[alohadave]

It would be more feasable to just hack it to have two seperate aperature curve settings - one for real A lenses and one for converted A lenses.

I'll have to look into the possibility of doing this once I get a chance.

That's pretty much what I meant, though you could simply use two settings, A and non-A. If and A lens is detected, use one, non-A use the other.

[Douglas_of_Sweden]

It should be noted that Av mode with a manual lens is not true Av mode. In Av mode with a modern lens you measure and focus wide open, but the camera sence the set apperture and calculate the exposure time for that apperture based on the full open measurement, and then close down when you expose. Av with a manual lens is actually so called closed down measurement, like on an old spotmatic. You have to first focus full open, then close down the apperture and expose.

[yusuf]

I guess it's still doable in software with auto calibration - camera can automatically find it out by stopping down and measuring the light that lever is linear to diameter or area.

It is certainly possible for software to compute different aperture lever motions for different type lenses, but I think there might be a few interface problems with the lens and also with the user.

As a critical example, an F1.2-F22 A type lens not set to "A" on the aperture ring looks to the camera exactly the same as all type K or M lenses (all pins grounded except "A"). How would this case be handled?

It might be that a single case of potential confusion like the above is the reason Pentax has not added more complete support for legacy lenses.

Dave

PS I've found it marginally useful to make my camera think all lenses are A type; this enables
P-TTL flash for all lenses (which I use infrequently).

[newarts]

M-mode & M-lens : On green button press it releases the aperture lever (the lens goes to the position set on the ring), meters and returns the lever to fully open. For some reason mirror goes up too while aperture is closed. Stop-down with exposure as well.

Av-mode & M-lens: continuous metering with aperture open, exposure results are the same regardless of aperture ring position although I can see aperture closing to the position set on the ring trough the lens when I press shutter fully. Green button has no effect (with its default setting at least).

With an M-lens other mode dial positions than M result in Av-mode being used. (Except movie, which somewhat surprisingly works with full aperture control with the ring :-)

Thanks, that's very clear and helpful.

Dave

[Lowell Goudge]

M-mode & M-lens : On green button press it releases the aperture lever (the lens goes to the position set on the ring), meters and returns the lever to fully open. For some reason mirror goes up too while aperture is closed. Stop-down with exposure as well.

Av-mode & M-lens: continuous metering with aperture open, exposure results are the same regardless of aperture ring position although I can see aperture closing to the position set on the ring trough the lens when I press shutter fully. Green button has no effect (with its default setting at least).

With an M-lens other mode dial positions than M result in Av-mode being used. (Except movie, which somewhat surprisingly works with full aperture control with the ring :-)

Note your last point is incorrect. With an M lens (I.e. K mount with no A or an A lens out of the A position) the camera will meter the scene for wide open aperture, and take the shot with the correct exposure for open aperture regardless of where the ring is set. The camera does not allow the aperture to close when shooting a manual aperture lens in any mode other than M (and bulb)

[Lowell Goudge]

won't move the lever on the lens. That's what confused the hell out of me by what Lowell said. He made it sound like to me M and K lenses were working like A lenses but just setting the f-stop on the aperture ring rather than the camera dials. But to the

Somewhere you are misreading things'

My specific statement was "As a result when put on a DSLR the camera will do nothing with them (assuming you set Use aperture ring) except shoot them wide open in every mode but manual. In manual the camera will let the lens stop down during exposure to the preset value on the aperture ring."

I do not understand how you can interpret "do nothing with them" as function like A lenses.

I believe you have mis resd my post, where I explain that for M42 lenses, because the lens stops down when you move the aperture ring, the camera meters the light the lens passes through and Av mode gives you the correct exposure, That was M42 not K mount lenses.

[Lowell Goudge]

I guess it's still doable in software with auto calibration - camera can automatically find it out by stopping down and measuring the light that lever is linear to diameter or area.

