[Masterbrew2]


I'm not sure what the most accepted term is. What I mean by Pixel Fill Factor is the ratio of the pixels to the space between pixels.

 

Some TVs have quite a lot of black between the individual pixels. What is the consequence of this to image quality?

 

 

 

Examples below:

 

The LG OLED has quite a bit:



 

The Sony X9 doesn't have much:

 

[KOF]

I'm also curious about the relationship between pixel fill-rate and perceived sharpness. Those that claim LCDs appear sharper than the plasmas say so because of bigger gaps in between pixels, but the F8500 has smaller gap than typical plasmas and yet appear sharper than the LCDs.

 

[Chronoptimist]

I would say that it's the opposite - a lower fill factor could appear to be sharper because the image will appear to be more aliased.

The better the fill factor is, the smoother the image will appear, and the better the clarity (not sharpness) will be.


It's the same thing people confuse about increased resolution. Increasing resolution does not give you a sharper image, it gives you a smoother, more natural image that contains more subtle details.

 

[Elix]

My theory about why panels with less fill factor (more black spaces between pixels) appear sharper is that from a certain distance given certain DPI where you can't definitely tell the gaps, our brain reconstructs detail where black spaces are. You have an analogy in moving images - a CRT display with scanning beam or an LCD display with backlight strobing will have better motion clarity/less motion blur than a simple sample-and-hold display with no black frames in-between images. Our brain actually reconstruct images in places of black frames to produce a visibly fluid motion. Don't take that as a fact though, it's just what I believe in until I find a better explanation.


For example, a JVC X70 D-ILA (LCoS) projector with extremely high pixel fill ratio, while resolving all pixels clearly due to high quality optics, is revered by many for its film-like image (among many other good characteristics) but may appear soft in comparison with an LCD projector.


JVC X70 D-ILA PJ fill ratio:



Sanyo Sanyo PLV-Z4000 LCD PJ fill ratio:

 

[8mile13]



Quote:
Originally Posted by KOF 

I'm also curious about the relationship between pixel fill-rate and perceived sharpness. Those that claim LCDs appear sharper than the plasmas say so because of bigger gaps in between pixels, but the F8500 has smaller gap than typical plasmas and yet appear sharper than the LCDs.

Pixel gaps on the F8500 and the F5300 are the same so the F8500 sharpness seems not to be pixels related.



SAMSUNG PN F5300



SAMSUNG PN F8500

 

[Masterbrew2]



Quote:
Originally Posted by Chronoptimist  /t/1519802/pixel-fill-factor-pixel-layout#post_24410071


a lower fill factor could appear to be sharper because the image will appear to be more aliased.

The better the fill factor is, the smoother the image will appear, and the better the clarity (not sharpness) will be.

 

 

That is a good point. In some ways it can function the same as an unintentional sharpening effect, exaggerating micro-contrast between two differently colored pixels.

 

[tgm1024]



Quote:
Originally Posted by Masterbrew2  /t/1519802/pixel-fill-factor-pixel-layout/0_60#post_24410794

 

Quote:
Originally Posted by Chronoptimist  /t/1519802/pixel-fill-factor-pixel-layout#post_24410071


a lower fill factor could appear to be sharper because the image will appear to be more aliased.

The better the fill factor is, the smoother the image will appear, and the better the clarity (not sharpness) will be.

 

 

That is a good point. In some ways it can function the same as an unintentional sharpening effect, exagerating micro-contrast between two differently colored pixels.

 

I believe that a lower fill factor will appear sharper because it better approximates a square wave.  And squarewaves by their nature are contructed by very high frequencies stacked up.

