[Wild Mark]

From time to time I have noted various commentary on these two lenses, with the faster of the two often the subject of scathing/ praising reviews. The slower is often regarded as super sharp and as such 'better' than the f2 version.

I am lucky to have both and actually enjoy using them in different ways. Yes I can say they are different lenses each requiring their own understanding and application. Here, I am going to attempt a succinct description of those differences with a photo comparison to kick off discussion.

Aperture
Yes, there is the obvious difference (f3.5 vs f2.0) but, when you look closer there are some other notable differences. The 35/2.0 has an extra stop between f4 and f5.6 in addition to the wider apertures f2 and f2.8. The f-stops available on each of these lenses is as follows

35/3.5 35/2.0
- 2
- 2.8
3.5 3.5
4 4
- 5?
5.6 5.6
6.7? 6.7
8 8
9.6? 9.6?
11 11
16 16

Note: ? = unmarked stop and I am unsure what the stop number is.

Set on a tripod I ran though each aperture setting for each lens using Av mode as I was interested to see what the camera metering would do. Surprise surprise, the camera metered each lens differently at commensurate aperture settings. The 35/2.0 version metered a faster shutter speed for f3.5 and f4. At f5.6 both lenses metered roughly the same (1/80 for the 35/2.0 versus 1/100 for the 35/3.5). From f6.7 on the 35/3.5 metered a faster shutter speed. At f8 metering was as follows: 35/3.5 - 1/50 35/2.0 - 1/30 ....... Curious.

Focusing
Interesting situation here. The 35/3.5 is acclaimed to be super sharp, and much sharper than the 35/2.0. Well, my results indicate a peculiar circumstance that may narrow the perceived disparity between the two lenses. At f8, both lenses are meant to be close to their sharpest (although results indicate sharpness is good through to f16). When using the contrast detect focusing mechanism for achieving focus, the 35/3.5 is by far the sharpest of the two lenses. And, if I didn't have a split prism focusing screen, I might have given up there and then (and thrown the 35/2.0 in the bin with disgust). The following photos are taken at f8 with shutter speed for both lenses set at 1/50 with the 35/2.0 using in M mode to achieve the faster shutter speed (as metering in Av mode was 1/30)

SMC Takumar 35/3.5 @f8 1/50




SMC Takumar 35/32.0 @f8 1/50



I have also included the Av metered 35/2.0 photo taken @ f8 for those who are interested. Over exposed!



When using contrast detect with the 35/2.0 I noted substantial focus inaccuracy through the viewfinder suggesting that the lens was not in focus. So I took another photo, this time using the focusing screen ...... I think the results speak for themselves.




Bokeh
Well, the 35/3.5 has fewer elements and aperture blades (5) when compared to the 35/2.0 which has 6 aperture blades. The above photos show that the 35/2.0 is far superior (IMHO) when it comes to bokeh. The aperture blades give a nicer blur when compared to the harsher pentagon shaped pattern of the 35/3.5. Also, it appears that the 35/2.0 handles the highlights much better than the 35/3.5 giving a smoother OOF. Check out the silverbeet in the lower left corner as an example.

Colour
Both lenses seem to render in focus colours the same way. However, it is clear to me that the 35/2.0 renders colour in OOF areas much much better. More saturation and pleasing to the eye.

Conclusion
If I want an easy to focus compact sharp manual lens I would reach for the 35/3.5. It will give dependable results in a variety of conditions and situations. An all rounder.

But, if one were to sacrifice convenience and take the time to focus the 35/2.0 (and get the metering correct) then the faster of teh two lenses is the one to reach for. The 35/2.0 is fickle in focusing and metering is erratic but, get these two elements right and you get fantastic sharpness (quite respectable) combined with great colour rendering and bokeh. The lens has better DoF capabilities giving opportunity to enhance 3D effects.

I think the 35/3.5 is a good utility landscape/ streetscape wide angle (normal), whereas the 35/2.0 is a specialty lens capable of much more, but harder to use.

