[ardour-dev] Gain automation twitchiness

[Joel White]

Twitching under automation also happens for me in beta22.  In this case, 
I have a guitar track that is brought from  -7.4dB to -12.0dB during a 
chorus.  During that chorus, the fader twitches to -7.4dB (which was the 
gain on the track before I added the automation).  There is no twiching 
during the parts of the song where the automation gain is -7.4dB.

Other observations about beta22 after about 5 min of use on a session 
recorded under beta19:

- All my tracks are now panned center (good thing I didn't save it! 
Panning is fine when I subsequently opened the session in beta19).

- For a sesssion that had about a 75% CPU load under beta19, the CPU 
load is now about 45% (all the tracks and plugins still seem to be 
active)!  Wow!  More headroom!  Fantastic!

Joel
(thinking about all the extra tracks and plugins I can now run...)

Joel White wrote:

> I see the twitching fader using beta19.  I also haven't heard a 
> problem, and there does seem to be some regularity to the twitch.  I 
> just built beta22, so I'll see if it happens there.
>
> Joel
>
>
> Mark Knecht wrote:
>
>> Joe,
>>   I've seen this also. I'm on something earlier than beta22. What I
>> see is the master fader jumping down to -inf and back to the proper
>> level. I do not hear any problems so I assume that it's jsut a
>> graphics problem.
>>
>>   It often also happen on my system in time with the music - i.e. -
>> at beat 0 in a new bar. Don't ask me why. Go figure... ;-)
>>
>> - Mark
>>
>>
>> On Wed, 22 Dec 2004 06:43:10 -0500, Joe Hartley <jh at brainiac.com> wrote:
>>  
>>
>>> I have 2 tracks in a session that have gain automation on them.  I
>>> noticed that beta22 introduced a "twitchiness" on playback/export.
>>>
>>> The levels will jump momentarily to levels they were never set to.
>>> As an example, I have a vocal track set to -2.2dB for a certain
>>> section of the song, and at -0.9dB for others.  The level jumps up
>>> to zero once or twice a second in some parts of the song!
>>>
>>> There also seems to me some co-mingling of automation data.  I have
>>> a gutar track that runs at -5.9dB for parts of the song, and I see
>>> that level being jumped to by the vocal track.  The vocal track
>>> ends up at that level every time, even though I just went and re-wrote
>>> the gain automation, and it was set at -0.9dB at the end.
>>>
>>> This may have something to do with the mis-assignment of gain 
>>> automation
>>> data that I'd noticed and reported last night, when the vocal 
>>> automation
>>> got assigned to my master track.
>>>
>>> Here's a cut-n-paste of the automation file for the vocal submix.
>>> There aren't nearly as many data points on here as there were 
>>> "twitches".
>>>
>>> version 1
>>> g 11264 1
>>> g 157696 1
>>> g 168960 0.902243
>>> g 180224 0.799211
>>> g 191488 0.722795
>>> g 214016 0.722795
>>> g 225280 0.799211
>>> g 236544 0.799211
>>> g 247808 0.741411
>>> g 270336 0.741411
>>> g 281600 0.779616
>>> g 4595712 0.779616
>>> g 4606976 0.799211
>>> g 4618240 0.819139
>>> g 4629504 0.839403
>>> g 4640768 0.839403
>>> g 4652032 0.860007
>>> g 4663296 0.880952
>>> g 4685824 0.880952
>>> g 4697088 0.902243
>>> g 6904832 0.902243
>>>
>>> -- 
>>> ======================================================================
>>>      Joe Hartley - UNIX/network Consultant - jh at brainiac.com
>>> Without deviation from the norm, "progress" is not possible. - FZappa
>>> _______________________________________________
>>> ardour-dev mailing list
>>> ardour-dev at lists.ardour.org
>>> http://lists.ardour.org/listinfo.cgi/ardour-dev-ardour.org
>>>
>>>   
>>
>> _______________________________________________
>> ardour-dev mailing list
>> ardour-dev at lists.ardour.org
>> http://lists.ardour.org/listinfo.cgi/ardour-dev-ardour.org
>>
>>  
>>
>
> _______________________________________________
> ardour-dev mailing list
> ardour-dev at lists.ardour.org
> http://lists.ardour.org/listinfo.cgi/ardour-dev-ardour.org
>