Bug, and a request

[jllaurado]

Hi everyone,
While i keep on honing my skills, both taking photos and later editing them, i´m becoming aware that i might have my monitor way off in regards to calibration.
In the photo i´m attaching here, i can see both color and contrast OK in my monitor, but if i see it in a TV set, a mobile phone, and my other computer (job), there´s no consistency at all, which is something to be expected, but maybe not that much.

For those of you having calibrated monitors, i´d appreciate if you can give me advice on how off do you estimate i have mine. Is the image too bright, too dark, too saturated? I´m planning on getting a hardware device to calibrate it, but i´d appreciate some input on the actual results of my edited photos in this regard.
Thanks.

Climbing Bug by jllaurado65, en Flickr

[Manfred M]

Unfortunately, it is very challenging to comment on the colours in this image as we have no reference here. My suspicion is that the overall image might be a bit warm, but that's purely a guess. Images with a lot of leaves often have a green and / or yellow colour cast.

The second part of your question is very complex and not easy to answer. It not only covers using a colorimeter or photospectrometer to validate the settings of your computer screen, it starts with your screen itself and unless it is sRGB or over 99% AdobeRGB compliant, calibrating and profiling it are going to give you mixed results. Your workspace (colours and ambient light levels come into play), the software you use (unless it is colour managed profiling and calibration are not going to work). Likewise, if you post on the internet, you have no control over what the downstream user is viewing your images on.

Your TV, phone, etc. are likely not worth worrying about as these devices are likely not designed to accurately reproduce colours. So far as I know, none of the companies that turn out profiling and calibration tools support Android devices and there is very limited software on the Apple hardware that is colour managed, even though some of the xRite products can be used to profile these devices.

And yes, I have an AdobeRGB compliant screen, it is calibrated and profiled, my workspace is setup (colour scheme and ambient light levels) is within the parameters most experts suggest will work well. I use an xRite i1 Display Pro.

[jllaurado]

Thank you, i guess the "bit warm" might come from a Velvia modifier i used. As in regards to the profiling in different environments, i understand what you mean, but as i´m not pursuing a professional career in photography, i just want my images to look "good enough" in the displays i might normally use to show them. What actually brought this question is that my job computer shows them much darker than i expected (not the case in the phone nor the TV set) and i have concerns wether my monitor might be too bright and so i would be editing the images in unproper ways relating exposure compensation or brightnes/contrast levels.
Thank you

[Geoff F]

Your 'bug' is actually a beetle.

Colours look about right to me but this isn't the best of scenes to make a firm decision.

Personally, with this image I would be tempted to slightly tone down the brighter areas on the right then give the whole image a tiny bit of highlight boost to produce a little bit more 'zip' to the image. But only a little tiny bit, it is so easy to go too far with these subtle tints.

[jllaurado]

Thanks for both details, the identificacion, and the color adjustments

[Manfred M]

Thank you, i guess the "bit warm" might come from a Velvia modifier i used. As in regards to the profiling in different environments, i understand what you mean, but as i´m not pursuing a professional career in photography, i just want my images to look "good enough" in the displays i might normally use to show them. What actually brought this question is that my job computer shows them much darker than i expected (not the case in the phone nor the TV set) and i have concerns wether my monitor might be too bright and so i would be editing the images in unproper ways relating exposure compensation or brightnes/contrast levels.
Thank you

A general rule of thumb is that you should be working on a computer screen with its output set to between 80 and 120 candela / square meter (the calibration / profiling hardware lets you do that) and have the ambient lighting in your workspace no higher than 70 lux. That will put you in the "sweet spot" so far as colour reproduction and contrast ratio go.

Unfortunately, we have no way of determining the setup of any of your viewing hardware unless you calibrate and profile it with a tool like the xRite i1 or Colormunki or Datacolor Spyder lines.

The members here who have made careful choices when it comes to computer screens and calibrating and profiling their hardware look at output quality requirements much like you do. Head off to the high end "pro" setups and the calibration and profiling hardware they use gets into the price range of a mid to high end camera.

[DanK]

Juan,

Manfred is writing about highly precise calibration. I think you can do much less, given that you aren't printing. I'm going to assume that your monitor can produce the sRGB color gamut. There is no reason for you to use a wide-gamut monitor (e.g., Adobe RGB) because you are interested in displaying your images on screens, and very few of them cover more than sRBG.

However, if you want reasonable color reproduction, you need to calibrate your monitor. True, many people will not have calibrated monitors, and your image may look different on those. Many will be brighter, and some will have a different color balance. Overly blue colors, for example, are common. However, if you don't calibrate your monitor, even displays with good color rendition will look wrong.

I've never tried to calibrate an image from someone else's computer, but I think it might be possible to estimate whether the whites are off, if you include a reference image. I don't think there would be any way to estimate brightness.

