Fix the sync issues now !! yes or no ?

[twisted-space]

yeah, it's not working in ALL situations. but the most important ones of a daw, it is. (i want it to be working everytime anyways, but the issue gets exagerated).

In who's opinion? Certainly not mine, or 3/4 of the users that voted in the poll.

[zee verkawound]

yeah, it's not working in ALL situations. but the most important ones of a daw, it is. (i want it to be working everytime anyways, but the issue gets exagerated).

In who's opinion? Certainly not mine, or 3/4 of the users that voted in the poll.

You'll get used to it. Dave states stuff constantly that's total unfounded BS. Like, "it IS a bug". He's just full of himself and I guess likes to feel as though he's important because he posts in defense of what exactly, no one knows.

[davepermen]

yeah, it's not working in ALL situations. but the most important ones of a daw, it is. (i want it to be working everytime anyways, but the issue gets exagerated).

In who's opinion? Certainly not mine, or 3/4 of the users that voted in the poll.

well, i voted yes, fix it, as i stated. but i don't have any situation where it bugs me, at all. i still like it to be fixed. and actually, i don't even know how anyone would not want it, unimportant of actual usage.

so. sync is unimportant for any daw situation where you just use one computer. sync is unimportant in any live act situation where the user just uses one computer. it is important if you combine some other devices like a 303 or something. it is important if you combine to other live computers. but these are not the main use cases for live. personally, i see about one in 10 to actually use live with external instruments that might need timing, or in a multi-computer setup that might need timing.

so it sure is not everyone having a need for this.

AND a lot have reported that they have successful sync setups. so it does work for some.

that reduces the size of the problem from everyone, to everyone that uses it in an actual sync situation, to SOME that use an actual sync situation. and this is much less, than everyone. much less.

and the vote does not represent anything meaningful.

[davepermen]

yeah, it's not working in ALL situations. but the most important ones of a daw, it is. (i want it to be working everytime anyways, but the issue gets exagerated).

In who's opinion? Certainly not mine, or 3/4 of the users that voted in the poll.

You'll get used to it. Dave states stuff constantly that's total unfounded BS. Like, "it IS a bug". He's just full of himself and I guess likes to feel as though he's important because he posts in defense of what exactly, no one knows.

yeah, so it is not a bug that live does not sync everywhere correctly? and it is not a bug that live crashed for 3phase with just some midi clips playing?

there is no unfounded BS from me. but have fun continuing attacking me uselessly.

[zee verkawound]

yeah, so it is not a bug that live does not sync everywhere correctly? and it is not a bug that live crashed for 3phase with just some midi clips playing?

there is no unfounded BS from me. but have fun continuing attacking me uselessly.

If it's a bug Dave, why wasn't it fixed three versions ago when the issue was just as paramount as it is today? It's an internal architectural issue Sir, that's why. Just like the fact the program does not support 64bit. Crashes don't always mean "bugs". This is NOT a bug fix. The sync issue does not even crash the program Dave. It's an internal process error that is the result of the program's architecture. Not a bug. So yes, you were spouting unfounded BS. As per usual.

[Dennis DeSantis]

Guys, please, relax.

We are investigating the issue. We can't say anything, as always, about when a problem might be identified and when/how a fix might be available.

ableton dont reacts on beeing polite..they react on to much noise in the user base..

Please note that neither of these are true.

Best,

[3phase]

Guys, please, relax.

We are investigating the issue. We can't say anything, as always, about when a problem might be identified and when/how a fix might be available.

ableton dont reacts on beeing polite..they react on to much noise in the user base..

Please note that neither of these are true.

Best,

Ok..sorry than that i got that impression...
maybe you just were too stressed out..

I anyway hope that you make a smarter algorythm happen.. You see.. we live in a time where you run multipple FFT processors in realtime but the syncing algorythm appears braindead compared to my alarmclock...
sorry..i just think that such an important issue can have some cpu cycles..at least you run that "plug in" only one time.
However i see that its a complex problem and one that easyly gets pushed along the timeline, in favor of more urgent tasks.
This thread was ment to show that it´s at least as urgent as any other feature you might work on, and defetly more urgent than serato bridge.

