when a client asks: are u doing any web video?

robfilms

Veteran
when a client asks: are u doing any web video?

what should i be asking? thinking? budgeting?

what should i be shooting with? lighting for? compressing standard?

sorry in advance for the newbie type questions.

up until now, all of my work has been for presentation events (projection, plasmas, crt) or broadcast. much of that work has been posted on websites but i have not been the one to "turn" my sd video into web material.

can u define "web video" for me?

any links where i can learn more?

thanks in advance for any info u care to share

be well

rob katz
harvest films
 
There is only 1 thing that differentiates a production for TV and a production for internet. the interlacing.
Internet content must be deinterlaced to avoid very ugly artifacts. If you can, shoot with a progressive camera.
For the rest, you only have to match the quite high standard that internet advertising offers. So just do a great job and it will be great in internet. (and remember that once you edited it in HD or SD, it will for sure look even better in internet because it will be a smaller format, hiding noise and other bad things that are visible in a large format.

Ask the client what format does his webmaster want: QT, WMV or Real.
 
There are actually several things you need to know, such as final destination and format. YouTube has it's own format and compression I believe. But if that isn't what they plan, you need to know what format, whether it be Flash, Avi, QuickTime, or Windows Media. Compression for the web also has problems with busy backgrounds. If you are shooting a talking head interview, make sure you have a solid or static background, as opposed to lots of people and things moving around behind them.
 
...And if they say "Real", convince them otherwise or walk away.



No really, there are only a very few differences in producing for the web.
Interlacing was allready covered, and that's about it for technical differences.

However; generally web encoding is done at lower resolutions and framerates that standard broadcast video. As a result, try to limit or reduce fades, and experiemen with pans, tilts & movement before getting too deep into the project so that you understand how they can affect compression.

Also, generally video is presented on a smaller screen online, so shoot just a little tighter, and be sure to use graphical elements such as text that are large enough to read when the image is scaled down.


And also be aware that video on a computer is presented as "underscan", the entire frame is shown, so if you shoot with the normal safe areas as you'd use for broadcast (which is a good idea), be aware that you will want to crop a bit from each side of the image when you encode to prevent catching "uglies" from the edge of the shot into the final video.


- Mikko
 
Nice way to clean up there Mikko! I also agree with the Real video being a no-no. While it may come out nice, I just really hate them, and the way they take over your PC. Come to think of it, aside from the nice saturation, I feel the same about Quicktime ;)
 
thanks to all who have shared their thoughts.

so let me get this right:

shoot progressive (dvx100b should do well)

stay in tight on talking heads

minimize movement when possible

keep away from busy backgrounds

check titles and title-safe

check with client as to delivery format
(by the way, since i use fcp i'm leaning towards quicktime)

there has been no mention of lighting contrast issues

what about sound? a non issue?

is there anything else that i should be thinking about?

thanks in advance for any other thoughts

be well

rob katz
harvest films
 
Use cuts unless you really need the dissolves. Web compression is much more noticeable with dissolves.
 
Use cuts unless you really need the dissolves. Web compression is much more noticeable with dissolves.


does this go for most dissolves? even 5 frame? 10 frame? 15 frame?

again, thanks for sharing your insights

be well

rob katz
harvest films
 
Yep.

Web compression works in chunks of frames (Groups Of Pictures actually, called "GOP"). Any changes withtin the group are somewhat based on the first and last frames of the group, so a clean cut looks good because the whole pictures changes quickly, but during a fade, the compression can have trouble and basically turn the picture to mush for the duretion of the fade/dissolve.

Also, at lower framerates, fancy transistions (like disolves) wont' be near as smooth, and may appear to stutter, whereas a cut simply can't stutter.

- Mikko
 
Back
Top