the id is stored in the seuqnce data of each instance. it's read via AEGP_EffectCallGeneric, which the effect can receive and send back anything you set it up to. note that you ABSOLUTELY can't call this from one instance to check out another instance of the same effect. it will call itself instead of the other instance... only an AEGP can call instances using this callback. ... View more

well... it's complicated... when saving the project while quitting, you might get a flatten call without getting a setdown call after it. in that case, you can of course free all derelict handles on global_setdown, so no biggie. it's a there's no robust instance identificaiton method, but that's what we have. ... View more

nope. no way of telling. it's an absolute nightmare. welcome to hell. for one of my plug-ins, where i needed to tell if an instant's sequence data belonged to it originally or if it's a copy, i'd scan the layer on the first non-seuqnece_data call (because in AE sequence data calls aren't always associated to a certain instance, and sometimes the original instance is called for handling the new instance's business...), and compare instances. if two have the same id in the sequence data, i know the latter of the two must be a new copy and then i create some new data for it.  ... View more

when i encounter such a problem, i start off in incemental steps and see where the problem occurs. some problems are well documented. the error you're getting, that's a first for me. ... View more

Ah. I see. I never saw this error before, so something with the renaming went horribly wrong... I'd suggest starting over with the checkout project which you're sure compiles and loads correctly. ... View more

that is indeed very strange. what happens if instead of re-opening the project you purge the ram cache? does the error correct then? ... View more

1. this might be silly, but did you do a clean and rebuild? on windows, the resource file is converted to a .rc file which in turn is the one included in he built file. the .rc file is only updated after a clean or a rebuild. on a regular build it's kept as is, so there might be a chnace you built the file with the skeleton's rc file.   2. does the skeleton sample (beofre renaming it) compile and load correctly? ... View more

i see. indeed a complicated situation. i'm not familiar with the OFX API, so some of what i would tell might not be applicable to your case.   1. sequence data. in AE, sequence data for the render threads is always separate from the one in the UI thread. you can have all render threads share one seq data handle or have each one have it's own, by using PF_OutFlag2_MUTABLE_RENDER_SEQUENCE_DATA_SLOWER. in the case of multiple render threads, a single OFX plugin shared between them may have asynchronous param values being updated (because of keyframed params, or expressions), and thefore produce wrong render results for a given frame. i don't know if that's a valid concern in the way your system is set up... perhpas you should consider the render threads copy of the OFX effect as a separate entity. it seems wasteful, but all it would mean that for every change on the UI instance, the render instances of the OFX are destroyed (on sequence_setdown) and rebuilt (on sequence_resetup) when syncing the ui and render thread seq data. if you're comfortable with that idea, then the nighmare trying to merge the render and ui seq datas into one persisant piece of memory is over. (and it is a nightmare, let me assure you)   2. initaiting a new instance. since in AE during a sequence_setup call the effect instance does not yet exist and is therefore inaccessible, i would suggest you mark the new OFX effect int he sequence data as "uninitiated", and then: - on the ui thread, use the first UPDATE_PARAMS_UI call and set the default params data then. - on the render thread, if a render call is triggered on an "uninitialized" OFX, just ignore the param values AE has, and use the deault ones in the OFX. that way, you'll get a correct render even before the UI thread has a chance to update the params.   3. sequence data is always available. is it not during xGPUFilterEntry? isn't PF_EffectSequenceDataSuite1 available there? ... View more

i'm not fluent in premier, but in AE during sequence setup/resetup the instance is not yet associated with an instance. AE knows it needs that data, so it calls upon your plugin to make that data for it, and only then does it create the instance where you can check your params.   i'll ask this: is the purpose of that process to "initialize" an effect instance's param state? if so, in AE you'll want to do that during UPDATE_PARAMS_UI, which is called immediately after an inastnace is applied. (if you set the falg in global data of course) that call, is intended only for cosmetics, i.e. showing/hiding/renaming. technically, you CAN change param values during that call, but there could be occurences where that call in not triggered (such as running AE without ui and loading a project, or automating it). if you want to change param values on application in AE (i'm assuming dynimically, otherwise you'd just set the defaults in param setup), is to flag your effect as "new" (by some hidden checkbox, or sequence data), and during the first idle_hook call, you can externally change any param value you wish.   if you'd like to describe your work scenario perhaps i can think of a more suiting solution. ... View more

amazing! thanks for sharing your excellent work! ... View more

here's a previous discussion here in the forum regarding the same issue. it has some code samples in it, though nothing complete to just copy paste. it does, however, contain a ton if information. https://community.adobe.com/t5/after-effects-discussions/non-custom-ui-drawing/m-p/4603725 ... View more

from what i saw, it uses bilinear in HQ and nearest neighbor in LQ. perhpas it only uses bilinear for upscaling, and uses a mipmap for downscaling. ... View more

no, the SDK doesn't do that. i personally use gdiplus on windows and quartz on mac, but many develoipers use cairo and qt. ... View more

