sound like real analog. you sure could replicate the timing circuits but its so much damned easier to not do that and as long as everybody thinks no timing = good timing its not going to happen i think.

.
.


well, according to your (above) posted paper, the random errors of a sequencer are unmusical... the tempo shifting have to have a certain linking between the notes to be musical, so this discard every drum machine that doesn't be programmed with this paradigm in mind.
I think there's no one, and for sure the 808/909/303/Linn/MPC doesn´t have it.

Quote:
The first has had completely random timing errors inserted, with no link between any previous timing error and the current one, and no link between the errors in different parts. The result sounds unmistakably unmusical and inhuman.
lilakmonoke
Quote:
Bernard Purdie is the man!!


they dont call him "bigfoot" for nothing! here he is with steely dan. all groove, no chops ;-) just bassdrum and snare does it. no drum machine can do that yet.



Quote:
the random errors of a sequencer are unmusical


exactly! the "errors" of an analog drum machine are not random at all, they are a result of a few interlocking and synching clocks that weirdly enough sound just right to me. nothing is random in analog and all the "humanizer" settings in daws are based on random and completely wrong. there is also nothing random about bernard purdie!
anselmi
lilakmonoke wrote:
its not a myth at all, here is a simple test that you can do: if you have a 606/808/909 record 3 minutes of a simple pattern without tempo synch and then take it in your daw. you can immediately see that the tempo drifts around an imaginary BPM in a breathing way, not random - which is exactly what real drummers do, they synch to the "pocket". thats why analog drum machines sound so organic.


I´m not saying that the drift is a myth, I´m saying that the relationship between the drift and something "special" is a myth

Quote:
here he is live with a band, listen to the timing and the crowd reaction:

http://vimeo.com/119763896


1. I don´t know why do you link the crowd reaction with human timing...I can show you a kraftwerk concert with people criying of joy while listening to straight machine-tempo music

2. he should put the delay under the same control

Quote:
finally here is an amazing mix by mark fell. this is new york house music from around 88-92 so no daw but synched machines recorded to tape. lots of roland drum machines, high end analog audio production and a totally different groove than all the contemporary "sample precise" ableton house garbage. now, if you cant hear the difference you are an engineer ;-)

https://www.mixcloud.com/secretthirteen/secret-thirteen-mix-073-mark-f ell/



again: it´s very hard to state that the time shifting of the machines used in this recording are what people likes

If you show me the same track with and without time shifting as the only difference and a blind test with a significant difference in preferences I´ll be convinced
if not, you´re a believer...which is OK, but don´t try to justify it


Quote:
anyways all this goes to prove that the analog roland drum machines were not only designed by engineers. im sure they tweaked them until the timing sounded just perfect - and that to me is the magic of these machines.


man, at the time this people barely knows what the customers want!
most of the roland X0X machines wasn´t a big success when released.
do you really believe that there´s an intentional clock tweak?

Quote:
im not sure that can be easily reproduced with a microprocesssor just like VA still doesnt sound like real analog. you sure could replicate the timing circuits but its so much damned easier to not do that and as long as everybody thinks no timing = good timing its not going to happen i think.


hmmm.....
seriously, i just don't get it
anselmi
lilakmonoke wrote:
Quote:
Bernard Purdie is the man!!


they dont call him "bigfoot" for nothing! here he is with steely dan. all groove, no chops ;-) just bassdrum and snare does it. (start at 1:19) no drum machine can do that yet.



Quote:
the random errors of a sequencer are unmusical


exactly! the "errors" of an analog drum machine are not random at all, they are a result of a few interlocking and synching clocks that weirdly enough sound just right to me. nothing is random in analog and all the "humanizer" settings in daws are based on random and completely wrong. there is also nothing random about bernard purdie!


