[rdenney]

Series Contents

For the general history of the Ukrainian PCS lenses, see Grand Lens Test 7.

Arsat is the generic brand name for Zavod Arsenal, the large defence manufacturer in Kiev that also made cameras and optics. The last lens they designed was this 55mm lens designed to provide coverage for movements. Thus, unlike the Hartblei conversions of regular lenses, this 55mm Arsat was purpose designed to be a shift lens. They were first shown at Photokina in 2000--the last gasp of the photographic innovation and production at Arsenal. There is a good description and history of these on Pentax Forums here.

Mine is serial number 885 from the year 2000.

These have a manual aperture, but unlike the Hartblei, they have an aperture switch ring that allows one to set the aperture, and the switch between that set aperture and wide open. It's not quite like a present lens, and it can also be switched using a cable release screwed into the side of the lens.

Barrel quality of these is decent and the lenses are thoroughly multicoated.

Mine is in a Pentacon Six mount and adapted to Pentax 645 using an adapter from Hartblei.

Here it is, shown at full shift:


These are reportedly excellent. A lens of normal focal length with shift is nearly as useful as a wide-angle lens, but it should be noted that this is a wide-angle lens on its 6x6 native format. In the 645z world, it has to compete with the superb 55/2.8 DFA AW SDM and all the rest of the alphabet, and the only advantage it is likely to offer (except for weapon value in case of a mugging) is its shift. But, nevertheless, that's not without value, so the test proceeds.

Here's the main full view, unshifted, from the Arsat, at f/11:


Let's compare with the 55/2.8 and see what looks different even at this degree of downsampling:


It seems to me that the Pentax 55 has a slightly shorter focal length than the Arsat, but they are close. The Pentax is contrastier, which is obvious to me despite the difference in exposure. And the camera clearly perceived the image a bit darker with the Pentax lens, metered wide open, than with the Arsat lens, metered stopped down.

Here's a center crop at f/11:


This is not too bad, to be sure. How does it compare to the 55/2.8?


Okay, the Arsat is not nearly as good. So, those who say the Arsat is excellent may need to redefine excellence. It's good (remember, at 100 pixel/inch of most monitors, these clips come from a print seven feet wide), but the Pentax lens is clearly better. The Pentax lens is also more contrasty, apparent even despite the differences in exposure between these shots.

At f/4.5, the Arsat doesn't look good at all. Here it is in the center:


Does this lens have curvature like the 45? Here's the left edge at 4.5:


Nope--the edge of the grass is about on the focal plane, and it's about as sharp as this lens gets wide open, while the foregound Azalea bush, which would be sharp no the 45mm Hartblei, is appropriatelly out of focus here.

Here's that left edge at f/11:


This is better, about as good as in the center.

And seeing how the stopped-down photos are a bit over-exposed compared to the wide-open photos, I begin to think that on this dim day, with the meter looking through a lens at f/11, it just wasn't reading as accurately as the same meter through f/4.5 or f/2.8.

One fault I am NOT seeing is lateral color. This lens is actually quite well corrected for color, which is most useful for a shift lens where adjusting for lateral color requires an off-center algorithm that defeats image software.

But it's a shift lens, so let's shift it. Here is the full image at f/8, shifted 12mm to the left:


Let's look at 1:1 crops from here and there around the image.

Here's the upper left corner at f/8:


No field curvature here. The leaves that are in focus are about in the focus plane. This isn't actually all that bad, considering the level of enlargement. It's not in the same class as the Pentax lens, of course, but this is actually quite usable. And some judicious sharpening would make a big difference, too.

Here's the left edge, again at f/8:


Clearly, the foreground foliage is out of focus, but in the focus plane it's not bad.

Here is the point at which the lens was focused before it was shifted (the focus confirmation does not work well with a shifted lens--a common problem with phase-detect autofocus):


It appears to me that the lens changed focus slightly as it was shifted, but it's not because the focus ring turned. The shift mechanism may be just slightly out of alignment. That's certainly the sort of problem Arsenal is well known for.

Here's the center of the shifted image at f/8. The focus plane seems definitely a bit behind where it was focused:


My conclusion is that this is a much better lens than the 45mm Hartblei, which is to be expected simply because the Hartblei was not designed to be a shfit lens. For the purposes of this lens, being soft wide open is no problem--most of the time it will have to be stopped down significantly to get acceptable depth of field. Thus, this is a tripod lens, and I will use it only when I actually need the shifting capability.

I don't think this one will go to Alaska. I'm just not sure it's worth the weight without that specialized need. Plus the DFA 55 has to go--it's an all-weather lens, it's by far the fastest focusing lens I own, and it's just too good to leave behind.

