I am somewhat confused about using Handbrake. What is the purpose of it outside of Vegas? Do you build your project in Handbrake? Do you build your project in Vegas and import to Handbrake? Does it really make the video look that much better?
Handbrake is an encoding application. It is not an authoring tool or editor.
Most people use it to encode DVD or Bluray to MP4 for compressed storage and playback on their home networks.
Many Vegas editors also use it to encode their finished videos to MP4 (x264) for website delivery, Youtube and Vimeo, and local PC playback.
Its advantages are good compression while retaining excellent quality.
If you are not doing any of these things, you don't need it.
Handbrake is not a Sony product, but a free open source project.
As far as its usage and configuration, their website is the place for you to start: http://handbrake.fr/
That is what I understand and have heard great things about it. My real question is related to the how...If I render a project in Vegas can I then move it to Handbrake to encode and if that is the working order, does it make the end project look better than the render from vegas? That is my real lack of understanding. I have seen proof that projects done with handbrake look better on the web. I just don't understand the workflow...
Handbrake doesn't accept Vegas projects. It accepts video files for re-encoding.
Handbrake x264 does not make the timeline video look better. It compresses it, while preserving excellent quality vs. compression when compared with the AVC encoders in Vegas at bitrates <10Mbs, in our tests. The other quality concerns are resizing and deinterlacing when applicable, which employ more advanced methods in Handbrake.
As Nick points out, rendering to DNxHD codec is an excellent visually lossless intermediate solution between Vegas and Handbrake. Handbrake does not accept frameserving, as in Nick's MeGUI methods. Best thing to do is download Handbrake, start with the settings we came up with, and get your feet wet. https://trac.handbrake.fr/wiki/HandBrakeGuide
Vegas is a video editor that can work at various resolutions, and render to resolutions and formats that are different than the source video. If you are editing a high definition video to high definition, your final result is very high quality and the render time isn't too long. The file will be too big to put up on Youtube or Vimeo though, and the format won't be right either. There is a good chance that high definition master will be interlaced as well and the Youtube or Vimeo encoder will make a mess of it.
Another option is to render to a format that is more optimized for upload to Youtube or Vimeo. This will probably be H264 format because the file will be smaller, easier to upload, and won't go over the maximum allowable Youtube or Vimeo limit. It will also probably be 1280 x 720 pixels and be a progressive image (as opposed to interlaced which looks good on a HD TV but terrible on a computer monitor). In order to do this from Vegas you need to deinterlace, resize, and compress your video into an H264 format. Vegas will do all this, but the quality won't be that great and it will take a long time for the render.
The option that many of us use (myself included) is to render a high definition master at full project size and format, with a high quality compression format, and then to rerender that master into a format that is better suited to Internet upload to Youtube or Vimeo. There are many programs that do this, but for most of us, our favorite is Handbrake because it is free and does a very high quality conversion.
You would think that these two steps (rendering a master, then converting with Handbrake) would take longer, but actually because Vegas (and other video editors) takes so much longer with the deinterlace and resize, it is actually quicker.
Handbrake has three advantages compared to Vegas when used to do this conversion:
1, It is faster. 2, It uses a Lancos 3 resize which looks better than the bicubic resize that Vegas does. 3 It does a decomb instead of a traditional deinterlace which looks much sharper.
Handbrake doesn't just beat Vegas in doing this conversion. It beats almost everything else as well. There is another way to do this which involves feeding the output of Vegas's master render (frameserving) into another program (called AVI Synth) and running a custom script that will do all the necessary steps required to deinterlace and resize according to the script, but what I find is that I like having a high quality HD master anyway, and that the quality of a Handbrake render as compared to the frameserved Avi Script approach is so close in quality that really it is just a matter of personal preference.
The Handbrake approach is much better (in my opinion) than doing it straight from Vegas (or any other NLE software) however. Videos rendered to an Internet uploadable format look fine in a Youtube or Vimeo window, but if you expand them to fill the screen, the Handbrake rendered quality just leaves the Vegas (or FCP or Premier Pro) render in the dust. That is why so many of us it.