Comments welcome.

[LensBeginner]

Very informative post.
I see that the 2.0 seem to also have some more CA, judging from the road sign..?
Also, taking pictures at f/8 is good to assess peak sharpness, but one of the test that should be done is shooting at or near wide open and at about one stop stopped down, so let's say f/3.5 (minimum common aperture) and f/4 - f/5.6 (f/4.5 would have been better but, as you pointed out, only one lens has it).
All in all, a nice review!

[Wild Mark]

Very informative post.
I see that the 2.0 seem to also have some more CA, judging from the road sign..?
Also, taking pictures at f/8 is good to assess peak sharpness, but one of the test that should be done is shooting at or near wide open and at about one stop stopped down, so let's say f/3.5 (minimum common aperture) and f/4 - f/5.6 (f/4.5 would have been better but, as you pointed out, only one lens has it).
All in all, a nice review!

Yes the 35/2.0 does show cyan colour shift/ fringing in wide open shots, with a tiny amount evident by f8. Colour shift is worst in the edges of the photo.

I have other shots and can post wide open for both lenses and f3.5 for the 35/2.0 lens. I didn't post because the focusing issue in the 35/2.0 was enhanced by the wide open setting.

Interestingly, the 35/2.0 carries nice'ish bokeh through to f16 - this appears to be a design objective in this lens.

[manntax]

Your results might be affected by many factors such as optical flaws of each lens, sample variation, errors in focusing, extreme lighting conditions, shake blur, subject blur ( wind ?), and so on.. the effort is noble , but results are meaningless really.. Even the choice of subject doesn't really help in determining the crucial factor - sharpness in corners and on edges, and the curvature of focal place - both exceedingly important for wide angle lenses - especially when measured against at f8.0 - which supposed to deliver top uniform sharpness , not only in the very center. Add to that the fact that instead of close focused distance - much more interesting would be to see how both lenses are resolving details at infinity distances.. Even bokeh is so hard to judge on these for simple reason that background is heavily blown out and uninteresting.
Nothing to worry about though, I've been there and done that :P - we all did those tests I believe...

The fact is : Until we know the *EXACT* condition of these two lenses, and their relative performance in comparison to others of the same kind - to rule out the sample variations factor - these results are telling me nothing more than one of YOUR lenses is better in some respects than the other in case of shooting heavily backlit subject. That's it - there is nothing more to it really ;-)

[Dartmoor Dave]

Since you specify SMC I assume that you're using the second version of the 35mm/2.0 in this comparison -- the one with the 49mm filter ring. I don't own that version yet, but I use the original version with the 67mm filter ring and love it.

The original version 35mm/2.0 has a very different colour rendering to the 3.5: very subtle and natural in comparison with the richly saturated results from the 3.5. For me, the choice of which one to use comes down entirely to the look I'm aiming for in the final print.

Also, the original version 35mm/2.0 is a lens that strongly favours micro-contrast over high edge contrast. This can fool people into thinking it's not that sharp, if they are used to the effect of pseudo-sharpness caused by the high edge contrast of modern lenses. It's only when you take the time to really look at shots with the original 2.0 that you realise how much detail it renders and how truly sharp it actually is.

I think you've identified one fundamental point here that can be applied to all fast manual focus lenses: In the majority of cases I've encountered, when somebody has complained about a lack of sharpness in a manual focus lens, the problem has turned out to be the user's inability to achieve proper focus. Your review confirms that the only way you can guarantee exact focus is to use a properly calibrated split-prism screen, and that's the main point that I hope everyone holds on to. Whenever anyone starts talking about the sharpness of any manual lens, the first question we should all ask is, "How did you focus?"

You've put together an excellent, thoughtful and thought-provoking comparison. Thanks for posting it.

