Like many here I'm sure, I've struggled with finding the correct brightness value to set my LCD display to for post production.
The manufacturers of calibration equipment will recommend anywhere from 90 cd/m2 to 140 cd/m2, and then many non-graphics LCDs today can't even go below 200 at their minimum brightness setting. The correct brightness depends on your viewing environment - a bright office requires a brighter screen. A dark studio will want a lower level. Therefore the correct answer to the question is it depends on your viewing environment.
Some calibration equipment have the ability to measure ambient light and then adjust your display (or instructs you to adjust) accordingly. But I've found this ability often doesn't work very well. The colormunki, in my experience, is terrible for determining LCD brightness based on ambient light level. Worse yet, the software doesn't display the cd/m2 it reads with which to experiment with.
So to match my LCD to my ambient viewing environment, I tried this with my Sekonic L-358 light meter, and it seems to work pretty well. It's not entirely scientific, but it seems to work well enough.
1. Display a completely uniform screen on your adequately warmed up monitor. I've been using white for this, but perhaps light gray would be better?
2. Set your light meter to EV mode (on the L-358, dipswitch 1 under the battery cover should be set to "ON"), and then set it take Ambient light readings.
3. With the dome extended, take several readings of the area where you view prints. Note these values. My studio averages 4.6 EV during a partly cloudy day.
4. Retract the dome. Then hold the meter in front of your face from exactly where you normally sit to work and take several readings of the display. The distance is important here.
5. Adjust the brightness of your monitor until the EV reading is close to those measured in your print viewing area.
6. For future reference, make note of what your calibration software declares this to be in cd/m2.
One note - it's impossible to convert your light meter reading directly to the cd/m2 that monitor calibrators use as there is a distance variable involved - the distance from the monitor to your eyes. Remember light falls off with the inverse square.
You might do this at several times during the day to acquire a decent average brightness for your display. Evening illumination provided by artificial light is going to be different than daylight through windows.
Any thoughts? Am I crazy or going about this the entirely wrong way?