This article is a concise test of the Ikelite eTTL2 conversion circuitry in a real-world dive scenario. The equipment used was a Canon 20D with 100mm macro lens, an Ikelite housing for the 20D, maunal focus capable flat port, and dual DS125's with a dual eTTL sync cord. The camera was set to Av (Aperture priority) so that the same background exposure will appear in each of the shots, i.e. the same amount of ambient light will influence each of the shots. This mixed light technique will also give us a feeling for how the system may perform when shooting mixed lighting scenes - wideangle for example. I feel this technique makes it easier to assess the system than just setting the shutter speed at 1/250th and firing the same series of shots. I could be wrong though, and if you feel a different testing technique should be used, and you can convince me why, I'd be happy to jump back into the 60 degree water and test it for you...:-)
The TTL controller is built into the camera tray and has flash compensation buttons built in.
I took a series of photos at each f-stop. A good TTL system should yield the same exposure for each of these shots - results are presented below.
f2.8 @ 1/90th of a second
f4 @ 1/45th of a second
f5.6 @ 1/20th of a second
f8 @ 1/10th of a second
f11 @ 1/6th of a second
f16 @ 1/3rd of a second
f22 @ 2/3rds of a second
Examining the metadata for each shot, the camera has recorded: "Flash Mode: On + red eye reduction." So, even without the flash popped up, the camera "knows" there is a flash connected.
The histogram from the photo taken at f2.8
The histogram from the photo taken at f22
Conclusion: The TTL seems to be working well. Besides the obvious motion blur in the longer exposure shots, the imags brightness is very close. For those who like quantitative data, I've provided the histograms from the shots taken at f2.8 and f22. I would say that the system is reproducing exposures that are within 1/2 stop of each other.