TTL Flash Problem - Can you duplicate

[dbara]

I am having a problem with S2 & flash pics(Happens with the pop-up and also if I use my SB80DX). The camere settings are as follows:

ISO200
pop-up flash up
"A"perature priority ( I like to change my depth of field when taking real portrait shots)
Speed defaults to 60
No custom camera settings.....all defaults
4 feet away (Manual says this is within the range of the pop-up flash for both f4 & f8)
Problem occurs with both 3D or center weighted set metering

I set up a kids toy against the wall and take 2 pictures. The first at F4 and the second at F8. With TTL I would assume the pics to be identical but the F8 is much darker.

I posted a xip file of the 2 pics for anyones review. Hope I did it right since it is my first try with this forum.

Is anyone else having this problem or does it work for you.

Am I doing something wrong?

Thanks in advance.

Don

[Tom Nolle]

It's not necessarily a problem; it depends on how much ambient light you have and how far any background objects are away from your subject. When you shoot flash, you really kind of take a "double exposure", one picture with ambient light and another with the flash. When the flash is on, meaning popped up, the camera sets a shutter speed that it can sync at, and the lens opening you select combines with that shutter speed to create an ambient light image. The flash generates its own light, and that creates an image of whatever is in its range. When you stop the lens down to f8, you cut the ambient light in half, so the parts of the picture your flash doesn't light will be darker.

Tom

[dbara]

Tom:

Thanks for the response but that is why I put the toy against the well. What you said is true if I had a subject and the background that was a ways away, but in either case the subject should not get dark if TTL is working correctly. The background is an inch away so I figure the flash should light the toy up (even at F8) but instead it is dark. Check out the pics I supplied to see what I am talking about.

Don

[Tom Nolle]

Sorry, Don, I wasn't clear on how far they were from the wall. I was on the road with a slow connection and couldn't load the pix!

I read somewhere, I think in Thom Hogan's guide, that the internal guide number was overstated for the S2. I can't find the reference right now, but might f8 have fallen outside the corrected range?

Tom

[dbara]

Tom:

Even if it is, if you step thru F4, f4.5, f56. etc. the pics gradually get darker & darker. I used the F4 & F8 pics because it was obvious.

Don

[Tom Nolle]

I'm afraid you've got me, then! I don't really use the internal flash much except (occasionally) for fill, because I've started shooting with the Quantum 2x2 and you don't use 123s with that, so can't use the internal flash. I've shot with the SB16 (real dinosaur, but it works) and my S2, and the flash seems to work as expected even out to fairly long distances (50 feet), but I don't do close flash work so I've never tried to duplicate your situation. The ONLY other thing I can think of is the metering mode; matrix, spot, center-weight or what?

Tom

[Tom Nolle]

Sorry, Don, I just noticed you answered the metering mode question in your first post!

[dbara]

Thanks again Tom. I also do not use the internal flash. I use my SB80DX. Using the SB80DX is where I first noticed the problem. I then tried the internal flash and got the same results so I knew my external SB80DX was not the problem but I thought it would be easier for people to test this using the internal flash of the S2 PRO.

Try this with your SB16. Put a subject 10-15 feet away against a wall (I used a colorful toy). Shoot at ISO 200 in Aperature priority mode once at F4 and once at F8. Did the wall and/or subject get darker. Mine does.

I do not have a SB16 but I would assume that F8 @ 10-15 feet away is within the range of the flash.

Thanks,

Don

[Tom Nolle]

You're right, it does.

OK, now let me give you my working hypothesis.

As I said in my first post to you, there are really two pictures taken, in a sense. One is taken by available light and one by flash. The two are additive in the camera, so what both of us have gotten is a kind of "ghost" with the ghost image and the flash image superimposed. When we set f4, we get some latent imaging with the ambient light (in my tests, I was probably about two stops underexposed without a flash on the wall that I used). When we stop down to f8, we've added a couple more f-stops to the underexposure. The result is that the iambient mage is darker; less light was present for the ambient image, so less lightness overall. If this is correct, we might be able to test it out by flashing a subject in a room too dark to develop an ambient image even at f4. Under those conditions, both shots should be the same. You might also try dialing in some flash compensation on your SB80; I think you can vary it from the standard fill value of 1.7.

Interesting stuff, anyway!

Tom

[dbara]

Glad to see someone else with the same issue. Yes it is interesting. I sent off the pics to FUJI and the dude I was talking to is suppose to send it up the chain because his S2 PRO did the same thing.

Thanks again,

Don

[Tom Nolle]

I've been fiddling with flash exposure issues for a little while now, and I'm getting ready to write a photo-tip for my website on it. The thing I guess I never realized was the way that ambient lighting and flash interplay to create the image. It's turned out to be a lot more complicated than I'd thought, largely because I really never tried to understand how to manipulate flash beyond the simple level. Your problem opened some new areas of exploration for me!

If I get a chance over the next week, I'm going to try to do a test shot in a very dark space to see if the results work out as I'd think they would. I'll post the result here.

Tom

[dbara]

Tom:

At about 10pm last night I turned the lights off & tested the S2 with an SB80dx flash. Same problem so it looks like ambient light has no effect on the subject in the pic. The F8 truned out darker than the F4, just like the other tests.

Let me know what your test shows. Good or bad.

Now I am really interested in seeing what Fuji thinks.

Thanks,

Don

[Tom Nolle]

I think I'll also try a more neutral background to see if there's any reflection issue playing into the picture. The thing that's interesting is that I have not had problems with flash exposure in real shooting situations. That makes me think that there's something about the setup that's causing the problem, and that the "something" isn't getting duplicated in the shots I take under normal conditions.

Tom

[dbara]

Tom:

With regards to your real shooting situations, do you use Aperature priority mode or do you use "P" mode when taking flash pics? Seems like when I use "P" mode the pics tend to look Ok. Probably because in "P" mode the aperature is usually around F4.5 with the 28-105 lens I use.

I found this issue shooting portraits in my basement. I wanted to have better depth of field on a shot so I went to F8 (from f4) and that is when it cropped up.

Anyway, let me know.

Thanks,

Don

[Tom Nolle]

I shoot everything in "A" mode, pretty much, flash or not. I use "M" in non-flash mode quite a bit, but not for flash. I almost never use "P" mode because I find I'll then forget to manage my shutter/aperture settings for my shooting.

With the exception of "snapshots" of family and friends, my flash shots tend to be shots of relatively distant things (10 to 50 feet) and thus I shoot at pretty wide apertures. My subjects are rarely close to background objects. If anything, I would probably tend to like the f8 version of the shots a little better because I usually try to expose my shots about a half-stop under. The S2 with RAW and EX provides good sensitivity compensation to +3 but doesn't go beyond -1, and color saturation is a bit better underexposed. For my taste, my f4 wall shot was just a tad light. Anyway, I'll run some tests by the end of the week-end and I'll post the results.

Tom

[dbara]

Tom:

Looking forward to your test results.

Don

[mike-C]

Mike Concannon, Galway, Ireland.
To the guy with the strange flash exposures.
Iv just bought a s2 and have been putting it through a series of tests as I could not work out why the results were not amazing from this camera. (except for the fine detail)
I fully agree that the camera seems to follow none of the rules of photography when working with a dedicated flash gun, in any mode including manual on the camera and auto on the flash (usualy safe flash mode on any camera). Go through a series of exposures with different f stops in a room with no ambient light, pointing the camera at a surface with average reflectance value, and the camera will give a different exposure for every picture. (The exposure value is theoreticaly the same for each shot)
I noticed this first when the first wedding I shot , all the indoor pics came out either over or underexposed.
If anybody has an s2 that does not do this then please post a message, I realy hope that this is a bad camera that I have managed to get hold of and that they are not all like this.
P.S. the problem happens with the built in flas also.

[dbara]

Mike:

I have doing a lot of testing with this issue and come up with the same results you do. I found this problem while shooting portraits and changing the F stop. Also, I had FUJI do the same test & their camera does the same thing so I do not htink it is your camera, I believe it is a flaw in the TTL flash mode. I will let ya know what FUJI tells me as it has been escelated up to the next level and I am waiting for a response.

Don

[mike-C]

Don.
Thanks for your reply. This problem identified, the next question to ask is, does the flash behave as it should for fill in? and are there more issues generaly with flash that we do not know about. I have had some strange over exposed results photographing large group shots against the church stonework using fill. I will be doing some fill in tests next week and will keep in touch with the results.

[dbara]

Mike:

Sorry, I can not help with the fill question as I generally do not shoot in that mode. Also, I have only had my S2 for a couple weeks so I would consider myself your basic beginner but I do feel the TTL is not working as advertised because what I am trying to do with the S2 I have done over & over again with my Nikon N80 film based camera and it works just fine on that camera.

Don

[sandman]

Dbara.don't try to compare the results of the N80 with the S2 same bodies but totally different exposure systems.

I had the european version F80, it's more like dealing with slide film. on trannies the lab compensates for exposure errors, the S2 does not. i'm like you..learning ..i've got a 80DX and found it's more critical than my old S28 when i was film based.

No answers ,just keep on learning.

