[Class A]

Mystery 1:
How can different lenses have different FF/BF characteristics on one body?

Mystery 2:
Why is it not possible to use non-centre AF points with a manual lens?

Mystery 3:
Why is it not possible to use multi-zone metering with a manual lens?

Regarding M1:
My understanding of AF is that the camera uses a feedback loop to achieve maximum sharpness. The lens should then not matter at all. My hypothesis is that lenses differ in chromatic aberrations and that the latter can throw off the prism based phase comparison. Another hypothesis is that there is no 100% feedback but that the camera always applies a small offset step after having found the correct point. This would be necessary if the optical path lengths between mount and AF unit and mount and sensor respectively where slightly different in each camera and instead of doing an optimal physical alignment, some software correction takes place. If lenses differ in how they actuate this final offset step then they would yield different FF/BF phenomena.


Regarding M2:
Centre focusing and then recomposing can introduce intolerable focusing errors when the DOF is very thin. It'll be nice to be able to use all AF points for focus confirmation. My hypothesis is that the camera doesn't know the aperture setting and if the aperture is stopped down too much, outer AF points may not work anymore. Is there a way, using information about the mount contacts, to make the camera believe that the manual lens can support multiple AF points? Not sure why even "A" lenses don't support them.

Regarding M3: I know that the camera needs to know the maximum and minimum apertures of the lens and manual lenses don't communicate them. But why not ask the user to enter this data? Again, perhaps, using information about the mount contacts, one could make a manual lens appear more capable.

With respect to M2 & M3 I could imagine that there are actually no technical obstacles but that Pentax wants people to buy their new lenses rather than being snug as bug in a rug with their old Takumar glass.

Although it is great that you can use manual lenses on your Pentax DSLR with just an extra button press added, I wonder why the following support hasn't been implemented:

Allow the user to enter an aperture value in Av mode. If it doesn't match the aperture set on the lens, tough. But if it does then you wouldn't need to dial in exposure compensation yourself. This shooting mode allows quicker shooting than "manual" + button press + shutter release.

Why can't there be a mode where the camera doesn't meter wide open and then exposes, but 1. stops down. 2. meters. 3. exposes? In other words, why can I not make the camera press the AE-L/green button itself every time I press the shutter release? Would the shutter lag be too large?

P.S.: If there really are no "technical obstacles" and Pentax indeed wants to entice users to buy new lenses, that'll be totally legitimate to me. Right now, Pentax is just the right brand for me and I'm grateful that I can play with the old glass at all. Just wondering why some things are not supported and if there is a technical explanation for it.

---

Last edited by Class A; 12-05-2008 at 06:41 PM.

[Lowell Goudge]

Mystery 1:
How can different lenses have different FF/BF characteristics on one body?

Mystery 2:
Why is it not possible to use non-centre AF points with a manual lens?

Mystery 3:
Why is it not possible to use multi-zone metering with a manual lens?

Regarding M1:
My understanding of AF is that the camera uses a feedback loop to achieve maximum sharpness. The lens should then not matter at all. My hypothesis is that lenses differ in chromatic aberrations and that the latter can throw off the prism based phase comparison. Another hypothesis is that there is no 100% feedback but that the camera always applies a small offset step after having found the correct point. This would be necessary if the optical path lengths between mount and AF unit and mount and sensor respectively where slightly different in each camera and instead of doing an optimal physical alignment, some software correction takes place. If lenses differ in how they actuate this final offset step then they would yield different FF/BF phenomena.


Regarding M2:
Centre focusing and then recomposing can introduce intolerable focusing errors when the DOF is very thin. It'll be nice to be able to use all AF points for focus confirmation. My hypothesis is that the camera doesn't know the aperture setting and if the aperture is stopped down too much, outer AF points may not work anymore. Is there a way, using information about the mount contacts, to make the camera believe that the manual lens can support multiple AF points? Not sure why even "A" lenses don't support them.

