I'm having a very hard time capturing quality narration. I've tried many different configurations. I've run cables across the house to tether my noisy desktop to a quiet capture environment. I've tried a variety of microphone placement.
No matter what I do, I hit a quality ceiling and simply do not sound, under any circumstances, like Morgan Freeman.
The current edit still contains some crappy narration although I've upgraded quite a bit of it. (The worst is my ROEI delivery. Yikes.)
I've narrated "Th" videos before, but I never tried to make it sound good. This is me trying to make it sound good. Apparently that is hard.
What is new here?
The very worst (scripted) narration that existed in previous edit was my assertion at 1h 38m 25s, ( https://youtu.be/ZsPOJkVZPtc?t=1h38m19s
)...
"This will fundamentally change our national conversation about the environment and energy independence."
The new dialog ( https://youtu.be/WvO1p6w4MZ0?t=1h55m38s )...
"The promise of abundant clean energy has already been made by solar and wind advocates. However those are diffuse and intermittent sources of energy."
...simply is less stupid. Yay for less stupid.
And the 2nd-worst issue (to me) was the LFTR-Branding-Pivot at 1h 05m 05s, ( https://youtu.be/ZsPOJkVZPtc?t=1h4m55s )...
"...you could catalyze the burning of Thorium indefinitely. This is a big deal because we have lots and lots of Thorium. -PIVOT- I was driving home from work in April..." [LFTR Branding].
The new pivot ( https://youtu.be/WvO1p6w4MZ0?t=1h04m00s )...
"...you could catalyze the burning of Thorium indefinitely. This is a big deal because we have lots of Thorium, and it has unique properties..."
And we make a massively different pivot into about 18m of BREEDER REACTORS in general.
So with this edit, the video is now more technical. I'm not worried about that one way or the other... technical is fine so long as the video flows, and I think this helps maintain the flow.
We pivot smoothly into breeders, and from 18m of breeders into branding.
I'm much happier with this. I get to talk about Pandora's Promise. I have Robert Stone talking about Pandora's Promise. That we've accomplished some amazing things yet the public has zero awareness of it brings up the industry's past inability to communicate and thus a great reason to talk about branding.
WHAT IS LEFT?
The biggest issue is narration quality. I'm chipping away at that, and intend to release a "good enough for the public" p01 as soon as possible. I can keep improving it after that along with the rest of the chapters, but if I wait until all chapters are complete then I'll be sitting on p01 for an unknown number of months. I'd rather not do that.
How this narration challenge plays out for the rest of the chapters though, I'm not sure. While I've found better-and-worse approaches to capturing narration, none are easy (for me). I'm even finding my ZOOM H4n cuts out when I capture sometimes. Which is just the icing on the cake when I can barely narrate in the first place. No idea if it is Windows audio drivers, the long USB2 cable, or PPro itself. I do know my MacBook could support the H4n without any problem, so I know the mic hardware is fine. But that doesn't tell me where the problem is other than "not the mic". Troubleshooting the cable is easy except that I need to pretend-record lots of dialog in noisy environment to see if the intermittent problem pops up... boo intermittent problems.
Obviously it would be great to have a non-Gord narrator, as that's a perfect task to outsource in terms of some guy talking into a microphone all day in an actual studio. However, I need the freedom to script the narration as an involved participant... -We- did this... -I- think that. Plus, I do get to take a Zoom H4n into the basement and speak any line that was poorly written... there's no external loop when my script sucks.
Zero-G Reactor
One excellent point that was raised was whether or not the Aircraft Rector Experiment was intended to operate without gravity. Well, no, not literally. Gravity lowered ARE control rods. ARE itself was not going to be literally dropped into a plane because they didn't get that far... the "reactor" which flew in a plane was not operating, they were just testing size & weight.
But powering an airplane was the ultimate (early days) goal, and likely did drive many decisions.
Non-pool MSR reactor designs (which do not rely on convection) lend themselves to non-Earth gravity operation. ARE wasn't gravity filled, it was vacuum filled. Liquids were forced through the operating ARE pipes by pump.
I still think the argument is strong enough to keep it in this video, as it opens up the discussion to what is a pool reactor how great it is to be able to take advantage of convection here on Earth.
But am eager to hear any/all arguments against so I can keep considering.
RGB 0-255 vs RGB 16-235
I just discovered why all earlier renders of TR2016 (and most of my "Th" videos) had white lines and boxes littered through-out them.
