Mitusents

Sorry I’ve been so quiet lately.  A combination of several things all going at once.  I haven’t felt well lately and have also been down.  Tied together or not, the result is the same.  However, I have made tiny progresses toward getting Treasure Hunt complete:

• I wrapped up the outtakes, though it’s rather long (about 30 minutes – a friend said that the few people who’d really enjoy it would prefer longer over shorter).
• I created an IMDB page for the movie.
• I submitted the film to the Skyline Indie Film Fest, which is less than an hour North of where I shot portions in Woodstock, VA.  Hopefully that’ll mean they’ll seriously consider accepting it.
• I converted the film into DCP format, though the viewer on my computer wasn’t very inspiring, as it constantly dropped frames (possibly because it’s really cheap).
• I purchased a portable drive that’s supposed to be compatible with the commercial digital projectors, though it sits on my bed just staring at me.
• I got some of the behind-the-scenes material from the young lady who filmed it, but seem to be missing the interviews she did.  Neither she nor her mother are great communicators, so I’ve been trying to get my wife to act as intermediary (mother and wife work together), yet have no progress yet to report.
• I (re)discovered the chemistry tests I did with the lead actors, so will throw that on the Bluray as well.  It’s probably not terribly interesting, but there’s some 50 GB of space available on a Bluray and the movie itself is only around 7 GB.

Psychologically, I think part of why I stalled is I can’t complete the Bluray without the behind-the-scenes and without the Bluray I can’t wrap up the crowdfunding incentives.  So I do nothing.  Not a very flattering portrait, but there you go.

I was planning on doing a private screening of the movie toward the end of April, coupled with trying to get local newspapers and regional TV to interview me about the movie, but that’s stalled.  I rationalize it to myself that I’m probably better off knowing if I’m going to be in a festival (I really should submit to more, just got to make the energy) or not, plus, I have some paperwork I need to tidy up before I can sell anything (e.g., release forms for a few actors, the composer and musicians as well as the locations (and probably a few other things)) and I’d like to have that done before I do the screening so I can offer DVD/Blurays for sale.  Tortured logic of a procrastinator or valid arguments, I can’t decide in the mental state I’m in right now.

Movie related, but not to Treasure Hunt, I submitted a screenplay to a couple of contests and decided on a next project.  Finalizing the screenplay was also rather draining and while I have a detailed synopsis written for the next project I don’t intend on writing until we go on a family vacation this summer since I have other, non-movie projects I’ve been neglecting.

Sorry I’ve been a deadbeat.

Yesterday I posted this.  Today I’m almost dancing in the street.  I tried a couple of the things I thought of yesterday morning and, lo and behold, they worked!  I was able to render the flick, with the audio in sync!  It took 2.5 hours, but at least I don’t have to be there or can work on other things.

(For those of you who might stumble across this searching Kdenlive, I turned off ‘proxy clips’ for all the subprojects and set each of the the render configurations for the highest quality.  I was/am using the appimage version, which is (or close to) version kdenlive-18.12.1b-x86_64.appimage.)

Now I need to get the music sorted out, plus check on a few things John mentioned when he last viewed it.  I also noticed, with the high quality render, that there were a couple of shots that seemed to be of poor quality.  Since the shots before and after seemed fine, I need to see if the original material has the same problem.  If so, well such is life.  If the original is clean, then I need to see if I can figure out how to fix it.

I re-rendered the trailer, after learning all these tricks.  It looks _so_ much better!  And rendered at 1080 highest quality it was still only around 65 MB.  I bet with the default quality (which, frankly, I can’t see any difference on my monitor) and at 720 (which is probably plenty large for a trailer for Youtube) it’ll be 1/10th that.  (In case you’re interested, the whole thing rendered, at it’s highest quality, at around 6.6 GB; at 720 and with the default quality level, it rendered to around 650MB.)  I still need to rejigger the trailer, I’m not happy with the middle where they’re singing, and I need to find out why the audio at the end was crappy (sometimes just re-rendering it will solve that issue), but I feel good now.

What a difference 24 hours can make!

Oh, I believe I found a near perfect festival to submit to:

https://www.skylineindiefilmfest.org/

It’s ‘right around the corner’ at about an hour North of our place in Winchester, Virginia, which is also a hotbed of the Civil War (a friend who’s deep in the Civil War tells me Winchester changed hands 72 times!).  Hopefully the topic, plus the fact that portions of it were shot in Woodstock, VA (about a half hour South of Winchester) will give me an edge getting accepted.

Nope.  I previously reported I was just about finished with the Treasure Hunt movie.  Well, the universe is not done f-ing around with me yet.  After getting all the chunks rendered in high quality (which takes much longer than before) I assembled them together, along with the audio, and now the lengths no longer match!

It’s possible I messed something up when I was ‘chunking’ things, I didn’t look closely as I was really upset when I discovered this, but I did very quickly try to sync things at the very end of the flick and wasn’t seeing things line up there.

