|
|||||||||
|
Thread Tools | Search this Thread |
April 10th, 2006, 12:25 PM | #1 |
New Boot
Join Date: Aug 2004
Location: France
Posts: 21
|
Pal resolution
Hi,
I don't want to start a new flame war. I just want to know, if you really get a significant better resolution with PAL model ? I really hesitate to buy the NTSC model for frame rates advantage. ps : I don't care to be compatible with my mother's TV. |
April 10th, 2006, 12:33 PM | #2 |
Inner Circle
Join Date: Feb 2004
Location: switzerland
Posts: 2,133
|
pal has more pixel, better color system, closer to 24p
no doubt that is an advantage. saying you could tell the difference just by looking at, probably only a trained eye really can. best of both world is HDV at 60i or 30p. same resolution for europe and US, while more picture for US standard. you can eventually downconvert to SD. |
April 10th, 2006, 01:18 PM | #3 | |
New Boot
Join Date: Aug 2004
Location: France
Posts: 21
|
Quote:
Better color system ? It is still 4:2:2. Never talk about HDV because I hate it. Did you read the story of Gigidu13 on the french forum : le repaire ? This guy was with his Z1 filming a road in our country and suddenly an ferrari was in front of him. He said : Whoaaaf cool ! He came back to home and looked at his recorded tape and he discovered than the ferrari was not there on the tape. After investigation, he concluded the GOP was guilty. This is a true story : http://www.repaire.net/forums/camesc...-200-a-16.html Never go in area51 with a HDV camcorder. nobody will believe you... |
|
April 10th, 2006, 04:03 PM | #4 |
Barry Wan Kenobi
Join Date: Jul 2003
Location: North Carolina
Posts: 3,863
|
In high definition mode they'll be indistinguishable. Only difference that matters is the 60hz vs. 50hz frame rate.
In standard-def mode, PAL will have higher resolution, but -- I mean, the question's academic, really. If you're wanting to make an NTSC DVD but shooting in PAL for the "higher resolution", you should understand that any advantage PAL had will be thrown out the window when you try to convert it to NTSC. |
April 10th, 2006, 04:23 PM | #5 |
Inner Circle
Join Date: Feb 2004
Location: switzerland
Posts: 2,133
|
Pal is not 4:2:2 but 4:2:0 , still better than 4:1:1 (almost the same, except samples are taken differently), you can check this by making a keying on a blue background for example.
NTSC will perform poorly versus PAL. I agree for HDV, that is why i am still waiting for some cheap camera offering 3 CCD 720p (shorter gop 6, instead 12) at reasonable price with real sensor (i mean 1280x720). I own a JVC HD1 that i only used for 1 hour, but it is not yet what i expect from HD camera. Anyway if your subject is normally static (moving at reasonable speed), you can expect HDV giving you the equivalent of 4:2:2 in SD. This is not really important for quality of picture in itself, but it give you a lot of margin when editing and applying filters on the picture. I made a movie with a FX1 and i have to admit that it gives decent picture. additionally, working HDV with some high end codec like cineform will give you the equivalent of a 10 bit 4:2:2 processing versus a poor 8 bit 4:2:0 DV codec. |
April 10th, 2006, 04:52 PM | #6 | |
New Boot
Join Date: Aug 2004
Location: France
Posts: 21
|
Quote:
Of course, I was talking about HVX-200 in DVCPRO-HD mode (720 and 1080). :o) |
|
April 10th, 2006, 04:58 PM | #7 | |
New Boot
Join Date: Aug 2004
Location: France
Posts: 21
|
Quote:
Can you really obtain from HDV 10 bit 4:2:2 with high end codec ? |
|
April 10th, 2006, 05:22 PM | #8 |
Inner Circle
Join Date: Feb 2004
Location: switzerland
Posts: 2,133
|
in fact this is not really the problem.
If you work in a space that just exactly fit the data, each time you are doing any calculation on the data, , you bump into the upper and lower limit and the data are cropped (lost). If you use a higher format (ideally uncompressed video at 4:4:4), you can generate any value and rework them as much as you need, while keeping the real information. simple math, if you workwith only one decimal. 1/3=0.3 multiply back by 3 --> .3x3=0.9 (and there is no way to know what was the original value) if you use a larger space (more decimal=more bits) 1/3=.333 -----------------> .333x3=.999 You got a better result and that was not depending from the orginal data. So it is not what you get from the camera that is important, it is the way you work on it. So if you got a good signal with plenty of information (like HDV) and it is treated properly, i am sure you can get amazing result. Unfortunately most people focuses only on on the data and forget about the processing. Working native HDV or DV is the worst thing i can imagine. By the way the cineform codec allows you to work SD too. There are other codecs (mainconcept DV50 for example) allowing to work SD at higher quality , but cineform is almost the only to capture and rescale the information instantly. |
| ||||||
|
|