| Author | Thread |
|
|
08/19/2013 11:19:14 AM · #26 |
Originally posted by JulietNN:
So back to the horse head, I can see the background and everything surrounding the head, so if I can see it, how is it a DQ??? |
Because if you can still see that stuff your monitor is adjusted way differently than the norm, more contrasty and a different gamma. I see none of that stuff on my monitor until pointed at it: when I look very closely I make out the ghost of the bars and no other detail. It's exactly the same issue you had with your gal: YOU could make out these details in the final image, just barely, but most of us couldn't.
It's worth pointing out, BTW, that there is no such thing as an "illegal tool" in advanced editing, just illegal RESULTS. IN basic, there are some tools we cannot use, but we can do what we want with the tools that ARE allowed.
Basic editing rules are "tool based", advanced editing rules are "results based".
Message edited by author 2013-08-19 11:22:22. |
|
|
|
08/19/2013 11:22:11 AM · #27 |
Exactly Bear, but even I can not see the items in my own photo at all. I would so DQ myself, in fact right now, wouldn't be sure I could bring back those details as it was so long ago I edited it. No my DQ was right and it is a good example of the rule set and why it should have been DQ'd
Edited to add: Even during the challenge I pointed out the barn stalls lol
Another edit: yes not sure why my screen is so different, but when I edit a photo, and I think I am finished, I always tilte my screen up and down to make sure I have not got something that I missed or would look incorrect.
Message edited by author 2013-08-19 11:25:30. |
|
|
|
08/19/2013 12:30:22 PM · #28 |
Originally posted by JulietNN:
... not sure why my screen is so different, but when I edit a photo, and I think I am finished, I always tilte my screen up and down to make sure I have not got something that I missed or would look incorrect. |
Many LCD monitors are very bright "out of the box." This can result in you seeing many things that others will not see. In addition, the fact that you are seeing substantial shifts with viewing angle tells me you probably have a 'TN" type screen. These are typical for some laptops and even some stand-alone displays. The display gamma is what shifts with viewing angle, and it can be very problematic for photo editing.
You want to make sure that your gamma and black point are set properly when the screen is viewed exactly straight-on, and then you need to take care that your viewing angle is correct during editing. This page is a great resource for setting up a display, although it is more than a little technical. |
|
|
|
08/19/2013 12:42:08 PM · #29 |
Yep, no I am totally set up correctly and with the Spyder calibrated as well, and in the voting page on the blacks I can see all of them bar the last two they are both blacks Same with whites. I do work with the moniture straight on, but I do tilt it at the end of my sessions, becuase there have been many mistakes that I can correct as well. I highly recomend it as a step in anyone workflow to be honest. There is a huge vast difference though when I go on the laptop that I got in Thailand, the black and white recently, looking at that on my moniture , I am surprised anyone can see anything but a big blob lol
Here is a perfect example of why to me, tilting the moniture is important. Now as far as I can see, no one mentioned it bar Harvey, if I look at it straight on I can not see it, but then tilting the screen. BAM huge great big black line. |
|
|
|
08/19/2013 01:00:49 PM · #30 |
| Now that one is pretty obvious, but honestly I would never have noticed it during voting. What's interesting about this example is that it's far more prevalent in the red channel than in the other two channels. Which brings up another technique I've often used to investigate editing technique & quality - viewing individual channels. Often, things that are not visible in an RGB image will "jump out" when a single channel is viewed. |
|
|
|
08/19/2013 01:10:06 PM · #31 |
So after all my rambling, what is your opinion on the horse DQ? (if indeed it was because of the background?)
Do you think that under the rule set of move, remove or duplicate any element of your photograph that would change a a typical viewer's description of the photograph it does in fact meet the rules if you can actually see the background but some may not be able to because of calibration or moniture?
I mean for all I know, the original could have a clown in it and we are not privy to see the original one, so there could be another explanation, but it seems that the majority are thinking it is because of the background being burnt out.