So, clearly, the open source Linux-based software I’m using (Kdenlive) is not ready for prime time.  In order to get the high quality renders, I had to get the latest ‘appimage‘ version, but that version couldn’t reread it’s own files after upgrading from the version I had used for all the editing.  Fortunately, the file is in (mostly) human-readable XML format, so I edited it so the stupid thing could read it correctly (imagine if I weren’t a professional programmer).

The core of the problem seems to be my initial decision to cut the library scene as an independent project. Since there’s no option to copy/paste (at least there wasn’t in the old version, I should try in the new version (multiple versions because all the various Linux distros update things asynchronously; yes, there can be an, at times very heavy, price to use Linux)), I was thinking I was going to have to reedit those scenes.  However, I found out I could include another project as a subproject.  The wrinkle?  The renders weren’t high quality for those subprojects for some reason (though I thought of something this morning I’ll try this afternoon).  If the copy/paste doesn’t magically start working, I don’t know how else to get things to come together.

Professionally, I used to ‘straddle the fence’ between Windows and the various flavors of UNIX, but when Microsoft decided to go to Vista from XP and renamed/moved so many of the administration applications, I decided, if I’m going to (re)learn a new operating system, it might as well be a free one.  Since then, I’ve only become more and more allergic to Windows.  Sadly, the most popular editing packages are all Windows-based, so I pretty much cut off my nose to spite my face before I even got started.  That being said, I never wanted to do editing, but was persuaded doing at least my first feature would be an invaluable learning experience.

The problem with bringing in someone else is they’d almost certainly be using some Windows-based software that’s totally incompatible with the software I’ve been using, which would most likely mean they’d start the edit all over again.  Not only can I not afford this financially (likely to be thousands of dollars for someone experienced), it’s likely to take several more months, and I want to put this puppy to bed.

Well, maybe I’ll get lucky and one of the ideas I came up with since I had to walk away last night will work.  I’ve managed to overcome all the editing obstacles up to this point…

I believe I’ve got all the editing issues worked out wrt getting Treasure Hunt all wrapped up.  I have the sound synced properly and found a way to preserve the DP’s original image quality.  I’ve got some music back from the composer and am working on placing it where I think it will add to the experience.  I really hope to have this completely wrapped up in the next couple of weeks.

My next task is to revamp the trailer and render it also in high quality.  Not dramatic changes, but there will be a number.  The bigger thing is to get the high quality render so when it gets downloaded from Youtube it doesn’t look like crap.

Then I get to start on figuring out how to get everything ready for the crowdfunding donors.  For a while I thought I’d see if I could figure out how to burn the DVD and Blurays myself, but I think there are too many edge cases and it will be better to hire professionals.  I need to get a place to host the electronic version as well. Vimeo seems like the best place as I can control who has access to it while Youtube doesn’t give that ability.

Then starts the search for which festivals to submit to and when.  Based on my research, there’s very little value in sending in a flick just before the deadline – most of the decisions are already made.  Instead, it seems the best approach is to submit within the first week, ideally as soon as submissions open, which may mean waiting a year for some to roll around again.

I started writing the first draft of the script February 25th of last year (that’d be 2018, for you looking at this in the far distant future).  In less than a month I’ll have spent a year on this project.  Hopefully I’ll look back on it in a few years with a warm glow of pride, but right now it’s just one big blob of color and I have trouble rewatching things for what seems like the 100th time.

Some good news: I got the second cut done for Treasure Hunt.  But (you knew there had to be one) the was a nightmare just around the corner.  After struggling for weeks on that damn singing scene, my mentor (and sound guy) John gave me some very good suggestions that allowed me to break out of the analysis paralysis.  I’m sure it could be made better with someone who has the patience to pick apart the singing and reassemble it so they’re all at the same time, but I think the scene works for what it needs to do.

The ‘but’ was when it came to render the gazillion clips of video and audio into a single coherent whole.  My wonderful software wasn’t keeping the audio in sync and by the end of the movie (it’s running 1 hour, 3 minutes and 45 seconds, with the opening and closing credits) it was several seconds off.

After much agony, cursing, walking away only to be drawn back in, I tried something I found via a Google and rendered the audio separate from the video.  I was told I’d probably have to stretch the audio somehow to make it work, but lo and behold, the rendered audio was the exact same length as the rendered video.  I had to have an extra step of re-rendering them together, but, at least insofar as my time is concerned, it’s minor (still takes about a half hour though).

The next PITA was rendering the video so it doesn’t look like crap.  While not huge, there are probably dozens of options to render the video and my research into them made not the tiniest bit smarter.  I eventually came up with a two-step rendering process, where I initially did it ‘lossless’ (a total lie, btw, as I can still see lots of compression artifacts that aren’t there in the original clips) and wound up with a 35 gigabyte (video only) file.  Then I combine that with the audio and tried a number of various options and found one that seems to work OK.  I’m not very happy with OK, I’ve seen what really seems much better on DVDs, which are a lot more compressed than what I’m currently working with, but, hopefully, will look OK to people not cursed with the eye that’s drawn to these artifacts.

