optimizing bitrate: what's effected?

s k r o o t a y p wrote on 3/19/2006, 5:50 PM
~i had a project ready to burn and the estimated project size was 123% of the available 4.7G disc (120 min. project/ 60 min disc). so i clicked optimize and pulled the bit-rate slider back until it was under 100%. DVD burned fine, fits disc, looks good.

what i want to know is what actually happens when you do that? are you giving up something somewhere else in exchange for getting it to fit the disc? is there a certain percentage range past 100% you should avoid trying to squeeze onto, say, an average 4.7G disc? i'd like to know. thanks!

Comments

s k r o o t a y p wrote on 3/19/2006, 6:37 PM
. . . also, when is right to check the override box?
Chienworks wrote on 3/19/2006, 6:52 PM
Were you feeding DVDA an AVI file or an MPEG2 file? If the former, all you did was specify what bitrate to use when encoding to MPEG2. Sure, a lower bitrate results in lower quality, but squeezing 123% into 100% isn't that much change.

If you're starting with an MPEG2 file then you're reencoding and recompressing an already encoded and compressed file. This is never a good idea. You would do better to go back to Vegas and render to AVI first.
s k r o o t a y p wrote on 3/19/2006, 7:11 PM
~the project was sent to DVDA as an AVI file. i am under the impression so far that MPEG(2) is to be mostly avoided until going to DVD. but, w/ regards to optimizing, what technically happens when you decrease bit-rate? is it like compression where a degree of dynamics are shaved off to gain space? and when would you say the percentage generally begins to push the envelope (w/ say, 4.7G discs)?
DrLumen wrote on 3/19/2006, 8:56 PM
A lower bitrate is equivalent to a higher compression. It's not really a compression like zip but in how the video is encoded and decoded. Technically, you would need to go through the math for encoding MPG2 (Discreet Cosine Transform), the I-frames and B-frames to "technically" know how the bitstream is affected. DCT is not a simple compression hence why it takes so long to encode...

For an example, take a graphic app like Corel, Photoshop, GIMP, etc and take a photo and then encode it as a JPG with a compression of 10% and then do it again with a compression of 90% (making the file smaller). You will easily see the artifacts generated. MPG does roughly the same. MPG also has to contend with change between frames so you will get the artifacts plus tearing or smearing in fast action frames.

I'm not sure how much help this was... Maybe I'm confused on how MPG works. :-)

intel i-4790k / Asus Z97 Pro / 32GB Crucial RAM / Nvidia GTX 560Ti / 500GB Samsung SSD / 256 GB Samsung SSD / 2-WDC 4TB Black HDD's / 2-WDC 1TB HDD's / 2-HP 23" Monitors / Various MIDI gear, controllers and audio interfaces

s k r o o t a y p wrote on 3/19/2006, 11:05 PM
~maybe i shouldn't have used the word "technically" : )

how about "lay person-technically" : )

thanks doc, i'm beginning to get the jist.

Chienworks wrote on 3/20/2006, 3:25 AM
You want a real simple example? Consider sending a telegram ... you have a 100 word letter to send, but you can't afford much, so you cut out words, abbreviate, take out maybe a whole sentence here or there. If you can afford to send a lot of words (high bitrate) then you can get your message accross relatively intact. However, if you have to go very cheap (low bitrate) you have to cut out more and more words. Your message gets harder and harder to understand. The person receiving it still gets words, but they might not know exactly what you mean. Eventually the message becomes a garbled mess when you reduce it too small.

Pretty much the same thing is happening with MPEG. The more bits you give it, the more faithfully it can encode and reproduce the original video quality. With fewer bits it has to start fudging the image more and more. Get the bit rate low enough and the image starts distorting to the point where you don't want to watch it.

What is the cut-off point? I'll tell you: i don't know. Neither does anyone else in this forum. It depends on you, your audience, the material, the expectations of the delivery format, and a host of other factors. Does the end result look ok to you and the customers? If so, you're ok. If you see an objectionable amount of artifacts and blockiness then you've lowered the bitrate too much. Scenes with low motion, large details, few sharp edges or contrast will compress to low bitrates very nicely. Fast action, small details, and high contrast edges will need a much higher bitrate to produce an acceptable image. I once taped a 2 hour 40 minute play that consisted almost entirely of people in drab clothing standing around on an empty stage. I had to reduce the bitrate to about 3200Kbps to fit it on one 4.7GB DVD. In "blind" comparisons no one could tell the difference between that and the original AVI files. On the other hand, a diving competition with water splashes and colorful swimsuits still looked noticeably bad at 7800Kbps.

So, what's the answer? It depends. Try a few experiments and see what happens.
s k r o o t a y p wrote on 3/20/2006, 5:59 PM
~thanks for the detail CW, and your continued replies to my oft tedious questions. appreciate it.