Vegas to Youtube, Vimeo, Web -- A New Look

Comments

musicvid10 wrote on 1/21/2011, 7:56 AM
Done. About 427MB zipped.
NickHope wrote on 1/21/2011, 8:53 AM
Mark, where's the thread where you were asking for footage for this? I want to see how the project got started and also want to be sure of permission to use the footage before I upload anything.
musicvid10 wrote on 1/21/2011, 9:29 AM
Nick, here is the first thread.

http://www.sonycreativesoftware.com/forums/ShowMessage.asp?ForumID=4&MessageID=742876

All of the footage used in the project came from Jerry (amendegw), Stringer, and Kimberly, and please feel free to check with them if you have questions.

Note that my tutorial is still in bits and pieces, and I will hope to have it complete by early spring. As always, your input is valuable and most welcome.
NickHope wrote on 1/21/2011, 9:47 AM
Thanks, I've sent them each an email to check. I'll try and get on it in the next few days, once Jerry's uploaded the bits.
musicvid10 wrote on 1/21/2011, 9:53 AM
And, if you want to add a HD clip of your own, with own credit of course, feel free to add to the project.
Just set video levels to strict 16-235, including text, overlays, and fades.
amendegw wrote on 1/21/2011, 11:47 AM
Okay, I've included a link on the project webpage to download musicvid's Vegas Project with source clips. Just to repeat the URL, it's here: http://www.jazzythedog.com/testing/dnxhd/getvideo.aspx

Enjoy!
...Jerry

System Model:     Alienware M18 R1
System:           Windows 11 Pro
Processor:        13th Gen Intel(R) Core(TM) i9-13980HX, 2200 Mhz, 24 Core(s), 32 Logical Processor(s)

Installed Memory: 64.0 GB
Display Adapter:  NVIDIA GeForce RTX 4090 Laptop GPU (16GB), Nvidia Studio Driver 566.14 Nov 2024
Overclock Off

Display:          1920x1200 240 hertz
Storage (8TB Total):
    OS Drive:       NVMe KIOXIA 4096GB
        Data Drive:     NVMe Samsung SSD 990 PRO 4TB
        Data Drive:     Glyph Blackbox Pro 14TB

Vegas Pro 22 Build 194

Cameras:
Canon R5 Mark II
Canon R3
Sony A9

bdg wrote on 1/21/2011, 4:05 PM
A little OT:
How do I stop Handbrake from displaying the filename at the start?
musicvid10 wrote on 1/21/2011, 4:12 PM
Handbrake doesn't do that afaik.
Are you using VLC to play the video?
VLC displays the filename at the start.
amendegw wrote on 1/21/2011, 4:14 PM
"How do I stop Handbrake from displaying the filename at the start?"Are you sure HandBrake is doing this? Are you playing in VLC as showing the filename is default there.

...Jerry

Edit: Ha! An over-posting. musicvid, you can type 2 minutes faster than I.

System Model:     Alienware M18 R1
System:           Windows 11 Pro
Processor:        13th Gen Intel(R) Core(TM) i9-13980HX, 2200 Mhz, 24 Core(s), 32 Logical Processor(s)

Installed Memory: 64.0 GB
Display Adapter:  NVIDIA GeForce RTX 4090 Laptop GPU (16GB), Nvidia Studio Driver 566.14 Nov 2024
Overclock Off

Display:          1920x1200 240 hertz
Storage (8TB Total):
    OS Drive:       NVMe KIOXIA 4096GB
        Data Drive:     NVMe Samsung SSD 990 PRO 4TB
        Data Drive:     Glyph Blackbox Pro 14TB

Vegas Pro 22 Build 194

Cameras:
Canon R5 Mark II
Canon R3
Sony A9

bdg wrote on 1/21/2011, 5:30 PM
Brilliant!!
Thanks a bunch guys.
I even found the preset in VLC to turn it off. Yay!
NickHope wrote on 1/23/2011, 9:34 PM
Hoping to get to my MeGUI test shortly. I have some stuff to clear first.

Just thinking aloud... Is it worth someone trying a test using Mike Crash's Smart Deinterlace Vegas filter and the x264vfw codec? Presumably YouTube and Vimeo can read the AVI file it creates. Then everything is done within Vegas. Also it looks like there are 64-bit versions of both available for those using Vegas 64-bit.
musicvid10 wrote on 1/23/2011, 9:44 PM
What you suggest would make an interesting test, but the x264vfw codec is not nearly as efficient as its mp4 counterpart (no b-frame support in AVI among other things).

I tested the Crash deinterlace against Handbrake with no resizing on 1080i source, and found Handbrake decomb to be sharper.

But now you've got my curiosity piqued so I'll give it a go this week sometime.
NickHope wrote on 1/23/2011, 11:41 PM
Actually I might have made a sweeping statement that YouTube and Vimeo can read such a file. Wouldn't they specifically need the x264vfw codec installed on their server? Might be worth doing a tiny test first to check.
craftech wrote on 1/25/2011, 12:07 PM
Nice work Laurence and Musicvid.