I’ve asked for some feedback and because today was a snow day got some right away.  For most of the conversations I just left in the off-screen character’s dialog that was captured by the boom mic from the on-screen character.  I sort of noticed that when I was working on it, but it didn’t rise to the level of consciousness until John pointed it out.  Now I can’t not hear it any longer, so there will be at least one more edit.  My intent, at this time, is to only make those dialog audio changes, though he did point out a few other spots I might tweak. However, I expect the length to stay exactly the same.  I’ve asked the composer if he could give me another run through, hopefully this time coming up with something less… creepy and more in fitting with a romcom.

I believe I see the light at the end of the tunnel.  I’ll probably let this sit for a week or so, to give my mind some distance, then tackle the dialog audio for what I hope is the last effort.  Could it be made better?  I’m sure with a professional editor it could, but not, I believe, better enough to justify the expense.  I feel, deep down, that I have something that’s watchable.  It’s a simple story, but I think the actors, DP and sound guy did a really good job so I had a relatively easy job editing.  If it wasn’t relatively easy, I doubt I could have done it.  I know, for absolute certain, I don’t want to be an editor.  I think writing/directing/producing is enough involvement (though I’ve learned I need a producing partner, as during actual production I can’t be an effective director if I’m also producing), I need to hand the project over to an editor for the project to be it’s best.

I spent a very good chunk of our vacation working on refining the edit for Treasure Hunt.  Replacing almost all the camera audio, tightening up some scenes and even rejiggering a couple.  But I’ve retained the perennial problem with the signing scene (though got some excellent suggestions from a friend yesterday) and am quite vexed at the unnecessary complexity surrounding credits.

Every movie ever made requires end credits, at an absolute minimum, and the vast majority will also require begin credits.  Why this isn’t a simple matter of adding in the text and selecting from a handful of default themes is beyond me, but it appears, after a number of hours of research, that that’s exactly the case.  And seemingly not just in the editor I chose, but this is common across editors.  Sure seems like an unmet need to me!

Anyway, as usual I didn’t get as much done as I wanted to, but I do think another week or so and I should have the second cut wrapped up.  Even after I’ve dealt with the singing scene and wrestled with the credits, I still have to watch the thing at least several times.  And it’s an hour long.  And find and fix any issues with the editor (I learned, for instance, when sticking in the opening credit place holder, that the user interface element that moves all the other clips up and down the timeline isn’t 100% perfect, necessitating manually repositioning of dozens of clips).

But I do feel it’s coming along.  With the new suggestions for editing the singing scene I believe I can finally get past this mental block and put it in shape for another review.  I expect to get back at least a handful of comments on the new version, so will have to carefully evaluate them and decide what to do, but my expectation is I’ll just be doing some final polishing.  Thus, my plan is to have the final cut wrapped up by the end of January.  There still remains the potential for color correction, though I feel I’ve addressed a number of those issues once I learned to adjust the brightness of each clip, as well as music.  Those issues may take another few weeks (or more), but I feel good that I’ll have the show wrapped up in the Feb/Mar timeframe.  Of course, there still exists the need to produce the extras and the behind-the-scenes, but I’m sure I’ll obsess less about the production values there.  Besides, I’ve come quite far as an editor in the last few months and don’t expect to take as long to get the same amount of work done.

Thank you all for your patience!  I do believe the light ahead isn’t an oncoming train, but the end of the tunnel 😉

I’ve posted the first (well, second, only a select few had to endure watching the first) attempt at a trailer for Treasure Hunt.  Here it be:

Any thoughts are welcome.  I’m going to turn my focus to reworking the rough cut of the flick for the rest of my time off (and beyond, if I don’t finish it by then).  I may revisit the trailer if I get any consensus on the feedback.

Over the weekend I became totally overwhelmed attempting to restart the momentum on editing my flick.  It seemed everywhere I looked I found more and more sound files.  Almost all with the same meaningless name (e.g., MONO-<some number>.wav).  I was sure I had duplicates, as there were several times I was too rushed to ensure I only copied the new files off the storage media and just copied everything, but I didn’t think they were of this magnitude.  That really bit me on the ass!

Being a programmer professionally, I eventually decided to write a series of scripts to help me find and locate the replicas.  That took a number of hours, but eventually I was able to find the original collection of 530 audio files and match them up with the written records I got from the sound guy.  I now feel 99% confident I have all the sound files.

Of course, the ones I have all have the meaningless names, so my next effort is to find out which ones the sound guy has already gone through to rename with the appropriate information.  I know he hasn’t got them all, because the second scene isn’t in his renamed collection.  One of the primary reasons I stalled so quickly when I finally started editing.

So the weekend wasn’t a 100% failure on my part, just, say, 99.9% failure.  Hopefully this week will be different and I can get momentum to move into next week, when we’re on vacation, and still achieve my goal of having the next version done by the end of the year.

I’ve been communicating with an experienced editor and he’s agreed to give it a once-over once I’ve recut it.  He’ll then let me know if he can help out in such a way to fit within my budget (which, frankly, is nigh on zero at this point).  It’ll be a little tricky getting him all the data files and I might go with sneakernet rather than attempt to upload it somehow.  Hopefully we can come to an accommodation and he can at least do the color correcting, to ensure the night scenes all look like night.