[mike-C]

Just posting this to ask you all out there to do this test and please, please let me know that I have a unique problem or that all s2's are like this.
Set your camera A or P or S and to any exposure mode. Then with the flash compensation button up near the shutter release, or buy using the control on the flash unit, turn the flash output down by two stops.This should give a nice fill flash.
Now take a picture of any area outdoors about 6--10 square, say a piece of the front of your house, have a flash meter pointing back at the camera, preferably one that reads daylight and flash simoultaniousley.
Does the meter read one or two stops more light have fallen on the subject than the little green leds had said in the viewfinder as you were taking the shot.
PLEASE DO THIS TEST IF YOU CAN AND POST A MESSAGE.
I NEED TO KNOW IF I HAVE A ONE OFF. BEFORE ACCEPTING A REPLACEMENT from fuji.

[dbara]

Mike:

I agree that they are 2 different exposure systems but by definition, TTL is TTL and the camera, N80 or S2 should adjust the flash based on your F-stop setting to make sure the subject is correctly exposed. Also, when I take portraits with my N80 I specifically ask the LAB to do no corrections to my film. My film base pics are correctly exposed as long as I stay within the range of the flash and the S2 is not.

Don

[dbara]

Oops, sorry Mike, the last reply was for sandman.

[Tom Nolle]

OK, forgive the long post, but I've run my flash tests and I want to provide everyone as much input as I can.

I ran the tests using a tripod, an SB16b, the S2 set at ASA200, shooting at a non-reflective wood background with a colored object set against it. Subject distance 15 feet. Power for the S2 was the Quantum 2x2; fresh AAs in the SB16. For all tests, SB16 in TTL mode.

First test; S2 in Aperture Priority, matrix metering (flash should be balanced fill mode), shooting f4, f56, f8, f11, f16. The shot at f5.6 was approximately correctly exposed. Exposure variation noted at all other f-stops.

Second test; S2 in Aperture Priority, spot metering, shooting the same range of f-stops (flash should be in straight TTL). Repeating the same f-stops, with approximately the same results.

Third test; S2 in M mode, spot metering, same f-stops, same result.

Fourth test; SB16 in "A" mode with indicated f-stop of f5.6 set on the S2 manually. Exposure correct.

My conclusion is that there's definitely something hokey in the flash setup with the external flash. The f16 shots should be underexposed at that range (which they were) but the exposure on the other f-stops was within the guidelines for the flash according to the SB16 manual and the S2 manual.

Looking closely, both the subject and the wood backdrop are affected by the exposure shift equally.

I've not really run into this with my standard shooting because I tend to use the SB16 mostly to light critters that are some distance from the camera and set about f5.6 to get a compromise between flash range and depth of field. I can see it would be a pain for generalized use, though.

The "indicated" f-stop for flash A mode as I said above happened to be the f-stop that worked correctly. It makes me wonder if the S2 is not sensing the TTL setting at all, but I don't see how the flash would know to compensate by using its internal meter. At 15 feet with ASA200 set, my "manual" f-stop should be about f27, which clearly wouldn't have given me an image at all.

I'm baffled!

Tom

[Tom Nolle]

Sorry, forgot to mention that the room was very dim (by design); focus was achieved manually before turning the room lights down. Ambient light showed considerable underexposure (three plus stops).

Tom

[dbara]

Tom:

Well its is nice to know I am not nuts. Sounds like you get the same results I do. I must have taken 100 pics today even using a light meter. Aggording to my light meter the TTL mode is pretty close around the F4.5 range but with other f-stops it is all screwed up. Maybe that is why when it is in "P" mode the aperature is always around f4 or so. Whats really weird is that in "P" with the pop-up it sets the aperature to f3.5, with my sb80dx it sets it to 4.8. Like it knows what dind of flash is on the camera. Who knows.

I still have not heard back from FUJI.

Any chace you could also call this problem into FUJI? Maybe if enough complain something can get done. For what this camera cost the TTL should work.

Thanks again for all your testing.

Don

[Rockyw]

I did not try that test, but I really had problems in TTL mode for a while. Last wedding we shot almost everything in apature priority mode and TTL on the SB80. We set the F stop at 4 and 4 1/2most the time. For closer shots went up to 5.6 and as high as f8. We find you have to vary the f stop to the distance of the subject, and thats with +1 comp on flash on the camera. I know TTL should not be like that but if you just set on one f stop and shot the resalts are dark and light pictures.

[Tom Nolle]

I'll report my test results to Fuji this morning. I agree with Rockyw that the flash setting can be adjusted to get spot-on pictures, and in fact since my own shots tend to be at a consistent f-stop I've lucked out in most of them, but optimum use of TTL requires that...well...that TTL works!

Tom

[Tom Nolle]

I did see your post, Theo. I also found that "A" mode worked with the SB16, but I have some concerns that the reach won't be as I need for my animal shots, and that the wider angle of the flash sensor will mess up telephoto shots. I'm hoping to be able to get TTL to work, even if I have to do a couple test shots to get the setting correct. That (according to the manual) would give me more range and would help insure my telephoto shots are better exposed.

Tom

[dbara]

Theo, yes you are correct, "A" mode on the flash does produce better results thru the f-stop range at close range but the point of this post is that TTL on the S2 PRO does not work as advertised. I agree with Tom that it needs to work to produce better shots with telephoto. I do not want to worry about forgetting to sync both the camera and flash for my shots. One of the reasons I bought this, over others was because of the ability to do standard TTL (No pre-flashes) and I expect Fuji to make it work.

Don

[Tom Nolle]

I'm not sure whether the SB16 adjusts to the zoom setting of the flash head from wide to telephoto, but the rated maximum distance in TTL is considerably further than for Auto mode. The problem I sometimes run into in Auto is that when tele-flashing, the flash illuminates a bunch of stuff in the foreground and if the lens doesn't pick it up but the flash sensor does, the exposure is wrong. Since there's no way the flash could know what telephoto I had on (or about the S2 focal multiplier) my concern is that I'm going to get too much foreground lighting in the metering range of the flash sensor.

The truth is that this isn't a major issue to me, as I've mentioned before, and I'm of the view that something this major couldn't have escaped notice. There is probably something you need to do that I'm not doing, and I would like Fuji to tell me what it is.

Tom

[dbara]

Theo:

I have the SB80DX and according to the manual, "A" mode is not for Digital SLR cameras (although not perfect it seems to work better than TTL when going thru the f-stop range). The manual says I am suppoose to use "AA" mode but I can not set it when it is on the S2. Anyway, like Tom said, I can not believe Fuji would produce a professional camera that says is does support TTL but does not.

Don

[Tom Nolle]

I reported the issue to Fuji yesterday, and I'm awaiting a response from them on what the problem might be. I'll keep everyone posted.

Tom

[CaptJR]

I may be showing my ignorance here, but here goes. I have use the 'P' for some flash pictures. When taking pictures inside, with the flash, in the 'P' mode, I don't remember ever seeing the camera picking a Aperture setting anywhere near as high(closed) as F8. Maybe I wasn't paying attention though.

In the 'A' mode without a flash, as the aperture goes up the shutter speed gets longer. In the tests you all are doing, is the shutter speed getting slower as the aperture gets smaller, or is having the flash turned on holding the shutter speed at 1/60th?

If it is holding the speed to 1/60th, then you are expecting the flash to flash brighter and or longer to compensate for the speed not getting slower. Am I following this whole thing correctly?

JR

[dbara]

JR:

When not using the flash and the camera is in "A"perature priority mode yes in deed the speed changes as the aperature changes to compensate and TTL works Ok.

When using the flash and the flash is in TTL mode and the camera is in "A"perature priority mode we see the problem. The speed pretty much stays constant at 60 when the aperature changes so the flash output should increase or decrease with the changing of the aperature. This is not happening correctly when you change the aperature on the camera thus causing over/under exposure problems.

Don

[CaptJR]

I'm not trying to rock the boat here. I'm just trying to understand flash photography better.

Effective range? When I think of effective range, I think of 1 or 2 limits put on something where it will or will not do the job. One would expect somewhere in the middle of an effective range, there would be an optimal range. As you work your way toward the end of the effective ranges, I would expect to see changes in the effectiveness.

Using the very first post in this series, Don was using F4 and F8. He said he used those because according to the manual they would both be effective at 4 feet. If you go to the chart in the manual you will see that at 4 feet if he went one more stop in either direction F2.8 or F11 he would be out of range.

The way I'm seeing it, F5.6 would be the optimal range. F4 and F8 would be the ends of the range. Thus F4 and F8 are both suppose to be acceptable, but being on opposite ends of the range they would be opposite. One being slightly lighter than the optimal F5.6 and one slightly darker than F5.6, but both acceptable.

[dbara]

JR:

By no means am I an expert but here is what I believe effective range for a flash means.

It means 2 things:

a) What is the minimum distance that the flash will correctly light the subject (Will not over power the subject with its light).

b) What is the maximum distance that the flash will correctly light the subject (Will not underexpose the subject because it does not have enough power).

