I keep waiting to meet the camera meter that can accurately handle high contrast scenes. Today shooting the house images above was the first time I've used the Matrix meter for a high DR image in... years! I use the Matrix meter when a scene offers less DR. Aperture Priority and Matrix meter works for me then...
Active focus sensor was just below and to the (scene) left of her left eye. With regard to active sensor choice and exposure skewing in Matrix metering it wouldn't have mattered where I placed the sensor in this scene. Easy peazy.
bitmaker wrote:
I've keep waiting to meet the camera meter that can accurately handle high contrast scenes. Today shooting the house images above was the first time I've used the Matrix meter for a high DR image in... years! I use the Matrix meter when a scene offers less DR. Aperture Priority and Matrix meter works for me then...
Active focus sensor was just below and to the (scene) left of her left eye. With regard to active sensor choice and exposure skewing in Matrix metering it wouldn't have mattered where I placed the sensor in this scene.
thanks Desmond for trying to untangle the mess and quirkiness of the voodoo in the Nikon flash system.
so in general terms when do you Not use BL?
in some basic testing around the house i do not find a lot of differences with my D3 and SB900 using either mode. but for years i avoided BL beginning with my D100 it was a disaster so i guess i have been not using it since then
jmcfadden wrote:
so in general terms when do you Not use BL?
J
Good question ! ................. actually I always use it now . Since wireless flash meters in TTL/BL mode and the SB400 meters in TTL/BL mode it is the way forward for Nikon flash and TTL can be compared to 'A' mode on the SB800 - it's there for people who have become comfortable with using it ...
With regard to your D100 - TTL/BL would work as explained in the 'official' TTL/BL site , the subject would need to be central , and if there was no bright background lighting it would fire at its weakest - I think that was the last Nikon that worked that way .
He says he doesn't want to 'go public' with the discussion but is happy to communicate directly as we exchange information .
At the moment the discussion is revolving around "" target="_blank" rel="nofollow">how TTL/BL changes its output depending on where the meter is sitting and which metering mode you are on .
He insists that TT/BL fires at the same output regardless of where the meter is and that TTL/BL measures the ambient itself regardless of what you do with the exposure while I am pretty sure that it changes its output when it detects that you are under-exposing the ambient as I saw in my tests . He told me I obviously did something wrong in those tests because my results can't be right so I did the tests again with all the settings to prove that the output does change .
In his first reply he concluded that the foreground must have been brighter than the background so this time I included a shot without flash .
Since he also insisted that wireless flash does not meter the same as TTL/BL but rather that it meters in TTL mode I included a shot with TTL flash and some with wireless flash . Though wireless fired weaker than all of them it showed the same characteristic of reacting to a change in the meter simply from changing metering modes .
In my mind these shots totally prove my theory that TTL/BL balances the flash with the existing ambient on the subject depending on how you choose to expose the ambient . For example , if you are shooting indoors and set a slow enough shutter speed , let's say1/25th sec , to catch the ambient TTL/BL will fire much weaker than if you are shooting at 1/200th and the meter showed that you were 4 stops under-exposed .
I'm going to do some tests now with a flash meter to see what variation we have and at what point of under-exposure TTL/BL 'takes over' completely .
I borrowed a flash meter and set my camera up in rear curtain mode at slow exposures so I could reset the flash meter after the pre-flash . What I found was quite interesting - in cases where the two systems meter equally [ central subject ] TTL and TTL/BL both gave the same results when the ambient was under-exposed by 4 stops ;
But when I had the ambient correctly exposed TTL gave the same output as we all know it would do which resulted in too much flash but TTL/BL automatically dropped back to the equivalent of TTL-1.7 which is a number many people aim for with fill-flash .
I'm slowly working my way into the iTTL system using multiple off-camera flashes and this was certainly an inducement to carry my studies further.
Dave
Thanks , just remember that though wireless doesn't give exactly the same results as the other modes it meters very much like TTL/BL and behaves the same in many ways , over-reacting to white in the frame and adjusting for ambient .
