One for J. Dillon

[Greg H]

I know you like ship paintings
Here's one by a local artist....

http://www.imagestation.com/picture/sraid148/p00fd024610e264b602b4e71c0e666886/f60adf01.jpg

[J. Dillon]

Thanks Greg H

Looks like a belly full of wind but some reason the back ground is all black. :( Maybe my screen :confused: but other images come in OK.

JD

[Greg H]

Hmmmm, should be misty blue grey to dark grey, the photograph doesn't quite get the light right. Looks better in person. That's Constellation off of the African coast.

[phiil]

What is it about marine artists, that they always seem to paint pennants streaming aftward, regardlass of the wind direction? (See mast tops)
Phil

[Mike Field]

.
Looks the same here too, Jack. But I think this is more like what it's supposed to be like --

http://www.woodenboatfittings.com.au/public/test-pic.jpg
.

Phiil -- not always --

http://www.woodenboatfittings.com.au/public/test-pic2.jpg

But very often, I'll grant you.
.

[Donn]

I think it has to do with monitors and display settings. On my 21" Trinitron, it's almost black, and I can barely see the ship. On my 12" LCD, it looks perfect. Mike's edit looks over-gamma'd on the LCD. I'll have to look at it on the Trinitron tomorrow.

[Greg H]

So everyones monitor is different, and no one knows what anything really looks like smile.gif

I looked at the photo the painting was based on... the pennants are flying at that angle, but have a dip in them and are not streaming straight out, I think the pic misses a lot of the nuance in the light values because it is done in glazes.

[Wild Wassa]

Donn, What does 'over gamma'd' mean? ... please tell me in detail ... as a person who studied both sensitometry and densitometry I don't know 'over gamma'd', D log E, I understand, slope and average gradient I understand. I'm totally confused. Where on a characteristic curve should I be looking Donn?

Gamma = a measure of the slope of the straight line of a characteristic curve?

Do you mean 'log brightness range' or 'subject tone separation'? ... or that there is poor tone separation in the shadows? ... and a loss of shadow detail.

Warren.

[ 12-05-2004, 12:26 PM: Message edited by: Wild Wassa ]

[Donn]

To me, Gamma is an ornamental grass, Tripsacum dactyloides.

Gamma is also a setting in many low-rent photo-manipulation programs, such as PolyView. Here's what they say about it, in their help files:

"The appearance of an image displayed on a computer screen, or any other media, is dependent on a complicated interaction between the way humans perceive colors and intensities and the way the particular media translates its input signals into light. Although a host of other factors contribute, these two factors dominate.

Human visual perception is sensitive to the ratios between light intensities. Consider a gray light intensity scale where 0 indicates a black image and 100 represents a white image. The three intensities 25, 50, and 75 will not be seen by the eye as linearly increasing in intensity, as you might expect. Instead, the intensity of 50 is twice as bright as 25, while 75 is only 1.5 times as bright as 50.

The rendering of an image on a computer display is a highly non-linear problem which is potentially different for every system. The intensity of a pixel depends on the characteristics of the screen phosphor and the characteristics of the electron beam that is exciting the phosphor to radiate light. Add to this the intensity response of human vision and the problem of correctly displaying the colors and brightness of an image becomes quite a challenge.

Gamma attempts to compensate for these effects by pre-adjusting the intensities of the pixels in an image according to the equation:

NewIntensity = Intensity**(1/Gamma)
where ** represents the exponentiation operation. The overall effect of this operation is that for Gamma settings greater than 1.0 there is a brightness boost in an image which tends to decrease as the intensity of a pixel increases. That is, the dark areas of an image increase more in brightness than do the light areas. Choosing the right gamma factor can result in the realistic display of an image.

The wide variety of image storage formats available treat gamma compensation with an equivalently wide variety of methods. Some formats, like Portable Network Graphics (PNG) format, offer the capability of storing the gamma compensation for an image along with the bits of the image. Others, like JPEG File Interchange Format (JFIF), simply specify that the images be stored with a gamma factor of 1.0 - leaving it up to the image viewers to appropriately compensate the image for display. In both cases, the specification and compensation of gamma is widely ignored. Many PNG images have no or incorrect gamma compensation, and many JPEG, GIF, and other formats are corrected to something other than 1.0.

Faced with these disparities in handling gamma, PolyView assumes that unless it is specified differently in the input file, an input image has been stored with a gamma factor of 1.0. For images where this assumption is invalid, the Gamma Adjustment dialog provides interactive adjustment of the gamma factor for an image. The dialog also allows the specification of a global gamma to be used to compensate all images as they are opened and displayed."

PS..Mike's edit looks even more over-gamma'd on the Trinitron.

[ 12-05-2004, 11:42 AM: Message edited by: Donn ]

[Wild Wassa]

Originally posted by Donn:
"The three intensities 25, 50, and 75 will not be seen by the eye as linearly increasing in intensity, as you might expect".

I don't expect ... I know. I didn't study this for four years not to know basic things.

Thankyou anyway.

Warren.

[Donn]

You're editing someone who wrote a help file for a 10 year old USD$20 piece of shareware. I imagine he/she assumed the lowest common denominator user, who really doesn't need 4 years of university to crop and adjust a snap of little Johnny before he/she emails it to Granny.

[Wild Wassa]

Well you posted it as your only bank of knowledge ... instructions for 10 year olds ... wow.

Warren.

[Donn]

Originally posted by Wild Wassa:
Well you posted it as your only bank of knowledge ... instructions for 10 year olds ... wow.

Warren.Don't like it? Don't bother me with your questions again.

[DerekW]

C'mon, play nice.

Charles Poynton is good on this stuff, IMO.

More about Gamma (http://www.poynton.com/GammaFAQ.html) than you really wanted to know...

Derek

[J. Dillon]

To get back to the painting. smile.gif

Phil, in this rendering of the USS Constellation she appears to be close hauled, the wind maybe fwd. of the beam. The pennants aloft at the mast head are streaming in the right direction. ;)

Mike, Thanks for "enlightening" the image. The viewer can see what is going on. :cool: BTW is that the Swedish flag on the stern of the gaffer ?

JD

[Mike Field]

.
For those who are interested, I gamma'd the picture as posted by about 2.8, added contrast, neutralised the colour balance, deepened the saturation, clarified the entire picture, increased the sharpness by a small amount, and levelled the horizon.

Without seeing the original, clearly I can't be sure how close I got to the effect the artist was after -- even on my own computer. But since Jack and I had similar problems with the picture as posted, we now presumably have good approximations of what it's supposed to look like. This is really all I'm concerned about.

On the other hand, since two of us had experienced the difficulty and it has now been resolved, you could conceivably mutliply this by a thousand to approximate the viewing fraternity who benefitted from my little bit of legerdemain....
_______

Jack -- sure looks like it (although the proportions mightn't be quite right.) And while the dills haven't used your robands, did you notice that they've laced the mainsail to the mast correctly? smile.gif
.