Now, when you factor in TTL (A function of the camera that adjusts the output of the flash to make a correct exposure) you should be able to change the f-stops on the camera, as long as you remain within the effective range of the flash, and no matter if is F4 or F8 the pics should look the same. There is no optimal f-stop because TTL adjusts the flash so each pic looks the same meaning each f-stop should produce the same pic.

Unfortunately the S2 is not doing that, it is grossly underexposing th F8 shot.

Hope this also helps.

Don

[Rockyw]

I believe also the S2 does very poor in TTL mode. To use this flash in TTL mode we have to be changing F stops all the time. f16 for a close hands shot, f4 for a shot of the first dance. The ISO and flash stay on the same setting but the flash cannot vary its self to the differant ranges. I also hope all this gets better. Good Luck

[dbara]

Hey Rockyw, call it in to Fuji as a problem. The more that call this issue in the better. Maybe Fuji will listen and fix the issue if enough of us call it it.

Thanks,
Don

[CaptJR]

Thank you

So the effective range is the beginning and end of where the flash can make the exposure just right. Not as what I was thinking.

You folks just don't know how stupid you made me feel. Now I've been doing weddings for a while now, but not nearly as long as most of you. (just officially in business a few months ago). I've been using digital for a long time, then Last year I got the Olympus E20N (my first SLR) and now I finally have the S2. I use manual mode most of the time. When I've been taking the formal pictures I've found it much easier to move the Aperture up or down a little rather than the flash compensation to get the picture a little darker or lighter. Stupid me, I didn't realize that it wasn't suppose to do that.

JR

[Rockyw]

dbara
I will call them today or tomarrow and talk to them also about this. I have tried the A mode on the SB80 and it works a little better but I would still have light and dark pictures if I didnt change f stops a lot.
When we use studio lighting in the church, we use manual, 60 shutter speed and ISO 160. We meter and set the S2 1/2 stop under the reading we get. The SB80 is set on manual also and set to 1/16 power, just to fire the whitel ightnings.
Ill post back what Fuji says .

[dbara]

Rockyw:

Thanks a bunch. A couple of us are waiting to here back from Fuji, I actually heard back from the second level but all he wanted to know was what my problem was. I told him & he said he would get back to me. That was 2 days ago & I am still waiting. How long can this take to test? Will post more info as I get it. The bummer was that he kept trying to point the finger at my sb80dx & I kept telling him it also happens with the internal flash and my second SB80DX. Not sure if that sunk in. Sure wish I was close to where they are so I could go over there and show them. Anyway, can only hope.

Don

[dbara]

JR:

Dont feel bad, it looks like that is what most of us were doing to correct the exposure. But yes, we should not have to do that and considering what this camera cost, TTL should work. If you have not called this in to Fuji, please do so. Maybe they will fix it if enough of us complain.

Thanks,
Don

[Tom Nolle]

I just got my first reply from Fuji, which was completely unhelpful; they told me to follow the instructions in the manual. I've made a more forthright response to their initial comment, which should (I hope) convey my lack of enthusiasm for a suggestion that my complaint clearly shows I'd already followed.

Re a previous question from CaptJR on effective range, the flash units fire a specific intensity depending on the flash tube, so to control exposure the flash unit varies how long the flash fires. If the aperture is too wide open, the minimum duration of the flash still overexposes the subject at close range, so pictures within a certain distance cannot be correctly exposed. The flash also has a maximum duration, so at distances beyond a certain point with too stopped down a lens setting, not enough light can be created to illuminate the subject. The effective range of the flash is the spread between the minimum distance the unit will flash at the smallest lens opening and the maximum distance it will flash at its largest (widest-open) setting.

Tom

[dbara]

Tom:

I am guessing you got an email back. If so that is the same garbage I got when I contacted them. I then called their 800 # an spoke to a warm body. I told him to get a S2 with a SB80DX flash and test it while I stayed on the line, reluctently he did. He was able to duplicate the issue so he sent the issue up the chain. 3 days later I got a call from the next person. I had to explain everything all over again (Like they do not talk to each other). He said he would test it (after trying to point the finger at the SB80DX). It's been 4 days now & I have not heard back from him. I will call them again on Monday. Try, try try.

The point is, you might want to call directly to speak to a warm body.

Don

[Tom Nolle]

Sorry to say, my next response from Fuji on this issue was even less satisfactory. In effect, they said that while they'd run the tests I outlined and gotten the same results (namely that TTL exposure varied by f-stop even though it's supposed to be metered through the lens and thus should be contstant within the allowed distance/f-stop range) it was normal!

My response was that their view was an embarrasment to them and the quality of the S2 overall; that there were any number of flash books available that would illustrate that TTL is NOT supposed to work that way. I told them I'd like them to buck the question up and provide a truly responsive comment, or tell me they stood by their original one--in which case I'd post it to this forum for you all to evaluate. I'm still waiting for their reply, and I'll post the response here.

Tom

[Tom Nolle]

That's pretty consistent with the tests that I ran, and that dbara also ran, Theo. The problem is that if we pick two or three lens openings that are within the allowed range for TTL, and if we have no ambient light (or very little) to mess things up, there should be no need to flash compensate with TTL metering for the flash set. The flash should be turned off on camera signal when the correct exposure is achieved. If you have to flash compensate, you are in effect changing the lens opening to secure the correct exposure. That's what manual mode is supposed to be, not TTL!

Tom

[Tom Nolle]

It's interesting that your "sweet spot" is about f4 and mine is about f5.6. It might have to do with the relative guide numbers of the flash units we're using. If Fuji doesn't have any constructive suggestions, I'll move to another test, which is to see if the sweet spot f-stop stays the same through the range of distances of the flash. If it does, then the problem's probably solvable by setting flash compensation based on what f-stop you have to pick (your test). If ot does not, then the thing's working exactly like manual flash.

Tom

[dhphy]

I bought my S2 back in February and was shocked at the ghastly colours and inconsistancy of the flash results with an SB28.
It was the fist time I had used the camera and it was a party indoors so everything was flash lit. The camera was on the default settings.

Using the SB28 with a F100 I have always got very good results, though I do appreciate that there will be some lab correction to negs.
These bad results from the S2 prompted me to do some tests as the camera was to be used for reportage work at weddings.

I did many tests photographing grey cards against neutral backgrounds with the flash in different metering modes at different aperature settings. ~All results were underexposed and also varied depending on aperture but interestingly not with the metering mode, at least on average scenes.
The best results were obtained from the SB 28 set in the flash guns "Auto mode". This should be the least consistant and least accurate method.
As I teach photography I borrowed the college S2 and compared them and got the same results.
The tests that various poster above have found are consistant with mine - ie the S2 does not do proper TTL

I posted my query on the dpreview forum and got many answers but not much help.
I felt I had explained the problem as well as possible but got many irrelevant and stupid responses, most suggesting that the S2 works better with an omni bounce? or a diffuser. This is true and it is a better light, but it is again not consistant and not what it is stated as doing.
They had all missed the point, which seems to be more focused here, that the TTL does not work as it should. That is my conclusion.
In fact you can get good results from the flash but only by constantly checking the histogram and retaking pictures varying the flash output. Not much help at weddings though.
I sent the camera into Fuji UK with all the test photographs in print and on CD with a detailed description of my findings.
(My findings were pretty much the same as others on this thread)
The camera was returned with no fault found.
In some respects I gave up on the camera as I found that this inconsistancy was too nerve racking, but have used it more recently for outdoor shots and use it with metz flash guns in the guns auto mode and allow a little underexposure to be corrected in PS
Would be interested in Fuji's response to the people who have posted to them.
Dave

[Tom Nolle]

It would be interesting to know if the N80/F80 had the same problem; in other words, is the problem in the Nikon body or in Fuji's electronics changes.

Tom

[dhphy]

We have an F80 and although I have never done extensive tests, it always seems to have worked as expected with TTL flash from an SB 24 and 28

There is definately something not quite right with the Fuji system.

It could be that Nikon are just trying to sell flash units with their D100 or it could be that they felt as they have stated that TTL, which for film cameras reflects off the film, wouldn't work reflecting off the CCD. That is the reason given by Nikon for the introduction of DX flash guns by Nikon
The D100 and S2 are based obn the same F80 camera body.

Fuji were (supposedly) able to overcome this problem, but it seems only half successfull. It does surprise me though. If it can get it right at one particular aperture, it is a straight forward mathematical computation to vary it for others.
Dave

[dbara]

First: I have a N80 and no problems taking flash shots with my SB80DX. I usually tell the lab to not make any adjustments to my negs and they came back supurb from flash shots with the SB80DX on the N80. The S2 sucks as far as TTL is concerned.

I can not believe FUJI does not seem willing to fix it, like U said it is a calculation. And if you have it right for one F-stop then the others should be a no brainer.

I take a lot of flash pics (99%) and I want it to work. I can not return the camera now so I am stuck with it otherwise I would have sent it back.

If the manual says it does TTL then why can we not force FUJI to fix it?

Wish I new a good lawyer!!!

There is no way I would ever recommend this camera for 2 reasons. 1: TTL SUCKS & 2: FUJI SUPPORT SUCKS.

