After months of waiting - it always seemed to be two weeks away - it seems jinfinance's adapter is now available: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=200401457777

Sadly, it only works on the 35, 45 and 90 - and I have the 45, 28, and 21 - but I'll still probably pick one up anyways.

His listing says there will probably be another version which will support older lenses but that likely will not be able to focus to infinity.

Thanks for the notice. Now I have another reason to pick-up a m4/3'rd. It's a shame, though about the 28 and 21mm G lenses not working. Both of these are excellent. However, the 45/2 is one of the sharpest around and would make an excellent portrait/short-tele lens on an m4/3. And the 90 a very nice longer tele option.

Not sure why the depth of the 28 and 21mm G lenses matter. The adapter will maintain the same flange to sensor(film) distance -- unless there is some internal baffle than prevents the width of the rear protrusion. Would like to know more about why they won't work. I wonder if it's the same for the Oly and Panny versions.

I will say that it is a clever way to provide manual focus with the G lenses. Well done.

The issue seems to be that the rear of the lens extends too far into the body, but I wonder how much more clearance is needed (the adapter alone should give about 9mm). I don't have G lenses anymore, but IIRC the 28's rear element is protected by a shield. I wonder if this shield is removed could the lens work? How far do the 21 and 28 lenses extend beyond the mount?

CVickery wrote:
The issue seems to be that the rear of the lens extends too far into the body, but I wonder how much more clearance is needed (the adapter alone should give about 9mm). I don't have G lenses anymore, but IIRC the 28's rear element is protected by a shield. I wonder if this shield is removed could the lens work? How far do the 21 and 28 lenses extend beyond the mount?


Yes, but what is the rear of the lens hitting? Clearly, it can't be the sensor. Here is a picture of rear of the Contax G's 28mm. The shields on the sides of the rear element extend past the rear element, but neither the shields or the rear element would hit the sensor.

As I recall it the G1 shutter is quite a bit in front of the sensor. Can that be it?

Jonas B wrote:
As I recall it the G1 shutter is quite a bit in front of the sensor. Can that be it?


That would seem to be an odd design if it were. Is it a focal plane shutter?

There are many Leica M and m39 mount lenses that also have protruding rear lens elements -- this is typical of rangefinder wide angle design. Can these be used with the m4/3'rds cameras, or is it something particular about the Contax G 28 and 21mm lenses or the design of the adapter causing the issue?

Lotusm50 wrote:
CVickery wrote:
The issue seems to be that the rear of the lens extends too far into the body, but I wonder how much more clearance is needed (the adapter alone should give about 9mm). I don't have G lenses anymore, but IIRC the 28's rear element is protected by a shield. I wonder if this shield is removed could the lens work? How far do the 21 and 28 lenses extend beyond the mount?


Yes, but what is the rear of the lens hitting? Clearly, it can't be the sensor. Here is a picture of rear of the Contax G's 28mm. The shields on the sides of the rear element extend past the rear element, but neither the shields or the rear element would hit the sensor...


I took a peek into my E-P1 and I suspect that it may be the diameter of the rear element that is the issue. There seems to be enough depth in the centre, but the baffles around the sensor might not allow a rear lens with a diameter of ~20mm to intrude the required ~5mm (these dimensions are my guess from the diagram of the 28 on the Zeiss website). The shields would need to be removed, but looking at your shot I would think that it's nothing that a Dremel couldn't handle.

Lotusm50 wrote:
Jonas B wrote:
As I recall it the G1 shutter is quite a bit in front of the sensor. Can that be it?


That would seem to be an odd design if it were. Is it a focal plane shutter?

There are many Leica M and m39 mount lenses that also have protruding rear lens elements -- this is typical of rangefinder wide angle design. Can these be used with the m4/3'rds cameras, or is it something particular about the Contax G 28 and 21mm lenses or the design of the adapter causing the issue?




Yes, it is a focal plane shutter. Between the shutter and the sensor there is this ultrasonic thing keeping the "sensor" free from dust. It is a busy place, the area between the lens flange and the sensor.

Anyway, I gave up on WA M-lenses on my G1 when I found out about the smeared edges when I tried the CV28/2. I fear a 21mm lens with its rear element even closer to the sensor would run into troubles as well.

