What is Sigma "S-TTL" For Flash Photography?

Hi, So it controls intensity of flash instead of or also duration.
--
Rick Wilkinson
From Gobblers Knob
http://www.pbase.com/short243

no problem, sorry for being so impatient

One last note, the one about "(this will not always work, more on that later)".

If you use a fast lens (lets say wide open aperture below f/3.5) the M / TTL method will work well for the lens wide open, stopping it down can lead to overexposure, it seems the exposure will only go back to normal stopped down further (f/8 to f/11) or anywhere close to the point where the flash runs out of power.

I got really bad results with the 30EX, but for example the maximum it is off with my 50EX in a test I did some days ago was depending on the f-stop between 1/3 stp underexposure and 1 f-stop overexposure. This is well within a range of exposure errors I can get in daylight as well and that SPP can deal with, or check your lenses and get a feeling for it.
--
http://www.pbase.com/dgross (work in progress)
http://www.pbase.com/sigmadslr/dominic_gross_sd10

Dominic Groß said:

It is certainly not as convinient as I-TTL or E-TTL2, but it is
from useless in all situations, not thinking about the exposure can
cause a lot of trouble in tricky situations and in this case it
will certainly get unusable.

Click to expand...

I was not saying it is useless. I do not know enough about. Search results say it is useless. I know about making money with cameras and perfect flash control is the bling bling kaching ching. I did not know why Sigma's cameras failed until now.

Dominic Groß said:

more here:
http://forums.dpreview.com/forums/read.asp?forum=1027&message=19557349

To show that it actually works I just took some pictures in my
room, flash on TTL, Camera M, 28EX, f/1.8. No corrections in SPP,
just downsized in irfan view, original exif should still be there.







--
http://www.pbase.com/dgross (work in progress)
http://www.pbase.com/sigmadslr/dominic_gross_sd10

Click to expand...

Dominic Groß said:

It is certainly not as convinient as I-TTL or E-TTL2, but it is
from useless in all situations, not thinking about the exposure can
cause a lot of trouble in tricky situations and in this case it
will certainly get unusable.

Click to expand...

......and it works a lot better than my experience in the past. I have a Nikon D-70 and not every flash shot with it is dead on either.

Fred R. Elias Jr.
Warren, MI.
[email protected]

I've had mixed results with my SD9 and the EF 500 ST. Not horrible, but mixed. I have a Nikon D50 that uses Nikon's i-TTL flash. Coupled with a Nikon SB600, it works great. If Sigma's S-TTL worked half as weel it would be a good thing.
--
My humble photo gallery: http://ntotrr.smugmug.com

Peter A. Stavrakoglou said:

I've had mixed results with my SD9 and the EF 500 ST. Not
horrible, but mixed. I have a Nikon D50 that uses Nikon's i-TTL
flash. Coupled with a Nikon SB600, it works great. If Sigma's
S-TTL worked half as weel it would be a good thing.
--
My humble photo gallery: http://ntotrr.smugmug.com

Click to expand...

I am satisfied with E-TTL II that no further improvements are necessary. E-TTL I was not always consistent but it worked well. Useless or random are not words that enter the vernacular. How do Sigma users work with slow lenses and three number ISOs without consistent take along lighting? Their next camera will have to fix these problems or again face rejection. The core minimum for professionally priced systems is an indoor wedding.

I have done some explanation earlier but could not find the message.

Fortunately Dominic got allmost everything correct except maybe one thing.

There are diffrences between SA flashes: They come with 1/3 (N) and 1/2 (no N) stop settings as come the SD cameras. For the S-TTL flash automatic to work as exspected on must have tha same aperture vivible on flash and on camera.

After that the camera is set to M mode and flash in TTL d/+- mode everything works fine. If exposure needs compensation it has to be done changing the flash settings.

some test photos
http://www.lumisoft.fi/gallery/thumbnails.php?album=64

If the flash is non N then one should stick to full stop values such as2.8/5.6/8.0/11.

M mode in camera just prevents automation on apertures and shutter speed as mentioned earlier.

All these things are covered in flash manual, maybe not the most user friendly way.

There are two things that have most impact on perfect flash photos:
1. Metering system, with SD cameras only spot metering is available.

