***[[[ Akai vs. JJ ]]]*** TIMING!!!

Hey all! i was reading a thread in the mpc5000 section and an interesting topic of sample playback timing came up, so i did some experiments using there idea and sample running Akai os 1.23 and JJ os 5.47. I got some interesting results!
please check it out! I would like to know how JJ os2 stands upto this test!

http://www.mpc-forums.com/viewtopic.php?t=89534

Peace!

at what specificity does the human ear stop hearing differences in timing?

No idea but its an interesting experiment, care to dabble nym? I know you would find some interesting results to! Please???

Peace!

Nym wrote:at what specificity does the human ear stop hearing differences in timing?

Depends.

If youre playing an instrument you can feel/hear the latency abocve 10-20ms

Groove is a strange thing, i think it works on a subconscious level and you have to concentrate and work with shifting notes to educate yourself so you can know the differences.
Also notes that are played a little bit too early stick out more (except if they are lourder than the rest, in which case the rest would be late).

When it comes to layering similar frequencies the human ear can detect the slightest differences in phase/frequency, thats because most of your ability to locate sounds in your natural environment is done by measuring the time difference between the sound arriving to your left and right ear (thats so fast that theres only a difference in the phase of the signal)


Will join the testing next week, no time till then.

Nym wrote:at what specificity does the human ear stop hearing differences in timing?

i actually think that timing is the reason that i can usually tell when a beat was made with fruity loops.

right, the midi timing, anyway. i hear the same kind of thing, from reason, too etc, but i've always tended to think that it's less the TIMING timing that i'm hearing and more the possibility that maybe the platform pushes its user in a different rhythmical direction than the mpc, which makes groove accessible. but i have no idea really.

the reason why i asked that question is because when it comes to notation and sequencing, if 2 samples (in the recorded sense) are 10 samples (in the notational sense) apart due to a machine's non-perfect timing, I CANNOT HEAR IT. 10 samples is a very, very, very small time, and unless those samples are phasing (layered same sound w/ self, etc) 10 samples is in all practical senses wholly inaudible.

that's why i wonder - exactly how powerful is the human ear, and at what point do we just split hairs about timing? by no means am i belittling the research done by rinse, trybtek and others, i am really thankful and interested in it.

http://img444.imageshack.us/my.php?imag ... lolmy2.jpg

Did i play right?

I think that "RYTHM" itself isn't something that is solely dependable on being "audible" - it also depends on feeling....

But like rinseout said - might not be able to point those timing differences out on a conscious level - BUT - you can "sense" them on a subconscious level....

If your ear can perceive that someone who is calling you from behind is 1feet away or 6 or 12 - you maybe can't tell the exact number but a approximate guess would be possible - and considering how fast sonics are - THIS stuff happens on a "unconscious" level.....

It's similar to the concept that the human brain on the unconscious level has no filters - so you're basically scanning everything at every second around - and I mean literally ANYTHING - just like a video-cam.....

BUT your consciousness is filtering stuff heavily out - because without it would be the same like watching 30 TV channels at the same time - overkill....

That's why a lot of autist's with asperger syndrom can do stuff like memorize whole telephone books - we could do that too BUT we have filters who are working against it - people who suffer from asperger basically JUST can't FORGET....

Stuff like that happens all the time in our brains - the most developed supercomputer on earth....

SO that is WHY people "hear/sense" groove differences between machines even if it's the same pattern - same samples - hard on quantize....
EVEN the slightest tolerance will be "sensed" - even though you might not be able to hear the difference....

Nym wrote:right, the midi timing, anyway. i hear the same kind of thing, from reason, too etc, but i've always tended to think that it's less the TIMING timing that i'm hearing and more the possibility that maybe the platform pushes its user in a different rhythmical direction than the mpc, which makes groove accessible. but i have no idea really.

the reason why i asked that question is because when it comes to notation and sequencing, if 2 samples (in the recorded sense) are 10 samples (in the notational sense) apart due to a machine's non-perfect timing, I CANNOT HEAR IT. 10 samples is a very, very, very small time, and unless those samples are phasing (layered same sound w/ self, etc) 10 samples is in all practical senses wholly inaudible.

that's why i wonder - exactly how powerful is the human ear, and at what point do we just split hairs about timing? by no means am i belittling the research done by rinse, trybtek and others, i am really thankful and interested in it.

then by this, the mpc 3000 had the worst timing out of all due to its rambling tempo changes. i still think the difference is soooo little that it only becomes audible as effects are piled up, long loops used, and/or a sample is layered with itself. i agree that the human senses experience important subtlety which changes our perception of the object being sensed, BUT if we could figure out just how long 10 samples is in TIME time we'd be able to judge whether or not that's the case here.

i think in this case it's like comparing quality of 2 wav files of such enormous sampling rate/bit depth that there's no discernible difference (not 16 and 24, i can hear that)

in other news 16 off the mpc1000's spdif is the new grit, i swear i can hear really sharp, beautiful nyquist tickling my eardrums recently

Nym wrote: BUT if we could figure out just how long 10 samples is in TIME time we'd be able to judge whether or not that's the case here.

Well, 44.1 kHz = 44100 samples per second, so that means each sample is .0000227 seconds long (rounded). So 10 samples would be .000227 seconds long... you be the judge.

Nym wrote:i swear i can hear really sharp, beautiful nyquist tickling my eardrums recently

You forgot to say no homo for the kiddies... they gonna jump on you for that one.

that's pretty gotdamn tiny
haha no homo was dumb as fk to begin with, folks really take it too far to the point where it makes no sense

"hey frank, have you seen cloverfield? no homo"

Nym wrote:haha no homo was dumb as fk to begin with,

Yep.