| Author | Thread |
|
|
07/14/2004 04:20:49 AM · #1 |
Hello all.
right, now, my advertisement shot:
came a respectable 83rd. i'm actually reasonably happy with that. what i do find interesting though is the fact that it chalked up 6 tens and 8 ones, with a pretty even spread each side of five.
taking on board the comments from my fellow voters, the score appeared to have depended on whether the said voter could actually see the image or not.
i suspect that my screen was set up incorrectly - but the tried and tested method of being able to differentiate between all the greys at the foot of each vote page kind of let me down because i CAN differentiate between the greys. perhaps an alternative might be useful - possibly something along the lines of my challenge entry? you know, something a bit more contrasty? thereby we can do eachother justice when it comes to voting particularly brightness susceptible shots?
any thoughts? or am i bitching unnecessarily? |
|
|
|
07/14/2004 05:38:35 AM · #2 |
You are not bitching unnecessarily. :D
You did not mention how your monitor is calibrated, but the bar at the bottom is of little value in determining if you are viewing the image the way the photographer wanted it viewed -- or from your perspective, that the voters are seeing the image the way you wanted them to see it.
Gamma is the biggest culprit in this problem, as an image edited on a monitor set at one gamma will appear lighter or darker (depending on which way the shift in gamma goes) on another monitor set to a different gamma -- even if all the bars at the bottom of the screen appear exactly the same on both monitors. An image edited on a monitor with a lower gamma will appear darker when views on a monitor set to a higher gamma.
The cross-platform agreed upon standard is a 2.2 gamma, if you do not know what gamma you monitor is set at, it needs calibrated. This is the easiest to follow sight I have seen that details the steps needed to properly calibrate your monitor -- with one exception. There is no mention of the temperature to set your monitor at. The temperature setting compensates for the various temperatures of the possible ambient lighting in the room the monitor is in -- it is very much like the white balance on cameras. The temperature the monitor is set at has a significant effect on the calibration, so it needs to be set first. Set your desktop to middle grey, or perhaps white, and adjust the temperature until it looks grey or white and not some other color. This is not the best, and is prone to error, but until I find a better way it is the best I have to offer. Then just follow the steps on the site listed above.
When you have finished, you will have a monitor you know is displaying the images, both while editing and voting, in a standard manner. And most importantly, you will know that those who have calibrated their monitors to the standard will be viewing your images the way you wanted them to be seen.
David
|
|
|
|
07/14/2004 01:27:18 PM · #3 |
Although what Britannica describes falls under the umbrella definition of calibration, it is really far from meeting the needs of anyone looking for a color managed workflow. The method described will get you in the ballpark, but that's about it: There are many variables, not the least of which is your own visual system (it adapts quickly to ambient light and whites).
That being said, it seems clear your goals are not CMW. Rather, you want assurance that the DPC voters see roughly what you see. The ballpark calibration method will give your very average results and unreliably so. |
|
|
|
07/16/2004 07:00:54 PM · #4 |
I checked your photo on three different monitors (two LCD, one CRT) and it looks gorgeous on all three. I also tried a few different gamma settings and it still looked good. I really think it's a shame when photos this good get voted down because some people simply don't see half the image. It is quite clear from the comments you got that a number of DPC'ers are using monitors that are drastically out of calibration.
How about requiring people to calibrate their monitors, at least roughly, in order to become members? Am I remembering correctly, isn't there already a "warm body" test in the registration process where you have to type the characters that you see in some funky font in a small image into a form? Well, how about making these fairly dark characters on a dark background, such that you can't see them unless your monitor is at least in the ball-park of good calibration?
I just realized that this idea belongs in the "site suggestions" threads, so I'll copy it over... |
|
|
|
07/17/2004 01:14:54 PM · #5 |
Okay, I was one of the ones who made a 'too dark" comment on that photo. I did see the other lettering but I felt if I had to squint to make it out, it was too dark. Now, I clicked on the link provideded for calibrarting one's monitor and I can see the difference between the 9 and 10 and the 95 and 100 but on first glance they would appear the same. After looking for a second, I saw the difference and now I can't not see the difference. My question is, how much of a difference should I be seeing? Is it a very subtle gradiation or a really obvious one? Logic tells me that it would be subtle. I have my monitor set to the recommendations for Apple monitors given an earlier DPC thread. I have my gamma settings at 1.80 which is what I was told was standard for Mac. Should I reset it for 2.2? White point is set at 6500 and ambient lighting is set to low because I live in a basement apartment and usually my office is only lit by the lighting from the adjacent kitchen and a small wall lamp several feet behind me.
As I said, I can see the difference between those areas but it is very subtle.
As for the gamma settings check, the only thing I can say is I can still see the difference between the smooth patch and the striped area but barely. I can't tell if the smooth patch is lighter or darker, what I notice is the perimeter of the smooth rectangle. If I squint it blurs into one.