Dan

PS: Geoff, In response to your quotation marks: in colloquial American English, "bug" refers to all insects, not just "true bugs" (Hemiptera). Some people will use it to refer to other arthropods as well, but that's generally considered incorrect, and you will hear people say things like "spiders aren't bugs." I think if you were to tell a random 100 Americans that beetles aren't bugs, 99 of them would think you had lost your marbles. By the same token, I would be surprised if more than 1 in 100 have any idea what "true bug" means. I have a hunch, but have no way of knowing, that this is why Hemiptera are called "true bugs"--to contrast them to the many bugs that aren't "true bugs".

I'm not suggesting that you use the term incorrectly, but the quotation marks could be seen as disparaging.

[xpatUSA]

... PS: Geoff, In response to your quotation marks: in colloquial American English, "bug" refers to all insects, not just "true bugs" (Hemiptera).

Quite so. Also, around here, anything long and thin that moves is called a 'worm'; and bugs that are quite obviously wasps to my English eye are called 'bees'. Funny world, ennit?

[Manfred M]

I don't think there would be any way to estimate brightness.

Actually there is and I keep forgetting to mention it.

Set up your computer screen to display a pure white image.

Set up your camera on a tripod and adjust it so that it sees as much of the screen as possible, but only the white sample.

Turn off all the lights in the room.

Set your camera to ISO 200 at set the exposure to 1/30th second. Keep adjusting until the camera shows a correct exposure at an aperture setting of f/6.3

That corresponds fairly well to 100 candela / square meter.

Also, I suggested that screens that are either 100% sRGB compatible OR close to 100% Adobe RGB compatible be used. Screens that do know have that either sRGB or Adobe RGB compatible unlikely to display all of the image colours correctly. I have tested a number of lower end screens and they tend to have a marked colour cast, even after calibration and profiling.

[xpatUSA]

Actually there is and I keep forgetting to mention it.

Set up your computer screen to display a pure white image.

Set up your camera on a tripod and adjust it so that it sees as much of the screen as possible, but only the white sample.

Turn off all the lights in the room.

Set your camera to ISO 200 at set the exposure to 1/30th second. Keep adjusting until the camera shows a correct exposure at an aperture setting of f/6.3.

That corresponds fairly well to 100 candela / square meter.

Very interesting, Manfred. Manual exposure, I assume, but why 200 ISO?

For that kind of test, I usually apply de-focus - usually "infinity".

[jllaurado]

Thank you, i´m going to do that.

[xpatUSA]

Very interesting, Manfred. Manual exposure, I assume, but why 200 ISO?

Out of curiosity, I referred to the Japanese CIPA DC-004. DC-004 because, unlike ISO stuff, it's free and likely applies to most modern Japanese digital cameras.

It says ISO = 10/Hm and therefore Hm (sensor exposure) = 10/ISO = 10/200 = 0.05 lx.s

Since the camera positioning appears to be trying to make the screen a Lambertian surface, then we can use the formula in DC-004 to see what Hm we should get at 100 cd/m^2 from the screen with Manfred's settings.

Hm = 0.65 x luminance x shutter time / aperture^2 = 0.65 x 100 x 1/30 / 6.5^2 = 0.055 lx.s

Close enough for Government Work !!

And, at the more common native ISO of 100, the camera setting would need to let in twice the light ...

[Manfred M]

Very interesting, Manfred. Manual exposure, I assume, but why 200 ISO?

For that kind of test, I usually apply de-focus - usually "infinity".

I suspect that this was used because a number of digital cameras have a base of ISO 200, so by using it allows just about anyone to use this method.

The method comes from the work of Dr M Robert Ito who is is the CAPA (Canadian Association for Photographic Art) Director of Photographic Imaging. Dr Ito is Professor Emeritus of Electrical Engineering at the University of British Columbia, in Vancouver, Canada.

[xpatUSA]

I suspect that this was used because a number of digital cameras have a base of ISO 200, so by using it allows just about anyone to use this method.

Thanks, makes sense.

The method comes from the work of Dr M Robert Ito who is is the CAPA (Canadian Association for Photographic Art) Director of Photographic Imaging. Dr Ito is Professor Emeritus of Electrical Engineering at the University of British Columbia, in Vancouver, Canada.

Thanks for that too ...

[xpatUSA]

Out of curiosity, I referred to the Japanese CIPA DC-004. DC-004 because, unlike ISO stuff, it's free and likely applies to most modern Japanese digital cameras.

It says ISO = 10/Hm and therefore Hm (sensor exposure) = 10/ISO = 10/200 = 0.05 lx.s

Since the camera positioning appears to be trying to make the screen a Lambertian surface, then we can use the formula in DC-004 to see what Hm we should get at 100 cd/m^2 from the screen with Manfred's settings.

Hm = 0.65 x luminance x shutter time / aperture^2 = 0.65 x 100 x 1/30 / 6.5^2 = 0.055 lx.s

That formula can also be re-arranged to measure the screen luminance directly, using the in-camera metering.

In aperture priority, my metering with no EC gave 1/10 sec at f/8 and 100 ISO. It came out as 98 cd/m^2, amazingly enough.

Won't bore you with the working ...