Analyze ALL footage for Warp Stablizer??

[1johnricca]

FCP 7 had a way to analyze every imported clip for the "smoothmotion" effect, which is Apple's identical twin to the "warp stabilizer" effect in Premiere Pro CS6.

Does anyone know a way to do this or a workaround to achieve the same basic thing? Simply making a sequence with every clip in it and adding the Warp to all of them makes my Warp stuff kind of glitch out. Clips lose their analyze data, warp stops functioning on some clips even if I re-analyze, and they don't always auto start the analyzing.

The way to do it in FCP was to actually add a new column in the project browser. There was kind of a hidden smoothmotion column. Add that and right click it, it gives you the option to analyze for smoothmotion in background. Select all clips, right-click, and boom. Leave it over night.

If someone could show me a way, you are a god!

[mjs4ever]

I completely agree with these sentiments. Even if there is no "Stabilize All" or "Analyze All" or if the rendering takes a week and a terabyte, I want the renderer to warn me "Not all clips have been Analyze - do you want me to Analyze remaining or Skip analysis?" I fail to see how a blue banner in the output file will make either rushed or perfectionist editors happy. Extra aggravating is that you can go through and click "Analyze" on several clips without any warnings and only later find that they have stopped Analysis mid clip. If Adobe thinks this is a good implementation, I want them to provide free button clickers that come to my computer and click "Analyze" on one clip after another. I'll provide a comfy chair. Thanks for the ProDAD suggestions - may provide relief for the surely miniscule minority of editors that find themselves with several shaky clips in their timeline and have other things to do than babysit the analysis process.

[Guest]

So your answer Should have been "No, not yet, hopefully soon."

Actually, Jim's answer should have been, 'No. Hope, will never happen, since it's silly to analyse extra data that don't need to be analysed at all'. Not to mention that stabilising with Warp Stabilizer requires sometimes some compositing work.

By the way, if you want to code something for PrPro, feel free to explore Premiere Pro SDK.

[1johnricca]

This community is a complete joke. I ask a question and get irrelevant answers and then an answer that hopefully the exact opposite of what I'm looking for happens.

I hope you're all proud of yourselves. A big pat on the back to all.

[Jim_Simon]

Your question was answered several times (which ultimately is the purpose of these forums).  You can't do what you want in PP.

Not liking the answer isn't going to change it, though.

[SFL46]

If your concern is with regarding to applying the effect to multiple clips, you could create a nested sequence and then apply the effect to the nested sequence.  But as others have specified, this will result in bloated project files and extensive processing time.  I have not worked with FCP, but I have worked with other third party warp stabilizer effects, and they have all worked the same way, you apply the effect, it analyzes the data, and then adjusts the image.

I disagree with your assessment of the community.  I have always received the support I needed when I've posted.   Remember that I and others who attempt to help are not paid Adobe employees.  If you think this is a feature that needs to be incorporated into Premiere, file a improvement request with Adobe.

[1johnricca]

I don't need to hear about the philosophies of how or why this feature should or should not exist. Simple answer, yes or no, and any input on how others attempt to analyze a bunch of clips in one go is what I clearly asked for.

I don't need to be told why I shouldn't do this or why adobe should never make this a feature. I am not attempting to analyze unnecessary data. There are times where a large group of shots are shaky from a windy day or just a bad assistant. I go in, make my selections, and then am left with a sequence of all the shots I hope to be usable. I would like to know if anyone has a workflow for this that doesn't involve doing it one at a time.

Nesting the entire sequence of clips and analyzing that would result in the most unusable disaster of a shot and a prompt that would say that warp can not be applied to the clip. You have to understand how these effects work to know what I mean, but trust me on it. If only it were that simple and easy.

And I'm a big boy. I don't need to be told why its a bad idea to analyze a bunch of clips. I pick all the shaky shots that I hope to see fixed and am looking for a way to put WS on all of those. The memory implications and file size issues are fully understood and I appreciate your "concern" for me over this issue.

[Guest]

I don't need to hear about the philosophies of how or why this feature should or should not exist

Yes, you do need, 'cos it's namely you who is arguing this feature should be implemented by all means.

I am not attempting to analyse unnecessary data

Yes, you are, since you're attempting to analyse whole footages in Project panel instead of particular clips in the timeline. In the timeline you are not prohibited from applying Warp Stabilizer onto as much clips as you want (and your spec allows) and leaving Warp Stabilizer running analysis phase overnight.