[Wild Mark]

Your results might be affected by many factors such as optical flaws of each lens, sample variation, errors in focusing, extreme lighting conditions, shake blur, subject blur ( wind ?), and so on.. the effort is noble , but results are meaningless really.. Even the choice of subject doesn't really help in determining the crucial factor - sharpness in corners and on edges, and the curvature of focal place - both exceedingly important for wide angle lenses - especially when measured against at f8.0 - which supposed to deliver top uniform sharpness , not only in the very center. Add to that the fact that instead of close focused distance - much more interesting would be to see how both lenses are resolving details at infinity distances.. Even bokeh is so hard to judge on these for simple reason that background is heavily blown out and uninteresting.
Nothing to worry about though, I've been there and done that :P - we all did those tests I believe...

The fact is : Until we know the *EXACT* condition of these two lenses, and their relative performance in comparison to others of the same kind - to rule out the sample variations factor - these results are telling me nothing more than one of YOUR lenses is better in some respects than the other in case of shooting heavily backlit subject. That's it - there is nothing more to it really ;-)

re: conditions - of course - I didn't go into that because I kept conditions between the two lenses the same. Kind of nullifies the issue in many respects. So, I disagree re: effort is meaningless (by the way I carry no bias on the lenses).

Anyway, if you want an indication of conditions .... then ... perhaps I could share this. Two kids under 10 years old on a Saturday afternoon. Getting more than a minute to think and do something with all angles covered under these conditions is, well, nigh on impossible.

Anyway this is why I aimed for f8, but I have a plethora of other shots to share.

If you must know, conditions are:

- Tripiod
- Shake reduction OFF
- Same camera (K30)
- same scene
- Photos taken as rapidly as possible for both lenses WITHOUT upsetting the above to keep light conditions similar
- A tiny bit of wind
- shot from the verandah because an adventure beyond there is not possible

The lenses ..... well both are mint with the 35/2.0 recently checked over by Eric.

Anyway, it wasn't meant to be scientific. For a relative comparison it was quite informative and that was the purpose. It was not meant to be an absolute appraisal.

Cheers

---------- Post added 11-22-14 at 09:31 PM ----------

Since you specify SMC I assume that you're using the second version of the 35mm/2.0 in this comparison -- the one with the 49mm filter ring. I don't own that version yet, but I use the original version with the 67mm filter ring and love it.

The original version 35mm/2.0 has a very different colour rendering to the 3.5: very subtle and natural in comparison with the richly saturated results from the 3.5. For me, the choice of which one to use comes down entirely to the look I'm aiming for in the final print.

Also, the original version 35mm/2.0 is a lens that strongly favours micro-contrast over high edge contrast. This can fool people into thinking it's not that sharp, if they are used to the effect of pseudo-sharpness caused by the high edge contrast of modern lenses. It's only when you take the time to really look at shots with the original 2.0 that you realise how much detail it renders and how truly sharp it actually is.

I think you've identified one fundamental point here that can be applied to all fast manual focus lenses: In the majority of cases I've encountered, when somebody has complained about a lack of sharpness in a manual focus lens, the problem has turned out to be the user's inability to achieve proper focus. Your review confirms that the only way you can guarantee exact focus is to use a properly calibrated split-prism screen, and that's the main point that I hope everyone holds on to. Whenever anyone starts talking about the sharpness of any manual lens, the first question we should all ask is, "How did you focus?"

You've put together an excellent, thoughtful and thought-provoking comparison. Thanks for posting it.

Yes indeed (second version); both lenses have the 49mm filter ring. And thanks for the spirit of your post. I was only attempting to compare two lenses that seem to get very differing reviews. Yes, focusing is the key and the 35/2.0 is very definitely a fickle lens to focus. i too think this is why many think this lens is soft when it really isn't. Can't wait to try this lens on my dad's A7r!

[manntax]

Anyway, it wasn't meant to be scientific. For a relative comparison it was quite informative and that was the purpose. It was not meant to be an absolute appraisal.

