April 2012 - Pro-apps-dev - lists.apple.com

renumbering parameter IDs / editing .moefs
by Paul Miller 30 Apr '12

30 Apr '12

But when I use the updated .moef none of my published parameters show up. What am I missing? _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com We have about 100 .moef templates which publish parameter IDs that start with 100. We'd like to renumber the parameter IDs to start at zero, without having to recreate all of the .moefs. I wrote a script that found all the target elements inside the publishSettings element, and renumbered the channel attribute to be in the new range.

1 0

Re: 10.0.4 - does it fix the parameter id 63 bug?
by Darrin Cardani 30 Apr '12

30 Apr '12

On Apr 30, 2012, at 7:03 AM, Paul Miller wrote: > Was wondering if the parameter-with-id-63 bug crash was fixed in 10.0.4? I'm at the position where we have an existing set of plugins that work in FCP 7 that all start at id=0, and I'd like the FCPX versions to match their IDs if at all possible. Yes, it was fixed 1 or 2 versions ago. Darrin _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

10.0.4 - does it fix the parameter id 63 bug?
by Paul Miller 30 Apr '12

30 Apr '12

_______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com Was wondering if the parameter-with-id-63 bug crash was fixed in 10.0.4? I'm at the position where we have an existing set of plugins that work in FCP 7 that all start at id=0, and I'd like the FCPX versions to match their IDs if at all possible.

1 0

Reset of Custom parameter = no work?
by Paul Miller 30 Apr '12

30 Apr '12

Okay this one I think is legit. Can anyone confirm this? _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com I have a custom parameter that I've initialized with a default "empty" NSData (0 length). Once I've assigned some data to the parameter (using a custom UI), if I press Reset, the custom parameter does not get its default value set back - it remains the last-modified value.

1 0

Re: alpha channel in FCPX?
by Paul Miller 30 Apr '12

30 Apr '12

On 4/30/2012 8:05 AM, Paul Miller wrote: I have a filter that makes some holes in the alpha, but in FCPX the holes are not getting blended correctly. I have the PreservesAlpha flag set to NO. Is there anything else I need to do? _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com Oh crap nevermind - looks like FCPX uses premul alpha and FCP uses straight alpha.

1 0

alpha channel in FCPX?
by Paul Miller 30 Apr '12

30 Apr '12

_______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com I have a filter that makes some holes in the alpha, but in FCPX the holes are not getting blended correctly. I have the PreservesAlpha flag set to NO. Is there anything else I need to do?

1 0

Re: FCP X parameterChanged bug?
by Darrin Cardani 26 Apr '12

26 Apr '12

Thanks so much! This is very helpful. Darrin On Apr 25, 2012, at 5:04 PM, Garrick Meeker wrote: > It looks like all the parameter get/set calls as well as the keyframe calls are mismatched (at least from parameterChanged). Image checkouts use the current frame. > > I put together a sample that makes this easy to reproduce, as well as the image checkout issues (8-bit only in parameterChanged, upside down in FCP X). I've filed two bugs, but it's the same sample. Click on the checkbox and you'll get a bunch of NSLog calls, and the behavior is different in FCP X from Motion. If the image is 8-bit, you'll get a message. You can compare the pixel values at (0,0) against the logged value from renderOutput. To make sure the correct image is retrieved, use a clip with a cut or obvious change at (0,0) and move to the first frame of the cut. > > Image checkout during parameterChanged: > Bug ID# 11322252 > Parameter and keyframe mismatch: > Bug ID# 11322373 > > Let me know if you need more information. > > On Apr 25, 2012, at 9:17 AM, Darrin Cardani wrote: > >> Garrick, >> I'll have to investigate more to figure out whether it's a bug or not, but please file a Radar anyway, as clearly it needs to be better documented. >> >> Darrin >> >> On Apr 24, 2012, at 1:34 PM, Garrick Meeker wrote: >> >>> I'm not sure if this is a bug or if I'm using the API incorrectly. This works in Motion but not FCP 10.0.4. In parameterChanged, I call startAction, startUndoGroup, and currentTime. Then I call: >>> >>> FxKeyframeInfo info; >>> FxInitKeyframeInfo(info); >>> info.time = t; >>> info.style = kFxKeyframeStyle_Linear; >>> [_keyframeAPI addKeyframe:info >>> toParam:parmId >>> andChannel:chan]; >>> >>> for chan in channelCount, followed by setFloatValue. The values and keyframes are changed on the PREVIOUS frame. If I'm at frame 60, the keyframes and value changes happen at 59 (but hasKeyframe will still report that there's a keyframe at 60). >>> >>> Is currentTime (and renderOutput) starting at 1 but the parameter and keyframe API's expect to start at 0? It seems like that's the case, but I'd like to get confirmation before I file a bug. >>> >>> >>> _______________________________________________ >>> Do not post admin requests to the list. They will be ignored. >>> Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) >>> Help/Unsubscribe/Update your Subscription: >>> https://lists.apple.com/mailman/options/pro-apps-dev/dcardani%40apple.com >>> >>> This email sent to dcardani(a)apple.com > _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: FCP X parameterChanged bug?
by Garrick Meeker 26 Apr '12