Don


Don

[dhphy]

Quote from Theo Lumens
"I'm not entirely sure, but I think users owning DX flash units are having the same problems with TTL on the S2pro."

Yes I think the same applies to the DX flash guns on the S2.
It would be interesting to know how well it works on the D100.
I know someone with a D100 but they don't have the flash gun.
They also use their D100 with a Metz or Nikon flash in the flash guns auto mode.

It isn't surprising that the DX guns don't work any better on the S2. Whatever Nikon did to the D100 circuitry to necessitate a DX flash gun, Fuji didn't do it to their S2. I would imagine their is no benefit using a DX gun on the S2.
Using a DX gun on other Nikons F100, F80, F65, Iwould imagine the gun performs as a normal flash gun for that system.
Only when coupled with a digital camera with the relevant circuitry will the DX circuitry installed in the camera combine with the flash and do whatever it is supposed to do.
If the flash doesn't sense the circuitry it will perform as an ordinary Nikon.
Therefor the S2 not having the necessary circuitry because Fuji "overcame" the reflectance problem some other way, DX flash guns will link to the camera and perform as an ordinary flash.

Quote from Don

If the manual says it does TTL then why can we not force FUJI to fix it?

I think Fuji would say that it does do TTL but within certain limits, though I agree it does not do it in the way I have come to understand it over the years.
Getting Fuji to fix it would take more than one good lawyer and would if succesfull bankrupt them. It is not as far as I can tell a firmware issue which might be possible to correct, but a hardware issue that can not be corrected.

All the above is of course supposition.
Dave

It might be worth checking out these sites if you don't now of them

http://www.xanga.com/home.aspx?user=garyfong1

http://www.bythom.com/

Thom Hogan does a guide to the S2 that might be more comprehensive, that I understand can be obtained free (I think) for fuji users in the States, might shed some light! Sorry not a good joke but need to keep a sence of perspective.
Dave

[CaptJR]

As I posted in the following thread, a week before this thread started - http://www.s2pro.com/forums/showthread.php?s=&threadid=1239

When I got my SB-80DX, the first thing I did was read the manual. When I attached it to the S2 I wondered why the DTTL mode didn't come up. {(D) to mean digital} It is suppose to have two TTL modes, DTTL and TTL. So I sort of expected that something was not going to work as expected. I figured that the DTTL mode was only going to work on Nikon digital cameras.

I'll try to make everyone feel better. I've done a lot of reading on other forums, including Nikon digital forums. Conclusion - TTL mode doesn't work very well on digital cameras. All you read about is people having to correct manually for over and under exposed flash pictures.

I'm going to give 'center wieghted metering' a shot although I'm not expecting any better results. Then I'll probably go to the Auto mode on the flash.

JR

[Tom Nolle]

My problem isn't so much the TTL flash bug as the condescending replies I've gotten. Yes, Fuji, I know that opening the aperture makes pictures lighter and stopping down makes them darker--I know that with respect to ambient-lit scenes. That's not how TTL flash works, though.

I'm going to wait through this week to see if anyone gets a useful response, then start another series of tests to see what workarounds will be best for my operation. I agree with the comment that this problem is probably very fundamental to the integration of digital logic to Nikons and not likely one that can be easily corrected--if there's no simple setting to fiddle with. Maybe centerweight, or "S" exposure mode? Anyway, I'll post progress here as it happens.

Tom

[Tom Nolle]

I'm finally in contact with what seems to be the right guy/level at Fuji support, and I'm going to send him some of the test shots on the topic. More when I hear it...

Tom

[dbara]

Well, you are doing better than me. I can not get the guy from the second level to call me back.

Good luck,
Don

[Tom Nolle]

Just sent him the extracted exif-jpgs from the EX 2.0 converter. I sent two series, each shot at 15 feet with aperture priority, at f4, f5.6 and f8. The first series was shot with matrix metering and should have set balanced multisensor mode, and the second with spot metering which should have set straight TTL mode. Both series look about the same in the converter output. He's examining them now.

Tom

[dbara]

Tom:

Well, maybe I will have to eat my previous words about FUJI. Looks like you are working with someone that is at least willing to check out the pics. Hope it works out.

I just sent off another email to FuJI asking for an update. All I can do is wait.

It will ve interesting to see what your guy says.

Don

[Tom Nolle]

This guy seems good; no defensiveness and willing to run his own tests pretty much on the spot. He's also willing to wade through six files of about 400kb each to look at the results of the test, and he's followed up on my questions on just what he needed very quickly--like in minutes. I'm hopeful.

Tom

[Tom Nolle]

Working with the Fuji guy, we found something interesting. On my SB16, the flash exposure that is correct in terms of set f-stop is the one that the camera picks automatically if you put it into "P" mode.

IMHO, what appears to happen is that the camera wants that optimum f-stop set and does its calculations based on it whether you've set it or not. That might also explain why fill flash works, because the metering is setting f-stop based on ambient light and if you're in A mode you have to fiddle f-stop to get the scene exposed properly.

Still working things through with this guy, but he's been in touch with Fuji Japan, Nikon, etc. so he's not sitting around. We may not like the response we end up with here, but at least we'll know what's happening.

FYI, his preliminary test in his studio with the SB28 don't show the variation I found with exposure on my SB16. Nikon, not surprisingly, says that I should just buy a new flash. However, a couple of you on this forum have tried this with the SB28 and say it doesn't work either. Could it be that he has a digital version or something?

Tom

Tom

[dbara]

Tom:

I have tried it with the Internal, SB80DX, & SB50DX all producing the same results.

Don

[Tom V]

I haven't paid much attention to the TTL flash metering on my S2 until I started reading this thread. I was pretty proficient with TTL using my F100 or N8008s with either my SB-25, SB-28, or both at once.

I have used both flashes at once with the S2, in a manual situation, and all turned out as expected. Using both flashes in TTL mode was entirely unpredictable on one commercial product shot (with the client looking over my shoulder at the LCD), so I made sure I got something usable in RAW format and pronounced it complete.

Today I did some simple tests using the built-in flash, and either of my Nikon Speedlights, and found the TTL system to be totally screwed up. The only shots that looked like appropriate, consistant exposures where made with the flash set to A, and the camera aperture set to the indicated aperture via camera exposure mode A or M.

The way TTL works (doesn't work) on the S2 makes no sense. Clearly, Fuji has screwed this up royally.

Overall, I still like the camera. Its biggest flaw is the flash systems: TTL, PC Terminal, slow sync speed, & Fuji not understanding TTL at all.

[Tom Nolle]

In what may well be the last word on the topic (at least as far as my own equipment tests can carry it), here's what I've heard.

First, Fuji says that the camera-to-flash interaction in P mode is more sophisticated than in other flash modes, meaning that more information is exchanged.

Second, they've run the tests I ran using my distance settings and using my EXIF data to set their test camera to match my conditions. The tests used an SB28 and a Metz 40MZ-3i, and in their test only the first shot at widest opening and the last (very stopped down to f16) were noticably off exposure. So in their studio, at least, they have two flashes that work with the camera.

I have absolutely no reason to think this guy is scamming me; he's been more than helpful and responsive, and he's contacted Fuji Japan and Nikon to try to get the details on how my SB16 might work with the camera. Thus, I have to accept that there are at least some flashes that work with the camera.

OK, that's the substance of what he said. Now I'll give my interpretation. I think that there is indeed an undocumented linkage between flash exposure and "P" mode. It's too much a coincidence that the aperture P mode selects happens to be the one that produces the best results when the aperture is set manually. The camera could be getting the "automatic" setting from the metering system even if that setting isn't the one the user sets manually.

Second, I think that this condition of "phantom f-stop" setting doesn't happen on all the cameras. Perhaps it's something that a firmware change fixed. I have a very early S2 (August of last year).

Third, I think that some flash units (like my venerable SB16) are more likely to show the problem than others.

Finally, the revelation that the "P" mode setting will work essentially solves my own problem since my shots with the SB16 are relatively long shots at twilight animals that require a large f-stop to illuminate the subject at the range involved, and since P mode will set that open f-stop under those shooting conditions anyway.

I can't test further to prove or disprove any of these theories because it's clear that the SB16 (which is over 15 years old) isn't the right flash for this camera, and that's not surprising. I'd encourage others who have more modern units, and in particular the SB28 or SB80, to continue to pursue their tracks with Fuji support. I find that despite the theoretical value of calling to get a real person, the email route has worked best for me because sometimes the real experts are people who you won't get on the spur of the moment in a call.

Tom

[dhphy]

Having a little bit of spare time I decided to do some further tests with the S2 and SB 28.
Firstly though someone on this thread seemed puzzled that the S2’s pop up flash in Program Mode (P) showed a different aperture setting than when a Nikon flash was connected.
That is so. I think that the camera communicates with the flash and if a more powerful model is connected then the aperture selected by the camera is smaller. In the same way, the aperture selected in (P) mode gets smaller as the ISO rating is increased. This does make sense.

In Aperture Priority mode (A) when a Nikon flash is connected the shutter speed defaults to 1/60 second if it senses that there is no ambient light assuming that the flash gun is turned on. That makes sense too.