This was of course a problem with PPro assets all being limited to the 16-235 range (broadcast range as captured by cameras), and PPro renders sometimes creating artifacts in the 235-255 range... like the crop line when I'm cutting off the edge of an image, for some reason PPro would make that brighter than the image itself.
My computer monitor was set to only display 16-235.
It has been like that for years. I didn't know that was a thing.
So going forward with this render, I'm both using a filter to cap 235-255 down to 235... it seems to remove the super-whites. And also... monitor now adjusted properly. Cough.
UPDATE 2016-03-13
I did set up a Yeti on a mic-stand, in my living-room environment (noise-free). This was me standing in front of HDTV with cables running upstairs. For whatever reason, the Yeti does NOT cut out the way H4n did... it is a reliable experience. That's a big deal, at least I'm not fighting the hardware/software I'm only fighting audio quality.
Unfortunately I have a crazy cold and could not utter a single sentence that I could use (I could barely speak). But it was clear that the capture was very different from the H4n.
While my voice isn't quite as broken as it was during the Yeti test, I'm still sounding wrong and am going to use my sick time to try quiet-ize my editing room. So whether I end up capturing in front of HTDV in living room, or in PC room, I'll have 2 options to pick from.
Reducing any echo in living room is hard because I need to look past the mic at the TV, so I can't sound-proof around it. The mic is basically between me and the TV (I try keep it high as I've heard that helps reduce S and P sounds). And the TV is a big flat service for sound to bounce off.
So all advice pertaining to mic-on-a-desk, yes, I'll be pursing that as well. I sort of need to be able to peer over the mic at my monitors so it can't be anything that stands too tall on the desk. I'll visit a pro-video store and see what exists, and/or make something from bed-foam. (Or order from monoprice.)
The big challenge and reason I never pursued sound-proofing the PC room (despite obvious work-flow advantages) is that the PC is crazy loud. There's 2 GPU and about 12 hard drives (1:1 local project backup) cooled by many-a tower fan.
I'm not sure if I'll be able to quiet it.
It is going to go in a closet. I'll try sound-proof the closet. I'll try sound-proof the whole room. I'll be closing the closet door when I'm capturing narration and otherwise opening the door back up so PC does not overheat.
If that works then it offers hope that chapters beyond "NASA" can be narrated in a time-efficient manner. Right now capturing narration is a thing that I try dedicate a full day to, when really it would be better to narrate at any time, to address any problem as it is found, or test out narration as it is written to see if it sounds stupid.
UPDATE 2016-03-17
I've now got a Yeti sitting on my desk, in a sound-absorbing 3-sided (+top) shield. There is a 2-mesh pop filter in front of the Yeti. Padding is on the walls. PC and RAID moved into closet. The closet door is opened normally, and closed when I want to capture audio.
Have bought some supposedly respectable over-the-ear Monoprice headphones, so that I'm not bleeding audio from headphones into the capture. The Yeti has a direct headphone/mic tap so I can actually hear my capture in realtime as I say it, without the slur-inducing 1/4 delay when the audio makes a round-trip from mic to PC to headphones while I'm talking. I can actually hear some of my mistakes as I'm making them, not only as I review my capture.
Am still recovering from the cold, so the audio I captured was p02 temp-track not p01 final-track. My voice is still audibly broken, although I can speak understandable words now.
The results are not perfect... I can't absorb enough noise to eliminate the closeted PC+RAID fans. But... the workflow is fantastic. Only needing to position my mic, close a closet door and drape some insulating bumpy foam in front of the closet is a big step up from me running wires into the living room and re-configuring my desktop layout... it makes the process of capturing audio a do-it-now process instead of a dedicated-afternoon process.
Also, the Yeti now serves as my permanent audio interface... by moving my keyboard, mouse and headphones downstairs, I needed to tell PPro to both change video outputs and audio outputs. It led to me occasionally unplugging my audio device while PPro was still dependent on it, which would in turn destabilize PPro. PPro does not need any help to be destabilized further thank you.
So I'm going to keep evolving this record-audio-in-editing-room solution... keep adding more sound absorption (inside and outside of closet), and also investigate use of Audible noise reduction to try improve my audio capture in post.
The Yeti really does sound great, I should have never presumed the ZOOM H4n was optimal for capturing narration.
Am currently focused on p02 while I wait for my voice to improve enough I can try improve p01. The p02 narrative was poor, I'm restructuring it.