"interlocking and synching clocks" in an analog drum machine? hmmm.....
please lead me to your source
lilakmonoke
Quote:
please lead me to your source


its in this paper that i posted a page ago. this is the 303 but all these machines have the same cpu:

http://sonic-potions.com/Documentation/Analysis_of_the_D650C-133_CPU_t iming.pdf

im no hardware expert but the way i understand this paper there is a masterclock which is an analog vco, so there is your drift because all analog vcos drift more or less. then there is one clock that runs the sequencer and one clock that runs the user interface which are interlocked by regular interrupts. these generate interference patterns and jitter.

all this is dependent on the machine (606 being the most stable, 909 the most organic), room temperature, the tempo of the sequencer, the user interface load and settings and probably the components and the mood of the operator. so any "groove templates" or audio loops for this are complete bullshit.

i swear every time i have this discussion somebody is trying to convince me im imagining things. ive been programming my own timing mechanics in pure data for a few years so i know exactly what im hearing and what not. believe it or not :-)
lilakmonoke
Quote:
it´s very hard to state that the time shifting of the machines used in this recording are what people likes


there is nothing wrong with grid based "mechanical" timing but its only a tiny fraction of whats possible in music. in electronic music production its unfortunately the norm because all the daws are based on it. personally im more into the timing of black sabbath so thats why i started experimenting.

here is a black sabbath track i did with the clock AND the delay drifting with 909 settings. its just a sh-09 and two patterns but to me that sounds like somebody played it manually .. or at least it doesnt sound like the computer it is. you hear that difference ?


[s]https://soundcloud.com/lilakmonoke/the-walking-leper-king[/s]
Flux302
I have the RD-808 here... I gotta say it is a ton of fun. I did this quick video just to kinda show how the unit sounds quickly (as instagram videos over @flux302 on IG just were not showing the low end properly ...
tenembre
Flux302 wrote:
I have the RD-808 here... I gotta say it is a ton of fun. I did this quick video just to kinda show how the unit sounds quickly (as instagram videos over @flux302 on IG just were not showing the low end properly ...


Nice video. Sounds like a party!
anselmi
lilakmonoke wrote:
Quote:
please lead me to your source


its in this paper that i posted a page ago. this is the 303 but all these machines have the same cpu:

http://sonic-potions.com/Documentation/Analysis_of_the_D650C-133_CPU_t iming.pdf

im no hardware expert but the way i understand this paper there is a masterclock which is an analog vco, so there is your drift because all analog vcos drift more or less. then there is one clock that runs the sequencer and one clock that runs the user interface which are interlocked by regular interrupts. these generate interference patterns and jitter.

all this is dependent on the machine (606 being the most stable, 909 the most organic), room temperature, the tempo of the sequencer, the user interface load and settings and probably the components and the mood of the operator. so any "groove templates" or audio loops for this are complete bullshit.

i swear every time i have this discussion somebody is trying to convince me im imagining things. ive been programming my own timing mechanics in pure data for a few years so i know exactly what im hearing and what not. believe it or not :-)



oh, man ... like I already said: yep, the papers exists, the drifts exists, but you are the one that concludes that this is something special

There is nothing in this paper that proves that this clock drift gives the 303, its mojo (EDIT: the paper also says it but without any prove, read my comments below)

I just wrote 2 times the kind of experiment that you have to run to prove this ... until this you can't relate both things

Quote:
0 INTRODUCTION
A lot of information about the inner workings of the 303
sequencer can be found on the internet [1], [2], but the ac-
curacy is sometimes questionable. To shed some light on
the timing of the original μPD650C-133 CPU and end the
speculations and misinformation exact measurements were
needed. In this paper we discuss the findings of analyzing
an original CPU. While taking measurements, it became
clear that the TB-303 has some properties that deviate from
an ideal theoretical timing, that are somewhat relevant to its
characteristic sound
.


too biased observation...you can describe the timing, but this statement isn´t scientific (trust me, I worked in biochemistry research for 15 years)

Quote:
8 CONCLUSION
The interaction of different clock sources and processing
times in the interrupt routine in the digital section of the
TB-303 brings a noticeable amount of fluctuations to the
timing of the TB-303 sequencer. Even if the nuances are
small, an emulation of the original TB-303 can not claim
to be complete without taking these variations into consid-
eration.


now this is better, but anyway, the last sentence would be wrong if you take human perception into the equation, that is exactly what this paper lacks
lilakmonoke
anselmi ... you are right there is no scientific proof that one is better than the other but you can experiment and see what you like better, provided you actually hear differences in the microsecond range which i definitely do.

ive asked myself the same question many times and there are a few simple theories in perceptive psychology and biology:

- the human brain is a highly sensitive complex acoustic signal tempo detector meaning we can can follow multiple signals and detect tiny tempo fluctuations. its apparently something that was needed to survive.