((and I am not being argumentative, I just find this interesting as each person see's things differently on their computers)
|
|
|
|
08/19/2013 01:26:58 PM · #32 |
Juliet, you have a pair of those eyes which doesn't miss a thing.
During voting I didn't notice the black line, but afterwards I did.
Now, I presumed it was a wire of some fencing.
But, no matter how I try, I just can't see the barn behind the horses head.
Except on your version, where it's obvious.
I don't see it on my MAC neither on my additional screen. And both are being calibrated on a regular bases with Spyder3.
As a matter of fact: just got the message that it's time again to calibrate and I won't give up until I see that barn on both of my screens.
Message edited by author 2013-08-19 14:00:41. |
|
|
|
08/19/2013 01:27:00 PM · #33 |
| Bottom line, I wouldn't feel comfortable offering any opinion without seeing the original. |
|
|
|
08/19/2013 01:41:58 PM · #34 |
So, Juliet, one thing that worries me about your monitor is that you have extremely limited dynamic range (two stops short, one on either end).
This could be either an issue with your monitor hardware (just not able to display the range it needs to) or it could be an issue with your calibration. Try lowering your contrast severely, and then see if you have a more acceptable level of dynamic range.
..
It might surprise you, but I have NEVER found a better monitor for editing than my Sony Bravia 55" LED 3D TV. Crazy right? But let me tell you why I love it so darn much.
1. Dynamic range is freaking amazing, that bar at the bottom? dead-black on the bottom, bright white at the top, and every shade in between is very distinct.
2. Auto-brightness adjustment - if I open a window, my monitor gets brighter, just exactly bright enough in fact.
3. Not certain, but there also seems to be an auto-white balance adjuster, so if I kick on a really yellow light in my room, the color shifts subtly to give me true rendering of color. Again, not certain of this, but I'm almost positive I'm seeing this happen.
..
The upshot of all this is that you might well consider a nice TV as an option instead of a monitor - mine seems to do much better than almost any monitor I've seen... The only one I've seen that stood up to this was an Apple screen that was made specifically for editing, and cost about 3x as much as my TV (and is about 1/4 of the display area)..
|
|
|
|
08/19/2013 01:56:04 PM · #35 |
| Each block on the gray scale represents something like a 5% increment, not "one stop", which would mean a doubling (or halving) of the neighboring value. |
|
|
|
08/19/2013 02:37:36 PM · #36 |
Originally posted by GeneralE:
Each block on the gray scale represents something like a 5% increment, not "one stop", which would mean a doubling (or halving) of the neighboring value. |
I don't think I like this explanation - I can't say for sure how many 'stops' that chart represents, but let's just have a little gedankenexperiment:
You are right of course, since it's a linear scale, it doesn't really translate easily.
But we can't actually compare percentages against stops, since they're completely unlinked, percentages are a fixed value, there are always 100 to be allocated. Stops are not fixed, and are very much dependant upon the starting point. Start at 0.001% and work up, and there's a hell of a lot more stops than when you start with 2%. This means that the values at the extremes are more important than those in the middle.
From that, I deduce that if you cannot see the black, or the white, end of the scale properly, you are losing the most important data, which is actually MORE than a stop.
Don't believe me? Do the math - assume that 'pure black' is 0.05% white and that white is 0.05% black. .. Now, how many 'stops' are there to that first 5% mark?
0.05% * 2= 0.1%
0.1% * 2 = 0.2%
0.2% * 2 = 0.4%
0.4% * 2 = 0.8%
0.8% * 2 = 1.6%
1.6% * 2 = 3.2%
3.2% * 2 = 6.4%
So, it takes us 7 stops to reach our first 5% ... How many more stops before 100%?
6.4% * 2 = 12.8%
12.8% *2 = ~25%
25 % * 2 = 50%
50 % * 2 = 100%
...
So between that first 5% or so, and the full 100% there are only an additional 4 stops of light.
..
Interestingly, as an aside, this is why it's important to expose to the right, since there is so much more difference in signal available when a stop is the difference between 5,000 photons and 10,000 photons, instead of at the lower register where perhaps the difference is more like a theoretically 20 vs 40 photons. Noise is far more detrimental when the sample size is so low.