When the flash gun is in Auto mode (A) i.e. you have to set the aperture on the flash gun to correspond to the aperture set on the camera, the camera and flash still exchange information. Changing the ISO on the camera changes the ISO on the flash. However, shooting a stop over and a stop under i.e. f5.6 set on camera and f8 set on flash produced an overexposed picture.
F5.6 set on camera and F4 set on flash produced an underexposed picture. Therefore the flash gun in auto is working independently of the camera and the light sensor on the flash and the flash circuitry only is quenching the flash when sufficient light has reached it (This is by the way 25 year old camera technology).

In these tests, the maximum and minimum distances did vary depending on what setting the zoom lens was on. However on the Program Mode tests where the camera selected F4.8 the distance range indicated was between 1 and 9 metres.
Manufactures of course assume that there is a reasonable amount of ambient light, but in these tests there was none.

I used a zoom lens to try as best as possible keep the subject in the frame consistent for each distance.
The backdrop was blue/grey, but about mid tone.
Although not here, I have done other tests with flash and the different metering settings and found no appreciable difference in them.
Also I have tried to keep the scene “average” and neutral. However my experiences with general flash photography with this camera and flash combination are the same as this test.

All the photographs are resized and compressed for the web but otherwise unaltered.

S2 in program mode set at ISO 200, matrix metering and SB28 at 6metres 4,3,2, and 1metre.
http://www.btinternet.com/~dhphy/Web/ft1.htm

My conclusion – All under exposed by 1 to 2 stops except at 1metre but at least reasonably consistent.
I have generally found that by upping the compensation in the flash causes saturated colours and hot spots when using this mode so I generally leave as is and expect to have to alter them in PS.


Camera 4 metres from subject in aperture Priority with varying aperture.
http://www.btinternet.com/~dhphy/Web/ft2.htm

My conclusion. All pictures get progressively darker. You wouldn’t want to use this mode.

Camera and flash working independently. No TTL, Flash guns light sensor and circuitry determine exposure.
http://www.btinternet.com/~dhphy/Web/ft3.htm

My conclusion. Pretty consistent and pretty accurate. Of course with that consistency you can put a bit extra or take a bit out of the flash on the compensation controls on the flash gun or of course fine tune it on the aperture.

Not that I need to, but you could shoot up to 1600 ISO with flash in this mode.
I did a few just around the house, not posted here and much more accurate than TTL

Dave

[Tom Nolle]

Nice summary! Your work seems to parallel that of some of the other SB28 users who've contributed to this thread. The problem is that Fuji, in their studio, runs the same aperture priority test and reports that the results are the same through the range of supported f-stops, using both the SB28 and a Metz unit. I don't get that with my old SB16, but pretty much everyone agrees that I should send that to the Smithsonian anyway.

The thing is, aperture priority is supposed to work and in Fuji's test reportedly does work! This is what makes me wonder whether some of the older S2s have a different firmware level or hardware wiring in how they handle flash.

I did a few additional tests in P mode, and I also found that the camera tends to select f4.8 as the optimum aperture for ASA200 shots. It makes me wonder if for some reason Fuji is picking up that f-stop as the computed flash f-stop even if the user, in aperture mode, sets something else.

Tom

[Tom Nolle]

I'm really sorry to hear that, Theo! I'd hoped that there was something I was running into because my S2 is fairly ancient by S2 standards. I'm completely at sea (apologies to CaptJR) with respect to why the Fuji test conducted in their studio by somebody who demonstrably took a heck of a lot of time to get the exact details of what I'd done and reproduce them, would produce correct results--and with two different flashes!

I never understood quantum physics. I never understood politics. I never understood women. Now am I to add yet another dimension--S2 flash photography--to my list of shortcomings???

Tom

[dhphy]

I wouldn't consider mine to be that old, bought in February.
It is fairly easy for anyone with an S2 and SB28 to do the aperture tests. Average scene at night in doors and about three to four meters, direct flash of course as that is what we are testing. Hopefully some others will try it.
I know every one suggests using bounce adapters or other forms of diffusion. I on occaisions use a small soft box. All of which are simply dropping the effective power of the flash by about 2 stops. That may be a workaround and may help the situation, but the point is, does the camera do TTL and I'm pleased that this thread is addressing the issue.
When I raised this issue on another forum

http://forums.dpreview.com/forums/read.asp?forum=1020&message=4920618

although to do primarily with the awfull skin tones I was getting if I upped the flash, nobody understood or was bothered to look at the issue and as everyone seemed happy with the results I assumed that I had a rougue camera despite it going back to Fuji and being pronounced OK.
Knowing what to expect in P mode (1-2 stops under and raised in PS) and using the flash in Auto allow me to get the results I want. Ultimately it doesn't matter whether it does TTL or not as far as I am concerned because I think I'll tend to use the flash in auto more often, particularly as I tried some fill flash shots in P mode and they were crap also.
Dave

[Ptwig]

SB28 Integration Problems ......

Not so long ago I purchased an S2 Pro camera in Australia and, although I'm really impressed with its performance, I also have a problem when using an SB28 flash unit with it. Set on TTL, and using A S or P modes on the camera, the images are consistently two or more stops underexposed.

The only useable set-up I've discovered so far is to set the flash to 'auto' (A), selecting an appropriate aperture (based on the output power/ISO combination) and then opening up the camera's aperture by half a stop more than the aperture setting of the SB 28.

Obviously this is a cumbersome process considering the sophistication of the camera itself, so does anyone out there know if there's a firmware update available that would fix this problem please?

[dhphy]

Quote from Tom

“Finally, the revelation that the "P" mode setting will work essentially solves my own problem since my shots with the SB16 are relatively long shots at twilight animals that require a large f-stop to illuminate the subject at the range involved, and since P mode will set that open f-stop under those shooting conditions anyway.”

Tom, I wouldn’t presume to go about telling you how to do your particular photography and have no wildlife photography experience, but even if TTL worked as I believe it should, you may still get spurious results. I’m thinking here of the two extremes, white animal or black animal at twilight.
Assuming that you have some time to set up your shots wouldn’t it be worth doing a few distance tests with the flash on manual. So for example subjects at 15m require F2 subjects at 12m require F4 etc. Manufactures assume some ambient light for their flash guns when they are rated and there would be little of that in your type of photography.
Backgrounds that are too light or too dark will also fool the best flash exposure system and as I believe, the Fuji doesn’t have one of them.
With a bit of practice and a few tests, you’d get much better flash shots as nothing in the subject would influence exposure.
Dave

[dhphy]

Quote from Ptwig

“Not so long ago I purchased an S2 Pro camera in Australia and, although I'm really impressed with its performance, I also have a problem when using an SB28 flash unit with it. Set on TTL, and using A S or P modes on the camera, the images are consistently two or more stops underexposed.”

I’m not finding two or more stops underexposure but nevertheless I do believe that the system is underexposing. When I did some research before I came across something that suggested that the Fuji maybe underexposing by design in P mode or any of the other modes. I can’t remember where I saw it but it seemed to suggest that it underexposed because it is possible to correct for underexposure but not for over. If the camera was designed to hit the correct exposure there is then no latitude for the system. By that I mean if it gets it only a bit wrong on the over side the picture is lost.
Do you have the facilities to post your findings on the web? It would be useful to see the results and the settings.

As an aside and it has always happened with camera equipment and other hardware, and I continue to find it peculiar and unsatisfactory, is that the people who have designed the equipment who obviously don’t just cobble the stuff together, must test it and refine it and must set specific parameters within which the equipment will do what they designed it to.
These people must know every peculiarity of their creation yet don’t seem to make this information available. It is left to others with no connection to the manufacturers to test the equipment and make guesses on how it functions based on insufficient information.

Dave

[Tom Nolle]

I agree with your comments about there being holes in the "P" solution, Dave, but I meant that with the P-and-f4.8 approach I could get what were pretty well exposed TTL-based shots. If I have an unusual condition of background-versus-subject reflectivity or something, I'll blow the picture with this workaround, but I'd have blown it if the S2 TTL flash worked in what I'll call the "correct" way, too.

Sometimes I can get a setup shot of an animal; polar bears at night from a tundra buggy aren't afraid of the flash and I could get a couple of shots to check histograms. For light animals (polar bears, arctic fox, etc.) I'll usually underexpose a tad to avoid burning out the highlights, and spot meter in any case.

The trouble with manual flash for night critter shots is that the distance-f-stop relationship has to be manually set and figuring even distance is a challenge because it's usually too dark to autofocus. Unless I have enough ambient light to manually focus accurately, check distance, set f-stop, then shoot, I'm going to miss something.

I'm making another run to Churchill in the late fall, and I'll be trying out some of the workarounds live there, but even before that trip I'm going to do some night shooting locally to see just what I can expect. Thanks for the comments!

Tom

[Ptwig]

Thanks for the suggestion dhphy, a good idea. Based on the old adage of "Don't tell me, show me!" I've put together a thumbnail sheet of tests I did an hour os so ago.

I think you will all find the results self explanatory in view of the fact that I've made no changes to the colour or density of any of the images - they are just as the camera recorded them (apart from pixel dimensions that is).

