[Jonathan Mac]

I've recently acquired my fourth K-mount 28mm manual lens, an M 28mm f/2.8 version 2 (different optics from the version 1, same optics as the A version) and decided to do a comparison between my manual 28mm lenses, as I have some of what are reputed to be the sharpest K-mount 28mms around. The list is:

• M 28mm f/2.8 version 2
• M 28mm f/3.5
• K 28mm f/3.5
• Vivitar 28mm f/2.0 (K01 in the Vivitar 28mm Bestiary, manufactured by Komine - Great Vivitar 28mm Bestiary: Details)

I've previously owned a version 1 M 28mm f/2.8 and wasn't impressed by it. My use so far of the second version seems to indicate it's much better.

The M & K f/3.5 lenses have different optical formulae, in case anyone thinks they are the same. The K is (typically for the line) much bigger and heavier than the M.

All my copies of these lenses are in good condition, in fact the three Pentax ones are very close to mint. There's some slight dust in the optics as to be expected for their age, the Vivitar is worst in this regard. I would expect the worst effect of this to be to lower the contrast slightly.

All shots were taken on a K200D (with Katzeye installed) on a tripod at ISO 100, daylight WB, focus at infinity, rubber lens hood, in manual mode and then processed in the Pentax (Silkypix) software. Some adjustment was made to exposure for the first series, but the adjustment was the same for each photo. All other settings were at their default. I would normally add contrast and maybe saturation when using old manual lenses (as I feel they often lack these qualities) but in this case I left them as they are.

The aperture and shutter settings were maintained, except for the Vivitar shot at f/8, where I needed a slower shutter speed to get the same exposure, indicating that the Vivitar lets in less light than the others when stopped down.

I compared the lenses with a wide aperture (f/3.5 for the two slower lenses, f/2.8 for the faster two) and a small aperture (f/8), so there are two shots from each. The full-sized shots are first, followed by 100% crops.

I noticed the dust spot at the top-right when I got home.

1. M28 2.8 vII, f/2.8, 1/2500s


2. M28 2.8 vII, f/8, 1/400s


3. M28 3.5, f/3.5, 1/1600s


4. M28 3.5, f/8, 1/400s


5. K28 3.5, f/3.5, 1/1600s


6. K28 3.5, f/8, 1/400s


7. Vivitar 28 2.0, f/2.8, 1/2500s


8. Vivitar 28 2.0, f/8, 1/250s


Crops (in same order):

1. M28 2.8 vII, f/2.8, 1/2500s


2. M28 2.8 vII, f/8, 1/400s


3. M28 3.5, f/3.5, 1/1600s


4. M28 3.5, f/8, 1/400s


5. K28 3.5, f/3.5, 1/1600s


6. K28 3.5, f/8, 1/400s


7. Vivitar 28 2.0, f/2.8, 1/2500s


8. Vivitar 28 2.0, f/8, 1/250s


The second set I did to give another different comparison shot. These were not shot on a tripod but the other settings were maintained. Focus was on the front wall, but when I went to do the 100% crops I found that sometimes the focus was slightly off, so I didn't bother. At normal viewing sizes they are fine though.

I kept these to wider apertures: f/3.5 for the two slower lenses and f/2.8 & f/4 for the two faster ones.

9. K28 3.5, f/3.5, 1/800s.


10. M28 3.5, f/3.5, 1/800s.


11. M28 2.8 vII, f/2.8, 1/1250s.


12. M28 2.8 vII, f/4, 1/640s.


13. Vivitar 28 2.0, f/2.8, 1/1250s.


14. Vivitar 28 2.0, f/4, 1/640s

---

Last edited by Jonathan Mac; 07-23-2012 at 12:32 AM.

[Jonathan Mac]

My conclusions.

1. All of the lenses are good but I'd say the Vivitar is the poorest performer. It's not as sharp at wider apertures and is lacking in contrast (though the internal dust might be partly to blame for this).

2. The two f/3.5 lenses are both excellent, very, very sharp even wide open. I don't know about everyone else, but I don't see a significant difference (if any!) in sharpness between the f/3.5 & f/8 shots.

