Grass Valley Forums Facebook   Twitter   YouTube  

Go Back   Grass Valley Forums > Editors > Editing with EDIUS

Reply
 
Thread Tools Display Modes
Old 02-18-2010, 07:05 PM   #111
SRsupport
Grass Cutter Gold
 
SRsupport's Avatar
 
Join Date: May 2007
Location: South America
Posts: 3,353
Default

Have you seen that what 4:2:0 means? That is the clue read what the sampling means then you will understand why you can convert it to 4:2:2.

It not only Canopus doing this you also have hardware chips doing this.
here you go:
http://www.fujitsu.com/downloads/COM...enation_wp.pdf
This will give you an idea what hardware does.

I think Brandon explained this before for Hq let me look if I have it....
__________________
Steve
EDIUS Trainer, Grass Cutter Gold
A proud EDIUS EDITOR
For more information on the Grass Cutter program please visit: http://www.grass-cutters.net

Last edited by SRsupport; 02-18-2010 at 07:14 PM.
SRsupport is offline   Reply With Quote
Old 02-19-2010, 09:15 AM   #112
zvit
Senior Member
 
Join Date: Sep 2007
Location: Israel
Posts: 2,145
Default

From the site alone, my question still stands. 4:2:2 is more color accurate than 4:2:0 so technically, you would have to guess to convert 4:2:0 to 4:2:2. Here's from the site:

Look at how much more accurate 4:2:2 color is! By my quick count, Id say that about 8 of the pixels are either an excellent match, or at least as good a match as how the 4:1:1 and 4:2:0 systems were performing. That means that in this (admittedly extreme) case, were getting up to four times as much color accuracy! In a real-world circumstance the difference in results would not be so exaggerated, because real-world video subjects usually dont have so much variation in color between pixels (unless, perhaps, youre shooting a tourist whos wearing an alarmingly colorful Hawaiian shirt, for example!)


But I'll read the PDF and see how they do it. I must be missing something. (I might skip some words though.... it's a little long :-)
__________________
Gigabyte Z77X-UP4 TH, Intel Core i7 3770k 3.5GHz, 4 Core, 8 Threads, 32GB RAM, NVIDIA GeForce GTX 660 2GB, 500GB SSD HDD for OS, 20 Gigabytes Usable Hard Drive Capacity, Window 10 PRO 64-bit Edius 9.5 WORKSTATION...
zvit is offline   Reply With Quote
Old 02-19-2010, 10:08 PM   #113
GrassValley_BH
Demystifier/Analogizer
 
Join Date: Apr 2007
Location: Hillsboro, OR
Posts: 5,785
Default

Using 4:2:2 instead of 4:2:0 is like calculating with more significant figures in the intermediary process.

The more precision you have in the calculations, the more accurate the result will be, regardless of the precision of the output because of rounding and truncations with less precision.
GrassValley_BH is offline   Reply With Quote
Old 02-19-2010, 11:28 PM   #114
zvit
Senior Member
 
Join Date: Sep 2007
Location: Israel
Posts: 2,145
Default

Thank you for trying to explain but I still don't understand.

If I pick 2 apples off a tree and give you 2 apples... I gave you two apples.
If I pick 4 apples off a tree and give you 2 apples... again, I gave you just 2 apples.

Did what I give you become more because I picked more? No.

The intermediary process is recording more accurate color numbers (4:2:2) but is only outputting to SxS less accurate color numbers (4:2:0).

Let's say you add a Gaussian Blur to a photo layer in photo shop, thus blurring it slightly (Averaging and blending pixels). Can you then do some math to restore the original pixels? I don't think so. To me this seems the same and is why I don't understand.
__________________
Gigabyte Z77X-UP4 TH, Intel Core i7 3770k 3.5GHz, 4 Core, 8 Threads, 32GB RAM, NVIDIA GeForce GTX 660 2GB, 500GB SSD HDD for OS, 20 Gigabytes Usable Hard Drive Capacity, Window 10 PRO 64-bit Edius 9.5 WORKSTATION...
zvit is offline   Reply With Quote
Old 02-19-2010, 11:36 PM   #115
GrassValley_BH
Demystifier/Analogizer
 
Join Date: Apr 2007
Location: Hillsboro, OR
Posts: 5,785
Default

Here's an easy example.
Let's go with input of 2 decimal places, and output of 2 decimal places.
2.37 * 1.99 + 3.67 * 5.28

If I do the calculations with 2 decimal places...
4.72 + 19.38 = 24.10

If I do the calculations with 3 decimal places...
4.716 + 19.378 = 24.09 (rounded down from 24.094)

Both answers are 2 decimal places, but the precision within the calculation stage makes a difference.

The degree of difference on output gets worse with more calculations as the "rounding error" compounds.
GrassValley_BH is offline   Reply With Quote
Old 02-20-2010, 01:50 AM   #116
zvit
Senior Member
 
Join Date: Sep 2007
Location: Israel
Posts: 2,145
Default

Found a very good explanation though I still won't fully understand as I will write below:
http://www.hometheaterhifi.com/volum...ug-4-2001.html

Here are some quotes from the page:

The highest resolution format is 4:4:4, which means that for every 4 samples of Y, there are also 4 samples of 'Cb' and 4 samples of 'Cr'. In other words, the color signal has the same resolution as the black and white signal. This format is generally only used internally within a device to avoid degradation during processing. When the image is recorded to a master tape like D1 or D5, it is reduced to 4:2:2 (see below).

