Exposure issues with older manual lenses


J2R

Link Posted 25/09/2013 - 20:10
I have a Samsung GX-10 (basically a rebadged K10D) which I originally bought because I had a decent collection of Pentax lenses from film cameras over the years (28mm, 35mm, 50mm, 135mm, 300mm). I have to say I've used these lenses rather less than I expected to, largely because I've found the exposure to be rather difficult to get right with them. It's a while since I used any of them, but my recollection is that round the middle of a given lens' aperture range, the exposure set by the camera was pretty good, but as one gets closer to the edges of the range, the exposure might be an f stop or more out (can't remember which way).

I'm contemplating getting a repair done on my camera (for another reason) but am weighing up getting a newer replacement instead, and one of the factors which might sway me is if this issue has been improved in any more recent models. Can anyone using older film-era primes tell me if there are still any exposure issues, and if not, what camera are you using? Thx.

Jonathan-Mac

Link Posted 25/09/2013 - 21:00
I understand that from the K7 onwards, this was improved. My K200D suffers from the same problem. I usually meter with the green button and then decrease the shutter speed by around 2/3 of a stop to get a good exposure. Of course, it depends on the situation and the aperture used.
Pentax hybrid user - Digital K3 & K200D, film 645 and 35mm SLR and Pentax (&other) lenses adapted to Fuji X digital
Fan of DA limited and old manual lenses

johnriley

Link Posted 25/09/2013 - 21:20
Older lenses have always been supported by Pentax, but there are obviously technical issues with some of them regarding metering. This was even true of the AF film cameras and the cautionary note in the operating manual has not changed since then.

It's just a matter of knowing how much compensation to apply and with practice it should become more intuitive as time goes by.
Best regards, John

froeschle

Link Posted 26/09/2013 - 09:33
The missing aperture ring coupler in the Pentax DSLRs is the main issue.
The problem thus is not due to the lens(es) but due to the crippled mount!
The metering with the help of the green button is not accurate.
Compensation strongly varies.
It seems to have improved somewhat over the years, but not really to an acceptable level, imho.

De-cripple the K-mount:
http://www.pentaxforums.com/forums/pentax-dslr-discussion/200075-pentax-ricoh-de...

J2R

Link Posted 26/09/2013 - 10:36
johnriley wrote:
It's just a matter of knowing how much compensation to apply and with practice it should become more intuitive as time goes by.

I wonder has anyone catalogued their own experiences with this somewhere? The lens I have the most problems with is my 300mm (SMC Pentax 1:4 300mm), which I use to take bird photos with from time to time. With other lenses, where I'm typically taking landscapes, I'm normally able to take my time and fire off a couple of test shots from which I can judge the necessary adjustments to make, but birds don't tend to wait around long enough for this process.

froeschle

Link Posted 26/09/2013 - 10:54
Quote:
I wonder has anyone catalogued their own experiences with this somewhere?

Yes, unfortunately the diagrams have been removed (in German dfn):
http://www.pentaxforums.com/forums/pentax-dslr-discussion/200075-pentax-ricoh-de...
Quote:
Test procedure: An image of a white area was taken with an A-lens and a K-5 at several f-stops (x-axis), the grey scale value of the image then was determined with Photoshop (y-axis).
Blue curve: Lens set to A, aperture value set in camera (which would be similar to the usage of an uncrippled mount).
Red curve: Aperture value set at lens, usage of green button (note the deviation from the blue curve and/or a constant value: the fluctuations correspond to about two f-stops -> predictiveness?)

http://forum.digitalfotonetz.de/viewtopic.php?t=82962
Fluctuations [!] of about two [!] f-stops.

Quote:
[...] test shots [...] birds [...]

Look at the histogram and stay at the same settings after optimisation...

J2R

Link Posted 26/09/2013 - 11:20
froeschle wrote:

Look at the histogram and stay at the same settings after optimisation...

Sorry, froeschle, I don't get what you mean here.

From the stuff you quote about the red curve and the blue curve, am I right in thinking that with A lenses, and the lens set at A, the exposures should be correct, and that the only problem is with M lenses (of which my 300mm is unfortunately an example)?

johnriley

Link Posted 26/09/2013 - 11:31
A lenses and later should be no problem. Earlier lenses require a bit more input from us!
Best regards, John

froeschle

Link Posted 26/09/2013 - 12:10
Quote:
Sorry, froeschle, I don't get what you mean here.

Try to find an optimal/balanced/compromised setting (aperture, shutter speed, ISO) in M mode by looking at the histogram. Keep this setting while shooting. Even this should be more reliable than the usage of the green button.
Quote:
From the stuff you quote about the red curve and the blue curve, am I right in thinking that with A lenses, and the lens set at A, the exposures should be correct, and that the only problem is with M lenses (of which my 300mm is unfortunately an example)?