Rick "Usable but not world-class" Denney

---

Last edited by rdenney; 06-30-2018 at 12:15 AM.

[torashi]

Have you considered further testing with an architectural subject?
Try metering wide open, unshifted and then adjusting manual exposure accordingly.

[kaseki]

Immediately below is an image taken with the subject lens while fairly close to the subject. (This was a casual experiment.) I recall having the lens at maximum offset, and my impression of the two wider posts upward divergence is that the correction is slightly too much. The lower photo was also with this lens with some unremembered correction (I was across the street), but I did strive for the house ends to be parallel. Scans were from Portra 160 (I think) by a Noritsu HS-1800.

[rdenney]

Have you considered further testing with an architectural subject?
Try metering wide open, unshifted and then adjusting manual exposure accordingly.

For me, trees are architectural.

Rick “who’ll try that metering approach” Denney

[alex2345]

Series Contents

For the general history of the Ukrainian PCS lenses, see Grand Lens Test 7.

Arsat is the generic brand name for Zavod Arsenal, the large defence manufacturer in Kiev that also made cameras and optics. The last lens they designed was this 55mm lens designed to provide coverage for movements. Thus, unlike the Hartblei conversions of regular lenses, this 55mm Arsat was purpose designed to be a shift lens. They were first shown at Photokina in 2000--the last gasp of the photographic innovation and production at Arsenal. There is a good description and history of these on Pentax Forums here.

Mine is serial number 885 from the year 2000.

These have a manual aperture, but unlike the Hartblei, they have an aperture switch ring that allows one to set the aperture, and the switch between that set aperture and wide open. It's not quite like a present lens, and it can also be switched using a cable release screwed into the side of the lens.

Barrel quality of these is decent and the lenses are thoroughly multicoated.

Mine is in a Pentacon Six mount and adapted to Pentax 645 using an adapter from Hartblei.

Here it is, shown at full shift:


These are reportedly excellent. A lens of normal focal length with shift is nearly as useful as a wide-angle lens, but it should be noted that this is a wide-angle lens on its 6x6 native format. In the 645z world, it has to compete with the superb 55/2.8 DFA AW SDM and all the rest of the alphabet, and the only advantage it is likely to offer (except for weapon value in case of a mugging) is its shift. But, nevertheless, that's not without value, so the test proceeds.

Here's the main full view, unshifted, from the Arsat, at f/11:


Let's compare with the 55/2.8 and see what looks different even at this degree of downsampling:


It seems to me that the Pentax 55 has a slightly shorter focal length than the Arsat, but they are close. The Pentax is contrastier, which is obvious to me despite the difference in exposure. And the camera clearly perceived the image a bit darker with the Pentax lens, metered wide open, than with the Arsat lens, metered stopped down.

Here's a center crop at f/11:


This is not too bad, to be sure. How does it compare to the 55/2.8?


Okay, the Arsat is not nearly as good. So, those who say the Arsat is excellent may need to redefine excellence. It's good (remember, at 100 pixel/inch of most monitors, these clips come from a print seven feet wide), but the Pentax lens is clearly better. The Pentax lens is also more contrasty, apparent even despite the differences in exposure between these shots.

At f/4.5, the Arsat doesn't look good at all. Here it is in the center:


Does this lens have curvature like the 45? Here's the left edge at 4.5:


Nope--the edge of the grass is about on the focal plane, and it's about as sharp as this lens gets wide open, while the foregound Azalea bush, which would be sharp no the 45mm Hartblei, is appropriatelly out of focus here.

Here's that left edge at f/11:


This is better, about as good as in the center.

And seeing how the stopped-down photos are a bit over-exposed compared to the wide-open photos, I begin to think that on this dim day, with the meter looking through a lens at f/11, it just wasn't reading as accurately as the same meter through f/4.5 or f/2.8.

One fault I am NOT seeing is lateral color. This lens is actually quite well corrected for color, which is most useful for a shift lens where adjusting for lateral color requires an off-center algorithm that defeats image software.

But it's a shift lens, so let's shift it. Here is the full image at f/8, shifted 12mm to the left:


Let's look at 1:1 crops from here and there around the image.

Here's the upper left corner at f/8:


No field curvature here. The leaves that are in focus are about in the focus plane. This isn't actually all that bad, considering the level of enlargement. It's not in the same class as the Pentax lens, of course, but this is actually quite usable. And some judicious sharpening would make a big difference, too.

Here's the left edge, again at f/8:


Clearly, the foreground foliage is out of focus, but in the focus plane it's not bad.