Next highest is 4:2:2, which means for every 4 samples of Y, there are 2 samples of 'Cb' and 2 samples of 'Cr'. There are still the same number of scan lines in the luma and the chroma, but the chroma signal has half as many samples on each scan line. When a 4:2:2 signal is decoded, the "missing" samples are effectively interpolated from the samples on either side.

The lowest resolution format, and the one used for DVD, is 4:2:0. This is a confusing designation, as it suggests that for every 4 Y samples, there are 2 'Cb' samples and 0 'Cr' samples, which is not the case at all. What 4:2:0 means is that there are half as many samples of 'Cb' and 'Cr' on each scan line, and half as many scan lines of 'Cb' and 'Cr' as there are of Y. In other words, the resolution for chroma is half that of luma in both the horizontal and vertical directions. If the full image resolution is 720 x 480, then the chroma information is only stored at 360 x 240. In 4:2:0, not only are missing samples interpolated on each scan line, entire scan lines of chroma must be interpolated from the scan lines above and below. The reason this is done is for efficient use of space on the DVD. Sure, 4:4:4 would be nice, but the DVD might have to be two feet in diameter. So, it's 4:2:0.

The tricky problem of 4:2:0 is that the method of interpreting the samples and interpolating new scan lines can be done two different ways. One way is used for interlaced images, and one for progressive images.


Lots more on the page.

I guess to understand an answer to my question, I'd have to be a mathematician or engineer. I'll ask one more time but don't expect an answer:

I graded 4 student's tests. The four grades were 10, 20, 30, 40.
If I want to know the average grade of my class I'd do:
(10+20+30+40) \ 4 = 25

Now, if I only had the average of the class, 25, how in the world can I know each of the student's grades?? It can be any 4 combination's that add up 100. Since in our case, every number (in the 4 pixels) is a unique color, guessing won't give you the original color. There must be a sophisticated mathematical formula to get the original numbers from the derived average.

Anyway, good talk. Thanks for the info and intel.
__________________
Gigabyte Z77X-UP4 TH, Intel Core i7 3770k 3.5GHz, 4 Core, 8 Threads, 32GB RAM, NVIDIA GeForce GTX 660 2GB, 500GB SSD HDD for OS, 20 Gigabytes Usable Hard Drive Capacity, Window 10 PRO 64-bit Edius 9.5 WORKSTATION...
zvit is offline   Reply With Quote
Old 02-22-2010, 09:38 PM   #117
GrassValley_BH
Demystifier/Analogizer
 
Join Date: Apr 2007
Location: Hillsboro, OR
Posts: 5,785
Default

There are actually two facets of your question...
1. How less color sampling is interpolated back to full color sampling.
Remember those math problems that went.
1, 2, 4, 8, 16, x
What is x?
It's like that.

Color sampling is based on location.

You can interpolate the missing values based on the adjacent values that you do have.

There are different strategies, and the simplest is to not interpolate at all and just duplicate the value from an adjacent pixel.

That's on the display and decoding side.


2. How more color samples affect the output, even if the output is a lower number of color samples.
On the processing side, it's like the example I posted about how many decimal points you use in the calculations.
By "preserving a spot" to put more of the calculated values in, you have a more precise result at the end, even if you have to "round it off" or remove samples.
The simplest strategy is to "pick what's in the corresponding box" from the working set, but some things employ more sophisticated methods.

The display interpolation and upsampling method can make the difference between fuzzy, discolored edges and sharp lines. Like with scaling, there's no universal answer - it really depends on the content of the footage. For example, in cell animation there are sharp transitions (one pixel can be full black while the adjacent pixel is full red) so blending neighboring values would result in fuzzed lines, while for natural scenes, blending rather than defined lines may be more favorable.
GrassValley_BH is offline   Reply With Quote
Old 02-23-2010, 08:33 AM   #118
zvit
Senior Member
 
Join Date: Sep 2007
Location: Israel
Posts: 2,145
Default

How do you explain this:

1: JPEG is a LOSSY compression.

From Wikipedia:

The ratios at which the downsampling can be done on JPEG are 4:4:4 (no downsampling), 4:2:2 (reduce by factor of 2 in horizontal direction), and most commonly 4:2:0 (reduce by factor of 2 in horizontal and vertical directions). For the rest of the compression process, Y, Cb and Cr are processed separately and in a very similar manner.

http://en.wikipedia.org/wiki/JPEG

We know that JPEG cannot be reconstructed back to the original photo (hence the name "lossy") but according to what we have been talking about, you SHOULD be able to, the same way you do with video.
__________________
Gigabyte Z77X-UP4 TH, Intel Core i7 3770k 3.5GHz, 4 Core, 8 Threads, 32GB RAM, NVIDIA GeForce GTX 660 2GB, 500GB SSD HDD for OS, 20 Gigabytes Usable Hard Drive Capacity, Window 10 PRO 64-bit Edius 9.5 WORKSTATION...
zvit is offline   Reply With Quote
Reply
 
Go Back   Grass Valley Forums > Editors > Editing with EDIUS
 

Bookmarks

Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
How do I increase the speed in Multicam mode on EDIUS 5.x ? nevodavid Editing with EDIUS 1 08-10-2009 11:44 PM
CPU Speed Increase - FREE zoolane Editing with EDIUS 87 11-21-2008 07:54 AM
Edius 5 Memory setting (buffer size) Albatros Editing with EDIUS 2 11-03-2008 05:44 PM
buffer settings RonnieMartin Editing with EDIUS 3 08-10-2007 12:35 AM

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are Off
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 05:49 AM.


Copyright 2014 Belden Inc. All rights reserved.