- in general slowdowns are interpreted as relaxing and speedups as requiring attention, and fluctuations in multiple signals represent feedback based synching or adaptation. in fact no tempo fluctuations do not exist in nature but probably mean nothing is happening.

- in other words a bassdrum without tempo fluctuation is a totally unnatural signal which is probably whats interesting about it. your heartbeat which usually fluctuates like crazy will be absolutely regular shortly before you die - which i find fascinating. so a straight BD actually represents DEATH?

- all grooves that i find interesting have elements that are straight and fluctuating, behind and before the beat at the same time.

frequencies and harmonics probably follow the same rules so there you have it, the magic of all music in a nutshell ;-)

here is a soundtrack im working on that has 3 clocks inside. only one is straight, the other two oscillate around it (+-30ms) and the clocks switch positions. its in 7x5 quintuplets so this is not based on any known rhythms. to me thats an alien planet soundtrack, i can listen to this all day.

https://clyp.it/xunslz31

.
cger
That's really what this is all about. Even with 100's of custom and processed 808 samples ready for your DAW pleasure, how much fun it is to just whip out the box and jam around. You really can't beat that. And if the sounds are 95%-98% or whatever "close enough", that is amazing enough.

https://www.youtube.com/watch?v=YKK7ygEq_m0


Flux302 wrote:
I have the RD-808 here... I gotta say it is a ton of fun. I did this quick video just to kinda show how the unit sounds quickly (as instagram videos over @flux302 on IG just were not showing the low end properly ...
Panason
Quote:
how much fun it is to just whip out the box and jam around. You really can't beat that.


Yeah it's fun but if you want to use the machine in a full track you 'll probably have to sample it anyway, so it's more of a toy if it doesn't have meaningful pattern control over MIDI (which it probably won't).

That flux video is disappointing. We need solo bass drum with full decay at various tunings.... without the distortion /transient shaper.
lilakmonoke
Quote:
that is amazing enough


i agree! and with that xtra distortion and filter circuit this thing can go from polite to all out balls and back. you might even get some oldschool jungle bass out of it.
kons
Really interesting discussion about the clocks and perception.
Some thought sabout it that I have

First of all, whatever way you lean on the importance of the perceived quality of the XOX sequencer timings. i don't find it likely that the Roland engineers actively designed them in. It was simply a lucky happenchance that resulted from the engineers making a component choice on cost basis. If they could have afforded a digital timing solution in 1981 then they would have used that. But an analog VCO clock was simply cheaper.

The average human is capable of perceiving sub ms changes in timing. We can all determine if a sound from behind us is to the right or left based (partly) on the sub ms differential in timing of the sound arriving at our left and right ears. Presumably this ability can be trained and improved. Maybe some people are just inherently better at 'hearing-perceiving-processing' tenths of a ms and maybe even 100ths of a ms...

That there may be an evolutionary advantage to responding to increasing tempos seems like a plausible possibility. The doppler effect indicating something approaching. and thus we respond to 'pushing' beats...

That the interaction of the imprecise clocks of the aerly 80's XOX rolands creates non-random sub-ms tempo changes that are somehow appealing to the human brain is also plausible. Modern DAW randomisations of tempo don't give me personally any mojo. The real things seem to- for me. But that is possibly just a placebo effect of actually sitting with what are essentially historic relics.

