Reid Review: 28 2.0 Ultron focus shift

[kevin m]

Sean has his review of the new 28/2.0 Ultron up. Evidently, it suffers from the same focus shift as the 35/1.4. It's very noticeable in his test shots, but his "real" pics seem fine. I wonder how much of an issue it is in practice.

 

[Tom A]

I have had the 28f2 Ultron since late May and shot extensively with it. On our flckr site there is a set tagged "Voigtlander Ultron 28mm f2.0". At the moment there are about 150 shots on it and more are spread around in the regular Flickr. Particularly, look at the shots from the Nikko Temple models - all shot at closest focus and @f2 (it was dimly lit)
I have found no focus shift problems with it. If anything, it has a formidable close focussing capability at f2 and 2.8. This is all shooting with film (Bessa R4M and Leica MP). A friend tried it out on his recently calibrated M8 and we did a side by side test of the Summicron 28/2 (6 bit coded). The Ultron did very well, center sharpness was better (marginally so) than the Summicron, but slightly softer in the corners. There was a bit of vignetting with the Ultron, but it was not 6 bit coded and that can account for that. Overall performance between the lenses is very close. The Ultron shows less flare, but the Summicron is an Aspherical and that can account for that. Sharpness from f2.8-4 is similar, neither one is better or worse than the other.
For all practical purposes, either lens will serve you well and in most cases it would be user error rather than lens performance that become the deciding factor as to which shot works and which didn't.

 

[P. Lynn Miller]

I will confirm what Tom A. has just written. I have used the Ultron 28/2 extensively in the past few months, and it is an incredible lens. I have detected no focus shift in 30+ plus rolls of film. The close focusing ability of this lens is mind-bending. I posted a large number of photos to this thread - The Voigtlander Ultron 28mm f2.0...

 

[back alley]

it begins ...

 

[Tom A]

This the Ultron 28mm f2.0 @f2 and 1/15 handheld. These are tax payments - the farmers had to bring these 60kg/125lb "bags" of rice in to the local office!
Takayama, Japan May 2008

P.S the dark shadow on the right is not fall off! There was a very large and heavy door cutting off the light.

 

[Tom A]

This is a scale model of the Nikko temple. You think you are good with Lego!! This 1/2 million pieces assembled! Ultron 28mm f2.0 at f2 and 0.7 meters. I think it was 1/8 sec handheld.

 

[Tom A]

Another shot from the Nikko temple model. I like model building and I have seen a lot of good ones, but this one is at the top of anything I have seen! Ultron 28mm f2.0 @f2 and 1/15 sec. handheld.

 

[Tom A]

Shiragawa ko, This is a UNESCO heritage site on the western part of Japan. Famous for the "praying hands" type thatched roof structures.
Ultron 28mm f2.0 @ either 5.6 or f8.

 

[Tom A]

Takayama Temple. Outside each temple there is a water well. This is probably at f4 with the Ultron 28f2.0.

 

[Tom A]

Sweden, September 2008. This is from the Autoseum in the small town of Simrishamn (southern tip of Sweden}. This is a post war Packard 8 Limo - formerly used by the Swedish Match Company. Three were brought into Sweden with the almost pre-war Packard radiators. Huge cars, 22 feet long and 2.5-3 tonnes.
Ultron 28mm f2.0 @2.8 and between 0.9m-1 distance. No focus shift to be seen!
As for resolution, look at the license plate - you can even read the small numbers on the top of it. Swedes like to keep track of things so this is the "code" for this specific plate!
This is with 400 asa bl/w film (all the shots are) and souped in Pyrocat HD, not super fine grain, but sharp enough to show up any shifts or misfocussing, be it optically or operator induced.

 

[Dan States]

Well that's an unfortunate blow for CV. Reviews like this hold a lot of sway in this limited market and this will not help sales one bit.

Separately I do wonder why Sean's 1.9 Ultron looks so flat and flare prone as mine has been flawless in all uses. I have compared it directly with my Biogon 28 and there was an amazingly small difference in total contrast.

These are the dangers of the cropped sensor digital world! The lens probably looks great on film at normal magnifications but at the pixel sniffing level it doesn't fare well.

I'll keep using and loving my Ultron 1.9. The attached shot was made at f2.4 on a cloudy day. No hood used.

Best wishes
Dan

 

[Dan States]

Here's a pixel sniffing crop of the above scan.

 

[gavinlg]

Focus shift is one of the biggest internet myth things I've ever seen. I recently bought a Canon 50mm 1.2L for my 5d, of which has a massive reputation on the internet as being a bad lens due to focus shift. The design of the lens itself causes this, and while it is true it DOES NOT AFFECT me during real life shooting. I get extremely high hit rates with the lens even in really dark conditions.