Here is the point at which the lens was focused before it was shifted (the focus confirmation does not work well with a shifted lens--a common problem with phase-detect autofocus):


It appears to me that the lens changed focus slightly as it was shifted, but it's not because the focus ring turned. The shift mechanism may be just slightly out of alignment. That's certainly the sort of problem Arsenal is well known for.

Here's the center of the shifted image at f/8. The focus plane seems definitely a bit behind where it was focused:


My conclusion is that this is a much better lens than the 45mm Hartblei, which is to be expected simply because the Hartblei was not designed to be a shfit lens. For the purposes of this lens, being soft wide open is no problem--most of the time it will have to be stopped down significantly to get acceptable depth of field. Thus, this is a tripod lens, and I will use it only when I actually need the shifting capability.

I don't think this one will go to Alaska. I'm just not sure it's worth the weight without that specialized need. Plus the DFA 55 has to go--it's an all-weather lens, it's by far the fastest focusing lens I own, and it's just too good to leave behind.

Rick "Usable but not world-class" Denney

Okay, the Arsat is not nearly as good. So, those who say the Arsat is excellent may need to redefine excellence. It's good (remember, at 100 pixel/inch of most monitors, these clips come from a print seven feet wide), but the Pentax lens is clearly better.

I hope that no one took it seriously, because you just got bad copy of Arsat PCS 55 lens. With my sample of this lens it is possible to shift up to 24mm along the longest side of 44x33 sensor with CA at the very edge of 24mm shift, which can be easily corrected in post process.

[rdenney]

I hope that no one took it seriously, because you just got bad copy of Arsat PCS 55 lens. With my sample of this lens it is possible to shift up to 24mm along the longest side of 44x33 sensor with CA at the very edge of 24mm shift, which can be easily corrected in post process.

I would invite you to show careful examples.

Rick “acknowledging the likelihood of sample variation” Denney

[alex2345]

at f4.5, at f8, center and panorama out of 5 images stitched at f11 (the same 33x44 sensor as in Pentax645z)

Attached Images

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

[rdenney]

I’m looking only on my phone at the moment, and I need to check on a computer monitor. But given that, I’m not seeing that your photos are particularly sharper than mine in the focus plane.

I said the lens was good, but it isn’t in the same class as the DFA 55/2.8, which is a superb lens. Have you done that comparison?

Edit: looking on my computer, I see the same. The lens looks good, but at 1:1 the DFA 55/2.8 outclasses it. I don't see that the new photos added change that conclusion--at screen resolution, they look great, but they show about the same acuity when enlarged. Note that my test photos are not sharpened in any way.

Rick “whose definition of sharpness may be closer to acuity” Denney

[alex2345]

I’m looking only on my phone at the moment, and I need to check on a computer monitor. But given that, I’m not seeing that your photos are particularly sharper than mine in the focus plane.

I said the lens was good, but it isn’t in the same class as the DFA 55/2.8, which is a superb lens. Have you done that comparison?

Edit: looking on my computer, I see the same. The lens looks good, but at 1:1 the DFA 55/2.8 outclasses it. I don't see that the new photos added change that conclusion--at screen resolution, they look great, but they show about the same acuity when enlarged. Note that my test photos are not sharpened in any way.

Rick “whose definition of sharpness may be closer to acuity” Denney

Wow, you use the best methods to judge the sharpness. The next would be to print out these images looking them under magnifying glass.
DFA 55/2.8 looks to you as superb lens, because your images made by PCS 55 do not have sharpness in the center and exhibit chromatic aberrations, whereas mine remain sharp even at 24mm shift resulting in 33x92mm panorama. In contrast, your system allows you only 33x68mm panorama with PCS 55 and no shift with DFA 55/2.8.
The real "Grand test" should be comparison of DFA 55/2.8 rather with Hasselblad Distagon 50/2.8 than with PCS 55/4.5, which is almost two stops darker and, regarding its purposes, comparable only with Super-Angulon PCS 55/4.5.

[rdenney]

Sorry, I have no idea what you are saying.

No, the DFA 55/2.8 does not shift. The Arsat 55 PCS does. If you must have shift, the choice is clear between those two.

I have made a panorama from images made with the Pentax lens, and gotten sharper results even after the geometric correction in software.

I am a collector of the Ukrainian equipment and have no shortage of experience with it. The Arsat wide-angle lenses usually have limited sharpness even though they make pleasing images if not printed too large. The 55 is the best of the Arsat rectilinear wides, even those reworked by Hartblei, by a healthy margin.

You’ll note from my review that I did not find any lateral chromatic aberration at appropriate apertures, which is truly excellent for a SLR shift lens of such coverage. But the sharpness is merely good.

Rick “not understanding your complaints” Denney