For my test I was working with a 4:05 movie edited in Final Cut Express and sourced in AVCHD on my Canon HF100. I started by exporting it to a 2.4 Gig QuickTime file for archive purposes. The main attributes of my first test, which turned out to be a pretty good recipe for YouTube, particularly when in a time crunch, were:
(Full specs on this later in the post, along with all the tests and a few notes on the results.)
A few top-line conclusions:
Deinterlace: Using the deinterlace filter in Compressor chewed up my text; using deinterlace under Frame controls worked like a charm.
Resolution: Dropping resolution by 50% hurt the image and softened text but didn't really speed things up or shrink the file much, at least going from 1280 x 720 down to 640 x 360. Of course, there will be situations where you have to drop the resolution, but both YouTube and Vimeo suggest leaving HD resolutions at 1280 for HD, so in this case there was no reason to downsample.
Bitrate: Changing the bitrate had a direct impact on file size (2500 = half the file size of 5000) and a noticeable impact on the image, though going down to 2500, the image still looked pretty good for web video. YouTube currently suggests not capping bitrate, Vimeo requests that you set a max of 5000, which is what I did for example #6, below.
Multi-pass: Multi-pass adds a lot to encoding time (4X in this case) but did give me a higher contrast image with richer colors. So if time allows, it looks like multipass is better, but in a time crunch, single pass can work.
]]> Default settings:File Extension: mov
Estimated size: 4.29 GB/hour of source
Audio Encoder
AAC, Stereo (L R), 44.100 kHz
Video Encoder
Format: QT
Width and Height: Up to 1280 x 720
Selected: 1280 x 720
Pixel aspect ratio: Square
Crop: None
Padding: Preserve source aspect ratio
(L: 0, T: 0, R: 0, B: 0)
Frame rate: (100% of source)
Frame Controls On:
Retiming: (Fast) Nearest Frame
Resize Filter: Linear Filter
Deinterlace Filter: Better (Motion Adaptive)
Adaptive Details: Off
Antialias: 0
Detail Level: 0
Field Output: Progressive
Codec Type: H.264
Multi-pass: Off, frame reorder: Off
Pixel depth: 24
Spatial quality: 75
Min. Spatial quality: 50
Temporal quality: 50
Min. temporal quality: 50
Average data rate: 10.24 (Mbps)
Test 1 = default (see specs above)
294 MB, 15 minutes to encode
notes: looks good and pretty smooth, 294 MB
Test 2 = like #1 but I switched deinterlace from Better to Best
notes: took much much longer; after 1 hour it was ~15% and I cancelled the encoding. Looked like it would take 6-7 hours for the total encode, or 24-28X the time. Yikes.
Test 3 = like #1 (so deinterlace was set back to "better"), but I toggled the deinterlace filter checkbox on
294 MB,19 minutes
notes: Same file size, but much much worse results - this is the one that bitmapped the text. I'm going to avoid this in the future.
Test 4 = like #1 but with multi-pass encode
314 MB, 52 minutes
notes: Slightly richer than #1 and slightly better contrast, though longer encoding time.
Test 5 = like #1 but with no bitrate limit
372 MB, 76 minutes
notes: Slightly richer than #1 and slightly better contrast, though longer encode; no big surprise -- this looks the best....
Test 6 = like #1 but 5000 bitrate
159 MB, 59 minutes
notes: Slightly less contrast/more faded than #1 but still pretty good.
Test 7 = like #1 but 2500 bitrate
66 minutes, 79 MB
note: main change is more chunkiness, mainly visible when the file is large. Still not bad though.
Test 8 = like #1 but 50% resolution
52 minutes, 285 MBs
note: Softness definitely noticeable, especially around the text. File size almost the same! So not much benefit to this....
Test 9 = like #1 but 29.97 resolution
60 minutes, 294 MB
note: No noticeable difference to my eye.
This one's called "I Cook It 2 It" and it's, well, it's sort of a hippity hop version of a recent blog post about me cooking.
Wearing a chef's hat when I do it,
-Cecil
time: 1:10 seconds; specs: 1.6M
Press Play to play.
In related news, Ira and I have also joined forces with the infamous Xian and the soon-to-be-infamous Ryan to forge a fearsome four-piece called "The Reuben Kincaid." Watch this space for TRK tracks TK.
Waiting for your family's big day to arrive,
-Cecil
time: 2:15 seconds; specs: 3.1M
Press Play to play.
So, for example, if I made toast, to which I wouldn't normally add Tabasco, that wouldn't count as cooking. (Or mebbe, if I added Tabasco to toast, that might count as cooking, but it'd count as bad cooking. I think. I need to try this.)
Similarly, if I made a can of soup, like some Wolfgang Puck's Thick and Hearty Lentil, and I didn't add Tabasco (which would be crazy) that also would not be cooking.
But, in contrast, if I made a can of soup, and I chose to add a few drops of Tabasco, in my heart it would feel like I had truly cooked that can of soup. Like the culinary choices (ok, choice) that I'd made had transformed that can of soup into a lentil-laden very personal and real expression of what I consider good eats.
Tonight, the lentil is hearty, yes. Thick too. I credit Wolfgang for this. But it's also slightly spicy, and I think I've made my point.
]]>Yap yap yap!
The whole enterprise was born from one simple question -- what if dogs, night-by-night, trick-by-trick, could learn to do everything humans do? Or, as many as could be covered in a dog's lifetime of weeknight TV shows?
It almost goes without saying, the first tricks involved filming and producing a TV show. Tricks Lucious had taught himself. After that came shows on how to make toast without burning your nose. How to open a pickle jar. How to make a pancake. Not all the episodes were about food, but the food shows got the best ratings.
About five years in, the focus switched to personal grooming. How to part your hair (extra tricky for stiff haired hounds). How to clip your own nails!
Now here they were on day 7,322. Daphne sitting in my rocking chair, nodding her doggy head, listening close, learning how to fix a carburetor. And that was when I walked in.
]]>
That way, when a customer walks in, they can say "I'll have the number one!" People love to say that phrase. Just look at how well McDonald's has done!
Now sit back and watch the money roll in.
]]>which can already grow up to 7 feet long
would need to be 11.76 times
larger than they actually are
and weigh 1,460 pounds.
1,460 pounds of Tundra Wolf.
Do you see what I'm saying?
]]>Prepare to get stirred!
(Written and filmed by The Whittlers -- East Bay cohorts in mini-movie mania.)
]]>"looks to me like you're not a fan of berries."
If he sees what I'm typing I'm a dead man.]]>
or to be
brought somewhere
for some larger porpoise?
Or,
what if the larger purpose
was to serve a smaller porpoise?
What then?
]]>