Prints are too dark

[Teetime]

About a year ago I ordered some prints from Bay Photo Lab and was disappointed in how much darker they were than my monitor images. In reading about this I determined my calibrated monitors were probably too bright. Today I decided to see if I could resolve this before ordering some more prints. Here is what I have done:

1. I recalibrated my two monitors (Dell ST2210 and LG IPS235) using my xrite I1Display with brightness set to 120 cd/m2 to be sure I had a known starting point.
2. I installed Bayshore's printer profile in the latest version of Lightroom Classic (running on latest version Windows 10).
3. In LR Soft Proof I opened one of the images I had sent to Bayshore and it indeed still looked brighter than the print I received from Bay Photo Lab.
4. Out of curiosity, I then turned the brightness on the primary monitor all the way down and using a Sekonic 308DC light meter took a reflective reading of a point of interest on the monitor image.
5. I then put the Bay Photo Lab print on the front of the monitor flush with its screen so it was subject to the same ambient lighting the monitor is and took a reflective reading of the same point of interest.
6. The reading on the monitor was 3.5 EV BRIGHTER than the print, even though the monitor brightness is now at it's lowest available setting (i.e., something well below 120 cd/m2).

Given all this I'm having trouble understanding how I can reliably get prints that match what I see on the screen. Any suggestions?

[Manfred M]

I run my screen at 80 cd/ sq meter and that is a good match for prints with 150 lux of light hitting them.

[xpatUSA]

4. Out of curiosity, I then turned the brightness on the primary monitor all the way down and using a Sekonic 308DC light meter took a reflective reading of a point of interest on the monitor image.

You can not take a reflective reading off something that is emitting light, sorry.

I would suggest that you measure the monitor's cd/m2 in a 100% white region and a 0% black region with no other lighting in the room, if the Sekonic can do it.

[Manfred M]

You can not take a reflective reading of something that is emitting light, sorry.

I would suggest that you measure the monitor's cd/m2 in a 100% white region and a 0% black region with no other lighting in the room, if the Sekonic can do it.

The i1 Display Pro (same as I use) can be used to set the computer screen to the appropriate output level (cd/ square meter)

I find that 120 cd/ sq meter is much too bright to get the printer output to be close to what the screen shows. The sRGB standard is a screen output of 80 cd/ sq meter and the Adobe RGB standard is 160 cd/ sq meter.

Ideally the workspace should have light levels below 40 lux for accurate print work, to ensure that the contrast ratio acceptably high.

[Astro]

+1 to Manfred's comment. I have my screen set to 90cd/m. However I do my own printing and (like Manfred), usually run a test print. From experience I can get a reasonable print first time, but every image is different and to get the 'best' there is usually a need to fine tune brightness, which some images takes more than a single proof to achieve what I want.

I do occasionally send images out for print (usually because I want sizes larger than I can print myself, or I want a print on a medium like Aluminium). In this instance I always proof print to get the best image I can before I pass it to the Lab.

An additional thought though, have you checked that when you transmit the image to the lab that the profile is included?..... though I would expect the Lab to query if it were not!

[DanK]

OK, I'm going to go out on a limb here.

First, I agree with Manfred and James: turn down the brightness of your monitor. It's easy to do this when calibrating. I use 100 cd/m^2, but I'll come back to that.

Second, like James, I print everything myself other than things I can't, e.g., metal prints, but I used Bay Photo in the past, and my experience is that they are a high quality lab that does good work. So I doubt the problem is there.

Now, time to to out on a limb: I think the idea that you can pick a setting that will make prints match the screen is a fallacy, for two reasons. First, a reflective surface and an emissive surface simply look different. Second, and more important, it depends on the ambient lighting.

Consider this thought experiment--or just try it. Sit in front of your computer with a print that you think more or less matches the screen. Then turn down the lights. They no longer look similar. The print obviously is darkened, while the screen isn't, but in addition, the dark surroundings will make you perceive the screen as even brighter.

So I think the better way to think about this is to ask how to set your monitor and ambient lighting so that prints look like you think they should.

I find that setting my wide gamut monitor to 100 cd/m^2 in the ambient lighting I use largely eliminates surprises. However, I don't just use the visual appearance of the image as a guide. I look at the histogram and the numerical white and black points frequently as I edit.

I wrote "largely", not "entirely." In response to this thread, I tried to think of recent times when a print didn't look like I thought it should. I thought of three instances. All three were photos of people, and my first reaction had been that the faces were too bright, not too dark. However, when I thought about it again this morning, that didn't seem likely because I track the white point and histogram. I had one sitting on a table here, and I don't think that's really the problem. I'd have to go back to the images on the computer, but I now suspect that the issue is misplaced curves. I think I accidentally compressed the tonal range in the faces.

