AM 250 lpi

[sornchai]

Hi. My standard is 175 -200 lpi with 300 dpi. If I want 250 lpi which dpi recommended ?
Another is how big dot sizes at 3% 250 lpi compared to 175 lpi and 200 lpi? Thank you

 

[Ulrich]

My standard is 175 -200 lpi with 300 dpi

really? Do you mean 300dpi resolution of a platesetter or ppi picture resolution?


sorry, i am not awakend enough in this morning to do the math with inch and mm... ;-)


You will find some - if not all -answers here on Gordo´s site:

The Print Guide

there:

How to calculate halftone dot sizes in microns


or in this (maybe a little bit dusty?) booklet for free:

http://old.gis.zcu.cz/studium/krp/screening_technology_eng.pdf

View these points in all there:

- 256 gray levels/supercells

- common limit of usually needed size 20 μ (=0,02mm) for a smallest dot-size for offset-printing technology with plates

- the fact, that the human eye is not able to see more details than a resulution of about 300ppi whithout any help of glasses at a distance of about 25-30 cm between the eye and a print for making sure that all your maths making sense... ;-):

(A formula for determining a usable, optical (!) "Minimum" image resolution is based on the viewing angle of the human eye (constant) at the point where an observer still has two points at a certain distance (variable 1 = sought "image resolution" ") can be differentiated / resolved depending on the viewing distance (variable 2).

This is done using a triangle calculation using the rule of three with an angle function (tangent)

Unfortunately, the constant - the viewing angle - is controversial or cannot be documented uniformly, especially since it is also subject to different eyesight, Wikipedia speaks of a minute of arc (roughly 0.0167 °)

Tangent angle alpha is the sum of the division of the opposite side by the adjacent side, where the opposite side is the image resolution sought and the adjacent side is the viewing distance. So you multiply the tangent of 0.0167 ° (which is then the value 0.0002915) by the viewing distance.)

Best Ulrich

 

[sornchai]

really? Do you mean 300dpi resolution of a platesetter or ppi picture resolution?


sorry, i am not awakend enough in this morning to do the math with inch and mm... ;-)


You will find some - if not all -answers here on Gordo´s site:

The Print Guide

there:

How to calculate halftone dot sizes in microns


or in this (maybe a little bit dusty?) booklet for free:

http://old.gis.zcu.cz/studium/krp/screening_technology_eng.pdf

View these points in all there:

- 256 gray levels/supercells

- common limit of usually needed size 20 μ (=0,002 mm) for a smallest dot-size for offset-printing technology with plates

- the fact, that the human eye is not able to see more details than a resulution of about 300ppi whithout any help of glasses at a distance of about 25-30 cm between the eye and a print for making sure that all your maths making sense... ;-):

(A formula for determining a usable, optical (!) "Minimum" image resolution is based on the viewing angle of the human eye (constant) at the point where an observer still has two points at a certain distance (variable 1 = sought "image resolution" ") can be differentiated / resolved depending on the viewing distance (variable 2).

This is done using a triangle calculation using the rule of three with an angle function (tangent)

Unfortunately, the constant - the viewing angle - is controversial or cannot be documented uniformly, especially since it is also subject to different eyesight, Wikipedia speaks of a minute of arc (roughly 0.0167 °)

Tangent angle alpha is the sum of the division of the opposite side by the adjacent side, where the opposite side is the image resolution sought and the adjacent side is the viewing distance. So you multiply the tangent of 0.0167 ° (which is then the value 0.0002915) by the viewing distance.)

Best Ulrich

300 DPI for images in files and 2,540 DPI Dot per inch for Ctp...Thank you very much

 

[Ulrich]

Hi Sornchai

sorry, (i had made it so often so long so wrong myself) i am hooking up here again:

dpi means dots, a dot means "bitmap" (either yes or no / on or off / black or white) and ppi means pixel. A pixel has a color depth (8 bit means 256 different values - 0-255 - in each channel) that is the sometimes important difference in the terminology...