My point, although somewhat rambling, is that there are actually an infinite number of 'stops' between 0% and 100%, and that % is a bad way to think of this problem, since the light and dark values are actually more important than the middle values when we discuss this.
By the way, if you're wondering why I chose those numbers, it's because 10-12 stops is a fairly 'normal' amount of dynamic range on a really good digital camera when you stretch the limits using RAW.
So, if you can't see the full scale, there's a good chance that you're really missing out on lots of detail in shadows or in highlights - arguably more than a couple of stops in fact.
Of course, I'll defer to Kirbic if he cares to disagree with my assessment here. :D
Message edited by author 2013-08-19 14:38:57. |
|
|
|
08/19/2013 03:12:23 PM · #37 |
Let's look at it a little differently...
A good LCD monitor has a contrast ratio of about 1000:1, which represents about 10 stops. That, of course, ignores all the "specsmanship" games that monitor manufacturers play to artificially enhance this number. If we are talking about an 8-bit space, then we are talking about mapping only 256 gray levels across the monitor's range, and we might assume that we *should* be able to see all the steps. Wrong. Refer to this page. A gamma of 2.2, which is the target value for computer displays, results in a much lower output for a given input in the shadows, so we do not necessarily see the difference between the first two levels. In fact, on a properly calibrated monitor, it can be almost impossible to tell the difference between the two darkest squares on the DPC "calibration bar."
Edit for clarity. The DPC Calibration bar increases in level by 9 between steps (10 for the last step on the bright end). So darkest step is 0,0,0, next is 9,9,9 and etc.
For most monitors, the demarcation between the last two steps will be nearly invisible, while the step between the second and third darkest bars will be visible.
Message edited by author 2013-08-19 15:25:38. |
|
|
|
08/19/2013 03:23:20 PM · #38 |
You guys are lucky that I have it on good authority that math doesn't suck or this stuff would really bug me ... ;-)
Over 30 years in graphic arts/commercial printing has made it so that I "naturally" think of shades of gray in percentages of black (which is what Photoshop's Info window does when viewed in Grayscale) and colors as CMYK blends.
I have red-green color-blindness, so I really try to rely on the Info window to know what color is displayed, rather than what I see on the monitor, this also from my pre-press training when we used Photoshop to adjust and "separate" photos scanned from prints or slides -- before there were digital cameras (at least as we know them) or (reliably) calibrated monitors.
Message edited by author 2013-08-19 15:24:04. |
|
|
|
08/19/2013 03:27:13 PM · #39 |
Originally posted by GeneralE:
You guys are lucky that I have it on good authority that math doesn't suck |
ROFL!
It's true, math doesn't suck... but it can and will bite you in the arse! |
|
|
|
08/19/2013 03:29:21 PM · #40 |
Originally posted by kirbic:
Originally posted by GeneralE:
You guys are lucky that I have it on good authority that math doesn't suck |
ROFL!
It's true, math doesn't suck... but it can and will bite you in the arse! |
That'd be a good name for a dog.... |
|
|
|
08/19/2013 03:41:21 PM · #41 |
Originally posted by Cory:
Originally posted by kirbic:
Originally posted by GeneralE:
You guys are lucky that I have it on good authority that math doesn't suck |
ROFL!
It's true, math doesn't suck... but it can and will bite you in the arse! |
That'd be a good name for a dog.... |
Standing in the park yelling "Math deoesn't suck! Come here! Math Doesn't Suck!" sounds like an excellent way to get a free three-night stay at your local psychiatric facility ... ;-) |
|
|
|
08/19/2013 03:42:52 PM · #42 |
Originally posted by GeneralE:
Originally posted by Cory:
Originally posted by kirbic:
Originally posted by GeneralE:
You guys are lucky that I have it on good authority that math doesn't suck |
ROFL!
