[Joe Marler]

In 1984, EditDroid was one of the first NLEs (PDF brochure):

typewritten.org/Articles/DroidWorks/ed-485.pdf

[Joe Marler]

Those are some long lists of things to improve. There are also 100+ items “needing improvement” on https://fcpx.tv/top.html.

In all of those, I don’t see any mention of fixing multicam so it works with stabilization, optical flow and audition. Also optical flow and tracking plugins like SliceX don’t work in multicam or sync’d clips.

This makes me wonder how many FCPX users actually use multicam? Or do they just never stabilize or use optical flow or audition or SliceX in multicam?

Premiere Pro CC has no problem stabilizing, doing optical flow smoothing, or motion tracking on multicam clips.

[Joe Marler]

[Shawn Liebling] “1080p at a fairly high bitrate (Panasonic GH3) and then edit/output in FCPX to 720p. Using FCPX, I edit in a 720p timeline and simply drop my 1080p footage into it. I prefer this method because it allows me to do pans/zooms on static shots, etc. I also assume that when applying image stabilization it might degrade the quality less (since the footage being cropped is also being downsized).

I have noticed that the final output video files tend to be somewhat blurry when compared to the original footage, no matter how I render the final file “

I have edited lots of H264 1080p material from various cameras including GH4 and exported 720p H264 without seeing a major decrease in sharpness or apparent resolution. I have done frame grabs of the output at 1080p vs 720p and there is not a huge visual difference. This is using 1080p content in 10.1.2 or later and edited in a 1080p timeline, exported at 720p vs exported at 1080p.

However I rarely edit in a 720p timeline. The procedure you describe should work without major problems, and I have done similar procedures between 4k and 1080p. I can’t explain your results.

I just finished a similar test comparing H264 4k material from an AG-DVX200 and A7RII in a 4k timeline, exported at 4k vs the same 4k content in a 1080p timeline exported at 1080p, where in both cases 2x cropping/zooming was done. Taking frame grabs of the final file played in Quicktime, I couldn’t see any visual difference at 2x zoom, which indicated FCPX was using the full-res 4k content when zooming into the 1080p frame. The purpose of this test was to see if any image quality loss happened when using a 4k timeline vs 1080p timeline, when original content is 4k and exporting in 1080p. Editing performance is faster if using 4k in a 1080p timeline vs a 4k timeline, thus the incentive for that method (which I’m sure you know).

In all cases I exported using Share>Master File, Format: Computer, Video codec: H.264 Faster Encode, Resolution either 4k, 1080p or 720p.

[Joe Marler]

[Marco Bortignon] “I will have to work occasionally with 4K videos and of course it will have to be via proxy..”

There is another option besides proxy. If your final delivery is 1080p, you can import 4k into a 1080p project. Editing and applying effects will be a lot faster, render and temp files will be a lot smaller, and FCPX is smart enough to use the underlying 4k resolution if you zoom into the frame. I turn off background rendering when I do this, and I can’t see any visual difference between a 1080p export using 4k material in a 4k project vs a 1080p project.

Obviously if final delivery is 4k then you’ll need to use a 4k project, and using proxy mode is a good way for a laptop to deal with this.

[Joe Marler]

[Tim Maloney] “I can scrub/play the clip in the viewer and cut using the keyboard or cursor ( which has changed to a hand when I hover over the cameras )

HOWEVER the cuts appear to cut but they don’t stick. When I play back – the camera cuts haven’t stuck.

With the same multicam clip double clicked into the timeline. When I hover over the angles the cursor doesn’t change to a hand and the keyboard doesn’t let the angles switch.”

When you scrub through a MC clip in the Event Browser, the cursor is a hand on the Angle Viewer. You can select a monitoring angle but not make an edit.

When you add that MC clip to a project in the timeline, the cursor in the Angle Viewer is a blade, indicating editing is active. Clicking on a camera will do an edit.

You can re-designate the monitoring angle in the Angle Viewer by doing OPT-click on the camera.

Due to the skimmer, it can be confusing which MC content is showing in the Angle Viewer — the MC timeline or the MC clip from the Event Browser. They can switch back and forth rapidly as you move your cursor.

It is actually easy to see where the Angle Viewer content is coming from: the upper left corner will either show a MC clip icon from the Event Browser, or a “Project Slate” icon that matches the timeline upper-left corner.

[Joe Marler]

[David Mathis] “…aggravates the (censored word of choice) out of me. Why does a cross dissolve affect both audio and video? Yes, there is a way to make a cross dissolve for audio only. Why can’t there be one for video only? It irritates me big time to split the audio from the video to do so. Then you have to trim the two separately to get them back together. Why must this be? I am ready to jump ship to Resolve, right now”

For a video-only cross dissolve, you just do CTRL-S to expand the audio and video, apply the video-only transition, then CTRL-S again to collapse them. You can do multiple clips at once — just select them, CTRL-S to expand, CTRL-T to apply video-only default cross-dissolve to all clips, then CTRL-S to collapse them. You can add a video-only cross dissolve to 20 clips at once, and it only takes a single keystroke to expand the clips, then a single keystroke to collapse them.

