[Rusty Rat]

Recently picked up a cheap, minty, old MF Pentax Converter-A 2X-S for use on my A series lenses. Took a few shots using my A* 200mm f2.8 and was blown away with the results -- for a converter.

I was using it on a K20D and the only proviso, the exposure compensation must be set to -2.0, otherwise the metering will be two stops off. The camera doesn't realize that the old manual 2X is there. With DA* lenses mounted, the camera still doesn't get it right. No matter, the exposure comp does the job.

This shot was on a tripod at f8 according to the camera, in reality f16. Used Program just to see how the metering etc. would be. AF assist seemed reliable.

http://home.mountaincable.net/~pcharles/IMGP2061.JPG

And this wasn't even 14mp RAW, it was shot at 10mp JPG.

Not a bad alternative for someone looking for a cheap converter, who doesn't mind going manual.

[imtheguy]

Sharp, oh yea! But I am suprised about the EV comp issue even though its one you can easily work around. I use both the 2xL and 1.4xL with my A*400/2.8 and no comp is required even though the registered focal length does not reflect the TC like the AF 1.7x does, which also requires no change in comp. Nor does my Tammy 1.4x Pz MC4. I am at a lose to explain why the 2xL and 2xS are so different. Time to hit up the experts on lens construction and http://www.bdimitrov.de/kmp/teleconverters/index.html to see what's cooking.

[Rusty Rat]

The EV issue was a surprise to me too as it had to work with the old ME Super Program. If the K20D can read the SMC-A apertures and the ME Super Program could do the same with the 2X-S attached, so why not the K20D?

You're right, calling experts . . .

[SOldBear]

... even though the registered focal length does not reflect the TC like the AF 1.7x does...

I know this is off-topic, but I can't help asking....

You seem to indicate that your AF 1.7x correctly modifies the signal so the camera knows that the actual focal length is 1.7 X the FL of the lens. Is my interpretation of your statement correct?

I'm curious because with my Tamron 1.4X and 2X TCs, the camera thinks the FL is the FL of the lens. With my Pentax 1.7X AFA, the camera requires me to enter the FL (like for a manual lens), no matter what lens is in use.

[imtheguy]

You are correct SOldBear. When I mount the AF 1.7x on my da*300/4 the camera reads f/6.7 wide open. When I mount on the A*400/2.8 it reads f/4.5 wide open (I think thats exactly it, close anyway). I have 4 other TC's and none have any effect like that.

[SOldBear]

You are correct SOldBear. When I mount the AF 1.7x on my da*300/4 the camera reads f/6.7 wide open. When I mount on the A*400/2.8 it reads f/4.5 wide open (I think thats exactly it, close anyway). I have 4 other TC's and none have any effect like that.

I realize that the Pentax 1.7X AFA tells the camera the correct aperture of the combo, but other TCs don't.

But your 1st post and my question were about focal length, not aperture.

[imtheguy]

Crapola....where are my pills?? You are correct again SOldBear and my neurological drugs must be to blame. To clarify, yes, the displayed aperture is affected by the AF1.7x but NOT the displayed focal length. edit - you have to input the FL just like with any "A" lens even using a da*300/4.

[SOldBear]

Thanks for the clarification. I was about to take my pills

[snostorm]

Hi RR,

I don't understand why it would overexpose by 2 stops. The only thing I think would cause this would be a faulty pass through link of the aperture control lever in the TC. It certainly wouldn't be caused by the light loss caused by the magnification, which really shouldn't make any difference to the meter. Light loss would cause underexposure.

The only time I'd think that improper Aperture value (the actual number) might make a difference is with P-TTL flash where (I believe) this value is used to calculate flash duration and power. In normal passive exposures, the numeric value doesn't matter, just the amount of light coming through the lens and the meter only calculates Av relative to wide open -- whatever that might be.

Take a look through the lens and see if the aperture blades are closing down properly at the time of exposure. Try the same aperture (a pretty small one) both with the RC A and without it. The blades should close down to the same point either way. Also try manually flicking the aperture lever on the lens alone, then with the RC A. If it's noticeably harder or more sluggish with the RCA, then that would give you some indication where to look for the problem.

Scott

[Lowell Goudge]

I know this is off-topic, but I can't help asking....

You seem to indicate that your AF 1.7x correctly modifies the signal so the camera knows that the actual focal length is 1.7 X the FL of the lens. Is my interpretation of your statement correct?

I'm curious because with my Tamron 1.4X and 2X TCs, the camera thinks the FL is the FL of the lens. With my Pentax 1.7X AFA, the camera requires me to enter the FL (like for a manual lens), no matter what lens is in use.

With the 1.7x AF TC yes it does modify the aperture but it does not feed any of the lens data into the camera because it is missing the 7th pin on the lens side of the TC. The TC substitutes its own data for focusing. There are some other TCs out there which also correct aperture, but I don't recall exactly which ones, I can hoewever advise that the SIGMA TCs and the Takumar A 2x DO NOT

Hi RR,

I don't understand why it would overexpose by 2 stops. The only thing I think would cause this would be a faulty pass through link of the aperture control lever in the TC. It certainly wouldn't be caused by the light loss caused by the magnification, which really shouldn't make any difference to the meter. Light loss would cause underexposure.