Message edited by author 2004-07-17 13:24:25.
|
|
|
|
07/17/2004 02:05:36 PM · #6 |
Originally posted by melismatica:
Okay, I was one of the ones who made a 'too dark" comment on that photo. I did see the other lettering but I felt if I had to squint to make it out, it was too dark. Now, I clicked on the link provideded for calibrarting one's monitor and I can see the difference between the 9 and 10 and the 95 and 100 but on first glance they would appear the same. After looking for a second, I saw the difference and now I can't not see the difference. My question is, how much of a difference should I be seeing? Is it a very subtle gradiation or a really obvious one? Logic tells me that it would be subtle. I have my monitor set to the recommendations for Apple monitors given an earlier DPC thread. I have my gamma settings at 1.80 which is what I was told was standard for Mac. Should I reset it for 2.2? White point is set at 6500 and ambient lighting is set to low because I live in a basement apartment and usually my office is only lit by the lighting from the adjacent kitchen and a small wall lamp several feet behind me.
As I said, I can see the difference between those areas but it is very subtle.
As for the gamma settings check, the only thing I can say is I can still see the difference between the smooth patch and the striped area but barely. I can't tell if the smooth patch is lighter or darker, what I notice is the perimeter of the smooth rectangle. If I squint it blurs into one. |
The difference between the last two squares should be as obvious as the difference between any other adjacent squares.
For setting the gamma, back up or squint enough you can not see the stripes. Backing up is best as squinting affects the amount of light getting too your eye. If, once you can not see the stripes, the shade appears the same -- your done.
What gamma? Yes the 1.8 is the default gamma for a Mac, and using it will produce images that look good at that gamma. However, there are a few things you should consider before deciding to stick to the default.
* sRGB is the standard for the web, and as been nearly universally adopted as a cross-platform default. (It is the most likely color-space any device will support.)
* sRGB is based on a 2.2 gamma. This is because at approximately 2.2 gamma the RGB color space has the greatest dynamic range. A reduction in gamma reduces the dynamic range (loosing detail). (The dynamic range is basicly how many stops it has; a greater dynamic range can display more stops than a lesser dynamic range can.)
* Most modern digital cameras encode the image files in sRGB. (RAW files excluded as they have not been encoded into a colorspace yet.)
My camera is not capable of producing RAW images, so the images it produces are encoded with the sRGB colorspace (at 2.2 gamma) and the primary means of displaying a digital image for me is on the internet (sRGB, at 2.2 gamma). It just makes good sense for me to remain at 2.2 gamma for as much of my work-flow as possible. The only thing I do with images that require a change at all is printing, and that requires a change in colorspace.
David
|
|
|
|
07/17/2004 02:30:15 PM · #7 |
I used PS 4 Color Settings to switch to sRGB profile which raised the gamma to 2.2. There is no marked difference in the black and white, which really seemed fine to begin with (I could see subtle differences from one end to the other) but now the gamma patch test is too bright at 10 feet away. The difference is quite obvious whereas at my previous settings it was barely noticible. When I use the sliding scale on the Monitor Gamma link the best setting appears to be 1.8.
I'm rarely one to give 'too dark' comments. The only other time I can think of was for a photo in the Team Sports Without Players which if memory serves placed last because everyone thought it was too dark.
So, should I stick with the sRGB profile or go back to my Mac profile? I haven't checked the Guinness photo yet. I have to reiterate in case I didn't mention it before, that I did see the first few letters of the label but that I thought it was too dark if I couldn't make it out immediately (since this was for an ad).
|
|
|
|
07/17/2004 02:35:39 PM · #8 |
I looked at the Guinness photo again and it doesn't look that much different to me. Of course, I can't do a side by side comparison. I have only my memory to rely on. Here is what I'm seeing:
The G is visible but appears rather grainy and the color isn't immediately apparent. The U is darker with the right side nearly invisible. The I is visible at the bottom but rather dark and nearly invisible at the top. The first N is very visible and the gold color and a bit of sparkle are becoming more apparent. The second N is very visible and is bisected vertically with a rather hot patch. The E appears pretty much the same as the first N. I am I seeing this correctly?
|
|
|
|
07/25/2004 04:02:42 AM · #9 |
Using your eyes to "calibrate" your monitor?
White's Illusion
Checker Shadow
Another interesting one:
Bulge Illusion |
|
|
|
07/25/2004 05:46:48 AM · #10 |
What do images specifically designed to fool the eye have to do with calibrating a monitor?
in each of those images, move your cursor over them and the portion of the image designed to fool the eye is removed -- and what do you know, the eye is no longer fooled. The eye is then capable of seeing the shades and such as they are.
Images used during calibration are not designed to fool the eye; but are instead designed to aid the eye in distinguishing small differences.
David
|
|
|
|
07/25/2004 06:19:10 AM · #11 |
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Prints! -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2024 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 04/16/2024 12:14:59 AM EDT.