If one only can take the build quality, focusing by wire, absence from distance scale and such, the Panasonic G20/1.7 seem to be a better choice.

/Jonas

I suspect that you may be right Jonas. If so, it's too bad, since I really liked the 28 when I owned it. The 35, 45 and 90 would be interesting to try though.

CVickery wrote:
I suspect that you may be right Jonas. If so, it's too bad, since I really liked the 28 when I owned it. The 35, 45 and 90 would be interesting to try though.

Yes, it would be too bad. And a 20 isn't a 28 either. I haven't used any Contax G equipment but I have liked the images I have seen. I also think the feeling of the gear is important. My best images have been taken with cameras and lenses I know well, and which I like to work with. This far I haven't got that feel for any of the Panasonic lenses.

No problem for mounting and shooting with G-Biogon 21 mm on Micro 4/3 camera, at least with GF1.

G-Biogon 21's rear protruding is 18.3 mm and diameter of the rear block is 23.5 mm
G-Biogon 28's rear protruding is 17.0 mm and diameter of the rear block is 24.0 mm
G-Hologon is iffy.

As long as you take out the rear lens protectors from these Biogons in the above
you have no problem to use them on G1, GF1. See the second row of http://www.pbase.com/kkawakami/inbox

KOJI, to me that is a surprisingly good result shown in that last image.

It seems to have no wide angle problem with the censor,
also I added an infinity shot too.

G-mount has 29 mm flange back, so there is enough space
between the last element of these Biogons to m4/3 censor.

Jonas B wrote:
KOJI, to me that is a surprisingly good result shown in that last image.


The Zeiss 21mm Biogon for Contax G is an outstanding lens. Easily the equal of the Zeiss 21mm Distagon. For many years, it was my favorite. On half-frame it can only be even better.

Why does it matter if you can't use the other excellent lenses? You could put a horrible lens on M4/3, and as long as it covers the image circle (which is smaller than a postage stamp) with something decent, it works.

Lotusm50 wrote:
Jonas B wrote:
KOJI, to me that is a surprisingly good result shown in that last image.


The Zeiss 21mm Biogon for Contax G is an outstanding lens. Easily the equal of the Zeiss 21mm Distagon. For many years, it was my favorite. On half-frame it can only be even better.



I never doubted that part.

But there is something going on that I don't understand with different WA rangefinder lenses on the G1. With some of them you get smeared borders and corners, with some of them you do not. I have suspected it to have something to do with the distance between the rear element and the sensor. That is about the same as saying I think it has to do with at what angle the light hits the micro lenses.

So, I thought the Zeiss Contax G 21/2.8 would be a hopeless case. Now I have seen one single image only from the combo... but it sure looks OK.

I wonder what it is that causes the smearing.

probably poop on the sensor from the factory.

Most (if not all) 4/3 sensors have filter assemblys on top of the sensor that are quite a bit higher than the pixel wells are wide - they don't take to well to fast glass and WA's that project light onto the sensor with an angle. They prefer to be "struck" from a 90º angle. There's a very solid reason why the 4/3 consortium recommends a maximum ray angle of 10º on to the sensor. That's roughly the angles that a F/1.8-lens gives you... :-/

Angles higher than this gives high pixel crosstalk, and as a result from this - low colour accuracy and low pixel-to-pixel contrast.

theSuede wrote:
(...)
Angles higher than this gives high pixel crosstalk, and as a result from this - low colour accuracy and low pixel-to-pixel contrast.

Yes, that's the problem we have seen with many WA rangefinder lenses on the G1.

Here is a CV28/2 Ultron sample, first the scene:


and then a combined crop, left and center:


(Imageevent seem to be slow today)

Now that is wide open and you may not care about anything but the center wide open. The problem doesn't go away until stopped down to f/5.6 and then there still was some CA there.

KOJI; what aperture did you use for the Contax G images you posted at pBase?

Grnmgfh... the crop factor is killing it. A lens with 75mm field of view and 35mm depth of field is not exactly the combination I was dreaming of.
What M4:3 needs is dedicated primes for the system, not legacy lenses.
But with proper focusing rings.
If zeiss made a small 14mm lens with a small image circle, infinity stop and distance markings I'd buy two M4:3 bodies.