3. The differences in colour cast become apparent with this test. The M 28/2.8 seems warmer than the others (which I like), the M28/3.5 seems cooler, the K seems the most neutral and the Vivitar has quite an ugly green cast to it.

4. Some more testing will be required before I choose an M or K f/3.5 lens to sell (the original purpose of the test)...

5. Of the two M lenses, the f/3.5 is doubtless sharper, but the f/2.8 seems no slouch, even wide open, and the warmer colour cast make it preferable to me. I also think the contrast is better, not surprising as it's newer than the others. My favourite type of photography is portraiture, so others may disagree as to the usefulness of the warmer colours.

Please let me know your opinions, I'll be interested to read them.

[lytrytyr]

The K is (typically for the line) much bigger and heavier than the M.

What we call the K lenses (for convenience) are not a uniform line.
They range from bayonet versions of the M42 lenses
through to predecessors of the A series.

The K24/2.8, for example, is an M lens in all but name.
Sure, it has the 52mm filter thread, but so does the M200/4.

[reeftool]

I checked out your thread because I'm using a Super Takumar 28/3.5 for the month in the Single in July challenge. Your shots all look pretty good with the Pentax lenses being slightly sharper and with noticably more contrast. One thing I have noticed with the Tak is resistance to flare. I don't have a hood for it and have been shooting in some very challenging light. Your M and K should have better lens coatings and perform even better. So far, I have been impressed with the Tak. I also have some other M42 28's but have never done a real test comparing them but the 28 Tak has found a semi permanent home on my K 10D. It's a very good size for a walk around and It's FOV is pretty close to normal on aps-c.

[stevebrot]

What we call the K lenses (for convenience) are not a uniform line.
They range from bayonet versions of the M42 lenses
through to predecessors of the A series.

The K24/2.8, for example, is an M lens in all but name.
Sure, it has the 52mm filter thread, but so does the M200/4.

What unifies the K series is the labeling and the fact that all were available for the K-series bodies. Many were direct descendents of the M42 SMC Takumars and some were the predecessors of the M-series lenses that followed. Many of the K-series lenses carried over unchanged (including labeling) to the M-series bodies due to adequate existing inventory, not that they represented some sort of proto-m variant. The K24/2.8 is a good example of this. Its size and build as well as its year of release (1977) is typical for a K-series lens. It was, however, a rather specialized and expensive lens and even though production ceased shortly after the M-series bodies hit the market, existing inventory was sufficient to meet demand for several years. It was only with the advent of the A-series mount that a new version was needed.



Steve

[stevebrot]

All of the lenses are good...

It has been said that it is pretty hard to find a bad 28mm...at least in terms of sharpness. The traditional points of failure for 28mm lenses on 35mm film are:

• Flatness of field (focus is not the same across the frame)
• Barrel distortion
• Vignette
• Resistance to flare

For use as a normal lens on an APS-c dSLR you can add nasty bokeh to the list. These lenses were designed for landscape and interior shots and shallow DOF shooting was not in the anticipated use cases.

It should be noted that the Pentax 28/2.8 was never a premium offering in any of its forms. For K- and M-series the lens was priced as a budget offering. As an example, in 1982 I paid twice as much for my Tamron 28/2.5 as I would have for a Pentax-M 28/2.8 in the same store and that was before springing for the adapter. A Vivitar similar to your K-01 would have been quite a bit more money than the Pentax based on its maximum aperture. Its poor showing may be due to compromises associated with that fast aperture.

As for optical formulas...you are so, so correct! It seems that Pentax could not decide how they were going to build a 28mm lens. Needless to say, most models are relatively inexpensive today and provide a nice "wide-normal" manual focus option for Pentax APS-c cameras.


Steve

[lytrytyr]

What unifies the K series is the labeling

There is no K labeling. For example,
whereas the M50/1.4 is labeled SMC PENTAX-M 1:1.4 50mm,
the K24/2.8 is just labeled SMC PENTAX 1:2.8 24mm.

and the fact that all were available for the K-series bodies.

The K-series bodies were introduced in 1975,
but the K24/2.8 was not introduced until 1977.