None of this argues against Manfred's advice that both the monitor brightness and the room brightness matter. But I'm suggesting that you matching the print to what you expect to see isn't quite the same as matching it to the monitor.

One of the advantages of printing yourself is that you can see right away whether you getting what you want, and you can make adjustments right then and there. That's a very big advantage, particularly when you are starting out. I've found that I am so used to my setup now that I often don't print small test prints, but I wouldn't recommend that when you are starting, and I don't recommend it when there is anything difficult in the image, e.g., out-of-gamut or borderline colors.

[rpcrowe]

I agree with Dan... As an example:

I had a canvas print done of a Venice Canal scene and was disappointed in that it seemed too dark. However, when I brought it to my photo club and hung it on the wall under a portrait light, the light did amazing things to the canvas print. It looked brighter and had a 3-D like effect. It was the perfect brightness for displaying under the portrait light. I don't think that brighter print would not have looked as rich.

As Dan mentioned, perhaps the moral of this tale is that there may not be one brightness that is appropriate for display in all environments...

IMO, the print surface also tends to require different brightness. A highly reflective print surface such as glossy or especially metal will look good with less brightness than a print done on a matte or especially a canvas surface.

[Teetime]

I agree with Dan... As an example:

I had a canvas print done of a Venice Canal scene and was disappointed in that it seemed too dark. However, when I brought it to my photo club and hung it on the wall under a portrait light, the light did amazing things to the canvas print. It looked brighter and had a 3-D like effect. It was the perfect brightness for displaying under the portrait light. I don't think that brighter print would not have looked as rich.

I have a similar example. I have a b/w canvas print from Scotland train station in my living room. On a sunny day it has beautifully rich tone and texture. On an overcast day it looks too dark and flat.

All of you have given me some excellent feedback. I just recalibrated the monitors using 80 cd/m^2 and they now look closer to the comparison print. Before I calibrated at 120 cd/m^2 and then just manually lowered the monitor brightness to its lowest value. Obviously, this threw everything off. I am going to do a test print with the new calibration and a fresh edit of my test photo to see if I get closer to my desired look.

Thanks to all for you valuable insights!

Edit: For some reason when I click on "thumbs up" for useful posts nothing happens.

[Manfred M]

I have a similar example. I have a b/w canvas print from Scotland train station in my living room. On a sunny day it has beautifully rich tone and texture. On an overcast day it looks too dark and flat.

All of you have given me some excellent feedback. I just recalibrated the monitors using 80 cd/m^2 and they now look closer to the comparison print. Before I calibrated at 120 cd/m^2 and then just manually lowered the monitor brightness to its lowest value. Obviously, this threw everything off. I am going to do a test print with the new calibration and a fresh edit of my test photo to see if I get closer to my desired look.

Thanks to all for you valuable insights!

Edit: For some reason when I click on "thumbs up" for useful posts nothing happens.

One of my printing mentors has worked as a commercial custom print maker for over 35 years and admits that he still has trouble with this aspect of the process.

Dan is right and what you see on your screen will never be exactly the same as what the print looks on the wall.

The "best practice" is to print to the light level that will be most common at the place that your print is displayed. If it is a place that is easy for you to access, then working with test prints and checking them at that location is usually the best approach. If you have a variable light source over your printer and can simulate the light level, that works reasonably well. It does require taking an incident light reading at the display space and getting your print lights to about the same level.

If you do not have this capability, a lot of printers have told me that they will print to a 150 lux illumination level unless the print is going to a gallery, where a 200 lux level is usually more appropriate.

I know another master printmaker who displays his work in galleries, but does not sell the print that is on display. Rather he custom prints a new print based on light levels that are more likely to be used in a home or office.

[DanK]

I almost never have the opportunity to print for a specific space, and I'm not sure I would have the patience to. However, the 150 lux idea is appealing. I think the xRite Display Pro pucks can display this information. If that's right, I'll play with it next time I calibrate my monitor.

[Manfred M]

I almost never have the opportunity to print for a specific space, and I'm not sure I would have the patience to. However, the 150 lux idea is appealing. I think the xRite Display Pro pucks can display this information. If that's right, I'll play with it next time I calibrate my monitor.

Dan - I use my incident light meter (Sekonic L-358) to check the evaluation light when I print. I am planning to set up a high CRI, variable output light right above my printer output tray.

That will be a fall project, I suspect, after the outdoor maintenance is done.

[DanK]

I don't have an incident meter. Maybe I'll look for one on eBay.