I think you'll find that when the camera meter is involved with the flash metering (especially matrix mode) it is not metering the brightest area or the darkest area, but rather any area that is not so dark (or far away) as to reflect "nothing" and not so bright (or close) as to reflect too much. A comparison is made between the ambient metering results and the pre-flash metering results, and an exposure for the flash is based on how much the lighting needs to be scaled up. This is what the Canon system does too and I figure the Nikon system does it better because it has more metering segments to work with (e.g. 1005 on the D700 vs just 21 to 63 on a Canon 1-series pro camera). Working this way prevents a reflection off a silver plate or mirror from shutting the flash output right down and allows the flash to work off a persons face instead of his black suit. Because the metering segments are smaller it is able to ignore a smaller portion of the entire image when there is a specular highlight, and so it tends to get the rest right more often.
Alan321 wrote:
I think you'll find that when the camera meter is involved with the flash metering (especially matrix mode) it is not metering the brightest area or the darkest area, but rather any area that is not so dark (or far away) as to reflect "nothing" and not so bright (or close) as to reflect too much. A comparison is made between the ambient metering results and the pre-flash metering results, and an exposure for the flash is based on how much the lighting needs to be scaled up. This is what the Canon system does too and I figure the Nikon system does it better because it has more metering segments to work with (e.g. 1005 on the D700 vs just 21 to 63 on a Canon 1-series pro camera). Working this way prevents a reflection off a silver plate or mirror from shutting the flash output right down and allows the flash to work off a persons face instead of his black suit. Because the metering segments are smaller it is able to ignore a smaller portion of the entire image when there is a specular highlight, and so it tends to get the rest right more often.
With TTL/BL , and pure "bounce flash , I have found that it responds strongly to any light coloured object in the frame . It's original purpose is ' fill flash' and it works hard to preserve highlights judging by its behaviour . So when it is the sole light source , using bounce flash , it over-reacts to white in the frame but is consistent in doing so , not progressive like with normal TTL flash .
When using TTL flash this is the metering pattern I see in my mind :
However with TTL/BL I see this pattern , a wider area that looks for the brightest object under a focus point , or surrounding area , receiving a pre-flash reading .... like lots of small 'spot meters' and not heavily weighted toward the centre of the frame .
When the flash head is tilted it loses the advantage of distance info and goes into what I call "blind mode" .
While TTL flash depends on an average reading affected by the size of the subject ...
and says " on the left is a bright reading - must be the closest object in the frame - we must not blow any highlights so back off to make that object average gray "
so in a case with 'pure bounce flash ' and whites in the frame I dial in + 1 1/3 ...
Just be warned that when you have the diffuser dome on or the bounce card out this compensation is not needed as much - the program still seems to keep distance in the back of its mind and adjusts output on the assumption that there is some light going directly forward .
Jay Parco wrote:
Best nikon flash write up I've seen so far
Thanks very much . I'm busy working on a "TTL vs TTL/BL" video but this time it will be a bit more professional rather than a hurried video with bad lighting .
I just need to fix my animated graph [ showing the relationship between ambient and flash output ] so that the 'stops' are more realistic and show a drop in half the light output with each stop of exposure lost .
Call me naive, but has anyone tried to ask the engineering dept. at Nikon for a clarification of the iTTL question? It's not like you are asking for the recipe for KFC.
jprin wrote:
Call me naive, but has anyone tried to ask the engineering dept. at Nikon for a clarification of the iTTL question? It's not like you are asking for the recipe for KFC.
I think you'll find that it is harder than getting the KFC recipe - Nikon have a superior flash system and I don't think they want to share the recipe with anyone . Russ Mc Donald thinks he has the recipe because he helped design the system in the 80's but I've proved him wrong on a number of points - Nikon have changed the system since the 80's and quite drastically too , and he hasn't been with the company for 10 years .
I've started a photography blog best of the forums and anyone is welcome to contribute . I've added a link to my TTL/BL studies and included a few tutorials on flash .
ANY TTL system by any manufacturer can, and most do use distance. just like thyrister systems they prefer to start off with a mathematic exposure. distance, light output, and sensitivity, aperature and a little love.
Easy peazy.
