If you're seeing weirdness in one composition but not the same
behaviour in another that you freshly create then it sounds like a
condition of the QTZ being corrupt/saved in an inconsistent state.
I've seen this happen especially when the clipboard use/plugins are
involved.
This is far from helpful but I've had a massive, complex composition
go bad after making a slight change. The only way I found to fix it
was to go back to an earlier version from backups and recreate the
changes from scratch. Second time round no problems.
Ps. Check out creativecommons.org/license
On 27 Jun 2009, at 08:00, Alastair Leith <[email protected]>
wrote:
Okay I'm going work on that and post it after this weekend. It isn't
easy. At one point I had a boolean splitter going straight to the
enable of 4 spheres and TRUE not enabling them. When I copy&pasted
the 5 patches (which had NO OTHER INPUTS) to a new blank comp, they
worked A-OK. LOol! That's the second head growing out of my
shoulder in mad frustration btw.
I intend this series of Marcos to be open source and maybe the basis
of a plugin if someone with the chops cares to do so. Just wanted to
keep my powder dry and all until I had them up and running. Can
anybody advise what License or © text to use so people are free to u
se and modify but not exploit in commercial ways (like sell without
my consent)?
One of many reason I use Macs is because I can get the gremlins
running (when I've used PCs in other peoples studios I've had their
PCs give kittens). I have been experiencing some weirdness in QC the
last month or so and am not sure it isn't my venerable Mac having a
laugh on me. If in knew more QC stuff I could say if it's abnormal
or not. Massive slow-down on save dialogue is one recent oddity.
Reboot solved it.
2009/6/27 Adrian Ward <[email protected]>
Hi Alastair,
Can you reduce your composition to the simplest possible state where
the problem is still observable? If so, post it to the list. I'd
certainly be interested in seeing why you're getting a different
output from an input and I'm sure you'll get some more help with an
actual example out there.
Sorry my suggestion wasn't much help - hope it's not driving you too
nuts.
Best,
Ade.
On 26 Jun 2009, at 09:50, Alastair Leith wrote:
Thanks Adrian
I've been re-patching the splitter when required (can be annoying
can't it). I have used iterator patch (iterations=0) to keep the
upstream patches "active" so I don't think it its the lazy
evaluation unless it's lazy evaluation going screwy (to be quite
technical!). Thanks though.
2009/6/26 Adrian Ward <[email protected]>
If you enable/disable consumer patches (purple ones) then this
affects all upstream patches. When disabled, anything before it
that doesn't need to be executed won't be, meaning it's output
value can become stale. I suspect you're seeing this when you spot
a diference between the input and output vales of a splitter.
Another thing could be that when you change the type of a splitter,
connections to/from it can be removed. I'm not sure why this is -
I'm tempted to file a bug report because it's only a nuisance and
not very functional behaviour. Perhaps you've inadvertantly severed
a connection somewhere without realising it?
A.
On 26 Jun 2009, at 06:42, Alastair Leith <[email protected]>
wrote:
I've got issues with a patch where Boolean values can be processed
thru a chain of patches okay but when the value type is changed to
any other type (say number) and then back to Boolean it will not
work.
I have the strange result of a True value on one side of an Input
Splitter Patch and a False value on the output node. I've noticed
this can occur at various points down the line of patches where it
is true in on a splitter and false out.
I'm working on a macro that can process any kind of value
(virtual) like image, number, QCstrucutre and so on. The values
pass through the macro to output un-changed but can be recorded
sent or substituted in various ways. All the values types work
fine except Boolean.
I remember cwright posting @Kineme there is an issue in QC with
the way Boolean are treated re: something else entirely, just
wondering if there is a know issue causing this mal-functioning
and if there's a known ways around it.
<true in.jpg>
<false out.jpg>
_______________________________________________
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/adrian%40clayinteractive.co.uk
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]