What do you lot out there think of the S2 Pro's TTL flash integration now?

[dhphy]

Ptwig
The set of photos with the window uncovered, were you attemting to test the flash "fill" ability. By that I mean trying to balance the interior levels with the flash light to the exterior levels.
If so, the metering mode selected, which you don't mention might influence the exposure. If the camera was set to spot or centre it might not be reading the darker areas and the flash may be, in that situation functioning correctly. If it was set to matrix it should do a better job of balancing than is shown in your pictures.
I only mention about the metering mode as pertenant as someone else will.
The photographs without daylight do show underexposure, and although you have said elsewhere you think over two stops, I think it looks more like 1 to 2. Nevertheless, the results look like what I would expect (Sorry I have come to expect) to get with my equipment which is exactly the same as yours. Thank you for posting them.

Quote from Tom
"First, Fuji says that the camera-to-flash interaction in P mode is more sophisticated than in other flash modes, meaning that more information is exchanged."

By this does the Fuji rep mean that in "P" mode more information is exchanged by the camera and flash and therefor is going to be more accurate than in "S" "A" of "M" modes.
OK that is quite possibly true, it would be nice if that were made clear by Fuji.

Quote from Tom
"Second, they've run the tests I ran using my distance settings and using my EXIF data to set their test camera to match my conditions. The tests used an SB28 and a Metz 40MZ-3i, and in their test only the first shot at widest opening and the last (very stopped down to f16) were noticably off exposure. So in their studio, at least, they have two flashes that work with the camera."

Tom I agree that he does sound like he was trying to be as helpful as he could. The tests I did in aperture priority however did show as you can see above that there is significant diference between shots at the same distance and different apertures using the SB28.
I find the statement that "P" mode is more sophisticated and there is more exchange of information to be quite telling. Does that mean we can expect more consistent exposures in "P" mode and less accurate in others. That is my findings (in "P"mode they are consistantly underexposed by about the same amount). Yet the Fuji rep in what I assume were aperture variation tests in TTL said that there was consistancy except for largest and smallest apertures. Does seem a contradiction here.

I would encourage others to post controlled examples of photographs in these various modes.
It would be interesting to know from Fuji exactly what the basis and rationale is of the TTL exposure system. Personally I cannot believe that it is a design fault but I do believe it underexposes in "P" mode and is inaccurate in the others.
Maybe if enough people gripe, the S3 might do a better job (be full frame with at least 250 second flash sync 500 would be better)
Dave

[Ptwig]

Dave,

Yes, sorry I should have mentioned the exposure mode, it was matrix and not spot or weighted. As you can see the ambient light/fill shots are deficient in proper fill, and the flash only light is (I agree) about 1 to 2 stops lower than it should be.

The first set was intended to test the fill balance level capabilities and I must reflect my disappointment at the results. When I use the same SB28 with my Nikon F5 the fill is always perfect in all modes. It's a great shame that Fuji haven't also come to grips with this concept, although perhaps all is not lost if they front up with a 'firmware' solution?

Peter

[Tom Nolle]

I agree with you, Dave. Your tests show what my tests showed, almost exactly. Yet I'm reluctant to conclude that my Fuji guy, after going to all the trouble he went to, then made up results. I have to conclude that his tests did in fact work. That means that there are some S2s that work, or some ways to set an S2 that makes it work, or something.

On the one hand, I'm tempted to run out and buy a nice new Metz 54. On the other hand, I'm half-sure it would do exactly what my old SB16 does. Maybe I will arrange to borrow a newer unit and run another set of tests.

It was my guy's conclusion that if I got a flash that he had and ran the same tests, and if they still showed inconsistency, I should return the camera for repair.

Tom

[Ptwig]

Dave and Tom,

I think I'm onto something, although it's a bit early to confirm it yet.

Try setting your flash compensation to 1 EV (Page 67 of the manual). I've just done a couple of quick tests on flash only, as it's now night-time here in Australia, and the results look very good.

Tomorrow I shall try the same setting with fill flash to see if that also improves the situation.

Peter

[dhphy]

Peter
You can set +1 on the camera and also on the flash (SB28)
The camera and flash only allow +1 but if you set it on both you get +2
That might help in "P" mode although when I tried it it seemed to muck up the colours. Won't be of any help in "A" mode because, at least in my findings, the exposure varied with each aperture.
I was interested in your comments about your F5 managing balanced exposures as in your tests and the S2 pro not.
Having gone this far I'm inclined to put a roll or film through an F80, what the Fuji is based on and an F100 to see how these cameras handle the same situation.
Dave

[Ptwig]

Dave,

Thanks for those tips, I'll have a play around to see what transpires.

It'll be interesting to see what results you get from the F80 plus SB28.

Peter

[Ptwig]

Well it has been officially confirmed at last, the SB28 is NOT fully compatible with the S2 Pro (read following email from Hanimex Australia):

Dear Peter,

Thankyou for contacting Hanimex Australia.

The Nikon SB28 is not 100% compatible with Digital SLR camera's.

This is why you are finding differences in your shooting log. Digital SLR cameras require a different type of flash gun e.g. the Nikon XD-50. (I think she might mean SB50DX)


The Nikon SB28 that you are using does not support the TTL function, however it will work when the flash is switched into Auto or Manual modes.

Donella Stroh
Digital Help Desk

Hanimex Pty Ltd
114 Old Pittwater Road
Brookvale NSW 2100

[dhphy]

Peter.
I assume Hannimax are distributors for Fuji or Nikon in Australia.
Not quite the horse’s mouth and I think a misleading statement by your correspondent.
The part about the SB28 not being wholly compatible with the S2 is probably correct.

However their statement about digital cameras needing DX flash guns is I think very misleading and sounds like the sort of statement you’d get from someone who knows the words but don’t know what they mean. I don’t think the DX series of flash guns are going to be any more compatible with the S2.
Others on this thread have commented that the problems experienced with SB28 and other non DX but integrated flash gun users are being experienced by users of DX flashes.
I suggested above that as the Fuji is not designed to work with DX flashes there is likely to be no benefit in using them. If I thought a DX flash gun would solve the problem I would get one without hesitation.

I would be interested to know if you have done any tests with the F5 along the lines of achieving balanced flash and daylight.
I did some more tests at home, smallish room, white ceiling and light walls and the performance of the camera in “P” and “A” modes was much better. Also put a roll of film through an F80 and F100, same ISO and same settings and distances, but yet to develop it.
Dave

[Ptwig]

Have explored another of the issues raised by other members of this forum, namely that there is a colour shift when using fill from an SB28 flash on the S2 Pro (compared with straight ambient light).

The attached pictures were taken sequentially .... one with flash, the other without. The colour-temperature was set to 'daylight' in all cases. The images that look as though they were shot with tungsten are those where fill flash was used?

Peter

[dhphy]

Peter
In the manual it says at the bottom of page 71

"When the flash fires, white balance settings other than AUTO and CUSTOM are ignored"

What I understood that to mean is that if you have daylight or tungsten setting on the white balance and then connected a flash, that WB setting is ignored. But ignored in favour of what?
I assume auto, but I have never tried testing it.
Very ambiguous and very unhelpful.
Lets see some comparisons with the flash and the F5 Peter.
Dave

[Ptwig]

Dave,

These images were taken with the F5, using SB28 fill.

Peter

[dhphy]

Peter
Very nice pictures and the fill flash looks spot on.
You might be interested in my latest tests
Photographing in a smallish room with white ceiling and light walls.
The Fuji S2 did quite well but still seems to be a general 1 stop underexposure though not by as much as the tests I did in the studio posted previously on this thread.
That might be due to more reflective surfaces.
First four photos are matrix metering, P mode at 4, 3, 2 and 1.2 metres
Next 6 are F2.8 to F16 at 4 metres in A mode
Next 6 are F2.8 to F16 at 2 metres in A mode

latest tests are here
http://www.btinternet.com/~dhphy/Web/ft4.htm

Whats really interesting is the shots with an F80 and F100.
The same 16 photos were taken with the same settings as above using the F80
Then rewound and the same film put through the F100. This eliminates variations due to developing /printing.
It is difficult to see from the scanned contact sheet but the F80 is definately darker in P mode than the F100 by about a stop.
The F80 is at least a stop underexposed and is less consistant than the F100 in A mode

http://www.btinternet.com/~dhphy/Web/ft5.htm

F80 body about £350
F100 body about £850

Apart from a tougher chasis and 1/250 flash sync, perhaps the extra £500 buys a more accurate (professional) metering system.
The really definitive test would be to use slide, but I think I'm pretty satisfied that the S2pro is not going to give me the consistancy and accuracy that I got on neg or slide with the F100.
Dave

[Ptwig]

Dave,

An interesting set of experiments you've done. However I did note there were variations between most of the exposures shown on that contact sheet, even those taken by the F100.

I copied the sheet into Photoshop and did 'K' readings (F8 key) of the 18% grey card that your model was holding up and these displayed significant differences. Perhaps we're trying to achieve the impossible with the equipment. After all the SB28 does alter its angle of coverage depending on the focal length setting of the lens. Also, with matrix metering, any slight movement of subject or camera will pick up different relective elements within the scene, which would change the flash output proportionally.