The only time I'd think that improper Aperture value (the actual number) might make a difference is with P-TTL flash where (I believe) this value is used to calculate flash duration and power. In normal passive exposures, the numeric value doesn't matter, just the amount of light coming through the lens and the meter only calculates Av relative to wide open -- whatever that might be.

Take a look through the lens and see if the aperture blades are closing down properly at the time of exposure. Try the same aperture (a pretty small one) both with the RC A and without it. The blades should close down to the same point either way. Also try manually flicking the aperture lever on the lens alone, then with the RC A. If it's noticeably harder or more sluggish with the RCA, then that would give you some indication where to look for the problem.

Scott

The issue of metering using a TC which does not correct aperture is not related at all to the light fall off of the TC, After all TTL metering does mean Through The Lens. The issue on the DSLRs and most predominantly the K10D and K20D is the way the light reflects of the microprism focusing screen. The metering is very non linear, and is heavily dependant on the aperture of the lens.

I have posted this chart several times,

The *istD is the most accurate metering body I have, the K7 is next and the K10D with stock screen is the worst. The K20 uses the same screen.

The issue with a TC and exposure will also be a function of the native aperture of the lens you put on, in terms of the number of stops of correction, but I usually use 1.3-1.7 stops for a 2x TC on an F2.8 lens

[Rusty Rat]

snostorm -- The aperture lever works freely with the TC in place. Checked the movement of the aperture blades and it appears to be stopping down precisely with the TC in place. Tried a number of lenses to be sure.

Lowell -- I have new screen on the way, in fact it should be here today. I'll see if that makes a difference.

[Rusty Rat]

After installing the new Katzeye screen, the exposure error dropped from -2 EV to -1.3 EV.

Curiouser and curiouser

[Rusty Rat]

More info:

Took shots in the garden with the converter on and off, using Program and -1.3 EV. With the TC in place, exposure were a tad light. Perhaps 1.7 or even -2.0 would be better. Took the TC off and fired away at the same subjects with the -1.3 EV still in place using Program and the shots were definitely a stop dark.

Unfortunately the light levels were up and down like a yo-yo thanks to the fast moving clouds. I'll have to wait until light levels are even to do a definitive test, but for now I'd say it's definitely TC and not the screen.

BTW, like the Katzeye screen, didn't like the install. Done plenty of screen changes in my day, but this was far and away the worst. The frame wouldn't snap back into place and it took two jeweller's screwdrivers to pry the frame back. The tab had to be lifted up and then pushed back down to accept the frame. Focusing is very accurate, easy even on just the ground glass portion. Speaks volumes about the poor manual focus capabilities of the factory screen. A treat to use vs. the guess focus before.

[imtheguy]

When I use a jinfinance dual-diagonal screen it generally causes overexposure of 2/3 to 1 stop. I swap screens in and out depending if its an SMC-A lens day or not. So, no surprise to me that your katzeye changes things by 2/3 stop. Its still the 2-stop issue with your lens that is the only mystery. Keep working on it.

[snostorm]

I have posted this chart several times,

The *istD is the most accurate metering body I have, the K7 is next and the K10D with stock screen is the worst. The K20 uses the same screen.

The issue with a TC and exposure will also be a function of the native aperture of the lens you put on, in terms of the number of stops of correction, but I usually use 1.3-1.7 stops for a 2x TC on an F2.8 lens

Hi Lowell,

This is interesting, and I would not have expected this, but I'm certainly not immune to making assumptions that aren't valid. . .

What is the vertical of the graph indicating? I'm assuming the luminance value of an image of a standard grey card with a consistent light source taken using stopped down metering.

So (if I actually understand this) this indicates that the metering systems are at least somewhat dependent on the max aperture (light gathering ability at widest aperture) of the lens, and that adding TC(s) would shift the baseline for the meter by decreasing the effective light gathering capability at max aperture. Greater magnification up to the limit of the AF system (for me since I use AF all the time, and consider the practical limit of the AF system to be @ f8) and the directly proportional light loss would result in an increasingly positive bias to the baseline of the metering system the longer I go (up to f8 effective at max aperture) assuming your cameras are typical.

I always thought that the metering system would essentially establish a linear baseline for all max apertures that at least fall in the range of AF capability -- since they're essentially mini computers and should be able to be tuned to offset variations in the analog sensors involved. Just shows (again) some of the difficulties involved in engineering a high performance DSLR. . .

Thanks for this insight. As you probably have guessed by now, I use TCs all the time, and this might explain some of the exposure anomalies that I've experienced, but just wrote off as being due to unusual lighting conditions in particular situations. Something else I'll have to take into consideration in the future. . .

Scott