[Guest]

You declared you are a good coder, being able to develop what you want within a day - you got the link to Premiere Pro SDK. Quite relevant.

You argued the feature you want should be implemented, but got completely opposite answer - you have an opportunity to analyse why those answers were opposite. When you have time, 'cos all the neccessary 'data' for such analysis, which are more than relevant, are included in this thread.

[CooperRoss]

Yes, for anyone working on a large project, it would silly to have EVERY clip in your project analyzed for warp stabilizer. But I believe most people on this thread are not aware of what former Final Cut Pro users such as myself had to "give up" when we made the switch to Premier Pro. Final Cut 7 had the option to analyze all clips in the project and it didn't bloat the project size or eat up enormous amounts of available memory. Those who are familiar with Final Cut X know that by default it analyzes all clips in the project in the background without the need to ask it to do so (add a filter). What this means it is, once the background analyzing is finished (and it analyzes very quickly) every portion of every clip is pre-analyzed for stabilization. So you don't have to wait for the blue analyze bar or the orange stabilizing bar to go away. This is a SIGNIFICANT timesaver for those of us who deal with terabytes of "run-and-gun" type footage in documentary/reality style projects. It's a shame there has been so much snarkiness on this thread because this is a huge advantage Final cut has always had and still has over Premier Pro.

[Ann Bens]

FCPx might be faster but the results are very poorly compared to Pr. And that is not to be snarky, this is from my own experience.

[Jeff Bellune]

I am editing this topic to remove posts and content that are too personal and/or aggressive in nature.  Please stick to debating the merits of the feature request and keep the nastiness out of it.

Jeff

[Ste777]

What about if you've applied the Warp Stabilizer effect to, say, 15 clips in a sequence and you want Premiere to Analyze them all, but you don't want to manually tell it to analyze 3 at a time (currently, it makes me go to each clip and click Analyze), you'd rather have Premiere analyze one clip after another until they're all analyzed (e.g., so you could leave it going overnight, say)?

Is there a way, in that sense, to set it to batch analyze? Telling Premiere to Render Effects In to Out doesn't seem to do this as it seems to think it's still missing that integral "Analyze" click.

I also fear putting all of the clips that require stabilization into one sequence and then adding the Warp Stabilizer to that because I reason that different clips require different degrees of stabilization and I don't want a relatively steady start of one clip affected by the shaky footage from the end of the clip that precedes it. Is that fair or am I crazy to think that?

Notes: I'm running Premiere Pro CC 2014. I responded to this topic instead of creating a new one because it was related and is the top result when searching for my question on Google using a bunch of different phrasings, hoping to help others like me--apologies if that isn't proper forum etiquette.

[Gnorman]

I completely agree with the OP about asking for a solution to this issue and getting none except snarky retorts and jabs about being wrong for asking the question in the first place. 

While the replies are accurate (I suppose...) they don't help solve the fundamental problem the poster has asked help with.

The worst response anyone can give in these forums is - don't even bother.  It really discourages anyone from using this resource or contributing to the community in the future. 

I came to this thread looking for an answer to the same problem and Ste777 outlines the reason for this question perfectly.

When I rendered out a long wedding recently I was horrified to find a bunch of Blue Banner error messages across multiple clips because I had either forgot to analyze the clip - or the analysis had stopped itself automatically - or that analysis data went out the window when I reopened the project - OR maybe I changed a transition and now there was more "frames to analyze".

I was very lucky that I review the work before showing it to clients, but it just sucked 30 minutes to an hour of extra time hunting and pecking for clips in my long multicam sequence to analyze - repetitively even because the analysis sometimes doesn't continue to completion for no reason.

Not a stupid question at all, a very necessary feature if you ask me to allow Adobe to keep its current market advantage over Apple's FCP X blunders, and one I hope someone will fix - especially the original poster with their coding skills! 

[andrewnoel]

I agree with 1johnricca this should have been a feature long ago - wether or not people choose to use it is up to them, but why hasn't anything been implemented? It seems this topic has been open and the requested feature submitted formally several times.

