4:2:2 , what does it mean?
AAndromeda
3 Jan 2009 21:44
Actually lpoland you could be right, I might have been mistaken.
Yeah the RED Scarlet looks interesting but so far RED seem like a whole lot of hype, I hope they can deliver.
Yeah the RED Scarlet looks interesting but so far RED seem like a whole lot of hype, I hope they can deliver.
ODesigns
4 Jan 2009 00:08
I know someone who owns 2 RED ONE cameras. He brought one over to my work one day and showed us it's features. Pretty nice. Think of it as shooting video RAW and manipulating it later. Kinda like shooting digital photos in RAW.
ionescu
6 Jan 2009 07:53
Hi!
OBDesign is partially right. First of all, miniDV is not a TV format, but a tape one. On a miniDV, one could record many kinds of signal, one of them being DVPAL, which is 4:2:0.
Then, the numbers in X:X:X scheme refer to frames, not pixels. Those coding systems exist long before digital, and in analog you had no pixels at all.
FWIMC:
http://en.wikipedia.org/wiki/4:4:4
http://www.atsc.org/
The last link has useful information for those interested in acquiring new equipment and willing to know the industry trends.
God bless us all!
Honestly,
Christian I. Ionescu
OBDesign is partially right. First of all, miniDV is not a TV format, but a tape one. On a miniDV, one could record many kinds of signal, one of them being DVPAL, which is 4:2:0.
Then, the numbers in X:X:X scheme refer to frames, not pixels. Those coding systems exist long before digital, and in analog you had no pixels at all.
FWIMC:
http://en.wikipedia.org/wiki/4:4:4
http://www.atsc.org/
The last link has useful information for those interested in acquiring new equipment and willing to know the industry trends.
God bless us all!
Honestly,
Christian I. Ionescu
ODesigns
6 Jan 2009 16:36
That's true. I use miniDV tapes in my HDV camera, which is 4:2:0.
However, I meant miniDV as used in a standard definition consumer DV camcorder.
I should have been more precise.
However, I meant miniDV as used in a standard definition consumer DV camcorder.
I should have been more precise.
stillgravity
15 Jan 2009 02:07
Hi,
The topic is complex. You have to separate in the discussion the medium or mode in which the information is registered (magnetic, optical, etc.) from the compression scheme applied to the signal.
From a sampling perspective component video signals can be divided in three mayor components: Luminance (as black and white signal with all details of the image) and two chrominance components encoding all the color information.
The manufacturer usually needs to introduce some kind of compression to “accommodate” the signal to the technical capabilities of the designed system and recording medium. Multiple schemes of chroma sub-sampling have been introduced to make better use of the available bandwidth sacrificing part of the information that in the designer’s opinion causes the minimal effect in the perceived quality of the image.
From a practical perspective accepting the maximum sampling quality as 4:4:4 if you find for example 4:2:2 implies that the color information has been “sacrificed” half in proportion to the luminance (which carries most of the image details information) and for that reason they tend to affect mainly chroma key processes. (4:1:1 and 4:2:0 are not good for chroma).
Hope this help. Best wishes.
Ricky
The topic is complex. You have to separate in the discussion the medium or mode in which the information is registered (magnetic, optical, etc.) from the compression scheme applied to the signal.
From a sampling perspective component video signals can be divided in three mayor components: Luminance (as black and white signal with all details of the image) and two chrominance components encoding all the color information.
The manufacturer usually needs to introduce some kind of compression to “accommodate” the signal to the technical capabilities of the designed system and recording medium. Multiple schemes of chroma sub-sampling have been introduced to make better use of the available bandwidth sacrificing part of the information that in the designer’s opinion causes the minimal effect in the perceived quality of the image.
From a practical perspective accepting the maximum sampling quality as 4:4:4 if you find for example 4:2:2 implies that the color information has been “sacrificed” half in proportion to the luminance (which carries most of the image details information) and for that reason they tend to affect mainly chroma key processes. (4:1:1 and 4:2:0 are not good for chroma).
Hope this help. Best wishes.
Ricky
FiveElement
19 Jan 2009 11:54
Another complication is whether the picture is progressive or interlaced. 4:2:0 on a progressive source is generally pretty good. Interlace on the other hand can cause a lot of issues.
It is also worth noting that PAL DVCAM is 4:2:0.
Another kicker is when it comes to cameras that use an H/V spatial offset such as the HVX200. You might think that you are recording a 4:2:2 signal to DVCProHD, but in actual fact the camera is not producing that much chroma information. The resolution increase gained by the pixel shifting is only for luminance, not for chroma.
Lastly Chroma can be interpolated pretty well using plugins such as those from Graeme Nattress. So if you have 4:2:0 footage and want to do green screen it would be worth getting one of those plugins first.
It is also worth noting that PAL DVCAM is 4:2:0.
Another kicker is when it comes to cameras that use an H/V spatial offset such as the HVX200. You might think that you are recording a 4:2:2 signal to DVCProHD, but in actual fact the camera is not producing that much chroma information. The resolution increase gained by the pixel shifting is only for luminance, not for chroma.
Lastly Chroma can be interpolated pretty well using plugins such as those from Graeme Nattress. So if you have 4:2:0 footage and want to do green screen it would be worth getting one of those plugins first.