2. The actual scene photographed. Ie in weddings the black velvet and pure white silk make up a challenging task whatever the metering system is.

Also some practise and common sense helps a lot.

Some flash photos









puppies
http://www.lumisoft.fi/gallery/thumbnails.php?album=115

edible



Hope this is of some help.

Aaro
--
Watching you



The man from Snowriver, Mies Lumijoelta
http://www.lumisoft.fi/gallery
http://www.pbase.com/aaro

You could have used better subjects in the first two photos
--
My humble photo gallery: http://ntotrr.smugmug.com

Dominic Groß wrote:
. . .

Dominic Groß said:

In M you can choose the shutterspeed (and thereby control how much
ambient light you want) and the aperture and it should output
enough light. At certain f-stops (stopped down slightly from wide
open it tends to overexpose) and depending on the brightness and
distance of the subject it won't get it right, if such errors occur
they are very systematic and can be dealt with with flash exposure
compensation.
. . .

Click to expand...

.. and in M mode the flash respects the exposure compensation setting. you can change this setting only in S, A or P mode, not M mode. nevertheless it changes the flash intesity in M mode as well.

Fred R. Elias Jr. said:

If you are shooting in manual where does the "Automatic" come in?

Click to expand...

The camera is in manual, but the flash is automatically set to to whatever power level is needed. For example, I usually set my flash for 180/sec and f8 to start with and then adjust the flash from there.

Fred R. Elias Jr. said:

The sensor for STTL is only a few sensor pixels in the dead center
of the field. This is a powerful manual flash but, the camera is
worthless to command it in any sort of a rational TTL/auto manner.

Click to expand...

You can also use the exposure adjustment on the flash to have the power output be what is desired for the primary subject, even though the TTL reading is from the center - or creative composition works as well to have something in the center that ends up metering about the same as your subject.

--
---> Kendall
http://InsideAperture.com
http://www.pbase.com/kgelner
http://www.pbase.com/sigmadslr/user_home

As a newbie I am totally bewildered by the comments concerning S-TTL. I think it is fabulous.

S-TTL does a better job than I can do manually.

It took me a while to figure out how it works though. I wish Sigma would have one of their experts explain it. I tried figuring things out here and other places on the Internet and eventually decided that all, or nearly all, that has been said about S-TTL is worthless.

So, I went to the patent database and spent an afternoon reading patents on TTL flash. Of course, this does not guarantee that S-TTL works the same way as the patents that I have read, but it at the least gives me some insight into the topic.

That insight, coupled with some experimentation, has lent to believe that this is how S-TTL works. Well, in part.....

The following comments pertain to a EF500 Super SA-N in a SD10.

With the flash on the camera turned off, the flash has no effect.

With the flash turned on in P mode the shutter speed is set to 200. i.e. the sync speed. The camera assumes that you want a flash picture and not a slow-sync picture. For a slow-sync picture simply use another mode. The camera's exposure is set to -2EV from where it would be if the flash was off.

In A and S modes the camera's exposure is set to -2EV from where it would be if the flash was turned off.

In M mode no correction is made to the camera's exposure metering.

In all cases there is an algorithm where the flash duration is used to complete the exposure. In manual mode this seems to result in an overexposure for most users as they do not set the exposure back enough to account for the flash's effect. In this case, the flash uses what precious little exposure latitude that might be left.

I am not convinced that there is center metering of flash only. I plan to do more experiments on that.

VWfan wrote:
...

VWfan said:

In A and S modes the camera's exposure is set to -2EV from where it
would be if the flash was turned off.

Click to expand...

. . .

with my EF-500 DG ST attached to a SD10 it seems to be -1EV in A and S modes.

sgalke - gee, that's nuts! What metering are you using?

I think my metering was set to spot when I did this.

I'll play around some more and see if I can duplicate a -1EV difference.

Happy 2007!

--
Happy Holidays!!!!! Oh look, here's Santa *

The camera's basic also known as "background" or ambient exposure is NOT changed when the flash is turned on or off. This behavior is different than many other flash systems e.g. Nikon.