I must say you've got a very patient model there.

The F5 shots I'd uploaded to this forum yesterday were all taken on Provia F trani-film as I considered they would have less variables than a neg/print chain.

Peter

[dhphy]

Peter I agree that the circumstances of these tests are far from ideal and not particularly controlled.
But I think they do show that the F80 is about a stop down on the F100 in A mode although both systems do vary at different apertures. This of course could be within the parameters of both cameras. The F80 shows about 1 stop down on the aperture tests at both distances so at least it is consistant and therefore could be compensated for.
I don't expect miracles and do accept varying exposures evan with TTL.
I've never used the F80 for on camera, flash lit reversal but have always been happy with the results of the F100 with flash although I don't do much slide stuff now.
Film asside and back to the S2, at least it is possible to review any flash photos taken so at least there are no unpleasant surprises.
I think I'd be inclined to only use the S2 in P mode and might look at the other metering modes.
I still believe as you and others have found that it is generally underexposing with flash in TTL and possibly using the Flash Auto system might be the prefered and more consistant method
of flash photography.
Dave

[Ptwig]

Dave,

I fully agree, in fact from now onwards I shall adopt those strategies with the SB28.

I'm tempted to go out and purchase a dedicated digital flash unit, although I'd have to see some convincing results first before splashing out the dosh.

Peter

[easternherp]

I think that everyone has missed the point from the very first post and that is that even the built in pop up flash doesn't work correctly. Has anyone asked Fuji to test that?
In the manual on p68 it says that the SB28 is compatible with this camera in ttl.

[Tom Nolle]

Pop-up flash should work in all modes, I agree. However, the reason I used an external flash in my own tests was that I use an external power pack and thus don't have the 123s installed. I can't flash with the pop-up under those conditions.

I'm not sure that the tests that we were doing would have been as easily conducted with the pop-up flash, anyway. The guide number on the internal flash is low, which means that the range of f-stops that it will support is also pretty thin. The difference in exposure that a bunch of us were noting is easier to see if you pick two different f-stops relatively far apart but still within the legal range of the flash unit. It's easier to do that with a powerful external.

I had a long dialog with Fuji on this problem, and they claim that on their camera the two different external flashes they tested an the pop-up all worked. I don't have either of the two they used, and I haven't gone back to repeat my own tests with the internal flash because I can't use it.

Tom

[KeithM]

I did a quick trial last night using the pop-up flash ( I don't own a compatible external flash unit ).

It seems to work as per the manual when engaging TTL mode by using spot-meter or manual mode - it gave consistent results when obeying the table giving distance for Aperture vs ISO rating.

( I set a number of apertures and shot a test subject over the ranges given in the table - exposure held good within the distance limits )

Given that the distance ranges in the table are roughly 3:1 ratio this would require about 9:1 ratio in flash power ( just over 3 stops ) and given that user selectable flash compensation available is 4 stops ( -3 to +1 ) this would seem about right.

My initial thought would be that no matter what mode you're using, the pop-up flash can only vary it's output over a 4 stop range so given a greater range of distances, you need to use an appropriate aperture to get within the flash's compensation range.

This does create a problem - when the TTL mode is used in 'A' or 'M' exposure mode ( which is when you want to control DoF )
you're constrained by subject distance to get correct exposure. The only variable is ISO which allows some variation in aperture,

I guess you could use an ND to allow wider apertures !

I have yet to try the balanced flash fill modes. The lens I'm using is a Sigma 17-35 D-type so it should offer all modes.

I have no experience of any other flash units or TTL flash metering modes. What I did was based on the S2 manual and my old understanding from using dumb flashguns many years ago. Please accept this input within these limitations ! :)

K.

[puglover]

Ok, I am going to show my ignorance, but i need help! Does the fill flash work at all with an external flash? I am using the SB80DX with the S2 and i am having absolutely no luck. Here is a tragic example of the wedding i shot last weekend in mexico. I finally had to just turn off the flash completely and deal the extreme backlighting issue and shadows (4pm wedding, sunny day). Since the flash won't sync at higher than 125/s what do you do? This was at F 6.7, 90/s, ISO 100, centerweighted metering (i tested this with the matrix metering and spot metering and same results). I know that this flash is incompatible with the S2, but what has anyone figured out about using it in harsh light for fill? Is it possible?

[puglover]

And just a side note, i realize at say F 13,16 this probably would have worked - but shouldn't i be able to use the fill (if there really is one...) at 5.6 or 6.7? Or is the problem that the camera won't sync at higher shutter speeds?

[Rockyw]

The SB80 works well outside with the S2. We use ISO 160 and P mode. Yup! P mode, it's the only time we use Program on the S2. Set the S2 to matrix metering and the SB80 to TTL. We use this for all outdoor shots. Try it, youll like it :)

[smunky]

I agree, and it works best with a bracket. Are you using a bracket or the flash directly on top of the cam?

[Rockyw]

Hi smunky If that question is to me, yes I use a Stroba-A-Frame ProRL. If it's to puglover, I agree any flash needs a bracket. It stops the horrid red eye and gives the image a better look as well.

[puglover]

well, i do have a bracket - so my only option is to use the 'fill' flash at an f-stop of 8 or more? that sucks, pardon my french. It would be really nice if they would come up with a solution to this highly annoying problem at Fuji. I love wide open lenses and do most of my work at very shallow DOF's (2.4-4) - i guess i am going to have to change my ways. double bugger.
:mad:

[ballroom_boy]

When I want to use a wide open aperature for fill flash on a bright day, I change the ISO to 100 and use ND filters (2 or 3 stops) - that way I can control the aperature more effectively. The highest flash sync speed on the S2 - until Fuji bases their future products on a pro-Nikon body (like an F5 - what a dream!), we have to live with that too.

Hope this might help a bit.

[Rockyw]

In P mode the f stop is normally around 4.8. You also can in P mode set the f stop you want, that gives you the option to control DOF. You might the next time also try camera manual and 11 or 16 for this type of shot, 125 shutter, ISO 160. Set the SB80 on A mode and a stop or two lower on the f stop than the camera ( flash set at f5.6 or 8) and we get good results. To get good images the S2 is in many Situations is not a point and shot camera. It takes time to learn it, good luck.

[puglover]

I never expected it to be a point and shoot - however i did expect the TTL and fill flash it advertised to work properly...so, i will learn to work around it.

[puglover]

And, when i went out earlier to test the camera - on P mode the (flash on...) with a shutter of 125 the F-stop was F-11. So maybe the ND filters are the way to go.

[Swampy]

Absolutely. If you have an extremely bright background, the camera will not be able to handle it without jacking the f stop way up due to the relatively slow sync speeds. Filters are going to be your only saviour until you get a new camera that can handle the higher sync speeds.

[ballroom_boy]

Hi gang,

I was lurking at Gary Fong's site and came across this post. I know the S2 doesn't do true TTL as advertised, however, this may explain when some of you get "blown out results" with external speedlights in P mode. I haven't tested this but I hope this helps.

"Just one more bit of info for those who do not use the Lithium batteries in the S2. We have found from our last wedding and some testing afterwards, that the lithium batteries do more than power the flash and aid in focusing...they also drive the functions of your dedicated flash! If you don't have the lithiums in there, your TTL will not work properly and you will find that your subjects get blasted out. We learned that the hard way, but luckily, we shoot film as well (until we get all this "learning curve" stuff out of the way).

[puglover]

well if that is true - Fuji should definitely have that in the manual. Extremely important information... of course, I have shot both ways, with and without the 123's and I don't remember it being much better with - but i'll have to try again. I did however use the P mode on camera and TTL matrix on my last wedding (night shots) and it worked much better. I didn't get to pick my apertures, but it kept the blow outs from happening - unless i got a little too close...

[Ansch]

Hi Puglover,

So the photo you posted was taken in M mode 1/90 f6.7?

IMHO, the issue with that photo is more a case of over exposed background than problem with the flash. What should happen is that you'd set the exposure to the background so that it is not blown out, and logically with that exposure the people would be underexposed because they have their back to the light. This is where the flash comes in. The 3Dmatrix fill flash should then calculate the amount of flash required to properly light the people. (the fuji system might not provide enough light because of the ttl problem which I can reproduce)

And the end result should be a "balanced" photo with the people and the background properly exposed. Yes, this might mean you're shooting at 1/125 f16. To enable yourself to shoot at f5.6 or lower, you need to reduce the light that reach the camera with filters as mentioned by the others. That is the limitation of the S2, based on the F80, it doesn't have the FP function that the F100 has.

Hope I haven't confused you more :)
Cheers..

[dhphy]

I haven't posted anything recently but have been watching these threads carefully.
I don't use TTL any more but use my metz and sometimes the SB 28 in auto mode with very good results.
I have been pondering this ND issue for fill flash outside.

If for example you are forced to F16 at 125 because of bright conditions, realistically the flash is only going to be effective up to about 2metres. Maybe a bit more as it is filling and probably blasting at full power, but if it is out of range the fill might be better if it is reduced due to insufficient power.

