I'm trying to further my understanding of how the CLS system works with wireless slave flashes using the SB800/S800/SU800 system. There are a few (!) points that bother me; maybe someone can clarify.
1) If I understand correctly, in the past, cameras handled automatic flash exposure by measuring the light being received by the camera during the exposure, and shutting off flash output when sufficient light was received for the exposure.
2) With the CLS system, a series of "monitor preflashes" are emited by the flash, and the reflected light is used by the camera to determine the correct flash output for that exposure.
3) Question 1: Does that mean that with CLS in iTTL mode, the monitor preflashes are used to set the exposure, and the camera then DOES NOT meter during the exposure to shut off the flash; and the flash output is preset by the preflashes before the exposure begins? Or, is there also metering during the exposure, and if so, what benefit does the preflash add? (I assume, based on what have written, that the preflashes are used to set the exposure, and there is no metering to adjust this during the actual exposure).
4) When the SB800 is used in the master mode, the monitor preflashes also use pulse-code modulation to transmit exposure information to each slave flash. What I don't understand is exactly how this is determined. The master obviously does not have any way to know how many slaves there are, as the wireless communication is one way, Master -> Slave. As a result, while the master can determine a correct exposure via the monitor preflashes, HOW exactly does it make the adjustment for the contribution of the slave flashes?
As a example, suppose I have one slave flash, set to a 0.0 exposure compensation. If I added in a second slave right next to the first, also at 0.0, then I've doubled the light coming from that source, and obviously that changes the exposure. Now, it would make no sense for the master to note that and reduce the output of each slave by 1/2 to maintain the same exposure, when presumably my intent was to double the light from that source.
From this, I am reasoning that the monitor preflashes from the master are used to select the correct exposure which is done fully independently of the presence, absence, number of slaves in the system, and only trial and error can really be used to adjust the compensation on the slaves.
If I added, for instance, a diffuser between the slave and the subject, and wanted the same amount of light to hit the subject from that flash, just more difuse, I would then need to increase the exposure compensation for that slave to account for the light loss from the diffuser, because the CLS system cannot do this automatically for me.
#1. "RE: Yet another question on how wireless CLS works!" In response to Reply # 0
Yes, the preflashes alone are used to determine the subsequent flash output for proper exposure.
Slaves are grouped into three groups. During preflash, the CLS master commands (via PCM, as you said) each group to preflash, and the camera measures the resulting exposure. The master then sends codes to each group to set the flash output level of that group for exposure, then commands all flashes to fire simultaneously once the shutter is open.
So, if you have multiple flashes in a group, they are all commanded to the same output level. If you don't want them commanded to the same level, you must put them in different groups. It is true that if you have multiple flashes in the same group at different distances from the subject, they will not contribute equally to the lighting of the subject, but you chose to have that happen by placing them at different distances but in the same CLS group. (Which can be a useful technique, of course.)
Since each group is preflashed and commanded independently by the master, if you put a diffuser screen between one remote and the subject, and if that remote is in its own group, CLS will compensate for the diffuser because during preflash, the camera will see how much of the flash's light is making it to the subject, and the CLS master will then command the flash to output accordingly.
#2. "RE: Yet another question on how wireless CLS works!" In response to Reply # 1
West Hartford, US
I realized after I wrote my question that the available documentation does say that with iTTL, all metering is done via the preflash and there is no metering during the actual exposure. I had missed that the first time through.
Where I was unclear was on whether the slave groups fire a preflash for exposure or not. I tried to figure this out from watching them in operation, but I wasn't really sure if the slaves were firing a preflash or not.
#4. "RE: Yet another question on how wireless CLS works!" In response to Reply # 3
Assuming that the master unit controls the remote(s) using its main flash head, I was wondering about the commander unit of the SB-R1C1. This doesn't have a flash head; just a red window which appears to be similar to the AF assist window of the SB-800. Does it control using IR? I've not been able to play with one to find out.
#6. "RE: Yet another question on how wireless CLS works!" In response to Reply # 4
The slaves use an IR sensor. A regular flash emits enough energy in the IR region to trigger the slave's IR sensor, but the R1C1's SU-800 commander unit uses IR directly. One reason why is because it is intended to be used for macro work, where even the relatively low-power flash from an SB-800 in master (only) mode could cause unwanted illumination of the close-in subject. Using IR for the master eliminates that problem.