Pitch is essentially frequency, which is at speed/tempo. Perception of pitch is a timing calculation in the human brain. Some people seem to be born with better pitch perception and some seem capable of training and improving their pitch perception. Rising pitches do seem to be 'alarming' as per tempos.
lilakmonoke
Quote:
Maybe some people are just inherently better at 'hearing-perceiving-processing' tenths of a ms and maybe even 100ths of a ms...


like ginger baker, the GOAT of rock drummers once said: "some people just have time!" 100ths of ms is pushing it but in relative impulse differences i think i can hear somewhere between 2 and 5 ms - which doesnt mean i have ginger baker time ;-)

more important i think is that there are big differences in how people process or feel time. the best groove bands often have a unique combination of different time processing members. my favourite example is late 60s black sabbath. their drummer came from jazz so he was behind the beat and slowing things down while the guitar played straight and pushed the beat and the bass wobbled something in between. out of this strange combination came the sludge metal groove ...

here is them in 1970 when they were young lads and not famous at all, there are like 50 people in the audience. amazing concert, keep in mind in 1970 there was NOTHING like this. :

tobb
anselmi wrote:
lilakmonoke wrote:
Quote:
please lead me to your source


its in this paper that i posted a page ago. this is the 303 but all these machines have the same cpu:

http://sonic-potions.com/Documentation/Analysis_of_the_D650C-133_CPU_t iming.pdf

im no hardware expert but the way i understand this paper there is a masterclock which is an analog vco, so there is your drift because all analog vcos drift more or less. then there is one clock that runs the sequencer and one clock that runs the user interface which are interlocked by regular interrupts. these generate interference patterns and jitter.

all this is dependent on the machine (606 being the most stable, 909 the most organic), room temperature, the tempo of the sequencer, the user interface load and settings and probably the components and the mood of the operator. so any "groove templates" or audio loops for this are complete bullshit.

i swear every time i have this discussion somebody is trying to convince me im imagining things. ive been programming my own timing mechanics in pure data for a few years so i know exactly what im hearing and what not. believe it or not :-)



oh, man ... like I already said: yep, the papers exists, the drifts exists, but you are the one that concludes that this is something special

There is nothing in this paper that proves that this clock drift gives the 303, its mojo (EDIT: the paper also says it but without any prove, read my comments below)

I just wrote 2 times the kind of experiment that you have to run to prove this ... until this you can't relate both things

Quote:
0 INTRODUCTION
A lot of information about the inner workings of the 303
sequencer can be found on the internet [1], [2], but the ac-
curacy is sometimes questionable. To shed some light on
the timing of the original μPD650C-133 CPU and end the
speculations and misinformation exact measurements were
needed. In this paper we discuss the findings of analyzing
an original CPU. While taking measurements, it became
clear that the TB-303 has some properties that deviate from
an ideal theoretical timing, that are somewhat relevant to its
characteristic sound
.


too biased observation...you can describe the timing, but this statement isn´t scientific (trust me, I worked in biochemistry research for 15 years)

Quote:
8 CONCLUSION
The interaction of different clock sources and processing
times in the interrupt routine in the digital section of the
TB-303 brings a noticeable amount of fluctuations to the
timing of the TB-303 sequencer. Even if the nuances are
small, an emulation of the original TB-303 can not claim
to be complete without taking these variations into consid-
eration.


now this is better, but anyway, the last sentence would be wrong if you take human perception into the equation, that is exactly what this paper lacks


At least ,your perception is special ! hihi
magnetize
lilakmonoke wrote:
keep in mind in 1970 there was NOTHING like this. :


except blue cheer 3 years earlier wink
lilakmonoke
Quote:
except blue cheer 3 years earlier wink


thats correct but they are not british enough and dont count. so is link wray who was 20 years ahead of his time but still doesnt count ;-)
Panason
Hmm, yeah but Black Sabbath and Mr Purdie were not doing electronic music...drum machines are popular because they don't sound like human drummers and most people don't want them to.

The rhythmic vibe of the 808 can only be because of regular deviations, aka swing, and of how the voices interact with each other and whatever compression is going on. The very subtle swing is probably a happy accident but maybe they added it in.
dannyrichardson
Dang, this is going to be awesome!
MUFF WIGGLER Forum Index -> General Gear Goto page Previous  1, 2, 3 ... 8, 9, 10 ... 13, 14, 15  Next [all]
Page 9 of 15
Powered by phpBB © phpBB Group