In theory yes, but let;s throw another curve at this (Pun Intended)

The camera metering on many later bodies is not linear, i.e. correctly reading the light and changing F-Stops accordingly. This was first noted on the K10D but exists on at least the K10D, K20D and K7D, I can't comment on the other bodies, because I don't have them to test.

As a result, if the camera does not know the maximum aperture as a starting point, it will have no way of determining the other apertures.

go back to this post in the thread

https://www.pentaxforums.com/forums/pentax-slr-lens-discussion/122264-how-mak...ml#post1263077

which shows the metering as a function of aperture with an F1.4 lens. I have taken lenses out of the A position and can confirm that they also exhibit this behavior, therefore unless the camera knows the aperture to begin with, it will not know how to apply the curve I measured, in order to determine the stopped down aperture.

[newarts]

What you really need to do is hack the firmware to allow the camera to change the aperture lever movement based on what lens is detected on the mount.

The problem is the camera can't tell the difference between an M lens and a F1.2-22 A lens . That's because both A F1.2-22 lenses (with the aperture ring *not* set on A) and M lenses have all pins shorted to ground.

I suppose the camera could ask the user if the lens is A or M when an F1.2-22 lens is detected; but users might get confused.

Users are easily confused I think.

Dave

PS I think the current Green Button solution is a good one, but I'd like to see P-TTL enabled for non-A lenses....maybe have the green button shoot the pre-flash or just do pre-flash at exposure time for all lens types? I do not understand the ramifications of such strategies.

I tricked my K-x into thinking all lenses are A type, that way P-TTL and the Green Button work with preset lenses like m42's.

[Lowell Goudge]

I think the current Green Button solution is a good one, but I'd like to see P-TTL enabled for non-A lenses....maybe have the green button shoot the pre-flash or just do pre-flash at exposure time for all lens types? I do not understand the ramifications of such strategies.

I have thought about this point a lot. and have gone from thinking you could use the green button to measure the change in aperture for P-TTL to calculate shooting aperture, only to realize on later bodies like the K10, things are so non linear the camera could not do it, to my current thinking.

The P-TTL system uses a preflash, wide open, and knowledge of the set aperture and the ambient light (without flash) to calculate the correct flash duration. If the order of functions was changed, such that the preflash was done after the lens stopped down, then the camera would see the impact of preflash lixed with ambient light at the shooting aperture, and base the exposure for non A lenses on that measurement, then the mirror would swing up and the camera would take the shot.

What is wring with this, Plenty.
- The measurement errors in the current line of cameras (which are calibrated out because the camera knows maximum aperture) mean the idea still requires knowledge of apertures (both open and shooting) to get the calculation right.
- the delay of stopping down the lens, before moving the mirror (since the mirror has to be in its normal position to reflect light to the focusing screen, where metering is done) will add an additional shutter lag, People already complain about the shutter lag of P-TTL

As far as I am concerned, the only way to get reliable flash, with K lenses is true TTL measuring light reflected off the sensor, During exposure. It would be great if pentax went back to this as a second flash control option (like it was on the *istD which supported both). Auto mode is the next best thing. and can work well in some cases.

[newarts]

What you really need to do is hack the firmware to allow the camera to change the aperture lever movement based on what lens is detected on the mount.

The problem is the camera can't tell the difference between an M lens and a F1.2-22 A lens . That's because both A F1.2-22 lenses (with the aperture ring *not* set on A) and M lenses have all pins shorted to ground.

I suppose the camera could ask the user if the lens is A or M when an F1.2-22 lens is detected; but users might get confused.

Users are easily confused I think.

Dave

PS I think the current Green Button solution is a good one, but I'd like to see P-TTL enabled for non-A lenses....maybe have the green button shoot the pre-flash or just do pre-flash at exposure time for all lens types? I do not understand the ramifications of such strategies.

I tricked my K-x into thinking all lenses are A type, that way P-TTL works with preset lenses like m42's, bellows, etc (but using M lenses is a pain in the neck - however that's ok with me, because I don't have any M type lenses, only A's & presets.)

