Ad
Ad
Ad
Pages: « 1 2 3 [4]   Bottom of Page
Print
Author Topic: I'm gettin' One of those "Wide Gamut" LCDs  (Read 30048 times)
Ethan_Hansen
Full Member
***
Offline Offline

Posts: 114


WWW
« Reply #60 on: December 21, 2009, 04:39:54 PM »
ReplyReply

Quote from: bjanes
Mark, if you look at the white paper by Karl Lang on the Adobe site, the best photographic or inkjet prints can have a CR of 275:1, but more typically 250:1. Of course, matt papers will have a lower DMax. My post was more in the order of a thought experiment and I haven't done such a calibration as I can not adjust the black point of my monitor. I would think that a monitor that displays an accurate rendering of the image at a CR of 288:1 would be adequate for soft proofing, but I would like to hear from Ethan Hansen or others who have actual experience in this area. Until someone convinces me otherwise, I think that a high CR is an advantage for a monitor.

Print contrast ratios depend on the ink set, printer (and driver), and paper used. Typical values for pigment ink printers range from 200:1 for high gloss stocks to 150:1 for lustre, semi-gloss, and satin, down to 40:1 and below for fine art papers. A few specialized stocks - usually polyester surfaces made for point-of-sale displays - can hit 400:1 or above. Dye ink printers achieve slightly higher DMax and, therefore, print contrast values than do pigment inks.

Silver-halide printers offer print contrast ranging between ~100:1 to 50:1 on standard surfaces depending on how much of the manufacturer's color management software is enabled. Specialized papers such as metallic stocks are usually 40:1 or below because of the less-than-white paper surface.

In actual use, I have not found the need to reduce monitor contrast in the interest of print matching unless working with prints made on very low contrast paper (examples below). Instead, use Photoshop's soft proofing tools and train your eyes to interpret what they show. Simulate Black Ink helps determine if your shadow details wil show up in print. If not, adjust. The Paper Color simulation does the same for highlights.

As Andrew mentioned above, the simulations are not exact. Your eyes "white balance" themselves to the brightest element in your central field of view. These include palettes, docks, taskbars, menus, or other UI elements unless you view the image in full screen mode or use a multiple monitor setup. Without distractions, the matching between print and screen can be very good. Do not make major tonal adjustments to the image with the simulation enabled until you have the chance to compare a variety of prints to what Photoshop displays. There is no point in trying to edit out a simulation artifact.

If, however, you are printing to a comparatively low contrast surface, reducing your monitor contrast can assist in previewing what you will see in print. You definitely do not want to match the print contrast. Doing so will make your monitor impossible to work on. When using a paper with maximum contrast in the 40:1 range (Epson's Hot and Cold press fine art papers, most satin or glossy canvases, Hahnemuhle Photo Rags or Bamboo, etc.) reducing your monitor contrast to 150:1 to 200:1 is useful. Standard Matte papers or matte canvases (CR of 30-35:1) the lower end of the range is more useful. For jobs destined for newsprint (CR of 6:1 if you're lucky), I'll go to a monitor CR of 100:1. Lower and you simply can not see enough to work. Don't do this on a laptop - even the best screens pose enough problems already.

What printer you use is also a factor. For example, Epson's desktop printers (e.g. 2400, 2880) still do not use the same quality ink linearization and limiting algorithms as do their wide format brethren. Our data, Epson's canned profiles, and even Andrew & Co's PixelGenius Exhibition Fiber profiles all show the desktop models producing ~20% lower print contrast on a given paper than do the wide format printers. A quick comparison of grayscale step wedges shows why. Poor linearization on the desktop printers, decent to very good linearization with the wide formats.

Semi-related note to Andrew: Somebody goofed with the PixelGenius profile labeling. The 7800 and 9800 profiles are for the same printer and the 7880 and 9880 profiles are identical as well. The other profiles all look reasonable.
Logged

digitaldog
Sr. Member
****
Offline Offline

Posts: 9188



WWW
« Reply #61 on: December 21, 2009, 05:03:54 PM »
ReplyReply

Quote from: Ethan_Hansen
Semi-related note to Andrew: Somebody goofed with the PixelGenius profile labeling. The 7800 and 9800 profiles are for the same printer and the 7880 and 9880 profiles are identical as well. The other profiles all look reasonable.

