Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

M42 Ultrawides - Can I improve on these Takumars?
Post new topic   Reply to topic
Home   Manual Focus Lenses
View previous topic :: View next topic  


PostPosted: Thu Sep 10, 2009 7:54 pm    Post subject: M42 Ultrawides - Can I improve on these Takumars? Reply with quote
Hi All,

I know, even from my brief time here, that this is an often-raised topic, so please forgive me for raising it again. I wonder if I can ask for some guidance among M42 wide-angle choices.

I took my K20D out shooting last weekend, and I brought along a few wide-angles: my SMC Takumar 24/3.5, SMC Takumar 20/4.5 and SMC Pentax 15/3.5. I was shooting an interesting stand of dead trees, and it seemed wise, in light of the focal lengths and the subject matter (and my old eyes' difficulty focusing wide-angles with the DSLR viewfinder!), to shoot at hyperfocal settings. Shutter speed was not an issue; I was using a tripod. So I set each of the three lenses to f/11, focused to the appropriate hyperfocal distance and snapped away.

Well, the shots from the 15mm were fine. But I was surprised by the lack of sharpness of the shots from the 20 and the 24. The images looked excellent at 25%, very good at 35% and passable at 50%. But at 100% they were disappointing - not really what I would call extremely sharp, to my eye at least, anywhere. So I repeated the experiment, this time in my backyard, using the same lenses, adapter and camera - and adding in a second copy of the 24mm I happen to own. The results were the same. Here they are (no sharpening, just a minor contrast adjustment and a resize in Photoshop). And I know that Dawn Redwood is dead, by the way; that's an ongoing controversy between myself and the nursery where I bought it.

24mm sample:


and a 100% crop:


20mm sample:


and a 100% crop:


These lenses were among my favorites in my film-shooting days, years ago, and I don't recall ever being unhappy with the results from them, at hyperfocal distances or otherwise; but their most demanding application back then was probably an 8x10 enlargement. This was their first time out on a DSLR - and my first opportunity to assess their performance at "100%".

My questions:

Is this the sort of performance I should be expecting from these lenses, when used on my DSLR? Does an impression of overall sharpness at 25%/35% sound about "right"?

If the answer to that question is "yes," then I would love to know how realistic it is to think I can do better (sticking with M42 mount) at these focal lengths. The default solution seems to be to buy a Flektogon (or two), and I'll gladly do that if necessary, when my budget permits. I gather the 21mm Yashinon (or even one of the Mir ultrawides) might be another option. If someone could point me to some 20mm/25mm Flektogon 100% crops, I'd be very grateful. I have no problem investing in some Fleks, but it would be awfully nice to have some reassurance that the improvement would be significant.

Thanks!

Jon


PostPosted: Thu Sep 10, 2009 8:02 pm    Post subject: Reply with quote
Whoa! Great collection of Takumars you have there.

Please also post examples of the 15mm; this way we have something to compare with. Is it the aspherical version, by the way?

I can also try my S-M-C 24/3.5 on my K-7, I must admit I haven't used it much lately.


PostPosted: Thu Sep 10, 2009 8:22 pm    Post subject: Reply with quote
talk about bragging ... LOL, impressive collection indeed - where are the Ultra Achromatic Takumars, left out intentionally??


PostPosted: Thu Sep 10, 2009 8:24 pm    Post subject: Reply with quote
Univer - I have a Pentax K10d and find focusing with wide angles to be a problem. It's not the camera or the lenses, it's the operator (i.e. me!)

Your lenses are all good, so discount errors from them. Also, forget the hyperfocal distance setting because in practice on a crop-frame DSLR, it just isn't reliable. And neither is the green light focus indicator.

What we all tend to forget is that with any lens there is only ever ONE critically sharp focus point - with a declining zone of clarity in front and behind. The depth of field scales are all worked out in relation to a particular degree of enlargement where the illusion of sharpness will be maintained over the zone indicated on the scale on the lens barrel. With a crop-frame DSLR the 'negative' is being enlarged much more than a 24x36 negative to give a similarly sized picture, so the illusion zone will be much smaller - i.e less depth of field.

I have a 20mm lens which I use on 35mm film, and believe me the depth of field is not infinite when you make enlargements or project slides.

I've beaten the problem by the following means:

For long distance pictures (over say 25 metres) set the lens to infinity.
For closer distances either focus VISUALLY on the screen (but make sure you've got the eyepiece adjustment optimnised to your vision) or if that's not possible then make the most accurate estimation you can and set that.
Do NOT rely on depth of field scale indicators at all.