Regarding M3: I know that the camera needs to know the maximum and minimum apertures of the lens and manual lenses don't communicate them. But why not ask the user to enter this data? Again, perhaps, using information about the mount contacts, one could make a manual lens appear more capable.

With respect to M2 & M3 I could imagine that there are actually no technical obstacles but that Pentax wants people to buy their new lenses rather than being snug as bug in a rug with their old Takumar glass.

Although it is great that you can use manual lenses on your Pentax DSLR with just an extra button press added, I wonder why the following support hasn't been implemented:

Allow the user to enter an aperture value in Av mode. If it doesn't match the aperture set on the lens, tough. But if it does then you wouldn't need to dial in exposure compensation yourself. This shooting mode allows quicker shooting than "manual" + button press + shutter release.

Why can't there be a mode where the camera doesn't meter wide open and then exposes, but 1. stops down. 2. meters. 3. exposes? In other words, why can I not make the camera press the AE-L/green button itself every time I press the shutter release? Would the shutter lag be too large?

P.S.: If there really are no "technical obstacles" and Pentax indeed wants to entice users to buy new lenses, that'll be totally legitimate to me. Right now, Pentax is just the right brand for me and I'm grateful that I can play with the old glass at all. Just wondering why some things are not supported and if there is a technical explanation for it.

I have been on to pentax about this for years with no movement, especially on entering apature (ax and min, and letting the camera control the apature like it does on A lenses. The issues are, they would have to rescale the movement because A lneses have the apature area change linearly with movement of the lever, non A lenses have the diameter change linearly with movement. it may be difficult to control smaller apatures, and it depends upon specific scaling and location of the levers, which did not need to be controlled on non A lenses as it was just an on off switch.

With respect to using the matrix metering, you could convert the lens to an A mount, I tried it, but the camera then tries to control it like an A lens and it does not expose properly,

with respect to the proposal to incorporate metering prior to shutter release, it would need to be before mirror up, because metering is off the focusing screen, and ther would be a lag. Overall if you can meter, shooting manual is not that bad. You also have to contend with the ultra bright viewing screens of the K10D and K20 D which screw up manual apature lenses and metering

[Ole]

There are some technical obstacles regarding M and K lenses. The camera cannot set the aperture of these lenses even if it knew the max and min aperture: On A and later lenses the aperture can be set by the camera by moving the stop down lever a calculated distance because these lenses are calibrated so that a given movement (say 1 mm) of the stop down lever closes the aperture down by a given value (say half a stop).

M and K lenses are not calibrated like that - the lever was just ment to close the aperture to the pre-set value just like on a Takumar screw mount lens.

So we can forget about using these lenses as if they were A lenses.

Now, exposure for M and K lenses could be used reliably via the 'green' button on the *istD series. That is no longer possible with K10D/K20D (I can't speak for the K100/K200) because the focusing screen has been "enhanced" to show a brighter image, but its brightness doesn't change propotional to the f-stop. And the exposure meter meassures the light off of the focusing screen (one solution is to exchange the screen for a *istD screen).

The ultimate solution might be a metering system like the LX thus cutting the focusing screen out of the loop.

[Lowell Goudge]

The ultimate solution might be a metering system like the LX thus cutting the focusing screen out of the loop.

while this might be the solution, i wold not want to wait for it to appear, unless a real pro camera comes out, and by the way re-introduces TTL metering.

WHile P-TTL has some interesting uses, true TTL is far superior for wild life photography, but it costs too much. That;s why it was dropped, it required a second analog inout into theprocessor for the reflected light off the sensor. It was argued that the sensor was not consistent to reflect light, but I never had a problem with my *istD, it is all about cost.

[Class A]

The issues are, they would have to rescale the movement because A lenses have the aperture area change linearly with movement of the lever, non A lenses have the diameter change linearly with movement.

I've read about this difference but didn't know that non-A vs A was the dividing line.