Yes, you can use the 7800 on the 9800 or vise versa but we were asked to supply a single named profile for each from the client.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Ethan_Hansen
Full Member
***
Offline Offline

Posts: 114


WWW
« Reply #62 on: December 21, 2009, 05:46:23 PM »
ReplyReply

Quote from: digitaldog
Yes, you can use the 7800 on the 9800 or vise versa but we were asked to supply a single named profile for each from the client.

Weird that they would have you make profiles for each individual printer in the lineup except for the 7xxx/9xxx pairs. Ah well, I have long since given up trying to make sense of Epson's decisions...
Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #63 on: December 22, 2009, 12:28:47 AM »
ReplyReply

Quote from: Ethan_Hansen
Print contrast ratios depend on the ink set, printer (and driver), and paper used. Typical values for pigment ink printers range from 200:1 for high gloss stocks to 150:1 for lustre, semi-gloss, and satin, down to 40:1 and below for fine art papers. A few specialized stocks - usually polyester surfaces made for point-of-sale displays - can hit 400:1 or above. Dye ink printers achieve slightly higher DMax and, therefore, print contrast values than do pigment inks.

Silver-halide printers offer print contrast ranging between ~100:1 to 50:1 on standard surfaces depending on how much of the manufacturer's color management software is enabled. Specialized papers such as metallic stocks are usually 40:1 or below because of the less-than-white paper surface.

Interesting that Andrew did not contest these figures. If the figures are true, then I'm seriously wondering if print media is the best method of displaying one's artistic efforts.

I mentioned earlier that I recently took delivery of a 65" Panasonic 12th generation Plasma display. My new house isn't ready for it, but this TV is currently in very short supply, so I bought the last one available in Australia (as far as I know). They're even out of stock of these TVs in New York, apparently.

This plasma set boasts a native contrast ratio of 40,000:1, a 'dynamic' contrast ratio (whatever that means) of 'greater than' 2,000,000:1, and 6144 gradations per channel amounting to 231 billion possible colors.

There's a particular moonlight shot I took a while ago with my 5D, that is problematic. The shadows are rather noisy and the prints I've made so far do not show the full detail in the darkest shadows, just as my calibrated monitor in proof mode does also not show detail in those deepest shadows.

In other words, there's no mismatch between print and monitor with proof colors enabled. Both the print and monitor lack that 'wow' factor. With proof colors unticked, the image on the monitor looks much better than the print.

I cropped this image to 16:9 ratio, converted to sRGB, reduced the file size to 6mb, transferred the image to an SD card and displayed it on my new plasma TV (which has a slot to accept SD cards). It was evening and I turned off all lights.

Wow! Wow! Wow! Phwoar! I've never seen such delicious and detailed blacks before. Just amazing! If I wasn't so modest I'd claim this is the best photograph I've ever seen. Sorry that most of you will not be able to appreciate it in it's full glory as I can.

Here's the full 2 megapixel HD shot. If you don't like it, tough!

[attachment=18824:01_3723_...filtered.jpg]
« Last Edit: December 22, 2009, 12:57:22 AM by Ray » Logged
Ethan_Hansen
Full Member
***
Offline Offline

Posts: 114


WWW
« Reply #64 on: December 22, 2009, 02:31:03 AM »
ReplyReply

Quote from: Ray
Interesting that Andrew did not contest these figures. If the figures are true, then I'm seriously wondering if print media is the best method of displaying one's artistic efforts.

Why would he contest them? If you don't believe our numbers (after all we have only built about 25000 profiles for silver-halide printers, many thousands of inkjet profiles, and even profiled 250 individual Epson 7880 printers), you can check the canned profiles provided by pritner and media vendors. I mentioned several Epson papers -- let's check them. Epson has profiles for their newer papers available for download without requiring you to install a printer driver. Most of these profiles were built with MonacoProfiler; using the profile black and white point values will get you very close to the actual print contrast. MP uses the measured paper white and RGB = (0, 0, 0) values for the profile wtpt and bkpt data.

Using the 4880 as an example -- I chose the 2880 dpi output mode for consistency (except for canvas where 1440 is the higfhest dpi) -- we get:

Code:
               Paper           Print Contrast
     ----------------------------------
     Hot Press Bright         45:1
     Hot Press Natural        43:1
     Cold Press Bright        44:1
     Cold Press Natural       42:1
     Premium Canvas Matte     36:1
     Premium Canvas Satin     41:1
     Exhibition Fiber        218:1

As to whether a print is the best output medium, this is a debate that has raged since the advent of photography. In the pre-digital era, many a photographer's head shook n dismay when comparing a dull print to the glory of a slide viewed with a loupe on a light box. Prints make sharing easier, however.

