Hi Ken, Dawid, Chuck, Jez and all,
From: Ken Norton <ken@xxxxxxxxxxx>
>A similar shot, taken last year at the Badlands National Park with the E-1
>is here:
>
>
http://zone-10.com/cmsm/index.php?option=com_content&task=view&id=191&Itemid=1
Beautiful picture! Does Ursa Major qualify as the bear you were looking for?
;^) BTW, your (geographic) latitude seems much higher than mine...
From: Dawid Loubser <dawidl@xxxxxxxxxxx>
>Also, a 1Dxxxx accurately auto-exposes up to 30sec
>at any ISO,
But, again... will it do so at any aperture -- I mean, WIDE OPEN too at max.
ISO?
>and a 30second linear exposure at ISO3200, though noisy,
>goes far beyond what is possible with
>any film ever made - you can leave the shutter open all night on your
>Delta3200/KodakP3200 it won't get there
>because of reciprocity which get very bad very fast with these fast
>films.
I respectfully disagree. As an amateur astrophotographer, I'm well used to
very long exposures, and reciprocity failure is not _that_ bad. Some older
films (like Tri-X) do need some compensation, which may look terrible on
paper (e.g. calculated 100 sec. should be... actual 1200 sec!!!) but in
practice you get acceptable results even without any compensation. Some
modern films (esp. Fuji Provia, Sensia...) have negligible reciprocity
failure.
From: Chuck Norcutt <chucknorcutt@xxxxxxxxxxxxxxxx>
>One thing I can never recall being discussed is the role of long OTF
>exposure and reciprocity failure. Somehow, I suspect that the 1970s and
>1980s electronics of the OM system do not compensate for reciprocity
>failure
You're right. And I don't think any lightmeter on any camera would.
>From data on
>the web, Provia 100F is remarkably immune to reciprocity failure out to
>3 minutes and the error is probably negligible to double that or longer.
I absolutely agree, according to my experience. The same goes for Provia and
Sensia 400. Can't tell about Velvia, because I never use it...
>For a less tolerant film how does one make a correct exposure? Take a
>test shot, consult the reciprocity table for the test exposure and lie
>to the camera via the ISO dial for the real exposure?
Seems the best method. However, you _may_ get decent results even without
compensation.
From: Ken Norton <ken@xxxxxxxxxxx>
>The meter
>sensitivity in modern cameras is not too shabby, but still within maybe an
>EV or two of what I had available--probably less and possibly worse.
Most likely.
>Let's see. The OM-4T(i) is able to OTF to -5 EV, and spotmeter to -7 EV.
>(ISO 100, 50/1.4).
Well, the -7 EV for spot metering is a "honest" error -- I'll explain about
it later.
>So, if it is true that the OM-4T(i) is good for auto-exposure down to -5
EV,
>what digital cameras meet or exceed this?
I'd say... none ;^)
The Pentax LX is as good or better, but in fact it's the only non-Oly OTF
metering system... The Nikon F3 (non-OTF, of course) is very good at this. I
can't recall the quoted specs, but with its unique system of microscopic
holes in the mirror, it can measure down to LV -4 or so.
>> EV 0-20 or 1-20 seem to be the norm for Nik and Can.
>This is interesting. I've been having difficulty finding the meter
>specifications with ISO/lens baseline information for recent digital
>cameras.
Now that I've mentioned LV... it's time to remind what the EV's are. The
"Exposure Value" is just the combination of shutter speed and lens aperture
in order to get a certain exposure on film -- the same value represents all
equivalent combos. The arbitrary reference (EV 0) is 1 second @ F1.0. Each
full-stop faster (speed) or stopped-down (aperture) increases EV in one.
Example: 1/125 @ f/5.6 (or equivalent!) equals EV 12, because:
(f/) -> 1.4 -> 2 -> 2.8 -> 4 -> 5.6 (stopped down 5 stops from reference)
(1/) -> 2 -> 4 -> 8 -> 15 -> 30 -> 60 -> 125 (7 stops faster than reference)
Thus 5+7 = EV 12.
But that's just the camera's settings. What about the LIGHT level? Here's
where the LV arises. You can use the EV scale _if_ related to a reference
film speed -- that's the ubiquous ISO 100/21º. So for ISO 100, LV = EV. But
if you use a faster film, you may use the same settings with a LOWER light
level, so:
@ISO 100/21º => EV = LV
@ISO 200/24º => EV = LV+1
@ISO 400/27º => EV = LV+2
and so on. LV is an absolute reference for light level, but the EV alone
_needs_ to specify the film speed.
Low-light metering performance is determined by two limits, whichever
happens "first": the minimum light required by the metering cell, and the
maximum exposure time -- either indicated by meter reading (manual) or
actual exposure (auto). At low film speeds, the latter is usually the
determining factor... but at higher ISO the first issue is dominating.
With TTL metering (be it OTF or not) the light reaching the cell depends not
only on the scene's lighting, but also on the lens speed, whose maximum
aperture should be known in order to specify metering performance. There's a
limiting LV for each camera-lens combo. Since we're cranking up the ISO, the
limiting EV will increase too -- eg. @3200, EV = LV+5. But since the specs
only mention ISO 100/21º, many people won't notice.
