Yep indeed that's the conclusion I came to as well - that the timebase was effectively rounding off some of the frame length. The timebase
My solution was lower tech - render at 25 fps, which results in a frame duration that doesn't require high precision. When I get around to it I'll do something along the lines of what you suggest to ensure the rendering doesn't get out of sync. (and commit it to google code) thanks for the tips, Rua HM. -- http://cartoonbeats.com http://haszaristwocents.blogspot.com http://myspace.com/haszari http://last.fm/music/Haszari ________________________________ From: Achim Breidenbach <[email protected]> To: Rua Haszard Morris <[email protected]> Cc: George Toledo <[email protected]>; [email protected] Sent: Sat, 2 October, 2010 7:45:08 PM Subject: Re: bpm-accurate LFO for quartz composer? Hello Rua, I looked into your code and come over this line: QTTime durationqt = QTMakeTimeWithTimeInterval(timeStep); The QTTime is a struct of two integers, which proberly don't reflect your "timeStep" when calculating the fritction timeValue/timeScale. The function QTMakeTimeWithTimeInterval() just picks a timeBase, unfortunately a "wrong" one. Check it with the debugger what exactly the values for timeValue and timeBase are. You may want to set them programmatically or even go the way back: timeStep = durationqt.timeValue / durationqt.timeScale; Another (more accurate) solution would be using the movie duration for the render time: QTTime currentMovieDuration = [mMovie duration]; [renderer renderAtTime: currentMovieDuration.timeValue / currentMovieDuration.timeScale arguments:nil]; (You may have to take your intro duration into account for the rendring time as well, I didn't looked at this part of your code) best, Achim Breidenbach Boinx Software On 02.10.2010, at 01:02, Rua Haszard Morris wrote: Aha - Indeed this is very accurate! (Nice animation by the way) > > >I tried this on one of my tunes using playback in quartz composer, and it >stuck >nicely to the kick. > > >However it drifted when I used my rendering app (qtzrendang.googlecode.com), >so >it's likely a timing bug in that or some other interaction. > > >What do you use to render your compositions & attach the audio? > > >thanks for the help >H > -- >http://cartoonbeats.com >http://haszaristwocents.blogspot.com >http://myspace.com/haszari >http://last.fm/music/Haszari > > > > > ________________________________ From: George Toledo <[email protected]> >To: Rua Haszard Morris <[email protected]> >Cc: [email protected] >Sent: Sat, 2 October, 2010 10:40:41 AM >Subject: Re: bpm-accurate LFO for quartz composer? > >Err, sorry for the string of responses, it's just that I re-read your note >about >every second beat. If you take the instance where I have a half note beat, and >offset phase by 180, then you will have it hit on the 2 and the 4 (eg., >typical >snare hit), instead of the 1 and 3. Using phase will control your offsets, and >"groove" (eg., slightly ahead or behind beat). > > >Also, I've capped the whole note at sequences of 4, but you could leave it >unranged, so that it becomes a measure count. If you change a chain to >generate >8th notes or 16th notes (or 3/4 time), you can adjust the associated range >patches accordingly, so that they actually keep track of that count. Just >throwing out some related thoughts that will hopefully help you in your quest >for tight synchronization. > > >-GT > > >On Fri, Oct 1, 2010 at 5:32 PM, George Toledo <[email protected]> wrote: > >BTW, please let me know if you do experience any drift in using this, as there >isn't any reason that I'm aware of that would cause it with this. I've run it >for a couple hours alongside a drum machine and had it keep lock. >> >> >>-GT >> >> >> >>On Fri, Oct 1, 2010 at 5:26 PM, George Toledo <[email protected]> wrote: >> >>As a matter of fact, I made something not that long ago for just this scenario. >>> >>> >>>I am attaching the "beat engine" that I used to synchronize >>>this: http://vimeo.com/15191953 >>> >>> >>>I'm using "square waves" so that a hard tick is generated on the counter, >>>but >>>you can change wave types to generate the other kinds of envelopes that >>>should >>>be familiar from synthesis via the LFO. >>> >>> >>> >>> >>> >>> >>> >>>On Fri, Oct 1, 2010 at 4:50 PM, Rua Haszard Morris <[email protected]> wrote: >>> >>>Hi list, >>>> >>>>I've been trying to animate in time with music by using LFOs running at bpm >>>>related rates. The bpms tend to be 128/130ish so are not nicely related to >>>frame >>>>rates (e.g. 120bpm would be!). >>>> >>>>I can't seem to get accurate sync between my audio (which is at a known >tempo) >>>>and the LFOs - the LFOs seem to drift. I think this is the accumulation of >>>>rounding error over the course of the animation. >>>> >>>>I wrote a BPM-LFO plugin so as to avoid rounding issues, supplying tempo in >>BPM >>>>instead of period, and this LFO seems to drift just as much. >>>> >>>>Has anyone used LFOs for roughly bpm-accurate synched animation over the >>course >>>>of about 5 minutes? If anyone's succeeded I'd love to know how to do it! >>>> >>>>BTW my needs are reasonably modest - things like an image that flashes every >>>>second beat over the course of a whole song - e.g. 64 times a minute, or a >>>>billboard that slides into view every 16 bars (64 beats) - but the LFOs >>>>need >>to >>>>ensure these elements stay in phase with the audio beat. >>>> >>>>thanks >>>>Rua HM. >>>> -- >>>>http://cartoonbeats.com >>>>http://haszaristwocents.blogspot.com >>>>http://myspace.com/haszari >>>>http://last.fm/music/Haszari >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>>Do not post admin requests to the list. They will be ignored. >>>>Quartzcomposer-dev mailing list ([email protected]) >>>>Help/Unsubscribe/Update your Subscription: >>>>http://lists.apple.com/mailman/options/quartzcomposer-dev/gtoledo3%40gmail.com > >>>> >>>>This email sent to [email protected] >>>> >>> >>> >>>-- >>> >>>George Toledo >>>[email protected] >>>www.georgetoledo.com >>> >>> >>>The information contained in this E-mail and any attachments may be >>>confidential. >>>If you have received this E-mail in error, please notify us immediately by >>>telephone or return E-mail. >>>You should not use or disclose the contents of this E-mail or any of the >>>attachments for any purpose or to any persons. >>> >> >> >>-- >> >>George Toledo >>[email protected] >>www.georgetoledo.com >> >> >>The information contained in this E-mail and any attachments may be >>confidential. >>If you have received this E-mail in error, please notify us immediately by >>telephone or return E-mail. >>You should not use or disclose the contents of this E-mail or any of the >>attachments for any purpose or to any persons. >> > > >-- > >George Toledo >[email protected] >www.georgetoledo.com > > >The information contained in this E-mail and any attachments may be >confidential. >If you have received this E-mail in error, please notify us immediately by >telephone or return E-mail. >You should not use or disclose the contents of this E-mail or any of the >attachments for any purpose or to any persons. > > _______________________________________________ >Do not post admin requests to the list. They will be ignored. >Quartzcomposer-dev mailing list ([email protected]) >Help/Unsubscribe/Update your Subscription: >http://lists.apple.com/mailman/options/quartzcomposer-dev/achim%40boinx.com > >This email sent to [email protected]
_______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to [email protected]