Quote from: Ray
This plasma set boasts a native contrast ratio of 40,000:1, a 'dynamic' contrast ratio (whatever that means) of 'greater than' 2,000,000:1, and 6144 gradations per channel amounting to 231 billion possible colors.
<<SNIP>>
Wow! Wow! Wow! Phwoar! I've never seen such delicious and detailed blacks before. Just amazing! If I wasn't so modest I'd claim this is the best photograph I've ever seen. Sorry that most of you will not be able to appreciate it in it's full glory as I can.

Digital signage is becoming more common. Mainly because one is no longer limited to stills, and even they can be switched regularly. Improved image pop does not hurt either.

Edit: I exaggerated in the numbers above. We have only built 24097 profiles for silver-halide printers.
Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #65 on: December 22, 2009, 03:29:31 AM »
ReplyReply

Quote from: Ethan_Hansen
As to whether a print is the best output medium, this is a debate that has raged since the advent of photography. In the pre-digital era, many a photographer's head shook n dismay when comparing a dull print to the glory of a slide viewed with a loupe on a light box. Prints make sharing easier, however.


Okay! I take your point. It's just that your figures for print CR seemed even lower than Andrew's.

The debate might have been raging for ages as to print versus slide show, but there's now a new display on the market which is far better (but not necessarily bigger) than either a slide show or a digital projector.

It's the Panasonic plasma display with an SD card slot, which 'knocks the socks' off everything else, including both print and projected slide show.

Slide shows were always a PITA. There was a risk of the cartridge jamming, the bulb fusing, and the inconvenience of the set-up procedures of screen and projector.

That's all now gone. As regard camera bodies, I need nothing more than a D3s. But lenses are a problem. (In fact, I possibly need nothing more than an upgrade to the Canon 3mp D30 with even lower high-ISO noise then the Nikon D3s, for personal requirements, although the higher resolution image may be necessary for sales.)
« Last Edit: December 22, 2009, 03:31:51 AM by Ray » Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2822



« Reply #66 on: December 22, 2009, 08:00:07 AM »
ReplyReply

Quote from: Ray
I cropped this image to 16:9 ratio, converted to sRGB, reduced the file size to 6mb, transferred the image to an SD card and displayed it on my new plasma TV (which has a slot to accept SD cards). It was evening and I turned off all lights.

Wow! Wow! Wow! Phwoar! I've never seen such delicious and detailed blacks before. Just amazing! If I wasn't so modest I'd claim this is the best photograph I've ever seen. Sorry that most of you will not be able to appreciate it in it's full glory as I can.

Here's the full 2 megapixel HD shot. If you don't like it, tough!

[attachment=18824:01_3723_...filtered.jpg]

Ray,

That is a striking and beautiful shot. Is the moon in this image taken with a telephoto lens real? The rays of light emanating from the moon look real.

Bill
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #67 on: December 23, 2009, 05:03:12 AM »
ReplyReply

Quote from: bjanes
Ray,

That is a striking and beautiful shot. Is the moon in this image taken with a telephoto lens real? The rays of light emanating from the moon look real.

Bill

Bill,
I wondered if someone would notice this   . As far as I remember, that's how the moon looked when I took the shot. But the dynamic range (or SBR) of this scene was so high and the exposures so long that bracketing of exposures was not successful for merging to HDR.

This particular shot was 30 seconds at F10 and ISO 800. The moon just wouldn't remain still for me, so I took a separate shot of the moon on another night, selected it and copied and pasted into the original scene.

You see I'm just not willing to misrepresent what I saw just because nature will not keep still, or because my camera has insufficient DR   .
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6972


WWW
« Reply #68 on: December 23, 2009, 09:22:04 AM »
ReplyReply

I noticed it too, have had the same kind of problem, knew you had most probably layered it in, and I have absolutely no qualms about doing stuff like that - it is simply overcoming a technical limitation in order to portray the scene as the photographer saw it. The only requirement is that it be done unobtrusively so that it would be "un-noticed" as much as possible. In this regard, I think a bit more smoothing around the circle (so that it all blends more seamlessly) would be a nice finishing touch.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
fike
Sr. Member
****
Offline Offline

Posts: 1373


Hiker Photographer


WWW
« Reply #69 on: December 23, 2009, 12:46:38 PM »
ReplyReply

I made a version of this shot using the same technique.  I spent a lot of time with HDR and blending layers and such, but those techniques just didn't work. The easiest/best thing to do was take another moon shot and scale it to fit on top.  I love your shot, but I just couldn't publicly present my shot with the layered moon effect.  It didn't fit with my photographic style.  

