Forgot Password
Pentax Camera Forums Home

Show Printable Version Search this Thread
08-14-2020, 08:22 AM   #16
swanlefitte's Avatar

Join Date: May 2015
Location: New Orleans
Photos: Gallery
Posts: 3,279
If 1 person starts typing a post then another person starts but finishes the post before the former. I am uncertain who would be first, and on what principal would we judge? Will the wording of this question affect your response? I know you are on the forums to answer but dodn't know your answer. Later I will know your answer but not where you are.

08-14-2020, 09:25 AM   #17
Not a Number's Avatar

Join Date: Mar 2012
Location: Venice, CA
Posts: 7,437
QuoteOriginally posted by StiffLegged Quote
First time I ever heard that film used pixels...
That's why I said "in a sense".... Pixel is short for PIcture Element or one point that makes up the screen image so by analogy or loose definition the grains of silver or dye clouds are the elements that make up the picture in film.

Confusion - he talks of a pixel being an array of pixels (let's call these dots even though they may be square), either 4 × 4 or 3 × 3. So is he saying a "square" pixel is an array of 4 × 4 or 3 × 3 of "dots" the same color and what not? If that's the case then isn't this the same in effect as using larger and less PPI? No wonder it looks blockier. Then aren't the triangular or rectangular arrays using the "native" "Dots" per inch? So if you not use the square array pixels and map to the "native" DPI you shouldn't get a "pixelated" image.

The advantage I see with using the triangular and rectangular arrays is storage. Instead of having to store each "dot" you can just store a description of each array.

BTW: Pixelated should not be confused with Pixilated or Pixilation is an animation stop motion technique using live actors as the frame-by-frame subject.
08-14-2020, 02:26 PM - 1 Like   #18
Site Supporter
Site Supporter

Join Date: Dec 2015
Location: Goldsboro North Carolina
Photos: Gallery | Albums
Posts: 1,746
QuoteOriginally posted by Paul the Sunman Quote
Yes, but WHEN were they there? (Nerdy quantum mechanics joke )
They were there when Schrödinger's cat was alive. Or dead. Or both.
08-15-2020, 08:05 AM - 1 Like   #19
Loyal Site Supporter
Loyal Site Supporter

Join Date: Mar 2012
Location: Rochester, NY
Photos: Gallery
Posts: 4,383
Stochastic imaging is where it is at. Here is a continuous grey scale. Unlike halftone printing where you use bigger dots where you want more ink, in stochastic printing all the dots are the same size but you put more dots where you want more ink. It requires a lot more processing power to so stochastic printing so it did not really take off until the late 1990's when RISC processors were developed. The modern CPU can handle these images without much effort.

The upsides are several. Better resolution, more crisp images and better ink coverage could lead to ink savings can be as high as 25%.

On the downside it required the presses to be in optimal condition. If presses are not properly maintained, the image can go south pretty quick.

08-29-2020, 07:45 AM   #20
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 10,830
QuoteOriginally posted by Fenwoodian Quote
In the above video he explains how "variable shaped pixels" are so much better than his original "square pixels".
First of all, the description "variable shaped" is misleading.

The pixels still all have regular shapes, except that he is using a "half-pixel" approach with eight possible shapes (four with vertical or horizontal divisions and four triangles).

Next, his comparisons don't make sense because he compares images which have roughly the same number of pixels when he really should be comparing images that use the same amount of data. By using pixels with a more fine-grained geometry he obviously achieves better approximations but he needs to invest more data. For B&W images it would be three times the amount of data and with colour images, his approach requires two colours to be stored per pixel as opposed to just one (plus the three bits per pixel to select the correct shape).

Finally, the whole idea is based on the premise that pixels can be reconfigured to have one of the eight shapes. If pixels cannot be reconfigured then you'd need smaller (invariable shape) pixels to display the shapes. Rotating masks don't work so pixels would really have to be reconfigurable in some way, perhaps similar to how LCD segment displays work (but colour complicates this quite a bit).

Without pixels that actually can assume a number of different shapes, the idea reduces to an image compression technique and I'm almost existing compression techniques, e.g., based on wavelet transforms (-> JPEG 2000), would outperform it handsomely.

Last edited by Class A; 08-29-2020 at 08:00 AM.

  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
photography, photoshop, square, square pixels
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Similar Threads
Thread Thread Starter Forum Replies Last Post
Small pixels, large pixels compared beholder3 General Photography 11 12-17-2019 07:38 AM
Architecture 1950 ignath Post Your Photos! 3 11-15-2019 12:03 PM
Machinery B & W version of 1950's Chevrolet pick-up @ MartinKP's request. RICHARD L. Post Your Photos! 5 11-01-2019 11:58 AM
More pixels or Bigger pixels - That is the question ... R. Wethereyet Pentax K-5 13 06-06-2018 10:49 AM
Abstract Circle, circle, square, square LensBeginner Post Your Photos! 2 11-19-2014 05:44 PM

All times are GMT -7. The time now is 03:48 PM. | See also:, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]