Nick, x264 CQ vs. Sony AVC?
Nothing else in the chain, x264 was faster in my preliminary tests some time back iirc.
And Vegas->DNxHD->Handbrake is about the same time as Mainconcept 2-pass.
Worth another head-to-head test.
If you're considering Handbrake you should check out Mediacoder. I've used both and in my opinion Mediacoder is much more flexible and gives MUCH better output at low bitrates than Handbrake using H264. The picture degrades "naturally" when the bitrate is too low, no massisive macroblocking anomolies that you'll see in Handbrake.
Mediacoder is more difficult to use but also much more flexible.
Sorry, I was thinking in terms of the whole workflow, for which Sony AVC wins hands down for speed and simplicity (NOT quality). But once I had my DNxHD, I found the render time about the same for Handbrake as for Sony AVC. These figures are on a 3-year old laptop rendering a 2:21 long project.
DNxHD 1080i - 25 mins
Handbrake 1080i>720p - 14 mins
Sony AVC 1080i>720p - 14 mins
MainConcept AVC 1080i>720p - 52 mins
1. .MXF is the way to go if you start with progressive clips - DNxHD (free version) seems unable to produce progressive output, and .MXF is OK for Handbrake (but better set it up for mono sound).
2. Use Vegas to properly prepare your project for compression. If you didn't shoot with good light, your video will have lots of noise. Clean it up - perhaps with Pixelan's Smart Blur - which preserves edge sharpness - or New Blue's skin smoother (if shooting against a green screen), or the free Mike Crash Smart Smoother (which works almost as well as Pixelan's).
Avoid shots with lots of detail and sharp edges. Tree leaves, for example, are notorious bandwidth hogs, for they may have lots of sun reflections' highlights (if the leaves are shiny, like rubber trees, for example), with lots of high-contrast deep shadows.
Nevertheless, most of my work is shot in the studio and I bring in the great outdoors - especially overhanging tree branches - as often as possible to give a more interesting and open feel to the scene.
For that reason, I shoot lots of green screen, then CK in order to get the outdoor background, which I blur. The blur not only reduces video noise in the dark areas of the tree, but also softens all the edges, reducing bandwidth requirements. As a bonus, the blurred background gives a desirable depth-of-field illusion.
By also avoiding panning except when really needed to move the story forward, avoiding unnecessary fades, etc., one can get good results with low bandwidth requirements.
At 46 seconds into the following clip you can see an example of greenscreen + trees (be sure to click the full-screen button): http://www.torealize.net/2physics.html. That clip also has ocean waves, fades, and rather fast pans, but nevertheless was encoded at only 500 kbps in handbrake (2 pass encoding).
As Amendegw pointed out to me on another forum, an even lower bitrate would be better yet, for connection speeds in internet cafes, etc., may be really slow. As a test, I encoded that project at 200 kbps and the image quality wasn't too bad; but I think it would be better to first reduce the size down from its current 1280x720 and then seek the lowest possible bitrate.
I was surprised to see that 640x360 still looked quite good. When I get time, I plan to post a slow-speed-connection version along with the above version for the videos on my website . . .
"MXF is the way to go if you start with progressive clips - DNxHD (free version) seems unable to produce progressive output"Use the following settings for rendering DNxHD to 1920x1080 60p (as musicvid pointed out to me, you can choose any of the progressive presets, but make sure you choose a 59.94 frame rate in the template):
That said, imho, MXF is a good intermediate format - advantage is that it produces a smaller, albeit lossy render.
What I find is that mpeg2 footage will only smart-render if it is flagged as interlaced. If it is flagged as native progressive it will do a lossy rerender throughout. What I do when I shoot progressive is to use the setting on my Z7U where the footage is progressive, but flagged as interlaced. Clips shot with this setting smart-render just fine. I often use a mix of 30p and 60i footage (all flagged as 60i). The decomb in Handbrake will leave the progressive bits alone for Internet renders.