My point is don't waste your time worrying about "focus shift", as it never rears it's head in reality - only maybe during tests with rulers or newspapers or whatever.

 

[Silva Lining]

I've got a Ultron 28/2 and I have used on a Bessa R3M, Leica M3 and M6 with absolutely no problems whatsoever. Focus is spot on. To my eye its a wide summicron-like lens, with good micro-contrast and great sharpness. (YMMV 🙂 )

Maybe the reviewer had a bad copy?

For the record, I've got a Canon 50/1.2 that many people rate as a good paperweight, I also find this a great lens, sharp and contrasty where you would expect, with a character all of its own at the widest apertures.

 

[CameraQuest]

After his first day of shooting the 28/2, Sean emailed me that the 28/2 appeared to be one of Cosina Voigtlander's best lenses, but that a lot of testing remained to be done. That was obviously before he found the focus shift.

I have never noticed focus shift with any lens. It is not one of the things I look for. If I like the overall look of the image, the bokeh, and the sharpness, that is generally all I am really looking for.

It should come as no surprise a $4000 lens should test better than a $579 one. However that is not really an issue with me, as there is no way I would pay $4000 or more for any new lens, Leica or not.

I wonder what Sean's review would have been if he had not tested for focus shift.

Stephen

 

[Freakscene]

You won't detect focus shift when shooting film unless you very carefully measure MTF or resolution. The thinnest of film emulsions are around 120 μm thick, which very effectively masks the focus shift.

On the M8, however, it all becomes MUCH more apparent, VERY quickly.

One reason that Leica have moved towards using aspherical elements is to minimise spherical aberration, which also minimises focus shift. The C-V lens shows some spherical aberration and therefore focus shift. For many uses, no problem. You can also learn to get around it. 'Sharpness' is over rated, after all.

Marty

 

[P. Lynn Miller]

it begins ...

Very thoughtful contribution to the discussion...

Back to the subject at hand, I actually cannot confirm or deny focus shift in the Ultron 28/2 since I never did any scientific testing. But I will say that in actual use, the supposed focus shift has not been visible and I have been seriously scrutinising the negatives under a 10x loupe, in very high-resolution scans, and in large prints.

If you have any misgivings about the Ultron 28/2, buy a lens from Mainline Photographics in Sydney, Scott normally gives a 14 day, no questions asked, satisfaction guaranteed or your money back on all purchases including lenses.

As for Cosina not using aspherical elements in their latest creations while Leica is, Mr. Kobayashi of Cosina has repeatedly made it known that he is not a fan of digital imaging. And Cosina does not have a digital body in its line-up, so it would make perfect sense for VC lenses to be optimised for film and not necessarily for digital. While Leica has made it broadly known that it is committed to digital imaging and is making its lenses accordingly.

Plus from memory there has been a lot of discussion on this very forum how aspherical lenses do not have the good 'bokeh' or 'paint' as well as spherical lenses. Maybe someone has been listening...

 

[back alley]

Originally Posted by back alley
it begins ...
Very thoughtful contribution to the discussion...

why pick on me?

 

[Avotius]

I dont think it will be a problem, and besides the CV lenses are such great value that it wouldnt mater much anyway. I have also had my paws on a Canon 50 1.2 EF lens and found it to be really awsome and that focus shift stuff didnt mean anything in real life shooting, even wide open and close up I was easily able to get eyes in focus and nothing else.

When it comes down to actually using the lens, dont even worry about it, most lenses have focus shift to some degree but its almost always covered up in DOF and in the real world away from newspapers tacked on brick walls, there is nothing in it. Heck even my beloved Zeiss 50 Planar is said to focus shift, does that seem to have influenced the results from my Planar thread?

Oh and btw, I completely understand Back Alley's comment, this is how stinks start, but really its still a usable lens.

 

[peter_n]

You won't detect focus shift when shooting film unless you very carefully measure MTF or resolution. The thinnest of film emulsions are around 120μm thick, which very effectively masks the focus shift.

On the M8, however, it all becomes MUCH more apparent, VERY quickly.

One reason that Leica have moved towards using aspherical elements is to minimise spherical aberration, which also minimises focus shift. The C-V lens shows some spherical aberration and therefore focus shift. For many uses, no problem. You can also learn to get around it. 'Sharpness' is over rated, after all.

A very accurate comment Marty. Over on the LUF, Reid points out that the four CV lenses that show little or no focus shift are all aspherical, and the three that do, including this 28/2, are not.