I don't think the noise in the shadows is bad.  It makes it feel a bit painerly.  Otherwise, I think if you darkened that section, it would make it easier to blend the noise away.  I presume that the noise comes from boosting the shadows a lot.  


Trailpixie.net: Singing Moon
« Last Edit: December 23, 2009, 12:47:43 PM by fike » Logged

Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #70 on: December 23, 2009, 01:01:56 PM »
ReplyReply

Quote from: MarkDS
I noticed it too, have had the same kind of problem, knew you had most probably layered it in, and I have absolutely no qualms about doing stuff like that - it is simply overcoming a technical limitation in order to portray the scene as the photographer saw it. The only requirement is that it be done unobtrusively so that it would be "un-noticed" as much as possible. In this regard, I think a bit more smoothing around the circle (so that it all blends more seamlessly) would be a nice finishing touch.

Good point, Mark. As it stands, I think the moon is perhaps a little too surreal. I think I'll try reworking it yet again, but so far this version seemed the best out of several attempts, which is perhaps a testament to my poor skills in Photoshop.

The extent of the brightness range in this scene did surprise me. A correct exposure for the moon is pretty close to a daylight exposure, whereas much of the rest of the scene is plain night.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #71 on: December 23, 2009, 01:20:45 PM »
ReplyReply

Quote from: fike
I made a version of this shot using the same technique.  I spent a lot of time with HDR and blending layers and such, but those techniques just didn't work. The easiest/best thing to do was take another moon shot and scale it to fit on top.  I love your shot, but I just couldn't publicly present my shot with the layered moon effect.  It didn't fit with my photographic style.  

I don't think the noise in the shadows is bad.  It makes it feel a bit painerly.  Otherwise, I think if you darkened that section, it would make it easier to blend the noise away.  I presume that the noise comes from boosting the shadows a lot.

Marc,
That's pretty much how the moon looked in my shot. I could darken the shadows, but this was a moonlit night and those shadows were not completely devoid of detail.

I have noticed on my plasma display that the noise in the shadows seems less prominent, I presume because the true blacks (ie. areas devoid of detail) really are blacker than they appear on my calibrated monitor.
Logged
fike
Sr. Member
****
Offline Offline

Posts: 1373


Hiker Photographer


WWW
« Reply #72 on: December 23, 2009, 01:40:00 PM »
ReplyReply

Quote from: Ray
Marc,
That's pretty much how the moon looked in my shot. I could darken the shadows, but this was a moonlit night and those shadows were not completely devoid of detail.

I have noticed on my plasma display that the noise in the shadows seems less prominent, I presume because the true blacks (ie. areas devoid of detail) really are blacker than they appear on my calibrated monitor.

Try making an hue and saturation adjustment layer and desaturating the noisy areas in the red, magenta, and blue channels.  Mask the adjustment and paint it in with a soft edged brush with a low transparency.  That will take some of the edge off the chroma noise without eliminating the detail.

As for toning down the moon, take the layer that the moon is in and set the transparency to something like 80%.  That will blend it a bit better by allowing some of the bright background layer to slip through.
« Last Edit: December 23, 2009, 01:47:54 PM by fike » Logged

Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #73 on: December 23, 2009, 02:15:22 PM »
ReplyReply

Quote from: fike
Try making an hue and saturation adjustment layer and desaturating the noisy areas in the red, magenta, and blue channels.  Mask the adjustment and paint it in with a soft edged brush with a low transparency.  That will take some of the edge off the chroma noise without eliminating the detail.

As for toning down the moon, take the layer that the moon is in and set the transparency to something like 80%.  That will blend it a bit better by allowing some of the bright background layer to slip through.