Not fully.
With an A lens, and the lens set at A, exposure is OK.
Now use one and the same lens.
With this A lens, and the lens now set at a fixed aperture (e.g. f11) and with usage of the green button, exposure is unreliable.
So, exposure is inconsistent if you set the aperture manually with the aperture ring (A, F, FA, DFA etc. lenses).
This is due to the fact that DSLRs are optimized to meter at maximum aperture.
In contrast, the green button method performs measurements at working aperture.
This results in inaccuracies (depending on lens, focal length, aperture, body, etc.).
So, the problem is not restricted to M lenses.
The cause is the missing aperture coupler in the bodies.
This problem could be fixed by de-crippling the K-mount in future cameras.

johnriley

Link Posted 26/09/2013 - 12:17
It is quibbling a bit though, isn't it? It isn't a problem with A lenses and later if they are used in the way that the bodies were designed for them to be used.

When I first started using lenses where the body controlled the aperture I thought it was a step backwards, but actually it's just the same thing. It doesn't matter what the method is as long as it's efficient.

Having said that, if the cameras allowed normal use of M and K lenses I would be more inclined to use them. As it is, I don't see the point as current lenses are as good, sometimes better. Unless of course we want to experiment, and that's something else entirely.
Best regards, John

froeschle

Link Posted 26/09/2013 - 12:41
Quote:
It is quibbling a bit though, isn't it? It isn't a problem with A lenses and later if they are used in the way that the bodies were designed for them to be used.

I don't think so. I agree with Bojidar Dimitrov:
http://kmp.bdimitrov.de/technology/K-mount/crippled_AF.html
Quote:
The only reason for the existence of this mount variation is cost cutting. This was achieved by leaving out the stop-down coupler of the body mount and the aperture ring of the lens mount. With this move Pentax sacrificed one of its strongest advantages, the unrivaled K-mount compatibility.

Quote:
When I first started using lenses where the body controlled the aperture I thought it was a step backwards, but actually it's just the same thing. It doesn't matter what the method is as long as it's efficient.

I like the "two wheel" approach (since the Z-1P) as a possible alternative, though.
E.g. setting the aperture at the aperture ring, the shutter speed with the front wheel and ISO with the rear wheel would also be a good option.
With an aperture coupler in the bodies one also could use P-TTL with M lenses.
So, the crippling of the mount leads to less options and restricts usage.
Quote:
Having said that, if the cameras allowed normal use of M and K lenses I would be more inclined to use them. As it is, I don't see the point as current lenses are as good, sometimes better. Unless of course we want to experiment, and that's something else entirely.

Turning this argument around, I would state that lots of M lenses are (at least) on par with current (zoom) lenses.
There should e.g. be no noticable difference in optic performance of M 50/1.7, A 50/1.7, FA 50/1.7, DA 50/1.8.
But there is a difference in price.
So, allowing full functionality (of M lenses) could harm sales of new lenses.

johnriley

Link Posted 26/09/2013 - 12:54
The page in question looks as though it was written a long time ago, when the first FAJ lenses appeared. This was the prelude to the digital age as the lenses came out with the new *ist film camera.

I think most current and certainly new users won't really care that they have to set lenses to "A" to use them fully - let's face it, that means lenses going back to the early 1980s can still be sued without any problems whatsoever, well over 30 years of lenses. Even older ones we have to accept some compromises, but they still work if you want them to.

You can't achieve that with Canon, there are quite a few restrictions with Nikon, Olympus restrict use to certain lenses at certain apertures....so Pentax come out pretty well in this. I don't think there's much to complain about.
Best regards, John

froeschle

Link Posted 26/09/2013 - 13:12
Old wisdom is still true .

A negligible investment would heal the problem.

Btw, the Nikon D7000[7100] offers this feature, which the K-5 [II(s)] still lacks.

geordie01

Link Posted 26/09/2013 - 13:15
I regularly use a K55-f1.8 and m28-f2.8 on my K-r and K30 and meter with the green button and it is never far away exposure wise and is easily sorted pp in silky pics if needed.

J2R

Link Posted 26/09/2013 - 13:16
johnriley wrote:
I think most current and certainly new users won't really care that they have to set lenses to "A" to use them fully

Well, actually I primarily chose the GX-10 as my first digital SLR because it could use the first-rate prime lenses I had acquired over the years for my Pentax and Ricoh (K-mount) film cameras. As mentioned above, these are 28mm, 35mm, 50mm, 135mm, and 300mm lenses, and of these only the 28mm is an A lens, all the others are M, so I do have to "accept some compromises", as you put it. Although I think the GX-10 is a great camera, I'm not sure I would have gone for it if I'd known about the metering problems with M lenses beforehand.

Maybe I just need a better workaround for using these M lenses.
Add a Comment
You must be registered or logged-in to comment.