First dives with Ike eTTL housing - my Review

After one year of diving without TTL, in January I sent my Ikelite Rebel housing back to Ikelite to have it updated to the eTTL version in preparation for 6 months of diving the Great Barrier Reef.

So Friday was my first dive in Australia and first dive with TTL. Wow, it really makes me never want to dive in the Caribbean again. I know that I must have been having way too much fun because only a handful of shots were in proper focus/exposure and I am not going to say anything about composition.

Not only was it my first dive out of the caribbean, it was my first dive with eTTL, dual strobes, and sigma 15 fisheye. I wasnÂ’t expecting great shots, it was more of a chance to get to know all my new gear, and to get used to diving in an new ocean.

I havenÂ’t seen many reviews of the eTTL housing so I thought I would post my experience with eTTL.

The camera is aware that the flash is connected. In the EXIF data it reports that the flash fired even when used in Manual mode (without the eTTL converter, the camera reports that the flash did not fire)

If eTTL converter is set to 0-FEC the in camera FEC works adjusts the exposure. If the eTTL converter is set to anything else, the in-camera setting is ignored by the strobes.

No matter what setting you have the eTTL converter set to (eTTL/manual and fec +/-) the in-camera FEC setting is what is reported in the EXIF data. So if you like to review the EXIF data to see what settings you used, make sure you use the in camera FEC adjustment rather than the converter's adjustment.

A small quirk is that you can not adjust FEC or manual power level while the strobe is recycling. When the batteries start getting low it can take a while before you can change the settings. But other than that moving control to the back of the housing is much more convenient than adjusting manual controllers.

The big question: does it work. - Well that depends on what lens you are using.

With a Wide angle lens, in my case sigma 15mm fisheye, even with FEC set to its maximum of +2 the converter grossly underexposes the flash exposure while setting the strobes manually gave the exposure I expected.

For example, both shots 1/40th sec, f/7.1, ISO 100

First shot eTTL FEC +2 - If I didnÂ’t see the flash fire, from the results I would swear the flash failed to fire.

Second shot strobes on manual control. What I was expecting from eTTL FEC=0

Conclusion, for wide angle eTTL is completely un-useable. If someone else has gotten it to work please speak up. I will probably give it one more try next weekend and if I get the same results I will quit trying.

On my second dive I was using my Macro lens. (100mm USM) I was not expecting much after the results I was getting with the 15mm lens.

Wow, what a difference. Every exposure right on. I have always had a tendency to underexpose most my macro shots so I'm loving this. Even the goby on sand shot which before always took me 3-4 attempts to not burn the sand came out great on the first shot. I'm never going back to manual for macro (unless my next camera/housing combo doesnÂ’t support TTL I guess)

F/14 1/160sec ISO 100

F/14 1/160sec ISO 100

F/18 1/160sec ISO 100

F/18 1/160sec ISO 100

Nice report, interesting results.

Welcome to the Pacific! Tis a world of difference eh? Actually, i have never dove in the Caribbean so i can't say for sure but i have no real desire to go there other than the Tigers and Hammers in the Bahamas.

Too many years in the Pacific have spoiled me i guess.

Anyhoo, did you ever shoot film underwater before going to digital?

I think you will find that the results you discovered with TTL is the same as film, works great for macro/fish portrait work where your main subject fills the frame. However, tis useless for the vast majority of WA because too much empty space/blue for the "camera brain" to figure out when to turn off the flash. What i do find interesting is that your WA was underexposed as opposed to overexposed which most people would get on film due to the camera giving a full dump flash.

Look forward to more results on your WA experiments but would push you to go with manual so you don't miss too much of the exotic Pacific life...

No I have never used a film camera underwater or above so I donÂ’t know what TTL was like on film.

What I do know is that eTTL works fine with a similar shot and a land-strobe.

Here are two shots taken with the 15 fish and Sigma EF-500 Super on my Digital Rebel.

As you can see in this shot, in which I have the flash turned off, the background is equivalent to the empty blue underwater - far away and brightly lit. The foreground is small and dark and needs to be lit by the flash.

Now in this second shot, I turned the flash on. The foreground which is only a small portion of the frame and is exposed properly. This shows that if Ike's conversion was as good as Sigma's, eTTL should be workable for wide angle underwater. Unless the problem has to do with the water for some reason?

Ok, after posting that it occured to me that I should try the coke can shot with Ike strobes on eTTL. And guess what - Proper expoure of the can... Anyone have any ideas why it works in the air and not underwater?

I still say if it doesnt work next time I'm going to give up on it for WA and be happy that it works for macro.

Thanks for the report William.

It's hard to tell from your images, but was the eTTL able to hold the highlight detail on the macro shots? They seem a little hot to me. What do the histograms look like? Are the highlights clipped off or do they slope all the way down to white? I'd love to see them? I don't know how to post an image of a histogram, maybe someone else here does and can tell us how to do it.

Danb

Ok, here are a bunch of DPP histograms.

First a random selection of shots in which FEC = 0.

