The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Contax G to MFT adapter - anyone?

Jonas

Active member
Hmmmm. I guess I understand what you're saying, but 1:1 looks different on different monitors, doesn't it? So isn't it a kind of poor universal unit of measurement? Like I said, I'm using a 1920x1200 monitor. So 1:1 magnification on my monitor is something like 1456x1120 for an M43 shot. If I were to use a 1650x1080 monitor, wouldn't the 1:1 magnification be a smaller area of the picture?
Maybe it is easier if you think of pixels rather than the absolute size of the image. The expression "100% crop" is by itself not very clear but there is a consensus of the meaning of it.

My G1 images are 3000x4000 pixels. They don't fit the screen. Whenever an application shows me the whole image it has been interpolated to a lesser dimension, counting the pixels.

Most image editors can be set to show the image at full size = 100% size = natural size = native size or whatever they call it. In Photoshop this is done by, for example, setting the image to 100% in the little Navigator window, or by choosing View-->Actual Pixels. Then one has to scroll around to see all parts of the image. Now every single pixel in the image corresponds to a single pixel on the monitor. That is the largest "magnification" that ever is useful for judging details in an image (pixel peeping indeed). If we set the image to more than 100% size it is enlarged and the image viewer interpolates what pixels between the native 12.000.000 should look like, similar to what it does when it shows a smaller version of the image.

So, sometimes we want to see a 100% crop. That is a part of the image only (saves bandwidth and downloading time), but we can see it just as when the image is set to be viewed in 100% size.

Look at my dirty B&W image above. It's resized, it doesn't tell us much about details. Before processing it I set it to 100% and marked a part of interest using the Rectangular Marquee Tool in Photoshop and copied that part to a new image. That is the color version of the girl's eye. A 100% crop. I could have marked a bigger or smaller part of her face, it still would have been a 100% crop.

I'm sorry if I explained things you already know, and for repeating what Jonathon already said, but I hope you now have the same idea about what a 100% crop is as everybody else.

regards,

/Jonas
 

Jonas

Active member
Thanks for your analysis Jonas. I have both the 20 1.7 and PL 45 2.8 and I agree with your assessment of relative sharpness of these two lenses. I like the PL 45 for its primary role as a macro lens, but I'm less enthusiastic about using it for portrait and most landscape work (the exception being landscape detail). The Contax G 45 seems to be readily available at reasonable prices so I may pick one up and see for myself how it compares with the PL 45.
Ah, not much of an analysis I'm afraid, just some early impressions. It will be interesting to learn about your take on this if you buy a Contax G 45. The Contax G lenses are an interesting group of legacy lenses as they are made so recently, so good and still available at reasonable prices. Unfortunately the G45/2 is the only one I'm interested in. Maybe the 90mm but it's a very long lens for me and so it wouldn't see a lot of use and I have a couple of Voigtländer lenses around that focal length.

/Jonas
 

lattiboy

New member
Maybe it is easier if you think of pixels rather than the absolute size of the image. The expression "100% crop" is by itself not very clear but there is a consensus of the meaning of it.

My G1 images are 3000x4000 pixels. They don't fit the screen. Whenever an application shows me the whole image it has been interpolated to a lesser dimension, counting the pixels.

Most image editors can be set to show the image at full size = 100% size = natural size = native size or whatever they call it. In Photoshop this is done by, for example, setting the image to 100% in the little Navigator window, or by choosing View-->Actual Pixels. Then one has to scroll around to see all parts of the image. Now every single pixel in the image corresponds to a single pixel on the monitor. That is the largest "magnification" that ever is useful for judging details in an image (pixel peeping indeed). If we set the image to more than 100% size it is enlarged and the image viewer interpolates what pixels between the native 12.000.000 should look like, similar to what it does when it shows a smaller version of the image.

So, sometimes we want to see a 100% crop. That is a part of the image only (saves bandwidth and downloading time), but we can see it just as when the image is set to be viewed in 100% size.

Look at my dirty B&W image above. It's resized, it doesn't tell us much about details. Before processing it I set it to 100% and marked a part of interest using the Rectangular Marquee Tool in Photoshop and copied that part to a new image. That is the color version of the girl's eye. A 100% crop. I could have marked a bigger or smaller part of her face, it still would have been a 100% crop.

I'm sorry if I explained things you already know, and for repeating what Jonathon already said, but I hope you now have the same idea about what a 100% crop is as everybody else.

regards,

/Jonas
Thanks for being very clear about the explanation; I do understand what both you and Jonathon are saying, but I guess this is more theoretical than practical question:

What if you had a 3000x4000 resolution monitor and clicked "native size" in Photoshop? Your 100% crop would be a full "view" of the picture, wouldn't it? If that is true, why wouldn't the inverse be true if you have a 1200x800 resolution monitor?