[tchan]

please guys, stay cool. the abes are actively working on the problem. i'd like to contribute more data to help solve the problem but i fear all the noise will bury my findings

anyway, here is more info on our test setup:

laptop 1 (master) / macbook (mid 2007) / http://bit.ly/a9ugzf
laptop 2 (slave 1) / unibody mbp-15 (late 2008) / http://bit.ly/c0OK2C
laptop 3 (slave 2) / unibody mbp-13 (mid 2009) / http://bit.ly/bqisKu

re: laptop performance

my own theory as to why the mbp-15 had a more consistent & lower tempo fluctuation is because of the architecture of the laptop. the mbp-15 has a discrete graphics processor while the mbp-13 has an integrated gpu which drains power from the main processor. here's some more info:

http://arstechnica.com/apple/news/2010/ ... pgrade.ars

the mbp-13 also had serato loaded up (although we weren't scratching and the bridge was disabled so we can sync live) so that may have added a little to the load. i also don't recall the live CPU load being too crazy (hovering at the 15% mark). anyway, i'll monitor more closely when i run further tests.

re: wireless network sync

i only have a primitive understanding of midi - don't even know what wordclock is let alone how to sync the interfaces! we only got the idea to clock sync the laptops via wifi two weeks ago after talking to the plastikman tech (shout out to ian!).

for the past month we've been fiddling with ad hoc wifi networks to control our live session with iPads in various performance situations (outdoor, gallery space, theatre, bar, etc). since we already had the network setup i thought i'd try to clock sync the laptops…and it worked! it's been almost 4 years since i last tried slaving live to external hardware (a korg electribe) and it definitely didn't work as well as network midi sync.

anyway, this thread has been great. i learned a lot about what to expect from reliable clock sync - thanks 3phase! i'll do some more tests and post my findings. hopefully this will help the cause

[deva]

the shouting kids usually get what they are aiming for..otherwise that reflex would have died out ages ago...

and do they deserve it for the bad behaviour? and no, so far, no kid ever got from me what they where aiming for, if they don't behave.

Fascist

[deva]

and the vote does not represent anything meaningful.

and if 10 out of 200 voted to fix sync, you would say otherwise...

[andydes]

Good God this thread is hard work to sift through.

Still, it seems to have served it's purpose. Maybe it's put the issue higher on the agenda, maybe not. It's at least gotten confirmation that it's being taken seriously. However, it's mess of personal attacks, and there's no coming back from that.

Maybe we need a new thread along the lines of the Live Performance thread, where we can have a defined format of how to compare results. I know what you're going to say 3Phase, that the problem is in the code, but some people certainly appear to have better results than others. Certainly Ableton need to work on a stable universal solution, but in the meantime we can gather some data and try to see what works best. Even if results don't help Ableton, we might be able to get a few people results they can work with. Would be best if everyone did their best to keep it strictly technical.

Sadly, I'm not the man to conduct such a thread, as I have little experience. I just intend to sync 2 laptops together in the near future.

Now I have to get a few things off my chest:

3Phase: I have no interest in personal attacks, but I question logic about how you phrase your posts. It's quite possible to be campaign strongly for your issues to be addressed without constantly questioning the competence of ableton staff. I don't see what you expect to achieve with that, but it will certainly make some people want to disagree with you and if they don't feel strongly about sync, maybe not vote with you. Even if you dismiss them as fanboys, you could use their support. Also a lot of people who don't enjoy a good internet brawl will simply give up on the thread.

I just wonder how the results would look, if you'd handled it differently? Better or worse? Guess we'll never know.

Davepermen: Although I agree with you a about 3Phase's attitude, trying to get him to stop isn't going to work, and by continually raising the issue, you're making yourself look just a bad. I know it's hard when he's using offensive nicknames and the like, but is it really worth the hassle?

Everybody: Like with all these big threads, we have the same people endlessly repeating the same points. Personal attacks, defending their position and technical points. It's like everyone has tourettes or something. When someone (3phase, I'm looking at you especially), types a long post with a couple of new pieces of information buried amongst things they've already said a thousand times, I for one will probably miss it, cause re-reading the same thing over and over again, is slowly driving me insane.

And now you buggers have me waffling on about all this crap. Anyway, it's not my place to tell you all how to conduct yourselves, and I doubt you'll listen to me anyway. You can tell me to f**k off, but I'm off to the pub anyway.

Have a good weekend.

[3phase]

Good God this thread is hard work to sift through.

"""""Would be best if everyone did their best to keep it strictly technical.

""""
3Phase: I have no interest in personal attacks, but I question logic about how you phrase your posts. It's quite possible to be campaign strongly for your issues to be addressed without constantly questioning the competence of ableton staff.

And now you buggers have me waffling on about all this crap. Anyway, it's not my place to tell you all how to conduct yourselves, and I doubt you'll listen to me anyway. You can tell me to f**k off, but I'm off to the pub anyway.

Have a good weekend.

good post. sure i am a grumpy old bastard and i know that the guys at ableton are very competend people with lots of programming expertise and practical music experiance..