Next is what happens when you Increase FEC. First Image FEC=0. Final shot was FEC=1.5.

Based on my experience with Ikelite/Olympus TTL with my C-5050, TTL works well for both wide angle as well as macro.

http://www.splashdowndivers.com/photo_gall...olympus_ttl.htm

And I expect Ikelite/Nikon iTTL to perform similarly with my D70 when it is available.

Thus, my take is that there is something wrong with your hardware or your settings, but no idea what.

Edited; Question: What metering did you use in the first picture e.g. for my D70 I will probably use center weighted or spot but not matrix as I want the exposure to be optimized for what I am aiming at?

With the Digital Rebel you get no choice of metering mode and no Canon dSLR short of 1Ds and 1D have spot metering.

In the picture I posted I was Manual exposure mode wich uses center weight average metering.

I also tryed using the camera in both Shutter and aperature priority which both use evaluative metering, I got similar results.

You should try the selective preflash next time. Move your object which will be lit by the strobe to the center (Set evaluation method to SPOT and select the center field), press the *-button to force the camera send a preflash. The camera will store this value, now reframe and shoot. Maybe this works better...

Good idea.

Set evaluation method to SPOT and select the center field

The rebel does not allow user selection of metering modes without the firmware hack. Metering method is always bound to the shooting mode.

But that's not the problem here I guess. We don't deal with a Nikonos 1-zone average metering :-)

In addition the Rebel does not feature spot metering at all. Like the 20D - no spot metering.

As you describe only reframing makes a difference when sending a preflash (*).

I think the centre AF point was used in the wideangle shots as well?

If yes: * preflash would not make any difference here.

Possible reason: I don't know how much light is send during the preflash. When it comes to wideangle shots to much light might be absorbed by the water. This could be the case when the preflash is very weak which could be necassary in order to match E-TTL system's criteria.

Just a loud guess as E-TTL works reliable for similar lightning scenarios topside.

By the way, manual strobe control rocks! :-)

Julian

William, another problem might be the settings you used for your wide angle pics i.e. "both shots 1/40th sec, f/7.1, ISO 100"

With that much natural light exposure in shallow, clear water, the camera may not have asked for much help from the strobes.

Thus, my suggestion would be to shoot manual with 1/100th sec. and f/9 to f/11 if you want more light from your strobes for shallow clear water shots.

Thanks for sharing your experiences William. I'm baffled. The macros look over exposed to me by about 1/2 a stop. And the wide angle is very wrong.

Like Peter I have used Ikelite TTL with the Olympus and found it worked very well.

The land shot looks perfect. one difference might be a minor flood in the cable/synch sockets that could be shorting out the strobe connections when it is underwater? Very small - so you wouldn't notice it - and it would dry after the dive - so the strobe would work on land.

What I suggest is to do a TTL test - shoot a range of test shots in the pool of a macro subject with different aperures (all at a fast shutter speed - so that the lighting isl flash dominated). If TTL is working they will have the same exposure. Here is an old example I did more than 2 years ago now on the D100:

The performance there might tell us a bit more about what is going on.

Alex

Ok, no SPOT metering. But you can use the AF-point selector and select the center one. Now, the ETTL uses this point to evaluate the exposure.

Example1: center spot ettl exposure saved with *-preflash first

example 2: moved camera to the right, *-preflash, reframed as above and fired

In addition the Rebel does not feature spot metering at all. Like the 20D - no spot metering.

Are you sure? I've played a bit with the 300D and when I "*" the objects first, it seems to be spot metering. When I move the target a very little bit out of the "*"ed field, the flash heavily overexposes. It's a "digital" response: perfect or extremely overexposed. Nothing in between. I would call this SPOT.

@Alex,

the Olympus TTL protocol is straight forward and simple with a normal preflash. So you get the same results as with the built in strobe in terms of exposure.

E-TTL is complicated and normally requires a modulated preflash.

This means: Ikelite Olympus TTL is as good or as bad as the camera TTL metering system. Canon E-TTL is as good or as bad as the camera + the conversion influence.

@Satura,

no spot metering: yes I am sure.

metering bound to AF point: As stated above I assume there was the centre point activated. However I doubt this being the reason for the problem.

Julian

forgot: and add the possibility of the weak preflash travel thru water.

@satura

300D:

No selection available, depends on exposure mode:

• Evaluative 35-zone (All modes except M)

• Partial 9% at center (Also with AE-Lock in P, Tv, Av, M, A-Dep)

• Center-weighted average (M only)

regards,

Julian

You should try the selective preflash next time. Move your object which will be lit by the strobe to the center (Set evaluation method to SPOT and select the center field), press the *-button to force the camera send a preflash. The camera will store this value, now reframe and shoot. Maybe this works better...

As I had allready said, and Jolly said a couple times since.

With the Digital Rebel you get no choice of metering mode and no Canon dSLR short of 1Ds and 1D have spot metering.  

 

In the picture I posted I was Manual exposure mode wich uses center weight average metering.    

 

I also tryed using the camera in both Shutter and aperature priority which both use evaluative metering, I got similar results.