[R Neil Haugen]

The decision process, within the Adobe programs as far as the user-base can see, is only partly driven by the program teams themselves. I've met most of the people involved with the back end of PrPro and SpeedGrade, and several other folks, at NAB in Vegas the last couple years. Every one of them does their own professional editing/grading work besides working for Adobe ... so they're heavy users of this stuff themselves. And rather nice people also.

But they're not the ones who call the shot for what bugs & features get included in each "cycle" ... there are people up above the teams who are tasked with studying the tabulated list of bug & feature reports to parse out how much of which section of the user base is affected; ... with doing market-research analyzation for what new features would get the most "buzz" or interest; ... and with determining the list of things that are worked on by the teams.

To get allocated budget, something has to be either way up the bug or feature report list, or catch the fancy of the marketing team. And then throw in that some "UI features" are decided even higher than that, that X programs across these areas all have to have the same color scheme or whatever ... at which point (unfortunately) the users of one program who intensely dislike the 'new' look of their program are cast against the users of a different program who love the new look.

Among the features talked about over the last three years of "this" topic, this issue has a very small request rate here on the forum. Would it be a useful feature? For some, yes. So, among many other things, ideally I'd like to see this implemented. For many editors, WS is a rarely used tool. For them, they'd rather have say a UI return to the "gold" text color by a long way. That's just Life ... your necessary tool isn't even involved in this other editor's tool kit. And what another needs you both look at and complain no one needs that more than you need your tools fixed. The arguments we've had here ... driven by being different humans with different ideas and needs.

Neil

[eschelar]

Whoever it is that is making the decisions at Adobe these days is entirely out of touch with a good half of their customer base.

I stopped recommending Adobe for video editing over a year ago, with a policy of "only if you have no other options available". In my case, I am Adobe-heavy on my workflow, so I still use it, but every day I use Premiere, my desire to switch back or maybe even try out some other new NLE's grows.

Honestly, CC marked a major downturn for Adobe obviously in pricing, but the bigger concern here is that without those iterative releases, many problems are just not getting dealt with or even looked at. It's all just one big endless stream of moving sideways, then moving back. I don't use CC, but 99% of the problems I have had with CC are exactly the same for CS6 as they are with current CC versions. This means that problems that existed 3 years ago are still there. It also means that quality decisions to make sensible improvements to things are becoming rare. Look at what they did to the forums. Sometimes it works OK, other times, it throws in unexpected extra lines and moves your cursor up a line or two... The decision to set this up cannot be motivated by a desire to make it work better - especially if the result is that it works worse and then leaves the "fix" queue because not enough people are complaining.

And once these decisions start happening more regularly, the type of person who reports these things starts to care less because they are less emotionally invested in the product. If I care less about Adobe and have less faith that they will behave responsibly with the information I provide, I'm very unlikely to put the time into creating a quality bug report. Or any bug report at all for that matter. This means that the quantity *and the quality* of bug reports will start to decrease. People who care and understand how things work will be less motivated to share their thoughts.

This shows that the system of "how many people are asking for it" as opposed to "is this a sensible/beneficial feature" is not well balanced. If Adobe isn't trying to make upgrades to make the product better, it's just trying to satisfy their most vocal customers and/or maximize profits. While that's a decent strategy for the short term, it just pisses off long term users (I've been an Adobe user for around 13 years and my satisfaction has dropped to an all-time low. 5 years ago, I started pushing everyone I know to stop pirating and "go legit", because this supports the company and keeps the give-take between the company and the customers happy. I no longer recommend this and since CC, I actively encourage people to not pay for Adobe software, preferably to just use CS6 - although most of them don't understand why they should have to pay full price for software that already hit its upgrade cycle 3 years ago. I have taught workshops at Universities at beginner and intermediate levels and I'd guess there's probably a good couple hundred students that I've recommended to not move to CC. For those students, the most sensible option for them is to pirate.

And my general impression is that 4-5 years ago, most discussions revolved around individual bugs and quirks of behavior (my product is broken). A far higher number of threads on the Adobe communities are now discussing problems that are systemic (my product doesn't work right/logically and still hasn't been fixed after years), resulting from a change in the way Adobe behaves towards their customers. This topic being a perfect example of that.

Products like Photoshop Touch - which had a good start, but was almost immediately dropped with several serious flaws and deficiencies and never "touched" again for 2 years until it was killed off - becoming the rather lackluster (and CC subscription requiring) LR Touch, which offers no significant benefit over previous iterations or any other software out there. It's a step sideways rather than a step forward and at the cost of good quality software with strong potential.

