How about starting a 50mm/1.5 lens test?

Raid - I second Howard's request. Maybe we can find the comparison on Roland's smugmug folder, but it would be nice t see here a side-by-side with the four early Sonnar variants: the Canon 1.5, the Nikkor, the CZJ and the best of the J-3s.
 
hofrench@mac.co said:
Raid,
These tests have been tremendously informative.
Can you show us anything more from the Canon 1.5?
I'd particularly like to see it compared in portrait crops against the Sonnar.
Cheers, Howard
Click to expand...


Hi Howard,

With so many lenses being tested [at specific conditions], it may not be easy or possible to match up pairs of lenses with extensive testing results. I will double check what I have in terms of images on thes etwo lenses when I return home after work.

Greetings,
 
Last edited:
jmkelly said:
Raid - I second Howard's request. Maybe we can find the comparison on Roland's smugmug folder, but it would be nice t see here a side-by-side with the four early Sonnar variants: the Canon 1.5, the Nikkor, the CZJ and the best of the J-3s.
Click to expand...

Hi John,

After I had completed my lens testing, I did two extra portrait sessions for twelve lenses only, with six lenses per session. You will not see identical images unless they are from the six lenses for each session. It was more an afterthought portrait session than a formal testing for all lenses.

I have results for all lenses from other lens test set-ups where I took photos of a doll.

I will check out what I have.
 
Sonnars only

Sonnars only

The best Sonnar only comparison I can come up with (missing the C-Sonnar though) is this, Raid (a subset of what I compiled above, all wide open):

337346957_f6xhK-X3.jpg


337346978_cwg76-X3.jpg


Your Canon 50/1.5 shot of Dana didn't come out well, IMO.

Raid might clarify via full serial number, but the best Jupiter is not necessarily early, John, when compared to the various Nikkors or the Opton, for instance.

Best,

Roland.
 
Last edited:
Thanks, Roland. You have given us more than I thought of.

My way of testing lenses has room for occasional error, and it would be better to have the entire testing replicated so that gliches can be compensated for by other back-up images. I did this with earlier lens tests when the number of lenses was around 12.
 
There is always the chance for focusing mistakes, the subject moving, etc., with these fast lenses, Raid, that's why I "filtered" when compiling. Totally normal and to be expected, IMO.

Roland.
 
I still don't know the full facts about having used such a light bulb in the test. Does it really open the door to causing differences between images based on the split second a photo is taken and based on the frequency of the light being emitted?
 
ferider said:
There is always the chance for focusing mistakes, the subject moving, etc., with these fast lenses, Raid, that's why I "filtered" when compiling. Totally normal and to be expected, IMO.

Roland.
Click to expand...

I did the same. In the last batch of cropped portraits, the images at 8.0 have one that is unsharp, so I did not show the aperture 8.0 images. I also have unsharp images for the Summilux, so I am not showing any images from it here. That is life. The different images for a given lens should be sufficient to shed some light on the performance.
 
Last edited:
raid said:
I still don't know the full facts about having used such a light bulb in the test. Does it really open the door to causing differences between images based on the split second a photo is taken and based on the frequency of the light being emitted?
Click to expand...

Not sure either, Raid. With 1/30sec exposure, in theory, you catch two cycles of the 1/60 HZ power source, and 4 cycles of the (twice as fast) light source. You should be fine if the shutter is exactly accurate. If it's a little off there should be variance, but I'm not sure if they matter as much as we see.

The candle pictures are much better, wrt flare, but then a candle is much darker than the light bulb.

Roland.
 
Whatever the reason may be, the Opton shows the light bulb the best defined in the crops above. I got it from Brian Sweeney. He is very particular about his lenses, so you cannot go wrong buying one those.
 
ferider said:
Raid might clarify via full serial number, but the best Jupiter is not necessarily early, John, when compared to the various Nikkors or the Opton, for instance.
Best, Roland.
Click to expand...
Roland, what I really know about the J-3 probably wouldn't amount to a paragraph. But correct me if I'm wrong - the design of the J-3 has been, umm... remarkably stable over the years. Design-, glass formulation- and coating-wise the Jupiter-3's from the 50's, 60's, 70's (& 80's?) are "contemporary" with the 5005 or Tokyo Nikkor, the Canon Serenar and the Zeiss Opton - nein?

Cheers!
 
John,

The J-3 variations seen here and there are most likely due to sample variations in the construction and in the maintenance resulting in wipe marks and internal dust. All examples are sharp.
 
Hi Raid,

thanks for the great labour of comparing all these lenses.
It looks to me that some of them have earnest focus issues in close distance, i.e. the Zeiss ZM Sonnar, and the Canon 50/1.5as well.

From personal expierence, I can speak for the Canon. I've used it the last years as my "universal fast normal lens" on rangefinder - but I do avoid very close distance shots due to these problems. But you will have a hard job to find a "better" lens for distances of about 3-5m or infinity. The Nikkor-S 50/1.4, in contradiction, is excellent close focus and shows a "dreamy" character with strong lights. And to tell the truth, if you want to use it for distances at 3+ meters, you better stop it down to f/11.
The Canon 50/1.4 is a great universal lens too (much better close focus) but I sometimes dislike it's modern (say: cold) character. In fact, even a Zeiss Planar 50/1.4 SLR lens (70's) gives a more "vintage" look at close distance wide open...

Most "tests" are done close focus exclusively. Results in medium distances or infinity can be very different. I agree also that small difference in exposure can have a huge effect in color rendition. The real speed difference of all these f/1.4-1.5 lenses will be big enough that these differences occur when all are tested with the same shutter speed/ f-stop ratio.

I wasn't surprised that the J-3 came out quite well, since the prewar Sonnar type where it's based on was such a great lens. But the problems seems to me to get a good example without cost-consuming adjusting it on an optical workbench..

cheers, Frank
 
Thank you, Frank.

The focus distance used for the test was pointed out to me before I did the test. Maybe it was you who suggested close/medium/far distance focusing? Anyways, it seems to be a factor. Once, someone suggested to me to use close focusing in such lens comparisons. I stuck with it.

I agree about the J-3; the Sonnar was a great design from the start, so lenses that copied this design came out as excellent lenses. Brian Sweeney has been a gentleman about helping me get my J-3's adjusted on a optical workbench.
 
You must log in or register to reply here.
Back
Top Bottom