But.. as often in the computerworld programmers sometimes are not so 100% practical thinking than the oily car mechanic.. and when this results in implementations that look like done by clueless incompetent people me bad guy is emphazising on that to get them at theire developers pride.
nasty? yes.. but when this results in a sync algo they can be proud of i am proud of them too...


For your problem with arguments going in circles.. you need to read such weeklong threads on one item rather from the end to the beginning than from the beginning to the end..

especially i tried at the end to give more constructiv input than doing ableton bashing..

but i think i ve said everything i had to say in that department..

i ve to sync wright now to another ableton guy on pc ..how we do that now ?? :-/

a cheap apple laptop with a cheasy pc..

anyway. rave on

[tchan]

another encouraging test session this afternoon
here are our latest notes:

10.08 SYNC TEST

Setup:
laptop 1 (master): macbook, 2.16ghz core 2 duo, osx 10.6.4, live 8.2, max for live 5.1.5, monome 128, jazzmutant lemur (mu patch), apogee duet interface

laptop 2 (slave 1): macbook pro-15, 2.4ghz core 2 duo, osx 10.6.4, live 8.2, max for live 5.1.5, iPad (touchAble control app), motu 828 mk2 interface

laptop 3 (slave 2): macbook pro-13, 2.26ghz core 2 duo, osx 10.6.4, live 8.2, max for live 5.1.5, iPad (touchAble control app), rane ttm57sl mixer


Setup Notes:
- all laptops and iPads connected to the same ad hoc wifi network
- laptops synced using network core midi in osx
- all 3 live sessions had a sample rate of 44.1khz w/ buffer setting of 128 samples
- we turned OFF serato SL for this test > with serato open you can only use the "serato virtual audio device" driver > with bridge disabled & live running on external sync, the tempo fluctuations were crazier than cee-lo…as in +/- 30bpm per half bar!
- with serato turned off, we could use rane's ttm57sl audio driver in live and external sync was working as expected
- i suspect we weren't running serato SL in our last sync test > i only marked down the timing fluctuations and was relaying the setup from my faulty memory


Sync Results (fluctuating tempo, duration = 785 bars)

[1] 107bpm > bar 0 - 385
mbp-15 (slave 1): tempo fluctuation from +/- .06 to 0.26bpm
mbp-13 (slave 2): tempo fluctuation from +/- .05 to 0.30bpm

[2] 120bpm > bar 385 - 465
mbp-15 (slave 1): tempo fluctuation from +/- .03 to 0.35bpm
mbp-13 (slave 2): tempo fluctuation from +/- .03 to 0.49bpm

[3] 125bpm > bar 465 - 564
mbp-15 (slave 1): tempo fluctuation from +/- .07 to 0.39bpm
mbp-13 (slave 2): tempo fluctuation from +/- .02 to 0.32bpm

[4] 82bpm > bar 564 - 661
mbp-15 (slave 1): tempo fluctuation from +/- .02 to 0.12bpm
mbp-13 (slave 2): tempo fluctuation from +/- .04 to 0.13bpm

[5] 94bpm > bar 661 - 735
mbp-15 (slave 1): tempo fluctuation from +/- .02 to 0.12bpm
mbp-13 (slave 2): tempo fluctuation from +/- .04 to 0.13bpm

[6] 82bpm > bar 735 - 785
mbp-15 (slave 1): tempo fluctuation from +/- .01 to 0.14bpm
mbp-13 (slave 2): tempo fluctuation from +/- .05 to 0.13bpm


Test Notes:
- we're really happy with our wireless clock sync so far
- it's definitely unsettling to see continual tempo fluctuation every half bar on the slave machines…but when you step back and listen, everything sounds good...and that's what matters!
- i used the lemur to adjust the tempo and the audio was quite smooth during all tempo changes > during the changes, all the audio from the slave machines stayed in time with no audible jitter…almost as if the clips were outputting from the same laptop
- the mbp-13 had marginally higher fluctuations but it wasn't audible > perhaps the lower processor specs come into play
- we won't know how the processing power affects clock sync until we start adding a few more resource hungry live devices and start messing around with the clips more
- CPU load varied for each laptop but hovered at the 15% range > once again, minimal effects and live devices were used in testing
- we ran another test that lasted over 1300 bars and the performance was great > there were a few audible glitches here and there but no show-stopping catastrophes

Future Testing:
- conduct testing at another location with another player to make sure this wasn't a fluke
- test at a downtown performance venue where there is sure to be more wireless interference

[3phase]

the result of the sync test between apple and windows:

fluctuations around +/- 0,7 bpm reguulr with peaks around +/- 1,5 bpm...

that seemed to be in sync "somehow" but all the tracks from the dj set just sounded dreadfull because that fluctuation is big enough to make the tracks sound fishy.. in repitch mode..
sends shivers up my spine

switching to beatwarp really kills the sound totaly you cant do that on ready mixed files for a dj set...
at least i cant do that , sends shivers down my spine..