Well, that's my free advice (and you know what they say about about the value of free advice !)


PostPosted: Thu Sep 10, 2009 8:27 pm    Post subject: Reply with quote
Nice collection!


PostPosted: Thu Sep 10, 2009 8:35 pm    Post subject: Reply with quote
>>>>Is this the sort of performance I should be expecting from these lenses, when used on my DSLR? Does an impression of overall sharpness at 25%/35% sound about "right"?<<<<<<

simply yes and no...

What i can see in your images is the reason why i sold my wideangle Takumars...

The performance on hyperfocal distances was poor wide open and needed to stopped down more than 3 steps to be useful.

For wideangle shots, take a Nikkor 3.5/20mm, a Flektogon 2.8/20mm on a Fullframe Body... that would be a far better choice...

Did own both lenses and also the Flek 4/25.
After compairing the lenses in a test, my favourit lens was the Nikkor AI 3.5/20mm.

I did own a 28mm Takumar, a 35mm Takumar and a Yashinon DS-M/Tomioka 2.8/24mm in this range. The first two lenses are gone after 2 weeks of using them intesively.
I also did own the M42 Flektogon 4/25.. which was very good in the center, but i shot with a FF Body (EOS 5D) and the edges where not as good as i expect... but if you use a crop cam, the 4/25 should be a very good performer for your needs..
Another very good lens is the Yashinon DS-M 2.8/24..

greetz
Hinnerker

Last edited by hinnerker on Thu Sep 10, 2009 8:46 pm; edited 2 times in total


PostPosted: Thu Sep 10, 2009 8:41 pm    Post subject: Reply with quote
I agree with Hinnerker completely. Easiest improvement start to use Slide film it will be improve a lot your crop DSLR.


PostPosted: Thu Sep 10, 2009 8:59 pm    Post subject: Reply with quote
hinnerker wrote:


What i can see in your images is the reason why i sold my wideangle Takumars...

The performance on hyperfocal distances was poor wide open and needed to stopped down more than 3 steps to be useful.


greetz
Hinnerker


Hinnerker - Sorry to say this, but the problem was with the factory's calculation of the depth of field scales - not the performance of the lenses. Depth of Field scales simply cannot always be trusted for anything other than small enlargements - 6x to 8x at the most. And with some makers even less. With DSLR crop-frames and 35mm format lenses the problem is even worse.


PostPosted: Thu Sep 10, 2009 9:43 pm    Post subject: Reply with quote
Not only the engravings are the problem, some formulas refers there optimum to infinity some have other reference optimum points. Thats a further aspect on that theme.

In addition to that often the infinity point of old lenses is not exactly, especialy if you use it on a DSLR with adaptors.

The Rollei QBM Mount is the best example for that. I've got some QBM Lenses, which you can adapt easy after file down the adapter and they reach infinity. The Rolleinar 2.8/105 i own fits perfect after file down 0,7mm to equalize the flange back Rollei to Canon EOS (44,7 > 44mm) and reached infinity.. The Rolleinar 4/80-200mm i own is on exactly the same Adapter not aible to reach Infinity.

If you mean this, you are right.

Thats the reason why i first try to look if a lens is exactly on the infinity point or not or behind the point.

Greetz
Hinnerker


PostPosted: Thu Sep 10, 2009 9:59 pm    Post subject: Reply with quote
Hinnerker - yes, you have some additional good points there. We are "singing from the same songsheet" I think!


PostPosted: Thu Sep 10, 2009 10:09 pm    Post subject: Reply with quote
The 24mm has a reputation of being a sharp lens but I have one and I have to say I seldom got sharp images with mine (this was 10 years ago that I bought it so I kinda put it away and forgot about it.) I must get it out again and try on my digital SLR. I still to this day do not know if it was the lens or me. The lens was so nicely made that I was reluctant to part with it. Having said this I have seen some images from this lens (probably on Flickr) that did look acceptably sharp to me.


PostPosted: Thu Sep 10, 2009 11:24 pm    Post subject: Reply with quote
Certainly, I would say you can't trust the depth of field marks on the lenses, because apart from anything else, the "circle of confusion" will be different if a full frame is cropped.

The concept of "circle of confusion" is a little bit arbitrary, but relates to the point at which a person perceives that part of the image is not in focus. The DOF marks on your Takumars are based on standard calculation for lenses of that focal length throwing an image onto 35mm film. And of course as previously mentioned, the ultimate size of the print of the photo.