 

The reason I mention this is because what it produces are an increase the recognition of jump discontinuities because the wells are now more apparent.  To take an extreme example, take a look at the following two "wave" forms:

Drops to zero:

                          ### ###
                      ### ### ###     ###
                      ### ### ### ### ###
                      ### ### ### ### ###
                      ### ### ### ### ###
                 ------------------------------
Jump discontinuities: ^ ^ ^ ^ ^ ^ ^ ^ ^ ^

100% fill:

                          ########
                      ############    ####
                      ####################
                      ####################
                      ####################
------------------------------
Jump discontinuities: ^   ^      ^    ^

 

[Masterbrew2]



Quote:
Originally Posted by tgm1024  /t/1519802/pixel-fill-factor-pixel-layout#post_24410919

 

 

I believe that a lower fill factor will appear sharper because it better approximates a square wave.  And squarewaves by their nature are contructed by very high frequencies stacked up.

 

The reason I mention this is because what it produces are an increase the recognition of jump discontinuities because the wells are now more apparent.  To take an extreme example, take a look at the following two "wave" forms:

Drops to zero:                          ### ###                      ### ### ###     ###                      ### ### ### ### ###                      ### ### ### ### ###                      ### ### ### ### ###                 ------------------------------Jump discontinuities: ^ ^ ^ ^ ^ ^ ^ ^ ^ ^100% fill:                          ########                      ############    ####                      ####################                      ####################                      ####################------------------------------Jump discontinuities: ^   ^      ^    ^

 

That makes sense. Good way to illustrate it.

 

[fafrd]



Quote:
Originally Posted by Masterbrew2  /t/1519802/pixel-fill-factor-pixel-layout#post_24412344

Quote:
Originally Posted by tgm1024  /t/1519802/pixel-fill-factor-pixel-layout#post_24410919


 


I believe that a lower fill factor will appear sharper because it better approximates a square wave.  And squarewaves by their nature are contructed by very high frequencies stacked up.


The reason I mention this is because what it produces are an increase the recognition of jump discontinuities because the wells are now more apparent.  To take an extreme example, take a look at the following two "wave" forms:
Code:

[CODE]Drops to zero:                          ### ###                      ### ### ###     ###                      ### ### ### ### ###                      ### ### ### ### ###                      ### ### ### ### ###                 ------------------------------Jump discontinuities: ^ ^ ^ ^ ^ ^ ^ ^ ^ ^100% fill:                          ########                      ############    ####                      ####################                      ####################                      ####################------------------------------Jump discontinuities: ^   ^      ^    ^

[/CODE]

That makes sense. Good way to illustrate it.

But not after it gets quoted

-fafrd

 

[tgm1024]



Quote:
Originally Posted by fafrd  /t/1519802/pixel-fill-factor-pixel-layout/0_60#post_24413164

 

Quote:
Originally Posted by Masterbrew2  /t/1519802/pixel-fill-factor-pixel-layout#post_24412344

 

Quote:
Originally Posted by tgm1024  /t/1519802/pixel-fill-factor-pixel-layout#post_24410919


 


I believe that a lower fill factor will appear sharper because it better approximates a square wave.  And squarewaves by their nature are contructed by very high frequencies stacked up.


The reason I mention this is because what it produces are an increase the recognition of jump discontinuities because the wells are now more apparent.  To take an extreme example, take a look at the following two "wave" forms:
Code:

[CODE]Drops to zero:                          ### ###                      ### ### ###     ###                      ### ### ### ### ###                      ### ### ### ### ###                      ### ### ### ### ###                 ------------------------------Jump discontinuities: ^ ^ ^ ^ ^ ^ ^ ^ ^ ^100% fill:                          ########                      ############    ####                      ####################                      ####################                      ####################------------------------------Jump discontinuities: ^   ^      ^    ^

[/CODE]

That makes sense. Good way to illustrate it.

But not after it gets quoted

-fafrd

 

Yeah, the "Formatted" Style type is horribly broken.  Not just from that either.  Try cutting and pasting lines into it.

 

[Mike99]

On a related note -


I remember years ago I was looking at VCRs & was at a store that had different VCRs each connected to its own TV. I remember up close display #1 had a nicer looking picture. Smooth & no graininess. TV #2 looked noisier or grainier & I'm guessing most people would choose #1 hands down.


Anyway when I moved away from the TV display, maybe 15' - 20' or so, then TV #2 looked sharper & better. This was one of my first learning experiences. TV #2 with the offensive looking high contrast edges looked coarse up close but those same sharp edges made for a better picture from a distance. And by comparison TV #1 did not look as sharp.