so..free sync.. it´´s nowhere as good as running two apples.. the drift is too high.. you need to resync constantly.. but better dont touch the nudge for that..
at least i get this much better done with traktor or fruity loops or a real record..

so next step..other audiointerface for the pc laptop and wordclock sync between the audio interfaces..

result..perfekt..

intersting: the pc had a much better start performance than the mac.. so my macbooks variying response to the start is maybe a bug or a flaw of the macs keyboard scanning?

anyway..with the stable macs clock out feeding all drummachines and hardware sequencers and the windows pc just wordclock synced to be on the same timebase woks perfect and actually much better as any try to involve midiclock, what just sounds dreadfull on +/- 0,7 bpm fluctuation allready..
And the start performace when resycing the external sequncers or whatever.. theese wooooahhh oooooa ennng in the first bar or even 2.. ..sorry.. i dont do that on stage...
and as mentioned earlier it is anyway not necessary that the slave has to do guess work on the tempo.. the user might just tell it the start tempo by setting it.

So as known before, just dont use clock slave sync.. newis...free sync works even better with a windows pc..

just dj sync.. with the good start performance of the windows version easy to achive.
And the wordclock syncs precisly the set tempo. you just need one little tip on the nudges and you get tight enoungh..actually thanks to the good start performance just stop/start again gives better results than the fast speeding ablton nudge.. and after beeing alligned that runs totaly stable for hours.. you just switch tracks and adjust with the offsets delays. what seems to be click free on pc..

so actual state for best quality in mac/pc syncing..

mac gives clock to external machines..windows machines cant do that without wobbeling the drummachines too much.
pc runs unsynced but wordclocked

wished optimistions for that free sync mode:
better nudging and better start performnace on macs if possible..


conclusion for actual state :

never try to sync via midi clock. use audio cards that have wordclock or spdif ports to establish samplerate sync.. adat ports might work aswell on rme intefaces but in general the adat protocol is not ment to be used as sync refference.. so you cant be sure that this gives desired results with any interface..
in general you should only buy devices with a wordclock I/O.. thats actually a quality proof for interfaces by design..

regarding the network sync..seems to give better results when 2 macs are involved..but only on newer probooks as it looks because i nevver jad such goood readings...and..
after seeing how horrible +/- 0,7 bpm sounds i am not to confident about 0,35 bpm fluctuations either..
intersting the wireless test above indicates that it gets worse on higher tempos..i wonder why? wther ist 60 or 200 bpm shouldnt bother the result of the tempo reading algorythm at all..except the ablton algorythm is really not rounding at all and just passes the jitter unfilterd to the reading.in this case the error is relativly bigger as faster the speed gets..

and..testing on 125 bpm gives often better results in tempo measurements because 125 bpm matches wth the standard samplerate.. its better to test on 120 bpm to see what can go wrong...


So ... my advice
wordclock sync is mandatory for max quality of the output of the multi laptop setup..
especially when pc laptops are involved... or., I have no way to test, .at least in the mixed mac/pc setup..

And the usefullness of wordclocksync for max quality will probaly still be the case when ableton got a stable tempo reading algorythm. the best resyncing algorythm is one that never needs to kick in thanks to the wordclock..

the best thing ..after all the hazzle my partner decided that its to troublesome for him to dj with live
:-/ so there wont be a dj/livset mixture just vinyl or live. ok..

another 4 hours wasted to confirm allready known things..

but the report might be usefull for others in the same situation.

[3phase]

one other observation..when beeing in EXT sync mode..its is possible to switch that off and your slave just runs allong with out iteruption.. great

but.. siwtching it on while running again even on the wordclocked sytem causes a glitch that allmost sets the slave in the offbeat.

i wonder if it wouldnt be very nice and usefull when the slave just smoothly adjusts to that external clock but keeps his actual timing offset up to the next restart event..this way one can prepare for slave operation by switching the EXT button without damaging the performance and the straight resyncing only happens on the next restart..this way the slave operator can decide itself at any time to get out of his slavery and back in without damage to the performance..after having ensalved himself again he might request a restart from the master..if necessary and gets some echo on. so the restart gap is covered from his side..

with set tempo recogition this would allow seemless operation

with song position pointer / song mode clock that could be even enhanced to a point where a user might load another projekt..switches the ext button and gets auto started at the next syncpoint without having the master restarted..

that would be comfi also for the allready blessed mac network syncers..

seems that the good results in the inhouse network syncing tests with theire apple probooks has kept ableton a bit unaware about the the problems users with other machines might have...