The rescaling should be a piece of cake for a DSLR. Its seems to be a matter of "don't want to" rather than "cannot be done".

Anyhow, it'll be already nice if the camera could only select the preset aperture but expose correctly with it in Av mode. AFAIC, that would be like a normal Av mode except that you would have to preset the aperture at the lens and enter the same value into the camera. No rescaling issues then (but no Tv mode either, of course).

With respect to using the matrix metering, you could convert the lens to an A mount, I tried it, but the camera then tries to control it like an A lens and it does not expose properly

I reckon because of the lack of rescaling of the aperture lever position.

I still don't understand why the camera cannot use matrix metering to determine the shutter speed and then just set the aperture to the preset value. Currently, the use of matrix metering is coupled with controlling the aperture in the style of an A lens, but I don't see why this coupling is necessary.

with respect to the proposal to incorporate metering prior to shutter release, it would need to be before mirror up, because metering is off the focusing screen, and ther would be a lag.

But maybe the lag would be acceptable. It would certainly be quicker then first manually forcing metering and then releasing the shutter.

So we can forget about using these lenses as if they were A lenses.

Without a rescaling of the aperture level position, yes. Although it would be nice to control aperture just from the body, I wouldn't terribly mind to set it at the lens, but not lose multiple focus points and matrix metering at the same time.

(I can't speak for the K100/K200)

I only read about K10D users complaining. My limited experiences with a K100D have been positive.

Thanks for all the input!

For all you mystery solvers out there: There is still work to do!

[kristoffon]

Mystery 2:
Why is it not possible to use non-centre AF points with a manual lens?

My K200D does it and pops the red spots on screen as it tries to check focus.

Mystery 3:
Why is it not possible to use multi-zone metering with a manual lens?[/b]

My K200D at least allows the selection, and shows the metering mode I select in the info screen when a manual lens is attached.

What I want to know is why, with a manual lens attached, will auto-metering consistently underexpose by 1.5 - 2 stops yet green button metering always does it perfectly?

[Class A]

My K200D does it and pops the red spots on screen as it tries to check focus.

My K200D at least allows the selection, and shows the metering mode I select in the info screen when a manual lens is attached.

What lens are you talking about?

Unless the K200D is different from the K100D, it appears your lens is either not a fully manual one or it has been tinkered with to tell the camera it is something more modern.

What I want to know is why, with a manual lens attached, will auto-metering consistently underexpose by 1.5 - 2 stops yet green button metering always does it perfectly?

That seems to indicate that it is indeed a lens where the aperture position is proportional to the diameter and not the area of the aperture. The camera seems to treat is as an "A" lens and hence doesn't use the correct aperture lever position.

Please check the mount of the lens. Do you see any changes done to it?

[kristoffon]

What lens are you talking about?

Unless the K200D is different from the K100D, it appears your lens is either not a fully manual one or it has been tinkered with to tell the camera it is something more modern.



That seems to indicate that it is indeed a lens where the aperture position is proportional to the diameter and not the area of the aperture. The camera seems to treat is as an "A" lens and hence doesn't use the correct aperture lever position.

Please check the mount of the lens. Do you see any changes done to it?

My manual lens is a 500mm Rokinon mirror with fixed f/6.3 aperture, screw-mounted with adapter. I don't think it gets any more manual than this

[Class A]

I don't think it gets any more manual than this

The focus could be stuck to 10m.

I reckon either the K200D is different or you have a funky M42 adapter.
Could you please check whether the adapter deviates from a normal M42 adapter (which is just a plain piece of a metal ring, I think, without any contacts.)?

[kristoffon]

The focus could be stuck to 10m.

I reckon either the K200D is different or you have a funky M42 adapter.
Could you please check whether the adapter deviates from a normal M42 adapter (which is just a plain piece of a metal ring, I think, without any contacts.)?

It's definitely just a plain piece of metal. Also, I think it's not M42 but T-Mount (or something with a similar name)