[newarts]

I have thought about this point a lot. and have gone from thinking you could use the green button to measure the change in aperture for P-TTL to calculate shooting aperture, only to realize on later bodies like the K10, things are so non linear the camera could not do it, to my current thinking.

The P-TTL system uses a preflash, wide open, and knowledge of the set aperture and the ambient light (without flash) to calculate the correct flash duration. If the order of functions was changed, such that the preflash was done after the lens stopped down, then the camera would see the impact of preflash lixed with ambient light at the shooting aperture, and base the exposure for non A lenses on that measurement, then the mirror would swing up and the camera would take the shot.

What is wring with this, Plenty.
- The measurement errors in the current line of cameras (which are calibrated out because the camera knows maximum aperture) mean the idea still requires knowledge of apertures (both open and shooting) to get the calculation right.
- the delay of stopping down the lens, before moving the mirror (since the mirror has to be in its normal position to reflect light to the focusing screen, where metering is done) will add an additional shutter lag, People already complain about the shutter lag of P-TTL

As far as I am concerned, the only way to get reliable flash, with K lenses is true TTL measuring light reflected off the sensor, During exposure. It would be great if pentax went back to this as a second flash control option (like it was on the *istD which supported both). Auto mode is the next best thing. and can work well in some cases.

Thanks, Lowell; good discussion.

Regarding metering non-linearity, I sure see it with the K-x. My faux-A preset lenses all appear to the K-x as A type f1.2-22 lenses regardless of the actual aperture setting on the lens. Logically I should tell the camera to keep the aperture at 1.2 for the exposure (so it doesn't think it must stop down during the exposure), but no; I must tell the camera to expose for f2.8....I can't imagine why. I don't think my K100D behaved this way.

BTW, faux-A preset lenses set with the K-x e-wheel at f2.8 work well with P-TTL regardless of the actual f-stop.

Dave

PS maybe in faux-A mode, K type lenses would work if one used different f-stop settings on the lens and with the e-wheel to compensate for non-linearity? *I'm not suggesting this as a useful strategy!*

[lguckert79]

wow this sounds like it is really not worth messing up a perfectlly good lens to get an A pin in it so with that being the case why do you think they put stop down in the camera bodies? and anouther thing i think could use inproved on is a good ttl light meter in the view finder so that when useing the K and M lenses you can use the apature ring with out always hiting the stop down button so with that ill talk to ya later

[newarts]

wow this sounds like it is really not worth messing up a perfectlly good lens to get an A pin in it so with that being the case why do you think they put stop down in the camera bodies? and anouther thing i think could use inproved on is a good ttl light meter in the view finder so that when useing the K and M lenses you can use the apature ring with out always hiting the stop down button so with that ill talk to ya later

No messing with the lens, just a small, reversible modification to the camera's mount.

This modification is easily reversible & increases the utility of my pre-set lenses (Macro-Tak 50:4, MIR 65:3.5 tilt/shift, bellows, Pentax 135:2.5, among others) a bit.

Dave

[cyclone3d]

It should be noted that Av mode with a manual lens is not true Av mode. In Av mode with a modern lens you measure and focus wide open, but the camera sence the set apperture and calculate the exposure time for that apperture based on the full open measurement, and then close down when you expose. Av with a manual lens is actually so called closed down measurement, like on an old spotmatic. You have to first focus full open, then close down the apperture and expose.

This is true IF you are using an m42 lens with an adapter.... however, if you are using a manual K-mount lens that the camera thinks is an "A" lens, it does the metering and focusing wide open and then stops down when you take the picture.

[blende8]

No messing with the lens, just a small, reversible modification to the camera's mount.

This modification is easily reversible & increases the utility of my pre-set lenses (Macro-Tak 50:4, MIR 65:3.5 tilt/shift, bellows, Pentax 135:2.5, among others) a bit.

I don't understand.
What does this help?