This is a disturbing pattern and it's happening for all Adobe software.

FWIW, I now use Mercalli for stabilization. There's even a decent workflow for doing in-place replace of footage if you have a timeline already built by holding alt (or is it ctrl?) when dragging footage in from the Source monitor.

[R Neil Haugen]

Understand. I'd add they haven't moved SpeedGrade forward since adding the new scopes and Direct Link back in the original CC or first CC2014. In some ways, it's feeling like it's gone backwards. Worries me. The new color workspace porting a fair amount of Sg's color science/controls into PrPro is in all a good thing, but parts of it aren't working exactly right (just filed bugs on the sat control bending the color left around the wheel and distorting a narrow "spike" of color into a wider diffused color mush) ... and they haven't done anything at all to announce in 2015 or the next partial-upgrade release either, as they've announced what changes are coming an nothing is slated for Sg.

The new color workspace in PrPro does mean a lot of basic work is easy enough to do it there & quickly, with a vastly better UI than say the three-way or fast color correctors ... to me, those are sick. Though they can be made to work. But that workspace is nowhere near the equivalent of a full grading program.

Neil

[BartonGarrett256]

I shoot and edit a lot of material that could benefit from stabilization.  Warp Stabilizer is a last resort. It takes forever(and I've got a $10K custom built editing computer), makes huge files and half the time it's jello. I find going into Ae and using the legacy tracker is simpler and gives better results, it's what I do most.

I sub-clip like crazy to keep the file sizes down. I can't imagine ever applying either stabilization tool to all my clips.  That would be computational and data storage suicide.  I wish the tracker worked faster and built smaller files, but I'll worry about other things, frame wider and not fret about it.  There is no free lunch.

‌

[betrion]

Even when I first used warp stabilizer years ago and saw blue lines with a message "this clip needs to be analysed" I thought it was unacceptable. Seeing this still happens in newest versions is just sad. Imagine if everything you do needs to be analyzed manually. Turned down that volume and didn't "analyse" it - sorry, back to default with you.

Why would anybody apply effect to a clip if it doesn't want it to be rendered? In that case you'd just disable/delete it. If effect is there it needs to be rendered - period. I just rendered a video for hours just to notice a blue bar on one of the clips.

Very silly especially since it should not be a problem to implement. Auto "analyze" on export is the least Adobe could do, even though "analyse everything" (maybe not even all at ones but in queue one by one or in pairs depending on processor/ram) would be nice for preview purposes.

This thread is from 2012

[Daniel Kelly Brown]

This thread is pretty junkey with a bunch of non sense.

I am in the same boat as OP.  We need a way to select all clips that need to be analyzed and have ability to analyze all rather then clicking though each one individually.

Also the dev at adobe working on this should consider making an auto nest feature.  Every time I have any effects applied to a clip that needs stabilization it give me a warning that I need to nest before stabilizing because of said effects.  Same goes for already stabilized clip when applying more effects.  Why not just make a feature that automatically nests clips rather then making us do one more silly step.

Cheers!

[Kevin-Monahan]

Hi Daniel Kelly Brown,

I am in the same boat as OP.  We need a way to select all clips that need to be analyzed and have ability to analyze all rather then clicking though each one individually.

Also the dev at adobe working on this should consider making an auto nest feature.

Create your feature requests here.

Thanks,
Kevin

[Daniel Kelly Brown]

Thanks Kevin!  You are the man! Will submit a few requests today.

You guys are constantly making a spectacular product better each and every update! 

Cheers!

[Kevin-Monahan · Accepted solution]

Hi JohnRicca,

Sorry, no. Create a feature request.

Thanks,
Kevin

[Guest]

Those future requests are bullshi*ts, even after 10y we still need analyze clip by clip. lol

[R Neil Haugen]

Quite a few feature requests do get added into the program. This one hasn't, but that doesn't make all the others crap. Get a life.

They haven't made it through all my fav requests either. But they've made some, as they have made changes for the 'user base' moving along. And yea, I'm frustrated at the ones I want that haven't made it yet.

As has been noted, talking with engineers they posit that allowing say 20 clips to be analyzed is going to lock up that computer maybe overnight. Well, if that's what the user wants, why not let them?

Maybe someday.

Neil