Of course I know it wasn't a scientific comparison - and as a fun exercise it has its uses and for some the results can mean something. For serious comparison that would allow to draw some meaningful conclusion you certainly would need to go a bit scientific course - especially have few lenses of each kind to rule out the sample variation issue, also the subject would need to be choose carefully and different for both close and infinity focusing. each shot would need to be attempted at least 2 times and then best out of series would go for comparison etc..

Despite all that I can admire the effort and hat off to you for that - but the conclusions in reality are of very limited value.

Cheers!!

[Wild Mark]

The challenge is set then. At some stage I will sit down and attempt the more scientific route. I have three copies of the 35/2 (two SMC and one Super) and two copies of the 35/3.5. Heck, I also have a K35/2 to consider. If you can outline the best conditions one can undertake the comparison then I would be greatly appreciative.

[manntax]

The challenge is set then. At some stage I will sit down and attempt the more scientific route. I have three copies of the 35/2 (two SMC and one Super) and two copies of the 35/3.5. Heck, I also have a K35/2 to consider. If you can outline the best conditions one can undertake the comparison then I would be greatly appreciative.

Noo come on no more tests - go out and shoot and then review at home which one looks or feels best :P you can spend life time on testing lenses - I nearly got caught up in this - now I enjoy shooting, occasionally I do some shoot outs outside my window to see how lens renders at infinity and on what plane sharpness sets in - other than that you will only loose time.

BUT if you are stil curious which lens is best in terms of sharpness I suggest going for :
1) nice day where light is fairly consistent and with no wind
2) distant scene - preferably of some raising ground, or toll trees where your position is roughly in the middle of height of your subject.
3) set tripod on stable ground, add some weight to it so it doesn't move
4) set your camera with lens on so that subject is fairly parallel to the film plane
5) use remote control ( 2sec. mirror is not enough for heaver/ longer lenses
6) use base ISO (or higher ISO for windy days - to keep the shutter faster )
7) focus your lens on the same subject in each case
8) set your exposure correctly from start - after that you simple adjust shutter speed by the aperture step ( if next stop is 1 EV down then adjust aperture by that, and so on )
9) take shot and quickly review - if it appears to be sharp, then refocus again - take shot, and again - I usually was taking 3 shots of each aperture. If shot appears to be not sharp - repeat that step - this can be tedious with bad lenses but good one will clearly be in focus even from wide open
10) go from wide open to fully closed down - to test across the board - that will show how diffraction sets in for each lens at f11 onwards
11) make a note of how many frames for what aperture you have for this lens - this is for comparing lenses with different aperture ring layout
12) change lens without disturbing the camera (keep the camera on tripod - just remove lens carefully )
13) repeat for each lens
14) do the same for close focusing / bokeh - this will need to have both subject and background lit with the same light - so both background and subject is well exposed

15) after you are done - go home , have a tea/beer/coffee and review at home on big screen (lightroom , library , spit view ) every photo for each aperture first to eliminate worse frames from each aperture setting for each lens . After you got rid of bad frames and you have single shot for each aperture of each lens - then review them side by side

16) check for central sharpness, borders, corners, all aberrations, distortions etc..

you are done - now you know for sure which lens is best from all IF you are not tired already, prepare report and post on Pentax Forums so other can appreciate your hard work.

NOTE : there is still chance that your results will be contaminated with sample variations , not obvious faults ( like slight decentering etc ) and the fact that among 10 even good lenses there will always be a star and a black sheep.

I went through this for the past few years, never got to making any reports but this cherry picking allowed me to know for sure which lenses ( from this particular flock - mind you!) are best and worth keeping. I found few stars - some lemons, and also learnt that some lenses are very very consistent and simple perform great. Pentax lenses are mainly just good - sometime bad ( I had one 28 mm f3.5 which was simple bad wide open, but better stopped down, but never reached excellent on borders -another copy was simple outstanding ). From other manufacturers I saw two Tokina 17mm lenses and one was star and other one was just good but first one was just star ( guess which I kept :P ) . So this all can serve some purpose but again now I simple take the lens out and review shots later - if I like the way it focuses , sharpness and CA I know this is a good lens - and I have loads of photos that are of some interest instead of 100's of the same test shots of my neighbourhood :P


Good luck !