>First of all, I will challenge the first set of specifications as I can
>attest that it goes far beyond 1 minute. In fact, my OM-4T probably goes
as
>far as 3 minutes!
As does mine, at least until ISO 250/25º. OTOH, the OM-2n has a constant
maximum exposure time, no matter the ISO: guaranteed two minutes, in
practice around three or more.
>Secondly, I think that the TTL spot metering exposure
>range of the specifications is accurate whereas the -7 EV as listed in the
>ESIF is wrong.
Yes, it is an error. So where does the -7 quote come from? Because of the
four minute exposure time in spot -- if 1 minute @ F1.4 is EV -5 (-> 4 -> 8
-> 16 -> 32 -> 64, round it to 60), then four minutes instead of one is 2
stops slower, thus EV -7... but NO LONGER VALID at f/1.4!!!
No OTF is available in spot metering, so performance is in par with other
makes (EV 0 @ ISO 100). Non-OTF metering read at full aperture, and
_calculate_ a longer exposure time if a smaller aperture is (pre)selected.
The real spot performance is EV 0 @ F1.4, that is two seconds at that
aperture. Four minutes, or 240 secs, is 120 times longer. This is fairly
close to 128, which is two to the power of seven -- IOW, 7 stops. But the
only way to get the full four-minute exposure in spot metering is by
stopping down the lens... 7 stops. In a F1.4 lens that'll be:
-> 2 -> 2.8 -> 4 -> 5.6 -> 8 -> 11 -> 16
Some other Zuikos have f/22 or smaller, but IIRC their maximum apertures are
f/2.8 or less, so they can't be stopped down the required seven stops...
>As such, I believe that the spot-metering low-light
>sensitivity of the Nikon D3 and the OM-4T(i) are close enough to be
>non-consequential. It only makes sense since the sensor
>technology/methodology used for spot metering is similar.
That is correct, no OTF available here.
>So it is possible that the specifications for the OM-4T(i)
>may be better than the current standard pro-level camera body, but for
>determining exposure prior to the shutter opening, I think the systems are
>very comparable.
Absolutely, starting at EV 0 like most SLRs.
>Provia doesn't have a big problem
>with reciprocity failure under two minutes, but it does red-shift something
>fierce.
How do you check these colour shifts due to reciprocity failure? Unless
you're doing pinhole photography, no sunlight will be available at such low
light levels, thus the illumiant may no longer be the expected 5500ºK...
One night there was a power failure -- the whole village was in total
darkness, safe for the (full) moonlight. I took (non-OM gear, bracketing in
B) several pictures on Kodak E200 -- the properly exposed ones look EXACTLY
(colour, shadows etc) as if they were taken at noon! ...safe for the bright
dots in the sky, a.k.a. stars ;^)
>I like digital for the
>shoot-chimp-shoot aspect, but dark-frame substraction means that you are
>doubling up on your shooting time and then you run into the nasty bit of
>night-blindness caused by looking at the bright LCD.
You're telling me! :-(
>If it wasn't for the
>OTF, I wouldn't have even bothered shooting because I didn't have a clue
>what the exposure should have been.
Good point.
>Frankly, I'd like to learn from others how to better shoot digital
>after the sun goes down.
Shoot film :-) :-) :-)
From: Jez Cunningham <jez@xxxxxxxxxxxxxx>
>Two OM-4 dawn shots on Velvia here:
>http://picasaweb.google.com/jez.cunningham/Temp02
Nice pics! The first, wider one (seems a very wide angle to me) does have a
huge tonal range, maybe increased by vignetting :-( so it's a very difficult
scene, no matter the metering system.
From: Chuck Norcutt <chucknorcutt@xxxxxxxxxxxxxxxx>
>OM-4Ti TTL avg. 5 to 19 (normal temp/humidity, ISO 100, 50mm f/1.4)
This is obviously a typo, since EV 5 (without the 'minus' sign) wouldn't
meter below 1/8 @F1.4...
>The OM-4
>manual says TTL avg. is -5 to 19. But it also says TTL spot is -0 to
>19. It's not clear why the 0 should be signed
You're right, it has little or no sense at all, so it's definitely another
typo.
>which makes me leery of
>the -5 in the OM-4 manual that's not repeated in the OM-4Ti manual.
It seems the 'minus' on the 0 is the 'minus' lacking from the 5 ;^)
>The OM-2s, however, does say -5 to 17.
Similar OTF performance. The _high_ limit should be lower because the 2S has
no 1/2000 speed... but that should account for one stop less (EV 18).
Electronics may put another limit at the top, who knows.
>The eSIF's OM-4Ti claim of -7 EV for
>spot is not to be found in any of the manuals.
And it shouldn't, it's just an extrapolation.
>The OM-3Ti manual only
>lists 0 to 19.
Manual camera, no OTF (except for flash), can't go lower than the indicated
exposure -- same performance as spot.
>The OM-40 lists 0 to 18 which does not agree with the
>eSIF's -0.5 to 18.
Maybe specified with a different lens? Some OM-2 leaflets specify EV -5.5
instead of -5, but with the 55mm F1.2 lens -- half a stop faster than the
usual F1.4.
Cheers,
--
Carlos J. Santisteban Salinas
IES Turaniana (Roquetas de Mar, Almeria)
<http://cjss.sytes.net/>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|