What is your latest technique for creating SD DVDs from HD camera footage?

Always looking for new suggestions to improve that.

John
musicvid10 wrote on 1/25/2011, 12:23 PM
I should live so long as to solve that one . . .
NickHope wrote on 1/25/2011, 1:15 PM
>> What is your latest technique for creating SD DVDs from HD camera footage? <<

I know you weren't asking me, John, but anyway...

Well, a key aspect of all this H.264 jiggery pokery is the deinterlacing/decombing, which isn't so relevant for DVD.

Personally for SD MPEG-2 I swear by Cinema Craft Encoder Basic for speed and quality and I haven't used anything else for years. I always thought Vegas' MC encoder was poor. Last time I was churning out DVDs (NTSC from 1080-60i) I would frameserve to CCE from the timeline with the following project settings, compress at 8Mbps CBR, and the resulting DVDs looked great:



However I've since bookmarked Laurence's thread on this subject and will experiment more next time, in particular with the "zero sharpen". John Meyer's workflow is also useful where there's cropping to be done.

Sorry to derail the thread and I will get around to my MeGUI test, I promise.
Laurence wrote on 1/25/2011, 1:54 PM
My experiments with 0 sharpening tell me that it is a worthwhile addition to the process. What I'm still not completely sure about is whether the zero sharpen filter should be pre or post downrez. I thought it looked better pre, but others here recommend post. Either way is an improvement over downrezzing without it.

I can't see any advantage of adding the sharpen to each video event over just adding it to the video tracks. Theoretically adding it to events should keep the sharpening from being added to crossfades which would exaggerate artifacts, but in practice I don't see this happening.

Adding the sharpening filter to text or animated photos doesn't look good so I wouldn't add it to the master video bus.

My observations were originally based upon recommendations from people in this forum, but I always do experiments so that I can see for myself which approach I like best. I really recommend experimentation. Until you see it for yourself, you don't really know for sure what you like. My experimentation tells me that these things make significant differences in the final quality one is able to achieve.
amendegw wrote on 1/25/2011, 2:53 PM
Okay, this "shows to go you" that I have too much time on my hands. I decided to create a webpage that displayed musicvid's test clip in various MediaPlayers: JW Player, FlowPlayer, SilverLight (open MediaPlayer), embedded YouTube & embedded Vimeo. The results are here:

http://www.jazzythedog.com/testing/dnxhd/mediaplayers.aspx

I'm not sure that I can come to any exciting conclusions from this exercise, except that the local players keep up with the download better however the quality is not as great - makes sense because the local video clips are 1/8 the bitrate of the YouTube/Vimeo. Edit: That statement is misleading as the clip was 1/8 the bitrate prior to uploading to YouTube/Vimeo. These hosting services then process these files and reduce the bitrate to their standands. See the discussion a few posts down.

Enjoy!
...Jerry

PS: I'm ignoring the discussion on HD -> SD as we might just re-start the following thread: http://www.sonycreativesoftware.com/forums/ShowMessage.asp?ForumID=4&MessageID=738210 {grin!}

System Model:     Alienware M18 R1
System:           Windows 11 Pro
Processor:        13th Gen Intel(R) Core(TM) i9-13980HX, 2200 Mhz, 24 Core(s), 32 Logical Processor(s)

Installed Memory: 64.0 GB
Display Adapter:  NVIDIA GeForce RTX 4090 Laptop GPU (16GB), Nvidia Studio Driver 566.14 Nov 2024
Overclock Off

Display:          1920x1200 240 hertz
Storage (8TB Total):
    OS Drive:       NVMe KIOXIA 4096GB
        Data Drive:     NVMe Samsung SSD 990 PRO 4TB
        Data Drive:     Glyph Blackbox Pro 14TB

Vegas Pro 22 Build 194

Cameras:
Canon R5 Mark II
Canon R3
Sony A9

musicvid10 wrote on 1/25/2011, 4:42 PM
So what makes Vimeo HD look so good?
Its processed video is 2Mbs VBR (7Mbs max), High@L3.1, and 4 ref frames (a clue?), 720p. I can't dig out any other information about its p's and b's.

Why no blocking in the transitions at that bitrate? Can this kind of clarity be duplicated in x264 at 2Mbs? If Vimeo can give us quality at this bitrate, from optimized 8Mbs uploads, why can't Youtube? I propose a local shootout against Vimeo.
http://vimeo.com/18690771
amendegw wrote on 1/25/2011, 4:54 PM
And the bonus is - Vimeo buffers much better than YouTube. Much less waiting for your player to catch up with the download. What is their secret?

...Jerry

Edit: I'll try a render at 2Mbps and post it on this Webpage. We'll see how it compares to Vimeo.