Wouldn't something like a set number of pixels (say 20% of the whole image, so 240,00 pixels on M43) be a better measurement of detail/sharpness/noise?
 
Last edited:

Jonas

Active member
(...)
What if you had a 3000x4000 resolution monitor and clicked "native size" in Photoshop? Your 100% crop would be a full "view" of the picture, wouldn't it? If that is true, why wouldn't the inverse be true if you have a 1200x800 resolution monitor?
Hmm. If I had such a monitor and viewed the image at 100% I wouldn't need to scroll around. As I would see everything at the same time it wouldn't be a crop.
With a 1200x800 monitor and a situation where I look at the whole image the image viewer has downsized the image from the original 3000x4000 pixels. Interpolation has taken place. I am not looking at one pixel from the image file corresponding to one pixel of the screen.

[/QUOTE]
Wouldn't something like a set number of pixels (say 20% of the whole image, so 240,00 pixels on M43) be a better measurement of detail/sharpness/noise?[/QUOTE]

That can be discussed for a long time. We choose 100% crops as the pixels are native, untouched, not interpolated. Interpolation will affect "sharpness". With a 100% crop we all know what we are looking at and we can download, resize a little an sharpen (if we want to, with some training it is easy to judge a 100% crop as either sharp or not).

Noise is best judged at print size. When comparing noise between cameras I think the best method is to resize all the images to the same size.

When i look at my own images at home I often prefer 50% crops as it is better compared to my print size than a 100% is.

that's my cents on this,

/Jonas
 

Jonathon Delacour

Subscriber Member
We choose 100% crops as the pixels are native, untouched, not interpolated. Interpolation will affect "sharpness". With a 100% crop we all know what we are looking at and we can download, resize a little an sharpen (if we want to, with some training it is easy to judge a 100% crop as either sharp or not).

Noise is best judged at print size. When comparing noise between cameras I think the best method is to resize all the images to the same size.

When i look at my own images at home I often prefer 50% crops as it is better compared to my print size than a 100% is.

Jonas
Exactly. Because a 100% crop is showing us native, non-interpolated pixels, we then have a common frame of reference for comparing images made with different lenses at the same aperture (or the same lens at different apertures). Many thanks, Jonas, for your clear and helpful contributions to this discussion.
 

lattiboy

New member
Thanks fellows. Nice discussion and explanation.

As to my other question, has anybody used both the 35mm and the 45mm Contax lenses? I've got a chance to get the 35mm for $230 with caps, hood, filter.......
 

ReeRay

Member
Thanks fellows. Nice discussion and explanation.

As to my other question, has anybody used both the 35mm and the 45mm Contax lenses? I've got a chance to get the 35mm for $230 with caps, hood, filter.......
I use the 35, 45 and 90 Contax G lenses on my GF1 with the RJ adapter

The latter two are awesome. Whilst the 35 is very good it is generally regarded as less sharp and less contrasty than the others and indeed this is what I have discovered.

This was shot wide open on the 90mm from the back of a very crowded area at 640 ISO




These links will show the 45mm detail (no artistic intent)

Full size
http://farm3.static.flickr.com/2711/4289461659_81ba979b59_o.jpg
100% Crop
http://farm5.static.flickr.com/4023/4290201474_00a5726a33_o.jpg

Hope this helps
 

monza

Active member
IMHO the 35mm has an undeserved internet reputation simply because the MTFs that Zeiss publishes are not quite to the high levels of the other G lenses. I would describe it as nothing less than excellent. Other G lenses are superb.

Geared adapter now in production in California. ;)
 
P

Punchman

Guest
That's great news Monza! When will actual products be available? Can't wait to read buyer reviews.
 

lattiboy

New member
Thanks Monza and ReeRay. I can't decide about the 35/2. I think the extra 10mm would make a big difference, but it might be a little too close to the 20mm to make sense for me. I'm not a super-serious shooter and don't want to re-develop a terrible case of G.A.S. Eh, we'll see....




PS ReeRay, those pictures of yours are what made me go after Contax stuff. Glad I saw them!
 

ReeRay

Member
Thanks Monza and ReeRay. I can't decide about the 35/2. I think the extra 10mm would make a big difference, but it might be a little too close to the 20mm to make sense for me. I'm not a super-serious shooter and don't want to re-develop a terrible case of G.A.S. Eh, we'll see....




PS ReeRay, those pictures of yours are what made me go after Contax stuff. Glad I saw them!
The 20mm equates to 40mm whereas the 45mm is the equivalent of 90mm (35mm format)

Don't forget to double everything!
 