The thing with the limitation of the resolution of the human eye on the one side and the neccessarity of a minimum resolution (pixel in a picture-file) for a finer screen regarding the nyquist-shannon-theorem (double sampling rate for error-free signal transmission which lead to the in the past teached formula: 150 lpi x 2 = 300 ppi) on the other side is one of the most often but very seldom thought to the end discussed topic in our business, because mostly mixed together and not understood and the last one moreover meanwhile - because the supercell- and precision and irrational screening-technologies (keyword smoothshade also) - not really up to date furthermore...

But please do not ask me to explain the special facts, that would cause the next fairytale ;-)

Even in print sometimes it could make sense to get the mostly possible information and there could be reasons for crossing the line of the ability of the human eye, think of pendants to microfilm for documentary or security marks for an example, but if there is no longer a need for regarding the 255 grey levels, there is for "normal" printing products no need also to "increase" the resolution of a picture up from 300 ppi, if a human can not see a difference between 300 and 450 ppi whithout any help like glasses...

Contradiction is welcome ;-)

Best

Ulrich

 

[Bret Hesler]

... if a human can not see a difference between 300 and 450 ppi whithout any help like glasses...

Contradiction is welcome ;-)

No contradiction here. The last place I worked printed everything with 300LPI hybrid screening (Stochastic below 5% and above 95%) We let 300ppi images go, and found there are very few instances where extra resolution was helpful.

 

[gordo]

Hi. My standard is 175 -200 lpi with 300 dpi. If I want 250 lpi which dpi recommended ?
Another is how big dot sizes at 3% 250 lpi compared to 175 lpi and 200 lpi? Thank you

Here's an uncomplicated way to look at it.
The finer the screen ( increasing lpi) the smaller the details in the image it can render. I.e. Finer screens = more detail.
If the dpi/ppi of the image is less than the lpi of the screen that's being used then you will likely reproduce the pixels that make up the image.
So, for example:
A 72 dpi/ppi image reproduced with a 175 lpi screen means that you will reproduce and see the actual pixels that make up the image. The image will appear "pixelated".
A 175 dpi/ppi image reproduced with a 175 lpi screen means that you will reproduce the image but you will probably not see the actual pixels that make up the image.
A 300 dpi/ppi image reproduced with a 175 lpi screen means that you will reproduce the image and you will not see the actual pixels that make up the image.

This is what that looks like:

75 dpi at 150lpi

100 dpi at 150lpi

150 dpi at 150lpi

300 dpi at 150lpi

Why anyone would use an AM screen at whatever lpi (especially above 175 lpi) is beyond my understanding - unless there was a very specific technical reason.

Your 3% 250 lpi is around 40 microns.

 

[SteveSuffRIT]

At 2400 spi addressability, 1 spot = 0.0004" or 10.6 microns (um).
Screen Lpi = 175, 200, 250
Dot row width = 0.006", 0.005", 0.004"
Spot Cell Grid = 14, 12, 10,
Total spots = 196, 144, 100,
3% Spots = 6, 4, 3,
3% Spot half-width = 2.5, 2, 1.7,
3% Dot width um = 25.4, 20.0, 18.0
3% Dot width " = .0010 .0008, .0007
All 3 Lpi have about the same spots diameter width of about 2 spots because you can't have partial spots, must round up to nearest whole spot.

 

[sornchai]

3% dot width um = 25.4 , 22.3, 18.0

Sorry Sir . Where are come from ?

 

[Ulrich]

3% dot width um = 25.4 , 22.3, 18.0

Sorry Sir . Where are come from ?