System Model:     Alienware M18 R1
System:           Windows 11 Pro
Processor:        13th Gen Intel(R) Core(TM) i9-13980HX, 2200 Mhz, 24 Core(s), 32 Logical Processor(s)

Installed Memory: 64.0 GB
Display Adapter:  NVIDIA GeForce RTX 4090 Laptop GPU (16GB), Nvidia Studio Driver 566.14 Nov 2024
Overclock Off

Display:          1920x1200 240 hertz
Storage (8TB Total):
    OS Drive:       NVMe KIOXIA 4096GB
        Data Drive:     NVMe Samsung SSD 990 PRO 4TB
        Data Drive:     Glyph Blackbox Pro 14TB

Vegas Pro 22 Build 194

Cameras:
Canon R5 Mark II
Canon R3
Sony A9

apit34356 wrote on 1/25/2011, 5:19 PM
The answer------- volume(number of users per second downloading). YouTube has massive server farms but the bandwidth becomes a problem as it routed through other network nodes. There are many processes, but in simple terms, large networks do a load balancing across it servers. There are network providers like Comcast that limit overall volume from sites like Youtube based on regionals,etc....
amendegw wrote on 1/25/2011, 5:38 PM
Okay, I made a 2Mbps render and added it as another JW Player selection. To my old eyes it looks as good as Vimeo.

Try it here: http://www.jazzythedog.com/testing/dnxhd/mediaplayers.aspx

...Jerry

System Model:     Alienware M18 R1
System:           Windows 11 Pro
Processor:        13th Gen Intel(R) Core(TM) i9-13980HX, 2200 Mhz, 24 Core(s), 32 Logical Processor(s)

Installed Memory: 64.0 GB
Display Adapter:  NVIDIA GeForce RTX 4090 Laptop GPU (16GB), Nvidia Studio Driver 566.14 Nov 2024
Overclock Off

Display:          1920x1200 240 hertz
Storage (8TB Total):
    OS Drive:       NVMe KIOXIA 4096GB
        Data Drive:     NVMe Samsung SSD 990 PRO 4TB
        Data Drive:     Glyph Blackbox Pro 14TB

Vegas Pro 22 Build 194

Cameras:
Canon R5 Mark II
Canon R3
Sony A9

NickHope wrote on 1/25/2011, 8:33 PM
>> makes sense because the local video clips are 1/8 the bitrate of the YouTube/Vimeo. <<
Jerry, the YouTube video isn't really 8Mbps is it? Are you confusing the bitrate of the video you upload with the bitrate of the video that YouTube/Vimeo re-render it to?

Last time I looked, a YouTube 720p download was 2Mbps and a Facebook HD download was 2.5Mbps. I didn't check a Vimeo one. You can dig a YouTube video out of your cache. I find this easier on IE than in Firefox. Delete temporary internet files first, then watch the Youtube file, then go into temporary internet files and sort by file size. The YouTube video should stick out at the top of the list and you can copy and rename it somewhere else.

Incidentally, for inspecting the characteristics of H.264 files you can use AVInaptic. The interface looks like it belongs on a 1994 Sun workstation but it's actually very useful. It gives you more info than Mediainfo etc..

>> And the bonus is - Vimeo buffers much better than YouTube. Much less waiting for your player to catch up with the download. What is their secret? <<

Besides what apit34356 said, I think this also depends where you are and how "close" on host's CDN (content delivery network) to you the file you are downloading is coming from. I'm in Thailand, and if I was a YouTube video with very low views, I often have to wait, I guess because the file hasn't yet been farmed out to the Singapore node. Videos with higher and regular views seem to play faster because the videos have reached all the nodes (p.s. this is uninformed guesswork, so take with a pinch of salt).
amendegw wrote on 1/26/2011, 2:36 AM
"Jerry, the YouTube video isn't really 8Mbps is it? Are you confusing the bitrate of the video you upload with the bitrate of the video that YouTube/Vimeo re-render it to?"Absolutely, correct. It was 8Mbps before upload and YouTube got its mitts on the the clip.

Furthermore, I'm sure apit34356 comments about server workload are also correct.

What was I thinking? I know better than that - this cold weather must be freezing my brain!

...Jerry

PS: The bottom line is that 2Mbps local MediaPlayer clips look and stream "pretty darn good" when compared to both YouTube & Vimeo. At least, here in the eastern USA. Obviously, YMMV and I'm very interested in what others see.

System Model:     Alienware M18 R1
System:           Windows 11 Pro
Processor:        13th Gen Intel(R) Core(TM) i9-13980HX, 2200 Mhz, 24 Core(s), 32 Logical Processor(s)

Installed Memory: 64.0 GB
Display Adapter:  NVIDIA GeForce RTX 4090 Laptop GPU (16GB), Nvidia Studio Driver 566.14 Nov 2024
Overclock Off

Display:          1920x1200 240 hertz
Storage (8TB Total):
    OS Drive:       NVMe KIOXIA 4096GB
        Data Drive:     NVMe Samsung SSD 990 PRO 4TB
        Data Drive:     Glyph Blackbox Pro 14TB

Vegas Pro 22 Build 194

Cameras:
Canon R5 Mark II
Canon R3
Sony A9