Howard

New member
Monza,

Thank you for the update. Now that the adapter is in production, when do you expect that it will be available for distribution to users?

Howard
 
F

FaradayCage

Guest
Geared adapter now in production in California.

Ready to hear details!
 

Jonas

Active member
More initial results. They aren't surprising but it is good to learn that all I've heard is true.

G1, Contax G Planar 45/2:


A Zeiss made in japan, very sharp, as it often is with Zeiss lenses. When running a series of images through IMA-test the lens performed as good as other classic very good lenses (comparable to OM90/2Macro (sharper than Leica Elmarit-R 100/2.8 Macro APO) and a tad better than the OM50/2Macro and ZD50/2 Macro also if not visible in small prints).
The borders are sharp at f/2 and f/2.8, then they are very sharp.
The corners are sharp at f/2.8 and very sharp at f/5.6.
As IMA-test targets are even more boring than USAF charts I post the later:




Very good​


The lens also passed a simple flare test. Not only passed, it performed among the best I have seen. When mounting a filter it got worse but not worse than one can predict. Below is the f/2 result, no filter:



Very very good​


In real life it means one doesn't need to worry. It's great with a late standard construction. Here is a simple sample from a short walk yesterday, shooting distance may have been around 8 meters or so:



Wide open, in a back yard of an office, overcast. I know, Barnack says "No statues" but...​

And finally, a 100% crop taken from the image above:

You judge​

It's like having a new reference lens.

regards,

/Jonas
 

scho

Well-known member
More initial results. They aren't surprising but it is good to learn that all I've heard is true.

G1, Contax G Planar 45/2:


A Zeiss made in japan, very sharp, as it often is with Zeiss lenses. When running a series of images through IMA-test the lens performed as good as other classic very good lenses (comparable to OM90/2Macro (sharper than Leica Elmarit-R 100/2.8 Macro APO) and a tad better than the OM50/2Macro and ZD50/2 Macro also if not visible in small prints).
The borders are sharp at f/2 and f/2.8, then they are very sharp.
The corners are sharp at f/2.8 and very sharp at f/5.6.
As IMA-test targets are even more boring than USAF charts I post the later:




Very good​


The lens also passed a simple flare test. Not only passed, it performed among the best I have seen. When mounting a filter it got worse but not worse than one can predict. Below is the f/2 result, no filter:



Very very good​


In real life it means one doesn't need to worry. It's great with a late standard construction. Here is a simple sample from a short walk yesterday, shooting distance may have been around 8 meters or so:



Wide open, in a back yard of an office, overcast. I know, Barnack says "No statues" but...​

And finally, a 100% crop taken from the image above:

You judge​

It's like having a new reference lens.

regards,

/Jonas
Very nice. Thanks for the lens evaluation Jonas. Here is an interesting discussion of lens testing with some reference to the G1 and kit lens.
http://technology.lenswork.com/2010/02/lens-testing.html
 

lattiboy

New member
Yeah Jonas, thanks for doing this in a more scientific manner than "OMFG SOOO SHARP!!", which is essentially what I was doing :) The only mark against this setup would be the bokeh. It's fine once you get to f/4, but it is quite busy beforehand. If I'm not mistaken, this is true of all Zeiss stuff.

Also, It figures I'm finally able to afford glass this good and I find the focusing to be so maddening.
 

Jonas

Active member
Very nice. Thanks for the lens evaluation Jonas. Here is an interesting discussion of lens testing (...)
Lol, asking for more?
But really, it is amazing how much one can learn from taking an image of a flat surface with lines on it. I liked the PDF file in that article, it gave some food for thoughts.
Thank you for the link.

Yeah Jonas, thanks for doing this in a more scientific manner than "OMFG SOOO SHARP!!" (...)
Thank you for looking/commenting. Pseudo scientific, but a way for me to learn about lenses I figure I may use a lot.
a) Yes, Zeiss is often like that.
x) My results can't be compared with results from other tests, be they made by a single person somewhere, or by Photozone or somebody else. But I can compare to my own results.

Here, for example, some flare results, all at f/2:

(So much for DPR not finding any flare in the G20/1.7)

regards,

/Jonas
 

CPWarner

Member
So any updates on the adapters? I see the kippon ones but am interested in the one by monza. I tried sending a PM, but his PM box is full.
 

monza

Active member
Just waiting for the first production...will post more when I have reliable data. :)

Jonas, how did you do those flare tests?
 

gmoe

New member
Hi monza, will the gear be bigger than the ones from the other adapter?

Thanks! And yes keep us all posted!
 
Top