well, that depends on what you mean by "draw pixels on screen". i can think of 2 meanings: 1. draw into the effect's render buffer. that buffer is used by a layer when rendering a comp's result. 2. drawing a UI on top of the comp, or in the ECW. such pixels are not used in the comp's result and are just there to guide the user.   drawbot is used only for UI. the drawbot suites will only operate on a ui buffer handed to your plug-in by AE with a special drawing content. you can't use drawbot for rendering to a "normal" render buffer.   in the "CCU" sample project, you can see how to access render buffer pixels directly. you can then rasterize text using an os buffer and os api (such as gdiplus or quartz), and then copy the pixels onto the effect buffer. you can also use 3d party libraries to rasterze the text or do some drawing and then copy the pixels over to AE's render buffer. ... View more

i don't understand. is it debugging correctly or is it not? can you set breakpoints and hit them? if you're debugging successfully, on what line do you get err code 3? are you positive the err is not gotten on a previous line? ... View more

it's hard to tell because the error codes are not check in the few calls before AEGP_GetNewKeyframeValue. please run through the debugger and watch step by step if an err is gotten somewhere. it's most likely you're arriving at AEGP_GetNewKeyframeValue with a bad StreamRef. ... View more

there are two steps here. 1. tell which keyframe is selected. for that, look into "collections". these are arrays of selected items. 2. get the keyframe value. for that, use AEGP_GetNewKeyframeValue(). ... View more

of course if you're using the native ECW idle calls you're better off using softer redraw methods such as PF_InvalidateRect or perhaps PF_UpdateParamUI (it's been a while, so forgive me if i got one of these wrong...). ... View more

well... that very much depends on what user experience you're aiming for. for example, AE's camera tracker does it's thing async in the background. however, the camera tracker creates tracking info, much like creating keyframes as opposed to processing the image. so user expectations differ. warp stabilazer does the same, but IS a filter that manipulates the pixels. it stores it's distortion data and then processes each frame with that data on request. (as opposed to storing processed frames and displaying them) storing very large data (such as multiple rendered frames) in-instance, be it in the sequence data or in an arb param, is VERY problematic in AE, is AE makes multiple copies of these data chunks and creates a huge bloat in ram usage. in such cases, it's probably best to write the frames to disk and either import them into ae, or have the plug-in read and display them directly. ... View more

ah, i see. using PF_ADD_NULL() you can rid of the stopwatch and reset buttons. you'll still have the twirly thigie. you can try to get rid of it (i've never tried it) using: 1. PF_PUI_TOPIC, to paint over the twrily triangle so it would become invisible. i don't remember if the AE's ui paints over the custom ui or the other way around. 2. if #1 works, you can make your banner param always twirl down during UPDATE_PARAMS_UI, in case the user accidentally clicked where the hidden twirly arrow is. ... View more

It's a custom ui implemented on some param. Take a look at the "color grid" sample project to see how an ecw ui is implemented. You can use a null param for the custom ui to spare your plugin some unused data being saved and handled. ... View more

1. you can pass a NULL instead of the plugin ID, if that simplifies things for you. 2. the panel class is a nice wrapper for initializing the panel and handling the window events. but by all means, you don't have to use that class. ... View more

good question. the compute cache's logenvity is while AE is open (at the most... i don't know if it might get dropped mid-session). so when the user re-launches AE, the data in the compute cache is gone. so now it depends on what the user is expecting of your plugin. some users (if not most, IMHO) use the disk cache, so they expect the project to be "as they left it" in terms of responsiveness thatnks to the cache still being there. in that case, losing your compute cache will cause your plug-in to be precieved as "slow" or "ugh that @#$%&* thing..." on the other hand, if your effect's outputs have been cached by AE's frame cache they'll still be available. but for that to happen, your effect needs to report the same value into the GuidMixInPtr() on pre-render. can it do that without the compute cache available? on the other other hand... having to relaunch the "process" button on every session... i think that's a big no no for users. they don't know the difference from that and from having to re-analyze wrap stabilizer on each session which we can agree would have been a deal braker. ... View more

a couple of thoughts: 1. keeping very large data (such as numerous fully rendered frames) has it's problems... AE makes A LOT of copies of the sequence data which it keeps around for it's own mysterious reasons, so storing 4 HD frames would give 100mb per instance, that can easily take up 1gb of ram for said extra copies. very large data is better off being wirtten to some temp files. that downside of that is that these files woun't be collected along with the project to be archived or moved to another machine.   2. if you did some background processing and you'd like to apply them to frames once done, you need to invalidate AE's cached results that rely on the previous output your plugin supplied. you can do that in a couple of ways: a. during pre-render (which AE regularly calls) put some new value into GuidMixInPtr(). b. during a UI event call, set the FORCE_RERENDER out flag. c. the most flexible way. have some invisible param on your effect, and change it's value to invalidate the effect's cached results. ... View more

I've always found rowbytes to be "whole" pixels. i.e. a buffer 10 pixels wide would have an extra 1 pixel worth of bytes. though i can't tell you if it's a rule or other sizes may occur... ... View more

start off with AEGP_GetNewCollectionFromCompSelection(). then traverse the collection using AEGP_GetCollectionNumItems() and AEGP_GetCollectionItemByIndex(). now that you've got the currently selected layer (there may be more than one selected...), use AEGP_GetLayerName(). ... View more

this is the answer i got. bummer. PF_GetCurrentState() is not implemented in the Premiere host   if you check the error return code (you should always check and respond to error codes!) it should return PF_Err_INVALID_CALLBACK   I am not sure an appropriate alternative for the same functionality on the Pr side, hopefully some other adobe person can weigh in here   --dmw   ... View more