I did something similar. The room where I do my editing has 6 BR30-size cans in the ceiling. I installed bulbs that are CRI=95 and R9=95 on a dimmer.

I suspect you know, but CRI is insufficient. The standard was developed during the era of fluorescent bulbs and for that reason does not include some ranges of reds. That's why it's important to find bulbs that have a high R9 value. In my experience, most manufacturers either don't provide it or just say R9=50, which I think may be a minimum standard in some jurisdictions.

[Manfred M]

I don't have an incident meter. Maybe I'll look for one on eBay.

I did something similar. The room where I do my editing has 6 BR30-size cans in the ceiling. I installed bulbs that are CRI=95 and R9=95 on a dimmer.

I suspect you know, but CRI is insufficient. The standard was developed during the era of fluorescent bulbs and for that reason does not include some ranges of reds. That's why it's important to find bulbs that have a high R9 value. In my experience, most manufacturers either don't provide it or just say R9=50, which I think may be a minimum standard in some jurisdictions.

There's a store in town that specializes in light bulbs (including tungsten, fluorescent and LED) and nothing else. Apparently very knowledgeable on the subject, so I will check with them.

[senojbor]

I use Saal digital for their ICC Profiles for printing. They are spot on for printing on their range of materials and finishes. Just don't embed them in the files. I never have issues with prints. You can download the profiles for free and there's a chart to help you. Use them for screen proofing
There's a bit about Monitor calibration as well.
https://www.saal-digital.co.uk/service/icc-profile/

[senojbor]

This is more useful for the answer to the original question.
https://www.saal-digital.co.uk/suppo...n-the-monitor/

[Manfred M]

I use Saal digital for their ICC Profiles for printing. They are spot on for printing on their range of materials and finishes. Just don't embed them in the files. I never have issues with prints. You can download the profiles for free and there's a chart to help you. Use them for screen proofing
There's a bit about Monitor calibration as well.
https://www.saal-digital.co.uk/service/icc-profile/

This is only useful if you use Saal for your print work as these are specific to their papers and printers.

For people like Dan and me, who do our own at home using inkjet photo printers, the information is less than useful. Using either the profile supplied by the paper manufacturer for the specific printer / paper. Once can (with appropriate hardware) have a custom profile made as well or have a third party create one.

Most quality print labs have this information available for their customers.

The information on the site is reasonably good, but I don't agree with it 100%. Much of it deals with basic post-processing, rather than print preparation, but with a non-technical clientele, simplification is a good approach.

Some of the information on the Saal site is a bit misleading, if one wants to get

[Manfred M]

This is more useful for the answer to the original question.
https://www.saal-digital.co.uk/suppo...n-the-monitor/

This article is quite misleading as the diagrams they show have a lot more to do with global contrast than image / print brightness. I see nothing about sharpening for prints in any of the materials.

I run my profiled and calibrated screen at 80 cd/sq m in a workspace where the light falling on it is below 40 lux. When evaluating my prints I either check the light level where I plan to hang the print or if it is an unknown I will evaluate under 150 lux illumination for most places and will bump that up to 200 lux for galleries.

[senojbor]

This article is quite misleading as the diagrams they show have a lot more to do with global contrast than image / print brightness. I see nothing about sharpening for prints in any of the materials.

I run my profiled and calibrated screen at 80 cd/sq m in a workspace where the light falling on it is below 40 lux. When evaluating my prints I either check the light level where I plan to hang the print or if it is an unknown I will evaluate under 150 lux illumination for most places and will bump that up to 200 lux for galleries.

I think the big issue here is to learn that prints whatever printer/lab you use you will be printed differently both in colour and luminosity than what is displayed on a PC monitor. I notice that there is a good tutorial on soft proofing and calibration on the main site on how to manage that.

It's a very in depth complex module on it's own but worth learning if the aim is to produce high end prints.
But it's fairly straight forward after a while when people get use to it.
I mainly use the better pro labs, although I print my own stuff which I only find fairly acceptable even on decent paper.
I find though printing my own stuff rather time consuming, limited and not really cost effective in the long run.

[DanK]

The OP isn't about the brightness of images in general; it's about the relative brightness of an image on the screen and on paper. The first part of the linked document has one of the most important factors: the brightness of the monitor. This takes some trial and error and won't be the same for every person in every environment, but the range they provide is a reasonable one. Manfred goes a tad darker, but I get good results using values of 11 or 100 cd/m^2, which is in that range.

However, the rest of the document, titled "tone correction check," is off track for the OP's question because it will brighten both the on-screen image and print.

[senojbor]

We never had this problem with film did we!