26 Apr '12

It looks like all the parameter get/set calls as well as the keyframe calls are mismatched (at least from parameterChanged). Image checkouts use the current frame. I put together a sample that makes this easy to reproduce, as well as the image checkout issues (8-bit only in parameterChanged, upside down in FCP X). I've filed two bugs, but it's the same sample. Click on the checkbox and you'll get a bunch of NSLog calls, and the behavior is different in FCP X from Motion. If the image is 8-bit, you'll get a message. You can compare the pixel values at (0,0) against the logged value from renderOutput. To make sure the correct image is retrieved, use a clip with a cut or obvious change at (0,0) and move to the first frame of the cut. Image checkout during parameterChanged: Bug ID# 11322252 Parameter and keyframe mismatch: Bug ID# 11322373 Let me know if you need more information. On Apr 25, 2012, at 9:17 AM, Darrin Cardani wrote: > Garrick, > I'll have to investigate more to figure out whether it's a bug or not, but please file a Radar anyway, as clearly it needs to be better documented. > > Darrin > > On Apr 24, 2012, at 1:34 PM, Garrick Meeker wrote: > >> I'm not sure if this is a bug or if I'm using the API incorrectly. This works in Motion but not FCP 10.0.4. In parameterChanged, I call startAction, startUndoGroup, and currentTime. Then I call: >> >> FxKeyframeInfo info; >> FxInitKeyframeInfo(info); >> info.time = t; >> info.style = kFxKeyframeStyle_Linear; >> [_keyframeAPI addKeyframe:info >> toParam:parmId >> andChannel:chan]; >> >> for chan in channelCount, followed by setFloatValue. The values and keyframes are changed on the PREVIOUS frame. If I'm at frame 60, the keyframes and value changes happen at 59 (but hasKeyframe will still report that there's a keyframe at 60). >> >> Is currentTime (and renderOutput) starting at 1 but the parameter and keyframe API's expect to start at 0? It seems like that's the case, but I'd like to get confirmation before I file a bug. >> >> >> _______________________________________________ >> Do not post admin requests to the list. They will be ignored. >> Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) >> Help/Unsubscribe/Update your Subscription: >> https://lists.apple.com/mailman/options/pro-apps-dev/dcardani%40apple.com >> >> This email sent to dcardani(a)apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: up-side-down image in Motion update
by Darrin Cardani 25 Apr '12

25 Apr '12

On Apr 25, 2012, at 10:03 AM, Paul Miller wrote: >> The planned fix is that all images returned via a Motion Effect will actually have their origin in the lower left, just like when running in the Motion.app, and they will all be marked as such. At least that's the plan right now. I can't guarantee it won't change (though it seems unlikely). I also can't guarantee that it will be in the next release - that's not decided by me. > > So we're talking about FCPX here right? Yes. When I said "via a Motion Effect" I meant a "Final Cut Pro Effect containing your FxPlug, created in Motion and applied to footage in FCPX." It's a shorthand we use sometimes. Darrin _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: up-side-down image in Motion update
by Paul Miller 25 Apr '12

25 Apr '12

The planned fix is that all images returned via a Motion Effect will actually have their origin in the lower left, just like when running in the Motion.app, and they will all be marked as such. At least that's the plan right now. I can't guarantee it won't change (though it seems unlikely). I also can't guarantee that it will be in the next release - that's not decided by me. So we're talking about FCPX here right? _______________________________________________ Do not post admin requests to the list. They will be ignored. Pro-apps-dev mailing list (Pro-apps-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/pro-apps-dev/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0