If you chose the hyperfocal distance on your lens for F11, then I would have thought that you would get an unsharp image. (which you did)
I would at least choose the hyperfocal distance for say F8, if the aperture is F11. This would yield a better result.


PostPosted: Fri Sep 11, 2009 12:05 am    Post subject: Reply with quote
dnas wrote:
The concept of "circle of confusion" is a little bit arbitrary, but relates to the point at which a person perceives that part of the image is not in focus. The DOF marks on your Takumars are based on standard calculation for lenses of that focal length throwing an image onto 35mm film.


The circle of confusion itself is arbitrary; it's just a name for each “blur circle” drawn by the lens. However, the hyperfocal distance itself depends on entirely the definition of “in focus”, i.e. the largest acceptable size of the circle of confusion that is still considered to be in focus. It is also important to realise that hyperfocal focusing makes things “at infinity” just barely “in focus”, and therefore one may be able to see an area of “even more in focus” if the definition for “in focus” was too lax.

The acceptable size of the circle of confusion that was used for computing the DoF scales on lenses is typically based on a fraction of the diagonal of the intended film format, i.e. the full frame. The idea is that if the image is enlarged, people would presumably step farther back to view it and therefore the image would supposedly look the same size. That may be a reasonable assumption in some cases, but it breaks down in many others (e.g. people sitting at the computer tend to position themselves at a fixed distance from the screen). This also breaks down on crop sensor cameras because the diagonal is smaller; to get to the intended original depth of field one should stop down at least one stop more than on full frame.

With digital cameras it is possible to define the “perfect” size for the circle of confusion based on the pixel pitch of the sensor in that camera – this will lead to “pixel perfect” depth of field. Unfortunately this leads to hyperfocal distances that are essentially “at infinity” for many lenses and settings, so one might just as well use infinity focus if one wants a “pixel perfect” DoF to infinity. (Many online hyperfocal calculators also fail to take diffraction into account; when the lens is stopped down enough, diffraction limits the resolution below that of the sensor and it no longer makes sense to compute the hyperfocal based on the unreachable sensor resolution.)


PostPosted: Fri Sep 11, 2009 7:43 pm    Post subject: Reply with quote
Hi,

Thank you all for the very helpful replies (and for not making fun of that poor sad tree). Thanks, too, for the kind words about my collection (more of an accumulation, I think). There's a few decades' worth of collecting there - although I freely admit that I have added significantly to the list since discovering this forum!

Attila, I acknowledge that shooting slides would take care of all these DSLR problems. Wink I haven't sold any of my film bodies (any that I care about, at least), so the option is there. Anything is possible!

Spotmatic, thank you for the kind words. I will post some shots from the 15mm as soon as I can. As for the aspherical question, my best recollection is that mine is not one of the rare(r) creatures, but I will double-check.

Klaus, I only wish I had one of the Ultra-Achromatics! The closest I'm likely to come is enjoying the photos you were kind enough to post not long ago; I very much enjoyed seeing the shots from the SMC and the UA side-by-side, thank you!

Stephen, hinnerker, dnas and Arkku - I really appreciate all your input regarding the use (and trustworthiness) of the lenses' DOF scales, particularly in connection with a DLSR. I will be more attentive next time - happily, the trees will still be there - and check my results as I go. (With a DLSR, there's really no excuse for not doing that in any case.)

I will say, just by way of additional information, that I also took a few shots that day, relying on the lens's DOF markings to determine hyperfocal distance, with a Schneider Curtagon 28/4 (via one of Alex's DKL->M42 adapters), and those shots, to my eye at least, were acceptably sharp.

Thanks again, all!

Jon


PostPosted: Sat Sep 12, 2009 5:57 pm    Post subject: My wide angle lenses. Reply with quote
Good night:

The lenses wide angle M 42 that I used with my Spotmatic and that gave me big results, are:

- Tamron SP 17 mm.
- Tokina 24 mm.
- Schneider Curtagón 28 mm, f:4,5.
- SMC Takumar 35 mm, f: 3,5.

Bests Regards: David


PostPosted: Sat Sep 12, 2009 6:52 pm    Post subject: Reply with quote
a lot of these old takumars including the 24 and 35 tend to have incorrect focusing scales. my 35/3.5 was so far out of alignment it wasnt funny... but there are instructions out there for adjusting them

http://orlygoingthirty.blogspot.com/2008/08/adjusting-infinity-focus-on-pentax.html
Post new topic   Reply to topic
Home   Manual Focus Lenses
Page 1 of 1