[Rimfiredude]

Thanks for the info. I always wanted to pick up an f3.5. Just might do it now.

[Dartmoor Dave]

Noo come on no more tests - go out and shoot and then review at home which one looks or feels best :P you can spend life time on testing lenses - I nearly got caught up in this - now I enjoy shooting, occasionally I do some shoot outs outside my window to see how lens renders at infinity and on what plane sharpness sets in - other than that you will only loose time.

+1 to that! The best way to test a lens is to go out and shoot with it in the type of conditions you're expecting to use it in. Get a feel for whether the colour rendering, contrast characteristics and sharpness suit your taste, and then decide whether the lens fits your needs or not.

Don't be one of those people who obsess over multi-lens shootouts and use terms like "IQ" and "performance" as if they think they actually mean something quantifiable.

[LensBeginner]

Noo come on no more tests - go out and shoot and then review at home which one looks or feels best :P
*snip*

My view exactly.
I usually do a comparison session with a test chart, get a general "feel" for the lenses, and then never worry again.

[Lowell Goudge]

Its funny, the SMC tak 35/2 was my first M42 lens, all the way back when I shot an *istD. I never had any trouble with manual focus and fast lenses. I think it is just not taking the time to shoot with them.. Sure a slower lens has more DOF, but for critical focus it is still only one plane of perfect focus.

[LensBeginner]

Its funny, the SMC tak 35/2 was my first M42 lens, all the way back when I shot an *istD. I never had any trouble with manual focus and fast lenses. I think it is just not taking the time to shoot with them.. Sure a slower lens has more DOF, but for critical focus it is still only one plane of perfect focus.

True.
Live view, 6x magnification, a steady hand and you can be at f/2 all the time you want!

[Wild Mark]

True.
Live view, 6x magnification, a steady hand and you can be at f/2 all the time you want!

I sort of agree. The 35/2 is not really all that sharp at f2. Anyway, i never really bought the lens for that express purpose - more its capacity to deliver good colour rendition and satuation.

My prior comments regarding these two lenses remain the same. Using prior testing guidance I took another series of photos (close focus) to test sharpness, contrast and colour rendering/ saturation. My additional notes/ observations are:

1. Light metering REQUIRES the A contacts to be shorted out when using the K30. After switching to M mode it became obvious as green button metering did not work. A strip of aluminium is required for both lenses.
2. The 35/2 is indeed better at delivering much better micro contrast. Greater detail is evident in the 35/2 as a consequence. In providing marginally less micro contrast, the 35/3.5 ends up delivering slightly less detail and as such can give the appearance of better edge sharpness. If I was going to use one of these lenses for a large print I would definitely use the 35/2 as it gives the detail needed.
3. The 35/2 gives better colour saturation throughout the image.
4. Bokeh is better in the 35/2
5. Colour shift/ fringing is worse in the 35/2 especially when shooting wide open. Little difference at f8 between the two lenses.

Ultimately, both lenses are very good and the differences are minor. They do offer slightly different characteristics and on that basis would probably be used differently on a shoot. The 35/3.5 would probably do better for landscapes while the 35/2 would do better in closer focus and even portrait on an APC-S (colour/ bokeh/ micro contrast). Neither are sharp enough to be adapted to macro use with tubes.

If you are looking at buying a 35mm Takumar I would first ask yourself the question of what you want to use the lens for. If tis is close to short range shooting then the 35/2 is a better choice. Landscape photography, I would go for the 35/3.5 simply because it is cheaper, smaller, easier to focus and more compact. The crops below were taken to examine for differences in micro contrast mostly but the other things mentioned are noticable. The left image is the 35/3.5 with the right being the 35/2.0