[Joe Marler]

The dwindling of Moore’s Law has been known for some time in the CPU design community. Even before this happened, so-called “Dennard Scaling” or frequency scaling plateaued around 2007 and there is no near-term solution for this:
https://en.wikipedia.org/wiki/Dennard_scaling

Moore’s Law itself is limited by the heat problem, which was discussed in a pivotal 2012 paper “Dark Silicon and the End of Multicore Scaling”: https://www.cc.gatech.edu/~hadi/doc/paper/2012-toppicks-dark_silicon.pdf

Essentially the heat problem is starting to affect how much of the chip can stay powered on. This can’t be fixed by a better chip cooler. Even though integration allows putting more systems on the chip, increasingly large portions must stay powered off (dark silicon).

Instructions Per Clock (IPC) improvements which benefit single-thread performance are also dwindling, since designers have already exploited all the straightforward methods. There are a few remaining tricks such as data speculation. Data speculation differs from control speculation, which is currently used to predict a branch. In theory data speculation could provide an additional 2x performance on single-threaded code, but it would require significant added complexity. See “Limits of Instruction Level Parallelism with Data Speculation”: https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.47.9196&rep=rep1&type=pdf

VLIW (Very Long Instruction Word) methods could give another 2x or so but would require new software and compilers. Intel’s unsuccessful Itanium was at attempt at this, but some researchers are still investigating the technique: https://millcomputing.com/

Going forward, performance improvements will come from heterogeneous processing features such as Quick Sync, AVX vector instructions, etc, which software must specifically access. This can be highly advantageous where it’s applicable. E.g, FCPX can export to H264 about 4x or 5x faster using Quick Sync than without it.

The GPU side is not quite as limited and there are significant future gains available from an architecture and fabrication standpoint. However that only helps if the highly parallel programming model can be leveraged. E.g, long-GOP encode/decode cannot be significantly accelerated via traditional GPU techniques. OTOH both nVidia and AMD have incorporated video encode/decode support using the NVENC and VCE APIs, but computers must have the specific card type and app developers write to the manufacturer-specific API.

Intel’s latest Xeon E7-8890 v3 has 18 cores and can do nearly 3 teraflops. However the heat problem limits the base frequency to 2.5Ghz and turbo to 3.3Ghz. In theory this could provide a single-socket 18-core Mac Pro, but the current price for this chip is $7,700 so Apple would need a big discount.

[Joe Marler]

BTW PluralEyes can do sync drift correction: https://www.redgiant.com/products/pluraleyes/

[Joe Marler]

[Dan Turner] “I’m a film lecturer currently researching what editing software is the most commonly used in the industry.

Id be grateful for any input on what you guys use, and what you’ve found in other post-houses or industry colleagues.

I appreciate its a market thats still influx, but just getting a sense of the position would be enormously helpful.”

In a narrow sense “the industry” means Hollywood. From that standpoint, Avid has at least 90% penetration, although it’s hard to give an exact figure. Despite that dominate position, Avid’s total annual revenue from software sales is only about $90 million (Avid was temporarily de-listed from NASDAQ in 2014), so the market for video editing software is a steep pyramid.

The Indian “Bollywood” film industry alone does over $2 billion annually, and I expect they probably use a higher % of Premiere and FCP than does Hollywood. What % I don’t know.

Due to the democratization of digital video production, a huge amount of corporate, documentary and non-profit content is now produced and distributed via internet. There are now video bloggers who make $10 million per year, and several make $40,000 per month. Most of them do not use Avid, and many are self-produced using Premiere or FCPX.

To my knowledge there is no good data about this. Researching what editing software is used in a given segment would require doing surveys.

[Joe Marler]

[Jenn Lindsay] “move the very unwieldy enormous library/project I referred to at the beginning of this post onto an external drive as well, but I’m scared that if I do that all the media will get disconnected and chaos will ensue.”

In Finder if you double-click on the library file it will open in FCPX, no matter where that library is located. If you move that library file someplace else on that hard drive or a different hard drive, you can still double-click and open it. FCPX will still know where your external media files are.

In Finder if you move your external media files to a different hard drive, they will initially show up in FCPX as missing. However as of 10.1.2, FCPX can relink all media in a project no matter what location on the drive volume.

If moving media between two hard drives, in theory you could move every media file and folder to a different relative location on the destination volume than it was on the source volume. You could even use different folder names on the destination volume, restructure the folder tree, etc. FCPX will search every folder on the new drive volume and relink all the media files.

This MacBreak Studio episode covers it starting around 04:10.
The specific relink procedure starts at about 08:00.

https://www.youtube.com/watch?v=MwKVVZixgk0

Some contents or functionalities here are not available due to your cookie preferences!

This happens because the functionality/content marked as “Google Youtube” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.