Thanks for the advice. I'll try that when I have the time, but I don't think I could ever make a print that would look as splendid as this shot looks on my plasma TV, nor as big using my 24" wide Epson 7600. I think my printer might now be relegated to the production of long, stitched panoramas which tend to look less impressive on a 16:9 screen with major bars top and bottom.
Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #74 on: July 11, 2010, 01:41:41 PM »
ReplyReply

Quote from: ErikKaffehr
Hi,

One thing to consider is that your screen should not be too bright. If you measure a white area on your screen and a white paper on your wall they should have similar brightness. A screen that is too bright cause you to print dark.

Best regards
Erik

One thing to keep in mind though is that considering how much printing costs and that they can't show as great or realistic a range of brightness, that the whole dimming the monitor thing should really only be done for print proofing and that they should be left to their best for regular usage and on screen viewing.
There is no reason to cripple the display ALL the time just because prints have a weak contrast range and sometimes poor blacks.
Personally I can't until one day we have 4-6k OLED displays.

And don't forget that people also used to project slides or view them on light boxes and they gave you some nice CR too.
« Last Edit: July 11, 2010, 02:00:35 PM by LarryBaum » Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #75 on: July 11, 2010, 01:44:49 PM »
ReplyReply

Quote from: fike
Obviously the Dell marketing people got a little excitable on this one, but if you read further in the review you will see that they do debunk this absurd claim, albeit in a considerate and non-inflammatory way.

[blockquote]While the DCR obviously worked to some extent, I've no idea where Dell got the figure of 80,000:1 from! ... I don't know where Dell picked this spec from?![/blockquote]

in the conclusion:
[blockquote]The dynamic contrast ratio was nowhere near reaching its supposed specification...[/blockquote]

As the review points out, the LCD color accuracy is good with custom, at home, calibration, even if the Adobe RGB and sRGB presets are substandard.  

So after asking the question here about the necessity of LCD calibration, I have read what people have to say and some more reviews and the consensus is that, particularly with the more economically priced 24" displays, calibration does result in substantial improvements in color accuracy and consistency.  

It's too bad that they don't have better factory calibration.  If factory calibration were done well, I can see them obsoleting calibration equipment.  I wonder if the manufacturers are under any pressure to improve their calibration quality.  Sometimes early in product development and manufacturing cycle they are still making improvements to quality and yield.  I'm not going to hold out too much hope that this will be the case here.

always make sure to turn dynamic contrast off, for sure when doing calibration

the U2410 actually has somewhat low contrast and somewaht high black levels from what I read (in the realistic static mode)

don't pay any attention to TFTCentral findings on wide gamut monitors when it comes to gamut and white balance they use an i1D2 which does NOT at all properly measure wide gamut monitors unless used with custom software
Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #76 on: July 11, 2010, 01:47:34 PM »
ReplyReply

Quote from: digitaldog
For the same reason having a 12 stop scene range and a 6 stop capture device is problematic. Or a 10,000:1 display contrast ratio trying to soft proof a print that has a 250:1 ratio.

I think you will be sorely disappointed if everything eventually goes OLED then....

Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #77 on: July 11, 2010, 01:52:42 PM »
ReplyReply

Quote from: Ethan_Hansen
-> (1) Is a 6 bit per channel monitor with a high contrast ratio better than a 6 bit monitor with a low CR?

Depends on your use. Most monitors achieve high contrast ratios by having the maximum luminance so high that you need wear sunglasses. LCD contrast ratio is governed by how little light leaks through when all the filters are active (black level) and how bright the backlight is when the filters are turned off (white level). Hitting 1000:1 or higher contrast ratios may well entail a white luminance of at least 200 cd/m2. That's bright.

A lot of LCD monitors can do 850:1 at only 110 cd/m^2. Some top recent regular old CCFL LCD HDTV can do a good 3500:1 at 100 cd/m^2.
Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #78 on: July 11, 2010, 01:54:37 PM »
ReplyReply

Quote from: Ray
Hey! Mark,

We all know here that 6 bits per channel is not ideal for photography. The issue is 'contrast ratio' and any disadvantages a specific and particularly high contrast ratio may have for photography and calibration purposes.

6bits doesn't give you a whole lot of room to swithc between say tone response: gamma 2.2, gamma 2.3, sRGB, gamma 1.9, L*, etc.

6bit panels are also pretty much all TN which also tend to have other issues as well, including severe wash out vertically
Logged
Pages: « 1 2 3 [4]   Top of Page
Print
Jump to:  

Ad
Ad
Ad