Seems almost counter-intuitive that if you fill in the all the gaps & provide more data & make the display have nice edge transitions that it might make the picture less sharp looking. That may not be a proper explanation but I hope you get the picture.

 

[Masterbrew2]



Quote:
Originally Posted by Mike99  /t/1519802/pixel-fill-factor-pixel-layout#post_24415415


On a related note -


I remember years ago I was looking at VCRs & was at a store that had different VCRs each connected to its own TV. I remember up close display #1 had a nicer looking picture. Smooth & no graininess. TV #2 looked noisier or grainier & I'm guessing most people would choose #1 hands down.


Anyway when I moved away from the TV display, maybe 15' - 20' or so, then TV #2 looked sharper & better. This was one of my first learning experiences. TV #2 with the offensive looking high contrast edges looked coarse up close but those same sharp edges made for a better picture from a distance. And by comparison TV #1 did not look as sharp.


Seems almost counter-intuitive that if you fill in the all the gaps & provide more data & make the display have nice edge transitions that it might make the picture less sharp looking. That may not be a proper explanation but I hope you get the picture.

 

I'm not a big fan of sharpness. The artifacts and aliasing it can create are too harsh for me. Being a somewhat enthusiastic photographer and photoshopper, I've tried and abused all kinds of effects - sharpness is one that I learned to be very careful with. It can easily ruin nature scenes, especially foilage detail can become unpleasant to look at.

 

[Elix]



Quote:
Originally Posted by Mike99  /t/1519802/pixel-fill-factor-pixel-layout#post_24415415


On a related note -


I remember years ago I was looking at VCRs & was at a store that had different VCRs each connected to its own TV. I remember up close display #1 had a nicer looking picture. Smooth & no graininess. TV #2 looked noisier or grainier & I'm guessing most people would choose #1 hands down.


Anyway when I moved away from the TV display, maybe 15' - 20' or so, then TV #2 looked sharper & better. This was one of my first learning experiences. TV #2 with the offensive looking high contrast edges looked coarse up close but those same sharp edges made for a better picture from a distance. And by comparison TV #1 did not look as sharp.

Does it add up with what I wrote, in your opinion? Or am I being delusional?

Quote:
Originally Posted by Elix  /t/1519802/pixel-fill-factor-pixel-layout#post_24410083


My theory about why panels with less fill factor (more black spaces between pixels) appear sharper is that from a certain distance given certain DPI where you can't definitely tell the gaps, our brain reconstructs detail where black spaces are. You have an analogy in moving images - a CRT display with scanning beam or an LCD display with backlight strobing will have better motion clarity/less motion blur than a simple sample-and-hold display with no black frames in-between images. Our brain actually reconstruct images in places of black frames to produce a visibly fluid motion.

Panels with small pixel gaps leave less space for our imagination.

 

[tgm1024]



Quote:
Originally Posted by Masterbrew2  /t/1519802/pixel-fill-factor-pixel-layout/0_60#post_24415451

 

Quote:
Originally Posted by Mike99  /t/1519802/pixel-fill-factor-pixel-layout#post_24415415


On a related note -


I remember years ago I was looking at VCRs & was at a store that had different VCRs each connected to its own TV. I remember up close display #1 had a nicer looking picture. Smooth & no graininess. TV #2 looked noisier or grainier & I'm guessing most people would choose #1 hands down.


Anyway when I moved away from the TV display, maybe 15' - 20' or so, then TV #2 looked sharper & better. This was one of my first learning experiences. TV #2 with the offensive looking high contrast edges looked coarse up close but those same sharp edges made for a better picture from a distance. And by comparison TV #1 did not look as sharp.


Seems almost counter-intuitive that if you fill in the all the gaps & provide more data & make the display have nice edge transitions that it might make the picture less sharp looking. That may not be a proper explanation but I hope you get the picture.