In P, S, and A modes automatic fill flash reduction is applied according to the LV of the ambient light metered in the scene. (LV = EV at iso 100) the transition from key-flash to fill-flash seems to occur in the range between 10 LV and 13 LV.

If the camera is in manual exposure mode and the flash is in STTL and the viewfinder display shows 0.0 or correct exposure you will overexpose by about one stop because you already have correct ambient exposure plus another correct flash exposure added to it. In other words you have doubled the correct exposure or over-exposed by 1 stop. In manual mode on the camera and assuming no correction set on the flash itself you always want to see -1.0 or less all the way down to -3.0 (not blinking) in the viewfinder. This is fill-IN flash as opposed to just fill-flash.

If you want to use full correct ambient or background exposure (0.0) you must manually set fill flash reduction on the flash. Ansel Adams reccommends the range from -1.0 to -2.0.

The first step in using the flash is deciding what you are trying to do. Are you taking a pure flash photo (key-flash)? Are you trying to make up for insufficient ambient light (fill-IN flash)? Are you trying to reduce the dynamic range of a scene thereby lightening overly dark shadows (fill-flash)? You have to decide this first before you can determine what settings to make on the camera and flash. This is not a point and shoot system you must take control. The system was also designed with the assumption that any minor exposure error up to about +or- 1 stop will be easily handled automatically by Sigma Photo Pro.

Take the SD9 for example the camera and flash are only adjustable to the nearest half stop. Thats a worst case 1 stop error. If you hit it within this range "you nailed it". It's still going to "look" up to one stop over or under exposed in the preview on the back of the camera. You can chimp if you want but I really don't usually find it to be worth it. In fact I often find the final result to be worse if you do.

Also the system tries very hard not to blow highlights. It will tend to underexpose assuming this can be fixed in SPP rather than allow a lot of highlight blowout. It doesn't attempt to determine whether a highlight is specular or diffuse and use dynamic matrix weighting to make finer corrections the way ETTL-II does in the Canon system for example.

If you go with the flow and work within it's limits and concept it works quite well. If you second guess it and/or try to make it work like a Canon or Nikon you will drive yourself nuts. Trust the meter and Sigma Photo Pro to give you a good final result. This is another of many reasons why I like Photo Pro better than other RAW converters.

It is also wise to learn the situations where matrix meters are likely to be fooled and know how to use exposure compensation to remedy the situation. Ken Rockwell ( I know boo hiss) has a nice little article on this topic. He's very capricious and opinionated but he really isn't a complete idiot.

The Super version of the flash is the one reccommended for the SD cameras. The ST "works" but I have yet to hear of anyone being very happy with it.

Ken

Actually it changes the background, ambient or basic exposure not the flash exposure.The flash exposure can only be adjusted by the buttons on the back of the flash.

You are correct however that the exposure compensation set in the automatic modes does carry over into the manual mode even though it cannot be set or adjusted from within manual mode.

Ken

With my SB800 Nikon speedlight, firstly I usually always meter w/o the flash on for my ambient light. I usually like to try and get as much ambient light into the scene as possible. This I do in manual mode, adjusting the Shutter, Aperture & ISO independent of the flash. Bearing in mind that my aperture value will play a role in my flash strength. I meter in spot , as this allows me to have full contol of the ambient light. If I meter in matrix or Evaluative, the camera & flash take over the metering for both the ambient light and flashlight...this is called TTL/BL...thru the lens balanced, and works very well sometimes.

I will usually adjust the flash strength down(-1EV) to play things safe, as I can very easily recoup some lost exposure in RAW.

I just took six quick shots in my playroom of the kids & wife...all exposed beautiful for flash & ambient light...all keepers.

Another really nice feature of the SB800 is FV Lock(fllash value lock). This simply sets perfect exposure for the flash strength on your subject. You activate this feature by pressing the AE-AF Lock button, and the camera sends Pre-flashes from the SB800 directly towards your subject for measurement...amazing feature.

I love the SB800 & Nikon's iTTL(intelligent TTL). I know that I will miss it, but if I eventually buy into the SD14 & Sigma speedlight, I would be very happy with 75% of the features that give me success with the Nikon system.

Mike

Thanks for the input.

