|
|||||||||
|
Thread Tools | Search this Thread |
December 14th, 2004, 02:52 AM | #31 |
Regular Crew
Join Date: Jul 2004
Location: Helsinki, Finland
Posts: 52
|
Cineframe24 sounds horrible. I'm glad we have cineframe25! It looks quite good. I have to try to capture some footage, as I've only been able to look at it on an SD television and an HD LCD monitor (the colors and motion we're poor, as usual for an LCD).
The cineframe25 at 1/50 has maybe a little bit more motion blur than progressive from DVX or XL2, but it does give it a nice film look. I used to work for a couple of months transferring film to video so I've seen a lot of real "film looks". I usually think that progressive material is missing that feel of a mechanical shutter. Currently I think that this cineframe25 gives a little bit like a mechanical shutter feel to the picture. Ofcourse when you raise the shutter speed, you can have that jerky progressive look, but it will cost some light. I have never before used a frame mode in a camera. I've always done some good deinterlacing in post. But it's really annoying to edit interlaced (you don't get the feel of the final movie), and this time I just might be willing to use cineframe25. But I'll just have to do some more tests and comparisons. I'm not that worried about losing resolution, (which the cineframe25 propably does), but we'll have to test if the resolution is lost only during motion. Is cineframe25 a form of adaptive deinterlacing, where only moving stuff is deinterlaced. Isn't the cineframe25 done in the same DSP (DXP) as the MPEG encoding? This might lead me to think that it would have the info about whether there's movement on each pixel, so that it might do some sort of adaptive deinterlacing. Any thoughts? Well, atleast everything that I've shot so far is cineframe25, cinematonegamma on, sharpness set to zero (I don't like digital sharpening). |
December 14th, 2004, 05:38 PM | #32 |
Major Player
Join Date: May 2004
Location: The Netherlands
Posts: 204
|
I think it's very important to look at the facts, all the facts and nothing but the facts if we want fair comparisons for the options of the fx1 in the film-simulation field (we already know that it can make good interlace footage).
We know that CF24 is not going to do it. But I don't think that because of that we have to write the fx1 off and certainly not the whole hdv-thing. Now for the ones that have not lost all their hopes after the CF24-dissapointment, there might be a ray of hope in the CF25 function, for PAL-landers already in the FX1E, and later also for the rest of the world in the Z1. Now I think there are some important things to know from this function; - Is it going to eat half your resolution, or is it smarter than that. I don't think we have had a definitive answer to that already. I am assuming (always dangerous!) that it works the same as CF30, but it might also not. We just need someone with a FX1E to make a good comparison test of a chart, and if you don't have one, just film something with a lot of detail with CF25 both ON and OFF (still shot - tripod). - How is the motion. Someone reasoned that when the normal interlace video is giving problems with pans etc., the image will only be further degraded when using CF25. While this might be very true, it's still an assumption, because someone else said that because the video is de-interlaced first and then compressed to mpeg-2 (because you have the CF25 effect also on uncompressed component-out), and since mpeg-2 is better in compressing progressive footage than interlaced footage (assumption?) that the image-quality is going to benefit from this. Bottom-line is, we need comparison from someone with a fx1E here as well. To answer this problem we need some moving shots (pans) as well as still shots with moving elements with CF25 ON and OFF again. I have the feeling that, although there was one or two posts about it, the interest to get some footage in CF25 was not very big. Probably because it's only the "pal" people who can really use this function right now, but because a lot of you Americans and other ntsc-ers are considering (or even pre-ordering) a Z1, that I assume they are also looking for some CF25 footage with great anticipation. So the real question is: which one of you fx1E-owners is going to be a hero and will post some footage? Hoping i'm not talking too long (or too much nonsense :) Steven |
December 14th, 2004, 07:46 PM | #33 |
Contributor
Join Date: Sep 2003
Location: Santiago, Chile
Posts: 932
|
> and since mpeg-2 is better in compressing progressive
> footage than interlaced footage (assumption?) Yes. 4:2:0 color encoding seems to have trouble with video, but handles progressive video better. I Wonder why HDV in interlaced mode is 4:2:0 instead of 4:1:1... perhaps just to keep hardware costs lower. Here is Adam Wilt on color sampling: http://www.adamwilt.com/DV-FAQ-tech.html#colorSampling
__________________
Ignacio Rodríguez in the third world. @micronauta on Twitter. Main hardware: brain, eyes, hands. |
| ||||||
|
Thread Tools | Search this Thread |
|