 

I'm not a big fan of sharpness. The artifacts and aliasing it can create are too harsh for me. Being a somewhat enthusiastic photographer and photoshopper, I've tried and abused all kinds of effects - sharpness is one that I learned to be very careful with. It can easily ruin nature scenes, especially foilage detail can become unpleasant to look at.

 

"Sharpness" is a term that has too many meanings colloquially, and that is part of the problem in discussions like these.

 

[Mike99]



Quote:
Originally Posted by Elix  /t/1519802/pixel-fill-factor-pixel-layout#post_24415700


Does it add up with what I wrote, in your opinion? Or am I being delusional?

If I look at your top picture from 5' away it looks almost like a solid gray image. However in the lower picture I can still the pixels, or details/edges.


Don't get me wrong, I'm not saying that fewer or larger pixels, or increased sharpness means a better picture. I was just pointing out some things that may affect a viewers perception of picture quality. FWIW my sharpness is turned up only a tiny bit because it just looked good to me.

 

[Mike99]



Quote:
Originally Posted by tgm1024  /t/1519802/pixel-fill-factor-pixel-layout#post_24415805


"Sharpness" is a term that has too many meanings colloquially, and that is part of the problem in discussions like these.

On another related note -

I have 2 versions of an old movie, one on Laser Disc & from an OTA broadcast. The LD is sharper. But I never realized how much the color was off until I saw the OTA version. One scene has a lady in dark purple blue dress. The OTA version shows dark buttons & a dark collar on the dark dress. The LD version looks almost black & I never even knew there were buttons on the dress, or a different color collar, because they don't show up.


The LD version shows more detail in respect to displaying fine lines in the image. The OTA version shows more detail in respect to displaying objects in the image. Now I'm comparing 2 different types of detail. If someone says one version shows more detail you would need some clarification on what they mean. Which one looks better is IMHO a personal preference. In this particular case the LD is not that much sharper or able to display very fine lines so I like the OTA version better because it displays more objects due to its better color.

 

[Masterbrew2]



Quote:
Originally Posted by tgm1024  /t/1519802/pixel-fill-factor-pixel-layout#post_24415805

 

 

"Sharpness" is a term that has too many meanings colloquially, and that is part of the problem in discussions like these.

Yeah, I guess you're right. I've come to think of sharpness as something perceived and unrelated to the actual amount of information or resolution in an image.

 

[John Mason]

"The sharpness of an image is evaluated when the coarse details are shown in high contrast."


Quote is from a brochure/pdf-paper by Arri's Dr. Hans Kiening, which has a section on resolution versus sharpness. Fig. 6 in the paper compares two photos of a movie camera, showing how higher contrast in one photo boosts sharpness while the other photo has more actual resolution. Fig. 8 has two graphs, each plotting resolution versus contrast of the two camera photos. The apparently sharper image has more area within its plotted curve. -- John

 

[tgm1024]



Quote:
Originally Posted by John Mason  /t/1519802/pixel-fill-factor-pixel-layout/0_60#post_24442160


"The sharpness of an image is evaluated when the coarse details are shown in high contrast."


Quote is from a brochure/pdf-paper by Arri's Dr. Hans Kiening, which has a section on resolution versus sharpness. Fig. 6 in the paper compares two photos of a movie camera, showing how higher contrast in one photo boosts sharpness while the other photo has more actual resolution. Fig. 8 has two graphs, each plotting resolution versus contrast of the two camera photos. The apparently sharper image has more area within its plotted curve. -- John

Doesn't matter.  My comment was about the colloquial usage of "sharpness".  And as such, it becomes very confusing, very fast.

 

[John Mason]



Quote:
Originally Posted by tgm1024  /t/1519802/pixel-fill-factor-pixel-layout#post_24442583

Quote:
Originally Posted by John Mason  /t/1519802/pixel-fill-factor-pixel-layout/0_60#post_24442160

Doesn't matter.  My comment was about the colloquial usage of "sharpness".  And as such, it becomes very confusing, very fast.

Agree. Find I sometimes slip into misuse of "sharp" to describe certain HD images. Just wanted to review and reference one of my favorite papers on the topic. -- John