Nikon's flash system rocks once one learns the ins and outs of it it which vary by camera model and flash model. I think it's probably the best dedicated flash system there is. I'm not an expert on it at all since I don't own a Nikon camera.

I've read a lot about both Canon and Nikon flash systems in order to try to piece together just what Sigma was up to with their system. Sigma's system is very primitive compared to Canon or Nikon but it does work as it was designed to.

It does not automate a lot of things that the others do and requires you to think. It also relies on the post processing to make up for it's shortcomings. It's hard to get used to the fact that the display on the back of a Sigma camera is giving you a view of the data at a point in the processing earlier that what you see on the back of a Canon or Nikon.

You get the illusion that the exposure is much more accurate on the Nikon or Canon. It is more accurate but not to the degree that the on camera previews indicate. The Canons and Nikons have already done a lot of processing before the preview is created. It is much more realistic to compare the on camera preview of a Canon or Nikon to the output of SPP in auto mode than to compare the previews directly.

I believe you hinder yourself if you try to judge the exposure of the SD cameras by the LCD on the camera, except in gross terms. I mean if the display is black you obviously have a problem. The histogram is very useful. It became much more useful to me once I proved to myself that the center mark is zone 6 and not zone 5

In my experience I get better photos if I let the meter and Photo Pro do their thing. Which usually means the meter gives SPP a little headroom through slight "under-exposure". Photo Pro seems to make use of this while working its Foveon magic.

The way I view it the process reminds me of volleyball. The camera's meter provides the setup (the optimum raw data set to work from) and SPP delivers the spike (massages the raw data into a good looking photo). Photo Pro's auto mode takes a good stab at creating an optimum photo but of course it does not know your subject or intent, so you have all of it's image controls to guide it where its assumptions may have gone astray and departed from your creative vision.

Ken

Central Scrutinizer said:

Thanks for the input.

Nikon's flash system rocks once one learns the ins and outs of it
it which vary by camera model and flash model. I think it's
probably the best dedicated flash system there is. I'm not an
expert on it at all since I don't own a Nikon camera.

I've read a lot about both Canon and Nikon flash systems in order
to try to piece together just what Sigma was up to with their
system. Sigma's system is very primitive compared to Canon or Nikon
but it does work as it was designed to.

It does not automate a lot of things that the others do and
requires you to think. It also relies on the post processing to
make up for it's shortcomings. It's hard to get used to the fact
that the display on the back of a Sigma camera is giving you a view
of the data at a point in the processing earlier that what you see
on the back of a Canon or Nikon.

Click to expand...

I guess what you are saying is that the data is kinda raw looking? How does it look? I would guess that on the SD14 you will see a jpeg basic on the LCD after the shot?

Central Scrutinizer said:

You get the illusion that the exposure is much more accurate on the
Nikon or Canon. It is more accurate but not to the degree that the
on camera previews indicate. The Canons and Nikons have already
done a lot of processing before the preview is created. It is much
more realistic to compare the on camera preview of a Canon or Nikon
to the output of SPP in auto mode than to compare the previews
directly.

Click to expand...

Well, even if you are shooting Raw, you are getting a 'basic' jpeg view on the LCD screen of my Nikon D70S after the shot. . This is automatically done. Of course you can use that jpeg, it is real. Or, you can also Post Process your Raw file.

Central Scrutinizer said:

I believe you hinder yourself if you try to judge the exposure of
the SD cameras by the LCD on the camera, except in gross terms. I
mean if the display is black you obviously have a problem. The
histogram is very useful. It became much more useful to me once I
proved to myself that the center mark is zone 6 and not zone 5

Click to expand...

Yes the histogram is your friend. However, the LCD screen is pretty accurate looking on my Nikon. Much more so than my recently departed Fuji S3.

Central Scrutinizer said:

In my experience I get better photos if I let the meter and Photo
Pro do their thing. Which usually means the meter gives SPP a
little headroom through slight "under-exposure". Photo Pro seems to
make use of this while working its Foveon magic.

Click to expand...

I like it! Almost sounds like my D70S and Nikon Capture.