250 lpi = 98,43 lpcm (=rounded 100 lpcm)
-> 10 lines per mm -> 1 line per 0,1mm
-> width of 1 line = rastercell is 0,1mm
->1 rastercell has 16x16 squares, because
8Bit (28=2x2x2x2x2x2x2x2)=16x16=256 different tonevalues,
the width of 1 square is (0,1mm rastercell/16=) 0,0063 mm

CMYK range 0-100% = 100 steps but 101 different values of cyan in range 0-255=256 different tonevalues
that leads to:
1% tonevalue cyan means 2,53 squares of/in 256 different tonevalues,
so 3% means (3x 2,53=) 7,59 (rounded up = 8) squares of 0,0063 mm width in a sheme three horizontal and three vertical what means also 3x 0,0063 mm = 0,019 mm (rounded up from 0,0189 mm = 18.0 um...)

next math is how to match this with the resolution of the different devices (1200dpi, 2400 dpi, 2540 dpi, 3000 dpi, 4000 dpi)

2540 dpi means 1000 dots per cm -> 100 dots per mm -> 10 dots per 0,1 mm and as smallest size 1 dot per 0,01 mm -> 10 um, because no good RIP will generate and no device will write only one dot (double signal rate...!) also a 1% tonevalue Cyan will be written with two or three squares and 2x 0,0063 mm = 0,0126 two square width will fit with 0,01mm smallest dot size but would be written with 2 dots with 0,02mm either in heigh or width...

 

[Ulrich]

Sorry, might be this :

...because no good RIP will generate and no device will write only one dot (double signal rate...!)

...is rubbish, fact is that 20um minimum size is a must have for the way from the most plate fabricats to the paper.

 

[sornchai]

250 lpi = 98,43 lpcm (=rounded 100 lpcm)
-> 10 lines per mm -> 1 line per 0,1mm
-> width of 1 line = rastercell is 0,1mm
->1 rastercell has 16x16 squares, because
8Bit (28=2x2x2x2x2x2x2x2)=16x16=256 different tonevalues,
the width of 1 square is (0,1mm rastercell/16=) 0,0063 mm

CMYK range 0-100% = 100 steps but 101 different values of cyan in range 0-255=256 different tonevalues
that leads to:
1% tonevalue cyan means 2,53 squares of/in 256 different tonevalues,
so 3% means (3x 2,53=) 7,59 (rounded up = 8) squares of 0,0063 mm width in a sheme three horizontal and three vertical what means also 3x 0,0063 mm = 0,019 mm (rounded up from 0,0189 mm = 18.0 um...)
View attachment 290368
next math is how to match this with the resolution of the different devices (1200dpi, 2400 dpi, 2540 dpi, 3000 dpi, 4000 dpi)

2540 dpi means 1000 dots per cm -> 100 dots per mm -> 10 dots per 0,1 mm and as smallest size 1 dot per 0,01 mm -> 10 um, because no good RIP will generate and no device will write only one dot (double signal rate...!) also a 1% tonevalue Cyan will be written with two or three squares and 2x 0,0063 mm = 0,0126 two square width will fit with 0,01mm smallest dot size but would be written with 2 dots with 0,02mm either in heigh or width...

Thank you very much

 

[sornchai]

2,53 squares of/in 256 different tonevalues,

Sorry . Can you explain more ?

 

[Ulrich]

256 different tonevalues divided through 101 (cmyk range 0% to 100%=100 steps but 101 values) = 2,53
;-)

You can work just with 100 also for what you are looking for.

And may be it is one step more correct if you take 257 because 16 x 16 = 256, but there is also to add a zero to 256 possible squares(=dots) i think, but than it does not fit so nice to the range 0-255 in photoshop anymore... ;-)

 

[sornchai]

256 different tonevalues divided through 101 (cmyk range 0% to 100%=100 steps but 101 values) = 2,53
;-)

You can work just with 100 also for what you are looking for.

And may be it is one step more correct if you take 257 because 16 x 16 = 256, but there is also to add a zero to 256 possible squares(=dots) i think, but than it does not fit so nice to the range 0-255 in photoshop anymore... ;-)

Thank you sir