Daniel Low
Forum Replies Created
-
Can you confirm that the clip is there if you export to a Quicktime Reference movie?
There is no reason why, or indeed how, compressor would simply miss out on one section in your whole clip.
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
Did you absolutely, 100% totally make sure that effect was completely 100% totally and utterly rendered in FCP?
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
It’s a good rule of thumb for any codec, however the H.264 standard is more flexible and is equally as efficient at 16×8, 8×16, 8×8, 8×4, 4×8, and 4×4.
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
The Canopus DV Codec is regarded to be one of the best, why would you want to use something else?
What one are wanting to use?
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
Well you say your audience is on a ‘fast connection’ and that could be anything. Fast to one person might be 1Mb/s, fast to me is over 8Mb/s but fast to people in other parts of Europe is over 20Mb/s.
If I were you I’d be delivering H.264 at a full frame rate with manually forced keyframes set at between 5 and 10 seconds, obviously alongside the automatic keyframes the codec will insert.
Neither I or Ed can give you a perfect profile, it’s up to you to experiment with your footage and your typical connection but we’re here to try and help tweak it after you’ve run a few tests.
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
Stick with the native, and excellent Prores422. There’s no point in changing it or indeed up-sampling to 10bit
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
but the image quality is not as good as using fractional frame rates when you have limited bandwidth (under 500-700kbs for example)
I was delivering live streams of news and sports at QCIF resolution at 100kb/s, 25fps to 1st generation 3G phones over 4 years ago. Picture quality needed to pass a number of different QA levels before being signed off and all fraction frame rates were presented. There was no need to reduce temporal resolution then and I personally haven’t had the need to do it since.
Your milage obviously varies.
When dealing with HTTP (Progressive download) delivered content, halving the frame rate rarely doubles the image quality, by doing so you are simply making the job a lot harder for the codec as the differences between frames is that much greater, this does not apply to static or near stationary content like talking heads of course.
As the title of the this thread asks for high quality, I consider that to be high quality in both the spatial and temporal domains. Fractional frames rates are instantly noticeable to me and reduce the quality of the viewing experience more so than the odd bit of blocking.
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
Google has the answer for you.
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
There is never a need to use fractional frame rates these days. Lower frame rates destroy motion in your clip and can actually make the job much harder for the codec. Even low powered cell phones can play back full frames rates.
The keyframe setting you apply is only saying tot he encoder that you want to force a keyframe every ‘n‘ seconds or frames. All modern codecs insert keyframes automatically when they need to, at scene changes or when the difference in frames passes a certain threshold based on calculations which take other factors of your encoding parameters in to account.
However Ed is correct in saying that the fewer keyframes there are the harder it is to scrub through a clip, accuratley but it’s more noticeable with certain formats, like Flash
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free! -
In my experience, black and white restore are best avoided unless you have the time to step through every frame of your clip looking for excesses in the restore processes, or the setting are at the absolute minimums
__________________________________________________________________
Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free!