Your struggles with the exposure foibles of the E-1 reminds me of an
article I read recently chastising our camera designers for not making
full use of the exposure measurement technology already available to
them. It covered more ground than ETTR but, since ETTR is of special
interest to me, it's what I paid most attention to. The author simply
pointed out that any camera with live view could easily implement an
ETTR exposure mode and guarantee that no highlights would be blown since
it knows the exposure state of every pixel.
Using the 5D's histogram or flashing highlights display usually works
fairly well for ETTR but sometimes is a real guessing game. It irks me
to sometimes have to underexpose a bit to be sure I've not blown one
channel. For the 5D the ETTR analysis has to be post exposure but any
live view camera should be able to make perfect ETTR into a point and
shoot exercise.
Chuck Norcutt
On 12/2/2011 10:52 AM, Ken Norton wrote:
>> Ken, I thank you... and my E-1 thanks you, too!
>
> Ditto in return.
>
> An analysis of the image is actually very instructive. This image could be
> a poster child for several E-1 issues.
>
> 1. The E-1's sensor is overly sensitive to near-IR. The problems seen with
> the Leica M8 are also present with the E-1, but just to a lesser extent.
> Typically, this is not a problem, but when using strobes it can be. Strobes
> have extensive IR and UV radiation. Caucasian skin will turn nice and pink
> when photographed with the E-1 using flash.
>
> 2. This extended near-IR sensitivity leads to the exposure lifting of
> caucasian skin--especially in the red channel. We are more likely to blow
> exposure in this case. If we use highlight-recovery to bring things back
> into line again, we get color shifts in the highlights as the four color
> array did not clip at the same point. Highlight recovery with the E-1 is
> very much a crap shoot. In comparison to the other digital cameras in my
> household--the exact same exposure with the E-1 will make caucasion skin
> look at least a half-stop lighter.
>
> 3. Because of 1 and 2, it is better to UNDEREXPOSE the image slightly and
> bring up the exposure during RAW conversion. With my E-1, and I believe
> this to be true with all E-1s, if you chimp the portrait on the LCD, if the
> skintones look "just right" you are about 2/3 to a full stop under. If you
> want the skintones to actually be properly exposed (for no pulling exposure
> in conversion) you have to expose the image so the skin looks a bit bright
> on screen. Generally speaking, I make sure I never have flashing (clipped)
> highlights and try to stay about a stop away from clipping. With the E-1's
> four-color array, you can be clipping in-camera and never know it. Anytime
> you are within a stop of clipping you are guaranteed to be clipping.
>
> 4. Just a side note, this child has extremely smooth skin.
>
> 5. When you get too close to your subject and you are bouncing the light,
> the in-flash auto-sensor of the T45 will overexpose because the exposure
> sensor isn't looking at the same point the camera lens is. Move farther
> away and this isn't a problem.
>
> 6. Not illustrated in this photograph because the ISO was set to 100, but
> when using a high-power on-axis flash and higher ISOs, the dreaded
> magenta-blacks will show up. A flash at full or near full power produces a
> ton of IR and the E-1's sensor in "boost" mode seems to be particularily
> susceptable to this problem.
>
> 7. Not related to the sensor or exposure is the lens. The 50/1.4 is bloody
> sharp. Notice that the proper focus point wasn't too difficult to achieve
> either. Even though this lens lacks that "wrap-around" characteristic of
> the 35-80, it still does a remarkable job for portraiture. The comparative
> 35-80 shots aren't as sharp because of the chromatic aberrations (see notes
> about extended near-IR sensitivity) and the fact I was at 1/60th of a
> second so I was inverted on the maximum handheld speed equation. The 50/1.4
> is easier to handhold than the 35-80.
>
> 8. Flash facing 90 degrees sideways is a fantastic technique. I bounced the
> light off the side wall. The images where she is looking "towards the
> light" are really really nice. Using the Rogue FlashBender to throw a
> little forward light works really well. Granted, in this case, I had a
> touch too much light thrown forward, but the theory is sound. The best
> shots were the ones where the FlashBender was bent back so much that it
> wasn't really in the flash's light path.
>
> 9. A bit of pride in the fact that my purchase decision a zillion years ago
> worked out so well this far down the line. At that time, I don't think many
> of us expected the E-1 to be viable past three years. Yet, for portraiture,
> this remains my go-to camera.
>
> 10. Just a random note about the exposure. I do believe the skin looks
> overexposed partially because of the proportion of the entire image area.
> Another shot of her and her mother had exactly the same brightness levels,
> but because they were "farther away" from the camera showed that they were
> exactly where we would want them to be. Oh, and they haven't been in the
> sun for a couple of months so their skin really is quite light.
>
> 11. Other than the blown highlights bit, (yes, the red sensels got blown
> out on this shot), this image also represents the 'old-school pre-Portra
> technique of underrating your film for flash exposure. To get the skin to
> lift enough and give good separation from the backgrounds it was not
> uncommon to overexpose the on-camera flash by a full stop. Granted, film
> was more graceful with blown highlights, but the images definitely had more
> snap. Snap that we make up for today by cranking up saturation, using LCE
> and aggressive curves adjustments.
>
> AG
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|