Live 5.0.3 starts in the middle of the beat

ak · Post by ak » Thu Dec 29, 2005 12:36 am

Hello,

maybe this is just a new feature and i don´t know how to switch it off.

When i stop then start, all 4 bar sequenced start at arround the 3rd bar.

like continous mode of a drummachine.

but i like to start at 1 not at 3 or in between. can this be switched off ?

loading the same sonf into 5.0.2 is fine, just with 5.0.3 this happens.

thanks

ak

Machinate · Post by **Machinate** » Thu Dec 29, 2005 12:42 am

yes, I have reported this bug as well.. Haven't heard from them yet.

It seems that if you retrigger the scene, the clips start from the proper place

eyeknow · Post by **eyeknow** » Thu Dec 29, 2005 2:23 am

it's a bug

my system, amd 64 4000+, 2 gigs ram, audiophile 192, too much shit actually!

Anyways, if you have not already tried, try first launching a sceene first then go about your buisiness (record and such) and it seems to help.

Alex/Ableton this is not 100% but happens allot in most recent release.

To reproduce,

Open any project
Use record of a clip as reference (someone elese add other instances if applicable)
Click record

Result, about 40% of the time the clip(s) start about halfway. If someone else has a better explaintion......list it

supster · Post by **supster** » Thu Dec 29, 2005 2:33 am

yes, same here.

you need to retrigger the clip manually to get it to start on the '1'.

arg .. guys ...

EDIT: yes the re-triggering is definitely fucked now.

- start two clips in session grid at 1.1.1.

- stop the transport before they cycle around their loop

- starting the transport should bring the clips back to 1.1.1.

it doesnt. the clips transport gets unlinked from the main transport and its starting from where it stopped

this makes the session grid pretty useless.

im back in 5.02 now - works fine. guess i have to stay with the previous version until this is fixed.

this goes back to what i was saying elsewhere about using the software:

please, please .. ableton guys ... please work with it for a little while - just work on some songs, play live for a while - play with it thoroughly on different systems and see what comes up before you give the software out to people.

this should have easily been caught before the update was released. it reall y should have ....
.

.

Machinate · Post by **Machinate** » Thu Dec 29, 2005 10:35 am

supster wrote:- start two clips in session grid at 1.1.1.
- stop the transport before they cycle around their loop
- starting the transport should bring the clips back to 1.1.1.
it doesnt. the clips transport gets unlinked from the main transport and its starting from where it stopped

I don't get this error. However:

- start a two bar clip.
- Retrigger it (with play button on clip) before the second bar.
- Stop it.
- Hit play again, the clip should start from 2.1.1, even though the arrangement position and start marker both read 1.1.1 .

Retriggering again with the play arrow on the clip does not solve the problem, but retriggering with the clip button works.

Comfy · Post by **Comfy** » Thu Dec 29, 2005 12:25 pm

Yup!, had the same problem last night. A 2 bar midi pattern which i recorded kept starting on beat 6 or 7.

ODD

Post by **Alex** » Thu Dec 29, 2005 1:24 pm

Hi folks,

we'll check that.
Just a bit patience, half of the crew is in holidays and the other half is ill

regards,
/Alex

Machinate · Post by **Machinate** » Thu Dec 29, 2005 1:27 pm

thank you alex! good stuff!

jethrosipho · Post by **jethrosipho** » Thu Dec 29, 2005 2:35 pm

I have noticed this too -- here's what's happening:

Samples in Session view are following where "Arrangement Selection Start" on the arrangement page. Session view clips should not be constrained by this, but they are at the moment.

As a temporary fix, be wary of where the yellow "Arrangement Selection Start" locator is on the arrangement side, and make sure it's set at 1,5,9, etc...

J

leisuremuffin · Post by **leisuremuffin** » Thu Dec 29, 2005 3:26 pm

supster wrote: this should have easily been caught before the update was released. it reall y should have ....
.

.

I blame you for not participating in the beta of 5.03.

.lm.

supster · Post by **supster** » Thu Dec 29, 2005 5:13 pm

leisuremuffin wrote: I blame you for not participating in the beta of 5.03.
.

.

eyeknow · Post by **eyeknow** » Thu Dec 29, 2005 11:37 pm

Alex wrote:Hi folks,

we'll check that.
Just a bit patience, half of the crew is in holidays and the other half is ill

regards,
/Alex

Oooooooo, I hope it's not that damn flu!

Stefan Haller · Post by **Stefan Haller** » Thu Jan 05, 2006 3:11 pm

Sorry for the delay, I'm just back from vacation.

What you are describing is not a bug; we consider the previous behaviour to be broken, and fixed it in 5.0.3, in our opinion.

The details may be a bit hard to follow, but I'll try to explain what's going on:

When you stopped and restarted a song (which implies a song position jump from where you stopped to where you are restarting from) in the old version, all running session clips would resume as if they had been fired at song time 1.1.1 (regardless of when they really had been fired). Which means that if you were restarting the song from the beginning, all clips would start at their start markers, which might seem reasonable in a way; however, if you started at a later position in the song, the clips would start somewhere in the middle at a more or less arbitrary position, which is completely broken. In either case, the clips would lose they relative offsets against each other, which we considered the biggest problem.

In 5.0.3, we changed the behaviour so that the clips retain their fire times (relative to the arranger) when the song is stopped and restarted. For example, if a clip was fired at bar 5, and you stop the song and restart it from bar 6, the clip will start playing one bar after its start marker. This does mean that clips may sometimes start in the middle (depending on when they were fired) even if you start the song from 1.1.1; on the other hand, all clips keep their relative offsets that way.

For those of you who complained about this change, I'd like to know why exactly it's a problem for you.

supster · Post by **supster** » Thu Jan 05, 2006 5:09 pm

In 5.0.3, we changed the behaviour so that the clips retain their fire times (relative to the arranger) when the song is stopped and restarted. For example, if a clip was fired at bar 5, and you stop the song and restart it from bar 6, the clip will start playing one bar after its start marker

this sounds very much like the "orange dot" paradigm. this is a way of keeping track of time that confuses a lot of people. honestly? to me it makes no sense, and i dont understand why you guys are still pursuing it.

if you are in session view, if you stop the transport, instinct tells you you expect the whole song to start back at the actual 1.1.1

OR - hold down a modifiyer key (like SHIFT) while you hit the spacebar to start exactly from the place you stopped it from.

thats it.

anything else makes no sense to me. if i stop at 6.1.1, why would i possibly ever want to restart my session song from 2.1.1???

since most of my projects in session view (like a lot of peoples i believe) are collections of small (4 or 8 ) bar clips, we never usually get far enough to where the transport ever restarted at anything other than 1.1.1.

we are testing arrangements, trying different effects, doing soundesign and modifying clips, and when we restart we just want to start at 1.1.1, so its predictable.

so this behavior seems totally brand new and confusing. again - as with the orange dot - how some of you came to the conclusion that this would be a useful and easily understood behavior is mystifying

its so bad that i went back to 5.02 because of it. for me this basically destroys the flow of session view; wrecks it ...

and now the only chance to make it functional again is to come back on this board and argue the point in a long threads justifying why the behavior never should have been changed in the first place?? agghhhh ... (heartburn)
.

Comfy · Post by **Comfy** » Thu Jan 05, 2006 5:28 pm

I'm with supster on this one!