It's true, math doesn't suck... but it can and will bite you in the arse! |
That'd be a good name for a dog.... |
Standing in the park yelling "Math deoesn't suck! Come here! Math Doesn't Suck!" sounds like an excellent way to get a free three-night stay at your local psychiatric facility ... ;-) |
Heh..
I was thinking something more like "Trig" or "Math".... And when people ask why you named your dog that, you can respond "Math doesn't suck, but it can will bite you in the arse."
:D |
|
|
|
08/19/2013 03:47:35 PM · #43 |
| I think my scenario is more likely to be mistaken for a Monty Python piece ... ;-) |
|
|
|
08/19/2013 04:04:26 PM · #44 |
After having recalibrated both my screens I must admit there's a difference.
A slight one, but there's a difference.
Before I saw only a - let's call it a pure black bg.
After calibration I see a trace of the edge of the door (?) next to the 7 or ... metal bars.
The grayscale underneath the entries to vote on shows me the left 2 white ones as 1 block and the right 3 black ones as 1 block.
*sigh* |
|
|
|
08/19/2013 04:09:02 PM · #45 |
Originally posted by Nadine_Vb:
The grayscale underneath the entries to vote on shows me the left 2 white ones as 1 block and the right 3 black ones as 1 block.
*sigh* |
Check your video card settings and monitor settings, what are the contrast values set to? Could be that you're just doubling up on adjustments - in which case, I'd say kill all the adjustments, reset everything to baseline, and start over, trying to adjust your monitor hardware first, then moving on to calibrating via video card if needed.
|
|
|
|
08/19/2013 04:48:07 PM · #46 |
Originally posted by Cory:
Originally posted by Nadine_Vb:
The grayscale underneath the entries to vote on shows me the left 2 white ones as 1 block and the right 3 black ones as 1 block.
*sigh* |
Check your video card settings and monitor settings, what are the contrast values set to? Could be that you're just doubling up on adjustments - in which case, I'd say kill all the adjustments, reset everything to baseline, and start over, trying to adjust your monitor hardware first, then moving on to calibrating via video card if needed. |
Hehe, thanks Cory :)
Looks a bit better now on my separate screen:
The 2 whites are fully different blocks. Only the 2 last black blocks are still one.
Gamma tested and ok - resolution of screen and card is optimal - contrast and brightness have been tested and are ok...
Still those 2 blacks...
Tomorrow I'll handle the Macbook
|
|
|
|
08/19/2013 05:00:57 PM · #47 |
Originally posted by Nadine_Vb:
Still those 2 blacks... |
If you can readily tell the difference between those last two blocks, you screen is probably too bright. They will normally be indistinguishable, or very nearly so.
ETA: Try enlarging the "strip" of blocks to, say, 200% and displaying it on a dark background. You will probably see the difference between the last two blocks. At smaller size, or against a light background, not so much.
Message edited by author 2013-08-19 17:04:19.
|
|
|
|
08/19/2013 05:04:12 PM · #48 |
Originally posted by kirbic:
Originally posted by Nadine_Vb:
Still those 2 blacks... |
If you can readily tell the difference between those last two blocks, you screen is probably too bright. They will normally be indistinguishable, or very nearly so. |
The 2 blacks ARE indistinguishable...
Which means, ladies and gentlemen, that this girl succeeded in getting at least 1 screen calibrated as it should :)
No more touching Nadine...keep your hands of and leave it as it is. |
|
|
|
08/19/2013 08:42:38 PM · #49 |
Lol Nadine!
It will be interesting to learn if the photo is indeed a DQ from the SC if it was the background that was the reason for the DQ. I await , with bated breath! =) |
|
|
|
08/20/2013 10:13:01 AM · #50 |
I have been thinking about this. If you take a darkly exposed shot and ramp it through photoshop shadow/highlight, you can use crime scene investigation techniques to discern detail in the shadows of nearly any photo.
Does that mean they are left out because the viewers monitor is too dark? What if the shot is underexposed and the processing is used to enhance this? This is certainly not the same as using a selective "burning" technique to remove items or a background. no?
Message edited by author 2013-08-20 10:13:29. |
|