If you put a ND filter on to reduce the aperture to, say F8, the flash will still be operating as if the aperture were F16 as it will also be reduced by the ND filter.
I can't see a benefit as far as flash is concerned although the reduced DOF might be beneficial.

FWIW I use the metz on auto two stops down from the main exposure for outdoor fill flash. This works OK.
Dave

[Ansch]

Hi Dave,

If you put a ND on the camera so that it meters at f8, the flash should know that the camera's on f8 if it's on ttl mode. If you're operating in M mode on your flash then you should be setting the flash to what the camera metered at anyway.

Cheers.

[dhphy]

Ansch
I agree
The Flash will know that the camera is on F8.
The flash will fire a burst of flash and the cameras circuitry will quench the flash when it has received enough light from the flash to correctly expose the film at F8.
However, the flash light has to continuing burning to overcome the 2x ND filter.

My point is that the flash has to fire the equivalent output of an F16 burst of light, because 2 stops of the flash light are absorbed by the ND filter.
I can't see how it can work any other way and therefore can't see the benefit of an ND filter in the context of trying to balance daylight and flash.
However I would be interested in others comments particularly if I have missed something obvious.
Dave

[Ansch]

you're right Dave, the flash will have to fire at a higher power to overcome the ND. The ND is purely for allowing a wider aperture and not to balance anything. The balance comes from exposing the background correctly and using the flash to fill in your subject. This is the general idea anyway but in real life, the limitation would be in the contrast between the background and the subject, and also the GN of your flash.

you're expecting the flash to compete with the sun afterall ;)

Cheers.

[rsvpgrfx]

everyone having TTL problems with your S2 camera please send me a private message with your contact info, I am extremely upset about this issue and ready to do battle with Fuji and would like your info for ammunition, all info will be kept confidential and if you would like just send your first name and camera serial number in the message

thanks

[Ansch]

ok.. this ttl problem is not really ideal but has anyone found a consistant way to work around this problem? For example, for a given image, I can either set the camera compensation to +0.5 to +1 or set the flash to +1 to yield a better result. But I suspect it's not a consistant thing and it depends on distance to subject and aperture setting. Might have to do more testing.

Any thoughts? :confused:

[frankgh]

I tried shooting the same spot on the wall centered on a light switch at 1/60 spot metering through the entire f/stop range. F4-5.6 about equally overexposed, 6.7-11 looked really good. F13-22 things started to fall off on the dark side. Looked to me like the TTL was working well within it's limits.

[Pixelfrank]

Hello all together,
as my first reply greetings to all members of the s2pro-forum!
I own a S2pro since 3 weeks.
The reason for buying the s2pro is the standard ttl-feature, so i think i can use my lovely Metz 50MZ-5 with SCA 3402.
What i find out is that the ttl-measuring works but not linear!
At f=11 or higher i must correct with +1 to +2 EV.
With an external flash i can make test-series with the same result as with the internal flash. I use a handheld Flashmeter to read the lightenergy. also with an NIKON SB-23 i get the same results.

regards Frank

[Gijs]

In reply to Don's original post and the many follow-ups.

I am a first time poster and a relative newby S2 user since I own my S2 only since August this year. From the start on I have had mixed feelings looking at the S2 pictures on which I either used the pop-up flash or the Nikon SB-25 flash. They often seemed quite off, especially compared to the results I was used to obtain with my F90. So looking at Don's pictures I kind of recognised the problems. Also the many other posts really seemed to confirm the problem with the TTL on the camera. So after spending a couple of hours reading this very long thread I decided to make a couple of pictures which I add in the next post.

Gijs

[Gijs]

All pictures were made using TTL, using 24 mm setting on a 24-120 mm Nikon lens and sensitivity was set to ISO200 and metering was matrix. Pictures 1, 2 and 3 were made using Aperture priority setting and bounce flash. Aperture settings were F4, F8, F16. Pictures 4, 5 were shot under the same conditions but now with direct (no bounce) flash. Apertures were F16 and F4 respectively. Finally picture 6 was shot using Program mode were the camera used F4.8. The pictures seem to confirm the trend earlier observed by other posts: the smaller the aperture the darker the picture defying the sole reason of TTL. So visually it doesn´t look very good. But are the darker pictures really so much underexposed? To answer that question I decided to take a look at the histograms of the pictures in Adobe Photoshop. Basically I determined the intensity values for which the histograms had declined to a just visible value. (I skip to the next post since I have this 1000 character limit)

Gijs

[Gijs]

Taking the maximum intensities, dividing by 255 and taking the 2log (since each illumination stop equals a factor of 2 difference in captured light energy) I find the amount of underexpose. You 'l be surprised: 0.12, 0.27, 0.65, 0.36, 0, 0.06 stops underexposure! That is far from bad considering the huge diference in aperture and flash used imposed. Incidentally, in contrast to analog overexposing is out of the question which means TTL should be biased to (small) underexposure anyway. Off course all people reading this thread will say: hey what does it help me to know when the numbers are right but the picture doesn't look good? I guess the answer must be the way we look a these pictures: primarily/only on the screen. Having a picture using only half of the possible pixel intensity values looks very dull. Still it is having between 7 and 9 EV dynamics for 8 bits representation, and between 11 and 13 EV dynamics considering the internal 12 bits representation. More.

Gijs

[Gijs]

Incidentally I believe the dynamic range of the human eye is about 40 dB, implying 6-7 EV. So with one stop underexposure the pictures obtained do not need to look awfull. What I expect to play a mayor role is the gamma setting of our computer screen which basically enlarges the dynamics of the brighter intensities and compresses the lower intenisities. To illustrate this point I took the worst picture (3) and adjusted the maximum level to coincide with value 255 in PS (2/3 stop compensation). As you can see the result is much better.
Concluding:
a) TTL works with underexposure of less than 2/3 stop in this experiment.
b) Due to gamma settings -1 EV does not look very good on a computer (did you notice the pictures do look good on the LCD screen of your S2Pro?!!)
c) Eventually we can still obtain very decent pictures out of the S2 flash photos (using the entire PS toolbox, but we wanted to do that anyway ;)
d) Fuji and all posters were right at the same time.

Cheers,
Gijs

[Gijs]

Theo,

I guess the human eye can cover the range of 15 stops only by adapting i.e. using its aperture. However, in one ´view´ this is only 40 dB (I used this fact way back when I was in optical telecom research). That is why all digital video has 8 bits (48 db) dynamic range for the luminance. As for the consistency of TTL, the highest underexposure I measured was 0.6 stops but this was at the maximum of the flash range since I used bounce flash and F16. All other exposures were within 1/3 stop underexposure which really isn´t bad in my view. Incidentally, when I were to be a camera designer and would have to optimise TTL operation I should use the maximum measured intensity in the matrix to signal the shut off of the flash for digital cameras but might be temped to use a (biased) average for analog since it can cope with 1-2 stops over-exposure. I.e. digital would always be biased toward underexposure whereas analog can be right on.

Cheers,
Gijs

[rsvpgrfx]

quote from above:
"ttl is only dead-on at the aperture the camera would select while in P-mode"

this means there is NO TTL flash compensation!

[Pixelfrank]

quote from above:

this means there is NO TTL flash compensation!


There is an TTL flash compensation, but not linear.

Frank

[rsvpgrfx]

^^^


wtf???

[puglover]

I am with you rsvpgrfx (and could you come up with an easier name?? ;)

can I say 'huh?' - for us of lesser intelligence, can you please explain this TTL but not linear comment? thanks :cool:

[Pixelfrank]

it relate to the post Metz Flashguns (http://www.s2pro.com/forums/showthread.php?s=&threadid=2734) .
Sorry for my bad english if there any misunderstandings.

I am just on to analysing the data-transmission between S2pro and flash. the data from S2pro contains values for all important
camera-settings. Apartur-value is for me the most important data.
when you drawdown the differences in exposure depending to apartur in an chart you can see a driftaway from the idealline.
my wish is to create an little blackbox that modifies the camera exposure correction. e.g. f2,8=-0,5EV; f3,5=0EV; f5,6=+0,5EV;f8=+1EV; ........ f22=+2,5EV .
the best way is that Metz creates a modified SCA-Adapter.
But when there is no order for a big amount of adapters, they have no interest.


Frank

[rsvpgrfx]

what you are proposing it seems with this black box is to accomplish the automatic TTL compensation which is NOT being done by the camera!

[Pixelfrank]

are you capable to measure the energy of light with an external handheld flashmeter? please make it. you will see when you close the apartur the energy of the ttl flash will increase. but not so many as necessary. It gives also user that have no problem with the ttl exposure, can be on the other hand it also, that your TTL work not at all.

Frank

[rsvpgrfx]

if the flash output decreases by half each time the aperture is opened by a stop, or conversely if the output doubles each time the aperture is closed by a stop then TTL works properly, anything else is useless

[Pixelfrank]

this is what i say, but the increase/decrease is in reality not so linear as you wrote and depends on the used lens.
i done many test-rows to measure the light-emmission for some lens/flash combinations.

Frank