Many were direct descendents of the M42 SMC Takumars and some were the predecessors of the M-series lenses that followed.

This is true, but does not cover the full range of what we are now calling the K series.

Many of the K-series lenses carried over unchanged (including labeling) to the M-series bodies due to adequate existing inventory, not that they represented some sort of proto-m variant. The K24/2.8 is a good example of this. Its size and build as well as its year of release (1977) is typical for a K-series lens. It was, however, a rather specialized and expensive lens and even though production ceased shortly after the M-series bodies hit the market, existing inventory was sufficient to meet demand for several years.

Bojidar Dimitrov, on his page

K 24/2.8

specifies 1977-1984 as the years of production for this lens,
exactly the same as for M lenses like the 20/4,
and thus all the way through to the introduction
of the A series bodies with the Super A in 1983.

[repaap]

Nice test, I was planning to do something similar with my s-tak. 28/3.5, pentax -F 28/2.8 and cosina cosinon 28/2.8. Just out of curiosity.

What I see from your shots is that K has really nice natural colours and sharpness, I'd propably prefer it over M. But there is size adwantage on M... But you have M 2.8 already.

[Schmidlapper]

Thanks for the comparison. Question what would be the largest common aperture of the lenses tested? f5.6 I'm guessing. I have a Vivitar K06 28/2.8 CF and the color cast is generally similar to your K01 out of camera, but when adjusted for daylight in Aperture 3 in post processing it tends to render vibrant colors especially red vs my k and m pentax's which do remain warm. In my own case I notice this when compared to my K24 and I actually cary the Vivitar more because of it.

[johnha]

Interesting comparison. I'm using an SMC-M 28/2.8 v1 at the moment (sometimes also a Tamron Adaptall-1 28/2.8 on a KA mount). I also have a Super-Tak 28/3.5 (58mm filter thread) which suffers from flare in back lit situations. I'm still deciding whether AF or KA functionality is important.

[abacus07]

Excellent test. It's nice to see how well the ordinary M28/2.8 holds it own against two very highly thought of lenses. I'm currently using my K24/2.8 most of the time at this focal length but I've always gotten good results with my M28.

I should do my own test sometime since I have a pile of lenses but I never seem interested in putting in the work.

[Jonathan Mac]

Thanks for the comparison. Question what would be the largest common aperture of the lenses tested? f5.6 I'm guessing. I have a Vivitar K06 28/2.8 CF and the color cast is generally similar to your K01 out of camera, but when adjusted for daylight in Aperture 3 in post processing it tends to render vibrant colors especially red vs my k and m pentax's which do remain warm. In my own case I notice this when compared to my K24 and I actually cary the Vivitar more because of it.

The largest common aperture should be f/4. The f/3.5 lenses have one click-stop between wide open and f/5.6, I understand that this is f/4, though I haven't tested it. I could have tested them all at this aperture, but I always like to know how a lens performs wide open. Obviously these kind of shots (landscapes) wouldn't normally be made wide open, but sharpness, colour and contrast are the same regardless of the subject (if not the shooting conditions).

I have been using the Vivitar on film a bit recently (though I haven't processed them yet), and I think on the results here I'll stop doing so. I normally PP my digital shots, but not so much with film, as the scans I've had done so far have been done in my local place and are hugely over-sharpened jpegs, so no RAW unlike the digital.

[whojammyflip]

Convolution....the difference between the lenses is going to be comparable to the difference between the lens resolution and sensor resolution...the sensor becomes a significant factor in this analysis.

[Jonathan Mac]

I have been using the Vivitar on film a bit recently (though I haven't processed them yet), and I think on the results here I'll stop doing so. I normally PP my digital shots, but not so much with film, as the scans I've had done so far have been done in my local place and are hugely over-sharpened jpegs, so no RAW unlike the digital.

I take this back, at least for B&W film. The shot below was taken with the Vivitar, I think at 1/30s f/2.4. Sharpness and contrast seem fine.

[jt_cph_dk]

The RMC Tokina II 28/2.8 (PKA-mount on LX) is my favorit compared to the SMC M 2.8/28 that rarely leaves my shelf. Kind regards and thank´s for the interesting test.