Central Scrutinizer said:

The way I view it the process reminds me of volleyball. The
camera's meter provides the setup (the optimum raw data set to work
from) and SPP delivers the spike (massages the raw data into a good
looking photo). Photo Pro's auto mode takes a good stab at creating
an optimum photo but of course it does not know your subject or
intent, so you have all of it's image controls to guide it where
its assumptions may have gone astray and departed from your
creative vision.

Click to expand...

Music to my ears!

Of course, the main reason I am considering the upcoming Sigma SD14, is what I am seeing and reading about the georgeous IQ. I am really interested in this sensor. The other reason is that I am a Raw Shooter. I tried out SPP briefly. And I do like it. I like the fact that Sigma is a RAW camera company, if you will. Of course we will now have jpegs. For me though, it would have been nicer for Sigma to have spent the money elsewhere on the camera rathe than on jpegs.

Mike

carauction said:

I guess what you are saying is that the data is kinda raw looking?

Click to expand...

Exactly!

carauction said:

How does it look?

Click to expand...

It's kind of like a 16 (or 12 if you prefer) bit linear output from a scanner. Only it has been Gamma corrected so it is not actually linear like unadjusted scanner output but has not had the sort of auto-leveling applied that a Nikon's jpeg has.

carauction said:

I would guess that on the SD14 you will see a
jpeg basic on the LCD after the shot?

Click to expand...

That's what I would assume although I have not used an SD14. There are others on this forum who have. I don't know if they are at liberty to give any details yet due to NDAs. Of course many people saw the cameras at Photokina so I would expect some of them could answer a question this basic without violating any trust.

carauction said:

Well, even if you are shooting Raw, you are getting a 'basic' jpeg
view on the LCD screen of my Nikon D70S after the shot. . This is
automatically done. Of course you can use that jpeg, it is real.
Or, you can also Post Process your Raw file.

Click to expand...

Precisely. I believe this misunderstanding is one reason why Sigma gets bashed a lot for its flash and exposure system. The view on the back of a Nikon is a fully processed jpeg. I think the view on the back of a Sigma is basically RAW with gamma encoding.

carauction said:

Yes the histogram is your friend. However, the LCD screen is

pretty accurate looking on my Nikon. Much more so than my > recently departed Fuji S3.

Click to expand...

The historgram on the SD is four channel RGBC RAW. It's very accurate and if you zoom in on the preview the histogram shows only the zoomed area so you can use it as a densitometer. A chimpers dream come true.

carauction said:

I like it! Almost sounds like my D70S and Nikon Capture.

Music to my ears!

Of course, the main reason I am considering the upcoming Sigma
SD14, is what I am seeing and reading about the georgeous IQ. I
am really interested in this sensor. The other reason is that I am
a Raw Shooter. I tried out SPP briefly. And I do like it. I like
the fact that Sigma is a RAW camera company, if you will. Of
course we will now have jpegs. For me though, it would have
been nicer for Sigma to have spent the money elsewhere on the > camera
rather than on jpegs.

Click to expand...

I agree completely. I think the reason for including jpeg is actually so that people will be able to output 14MP jpegs, print them, look at the result and see that the SD14 truly is approx a 14 MP camera in output quality.

I fully expect it to give the Canon 5d and even 1ds MkII a run for the money in image quality. It will get it's butt kicked in handling, speed and features but thats not what Sigma is working on at present. Or more correctly it probably is what they are working on but have a ways to go yet.

I think just adding flash exposure lock as well as dedicated focus lock and exposure lock controls would do wonders to enhance the system. Also the option to let the shutter speed float as low as 1/60 in program mode with flash like the Nikon would be welcome.

Ken

Dominik,

one of your messages is that depending on the setting of the camera the flash is set to fill light or main light. This is not obvious at all. The flash system is expensive enough to justify a good user interface - e.g. set it to fill or main light on the flash itself.

The TTL system as implemented with the Sigmas - Center spot measuring is mainly not of much benefit. There had been auto measurement with flashs before TTL measurement and for me this works much better with a Nikon SB28 than EF500 ST.

You asked me once why I wish distance readout (which is prepared in the X3F) for lenses. Well, then you could control the flash with this information.

Regards

Wolfgang

My Nikon SB-28 example: