Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
04-05-2014, 03:37 PM   #1
Inactive Account




Join Date: Oct 2011
Posts: 147
Comparing Adapted Lenses on Q

CAUTION: Skip if ‘testing’ bores you! This will..

The ‘Adapted Lens’ thread is a great resource for checking out how various legacy/other lenses perform ‘real world’ on the Q. It has helped me many times. But I do find it a bit hard sometimes to compare lenses, because every photo example is different. So this thread.. not to attempt to replace it, but to possibly complement it.

Background – until I got the Q, I shot a K-01 with M42 manual legacy Takumars exclusively – enjoyed doing so – and still do. Then when I got the Q, I found out – as many before me, the Q is brutal on legacy glass. If there is a hint of CA/PF, or lack of sharpness or contrast, the Q will show it in spades.

Recently I wanted to ‘thin out’ my M42 collection, reducing by eliminating duplicates and any ‘not so great’ lenses accumulated over the years. I found an easy to understand and implement testing approach by a Canon guy, Bob Atkins: Camera Lens Testing- Sharpness, Chromatic Aberration and Distortion - Bob Atkins Photography

For my K-01, an APS-C like the Canon Atkins used, I prepared the same 25x test chart 15” x22” with 5 images of his 4x6 test pattern. Ran tests with a number of my M42s – and got a lot of info on Contrast and CA/PF. Well worth it. But as Atkins explained after I went back to reread it, this test is not as definitive on Sharpness. It will certainly point out a really bad lens (Ex: Camera Lens Testing - Some example images - Bob Atkins Photography) but many times the resolution results seemed almost identical. Why?

Atkins explained it understandably: In the case of APS-C, the sensor resolution was limiting, not the lens. Using his Canon example rather than the slightly different numbers for the K-01 APS-C, he takes the 175pixels/mm resolution of the Canon sensor and shows that the most line-pairs you could ever expect to resolve with that sensor and any lens is at most half of that: or 87.5 lp/mm, and realistically – 75 lp/mm no matter how good the lens is. Trouble is: Many good lenses resolve much more than that. So it takes a truly bad lens to show significant differences with this test. (See his Test Examples above)

So – what does all this lead to? It leads to the Q being a great lens resolution test platform! Because the Q is brutal with its pixel density – so brutal that the Q sensor is likely never going to be the limiting factor in a lens resolution test. So I decided to modify Atkins approach and fit it to the Q:

First, the Q sensor is 6.17mm x 4.55mm, and 4000 x 3000 pixels. This in turn gives a pixel density of 648pixels/mm or 659pixels/mm – say 650pixels/mm. Half of that – for a maximum lp/mm resolution would be 325 lp/mm – over 4x the max on APS-C (16mp). Perhaps realistically it might only be 250 lp/mm or 200 – but whatever, its substantially greater.

Atkins charts are designed to be printed on a typical 600 x600 printer, and used at 25X. If your printer is like mine, it won’t resolve anything more than the 5lp bar set – or 125lp/mm (25x). However, the solution appears simple (to me anyway, correct me if I am wrong) – use Atkins charts at 50x, not 25x. Then, the 5lp bar is going to be 250lp/mm – not 125 – and so on for all the bars. 2X original.

Since 1” equals 25.4mm, sizing the chart is easy – approximately the sensor size in mm turned to inches. i.e a 25x for Q would be 6.17” x 4.55” – and double that for 50x would be about 12 3/8 x 9 1/8. I made one that size, cutting down Atkins 4x6 to fit. (See first JPG attachment).

Now the first tests: I started with one on my sharpest lenses (from the K-01 tests) – a SMCT 55mm F1.8 M42 – late, with the open aperture gizmos on the back. The 2nd attached JPG shows center resolution at F5.6 – the sharpest setting for this lens. To me it looks like the 4.0lp is just resolved. At 50x – this is 200lp/mm. (Edge resolution was virtually identical – not surprising, given ‘its really not the edge’ for anything other than a Q. So I did not attach them)

I then took a real beast I though was junk after using it once (at 200mm handheld – ugh! That was the problem apparently…). It’s a Vivitar Kobori 28-200mm F3.5-5.3 - in a rather useless Minolta M/MD mount. But Mint otherwise. I bought it for $20 (they are all over ebay around that in MD/M or Canon mounts). I spent as much on the Fotodiox adapter. The reviews here on PF aren’t really bad: Vivitar 28-200mm f3.5-5.3 Lens Reviews - Vivitar Lenses - Pentax Lens Review Database. The MD mount means its really cheap – but not unfortunately usable on the K-01. Multi-Coated. c1985. Set at 55mm, and F5.6, the center test shot is 3rd below. I think the 3.2lp is definitely resolved – maybe 3.6lp. So its 150 or so lp/mm. Not as good as the SMCT for sure – but not a Canon Samyang. Poorer contrast definitely. More aberrations too. (Look at the Blocks). But to call it ‘complete junk’? No…. Perhaps OK for Video, where you want to be able to Zoom. Or a situation where you can’t change between primes. ‘Useable’.

To wrap up, the SMCT vs. Kobori example showed what I was looking for in this Q-based lens resolution test – real differences. It seems this test, which is at least easily reproducible, might be one way to additionally compare adapted lenses – at least for center resolution, contrast, and CA.

Why not show us some results from your favorite adapted lenses?



PS - I thought it interesting to add the last JPG -Atkins center from his image shot using the EOS 40D and the Canon EF 300/4L (non-IS version), wide open at f4. This is not a cheap combo! And remember the 25x vs. 50x means the Canon resolution - at the same lp, is half of the SMCT. Yes - the SMCT is 5.6 vs. Canon F4, but the SMCT @ F4 is very similar to 5.6. And yes again, I did forget the EF 300/4L is effectively 450mm on APS-C, not 300. (Perhaps I should have used my M42 100mm F4 MACRO Tak...)

But I still think this confirms the Q with adapted lenses like the SMCT F1.8 55mm, and likely M42 100mm MACRO Tak, plus many others, can be very much the equal of much more expensive APS-C 300mm 'native' combos.

Attached Images
View Picture EXIF
PENTAX Q10  Photo 
View Picture EXIF
PENTAX Q10  Photo 
View Picture EXIF
PENTAX Q10  Photo   

Last edited by cahudson42; 04-06-2014 at 12:07 AM.
04-05-2014, 06:16 PM   #2
Site Supporter




Join Date: Sep 2006
Location: midwest, United States
Photos: Gallery
Posts: 3,225
Great tests and explanation. As you point out, the Q sensor stresses the adapted lens a great deal. Your test is a worthy addition to our lens testing arsenal. Good job. Your hard work is appreciated.

thanks
barondla
04-06-2014, 10:50 AM   #3
Inactive Account




Join Date: Oct 2011
Posts: 147
Original Poster
Selecting a M42 50_55mm Tak for Q

Thanks Barondla!

I thought this morning I'd try to use this approach to help understand the on-Q differences between three popular M42 legacy Takumars - the SMCT 55mm F1.8, the SMCT 50mm F1.4, and the MacroTak50mm F4.0. (Approximately $45, $90, and $120 for 'mint' examples) All three were previously checked out using my K-01 and my original 22" x15" target, and selected as 'keepers'..

Four shots follow below. The first to compare is shown in the Thread Starter - the 1.8 @ 5.6:

SMCT55_1.8_5.6Ctr (In thread starter - 2nd image. Can't duplicate here...)
SMCT50_1.4_5.6Ctr (Image 1 below...)


I have always suspected the least expensive SMCT 55 1.8 was sharper stopped down than the 1.4, while the edges were a bit worse. These two seem to confirm the center sharpness. The SMCT55 is slightly sharper, and has a bit more contrast, IMO. So for the Q, the 1.8 seems a hair better. But for the K-01 (edges) - I don't know anything new.

Now lets look at these at F2.0:

SMCT55_1.8_1.8plus1click (Image 2 below...)
SMCT50_1.4_2.0Ctr (Image 3 below..)

Situation reverses. The 1.4 @2.0 has better center contrast and slightly better resolution than the 1.8.

MacroTak50_F4.0_F5.6Ctr (Image 4 below..Last here)

Simply the best contrast and best resolution of them all.

What to conclude? Well the resolution differences are often not great. Differences can be effected by my focus, lighting, and lens sample variation.

It does seem the 1.8, wide open, has more of a contrast problem than the others. But I think any of the three are reasonable legacy 50-55's to use on the Q - until I see something better..

When I know I will have oodles of light, I'll start with the MacroTak. When I'm concerned about light, or DOF, the 1.4. But for the $, I can't complain about the 1.8 stopped down either - remembering that I need to stop it down.

AS always, YMMV
Attached Images
View Picture EXIF
PENTAX Q10  Photo 
View Picture EXIF
PENTAX Q10  Photo 
View Picture EXIF
PENTAX Q10  Photo 
View Picture EXIF
PENTAX Q10  Photo 
04-06-2014, 05:34 PM   #4
Pentaxian
hnikesch's Avatar

Join Date: May 2009
Location: Michigan, USA
Photos: Albums
Posts: 1,246
Great info, and the Macro Tak looks great but I never looked at my 50's because the Q with 06 or the K30 with 55-300 fill that slot for me. I find I only use adapted lenses when I need reach so 100, 135, 200 and 300 are the lenses I use adapted. One thing I didn't quite understand on your test chart is what size you printed the chart, the last image you posted what size was that part of the chart printed. To get any valid comparison we would all need to print the chart exactly the same size. I would like to include several lenses to this post in the range I like to adapt.

04-06-2014, 06:00 PM   #5
Inactive Account




Join Date: Oct 2011
Posts: 147
Original Poster
QuoteOriginally posted by hnikesch Quote
Great info, and the Macro Tak looks great but I never looked at my 50's because the Q with 06 or the K30 with 55-300 fill that slot for me. I find I only use adapted lenses when I need reach so 100, 135, 200 and 300 are the lenses I use adapted. One thing I didn't quite understand on your test chart is what size you printed the chart, the last image you posted what size was that part of the chart printed. To get any valid comparison we would all need to print the chart exactly the same size. I would like to include several lenses to this post in the range I like to adapt.
Hi Hans,

The charts were printed at 4" by 6" - the size provided by Atkins. What changed was the size of the 5-images rectangle - not 15" by 22" for APS-c. but the 12" or so 9" or so based on the Q-sensor size - giving 50x rather than Atkins 25x at full screen in live view..

Supposedly, if you use this 5-images 12" x 9" so that it is 'full screen' on the Q Live View, the numbers on the charts multiplied by 50 will give the resolution in lp/mm.

The images posted are center crops similar to those used by Atkins. They may be slightly different areas of the original. but they were all cropped at the same original 50x. So looking at the lp resolution on each, they should be comparable.

If I did not understand the question, or did not answer it sufficiently, let me know.

Would be interesting to see the 06 at 45 in comparison! (That is: Should I consider getting one??)

Last edited by cahudson42; 04-06-2014 at 11:10 PM.
04-07-2014, 05:13 AM   #6
Pentaxian
hnikesch's Avatar

Join Date: May 2009
Location: Michigan, USA
Photos: Albums
Posts: 1,246
Ok got it, today's forecast is rain so unless my wife has something scheduled I should be able to shoot some charts
04-07-2014, 06:13 AM   #7
Inactive Account




Join Date: Oct 2011
Posts: 147
Original Poster
QuoteOriginally posted by hnikesch Quote
Ok got it, today's forecast is rain so unless my wife has something scheduled I should be able to shoot some charts
Great Hans!

Couple thoughts with my having tried Atkins stuff: Its critical to get the chart parallel with the sensor plane. I used white posterboard for the background, and packing-taped it to a sliding glass door going to the Porch. I did make up Atkin's 'sticking out dowel' thing for setup, taking a few shots and checking the corners to see they were the same, before I went off taking lots of shots. If you don't have any dowel, a combination square, empty tin can, or anything that can stick out perpendicular to the chart probably will also work.

Also, if you are going to try native Q lenses like the 06, you might show us the corners in those cases. I skipped the corners on the legacy Taks, because being designed for FF, I found the 'corners' on the Q were virtually identical to the center (because considering FF, the Q 'corners' are still virtually at the center..)

This may not be a good assumption for lenses specifically designed for the Q/Q7 sensor - there easily could be quite different edge effects than what I saw with the Taks.

Have Fun!
Chris
04-07-2014, 02:47 PM   #8
GUB
Loyal Site Supporter
GUB's Avatar

Join Date: Aug 2012
Location: Wanganui
Photos: Gallery | Albums
Posts: 2,280
Great stuff!
I may have a go at this in winter (Autumn here now)
I have taken the liberty to crop your images and stitch the crops together in the image below. I hope you don't mind.
I think it makes it easier for an outsider to absorb the info.
Maybe there is a more relevant section of the test image to crop to?
It is as simple as doing a 300pix crop , captioning it and copy pasting onto a blank page that is no bigger than 1200x800 landscape.
One of the tests I would like to do is take a lens I have a lot of duplicates of and test them all to demonstrate the variability of individual lens.
(I have 5 x M50_1.7 and 2x A50_1.7 which together would be a good comparison.)
Actually your 55mm at 5.6 is very close to the Macro tak.

Attached Images
 
04-07-2014, 03:28 PM   #9
Pentaxian
hnikesch's Avatar

Join Date: May 2009
Location: Michigan, USA
Photos: Albums
Posts: 1,246
First test I shot my 06 and my (2) 50's I never adapt my 50 because I assumed the 06 was a better and easier solution in that range, I tested the DA 50 f1.8, the SMC A 50 f1.7 and the 06 at 45mm then just for grins I also shot my SMC F 80-200 zoom and shot test shots at 80 and 150mm. The 06 did out perform the (2) 50's and at a larger aperture. I was also happy with the performance of the zoom. All were shot with the original Q with SR off and triggered with a timer the 50's with 2 sec and the zoom with 12 sec. too much mass and shake at the longer FL. That little 06 is a great little lens and it has AF

Pentax 06 @ 45mm no PP f4



Pentax SMC A 50 f1.7 no PP f5.6



Pentax DA 50 f1.8 no PP f5.6



Pentax SMC F 80-200 @80 no PP f5.6



Pentax SMC F 80-200 @150 no PP f8



---------- Post added 04-07-2014 at 06:42 PM ----------

Just for the hell of it I PP'ed the above shots in my normal Q process in LR5

Pentax 06 @ 45mm



Pentax SMC A 50 f1.7



Pentax DA 50 f1.8



Pentax SMC F 80-200 @80



Pentax SMC F 80-200 @150


Last edited by hnikesch; 04-07-2014 at 07:08 PM.
04-07-2014, 03:45 PM   #10
Inactive Account




Join Date: Oct 2011
Posts: 147
Original Poster
QuoteOriginally posted by hnikesch Quote
First test I shot my 06 and my (2) 50's I never adapt my 50 because I assumed the 06 was a better and easier solution in that range, I tested the DA 50 f1.5, the SMC A 50 f1.7 and the 06 at 45mm then just for grins I also shot my SMC F 80-200 zoom and shot test shots at 80 and 150mm. The 06 did out perform the (2) 50's and at a larger aperture. I was also happy with the performance of the zoom. All were shot with the original Q with SR off and triggered with a timer the 50's with 2 sec and the zoom with 12 sec. too much mass and shake at the longer FL. That little 06 is a great little lens and it has AF..snip
Nice work Hans. The 06 is at Center, is that correct? You made a 12" x 9" with one of Atkins 4x6 in the middle (and maybe 4 more in the corners of the 12 x 9?), moved the tripod until the 4 outside corners of the 12 x 9 were at the edges of Live View, and shot?

(I ask because the resolution is really outstanding - better than anything I have seen. Just wanted to check we did it the same way.)

Curious how the 06 held up in the corners? Still beat the legacy glass?

Now you have me looking up prices on the 06...

Nice set of shots...

---------- Post added 04-07-14 at 04:12 PM ----------

QuoteOriginally posted by GUB Quote
Great stuff!
I may have a go at this in winter (Autumn here now)
I have taken the liberty to crop your images and stitch the crops together in the image below. I hope you don't mind.
I think it makes it easier for an outsider to absorb the info.
Maybe there is a more relevant section of the test image to crop to?
It is as simple as doing a 300pix crop , captioning it and copy pasting onto a blank page that is no bigger than 1200x800 landscape.
One of the tests I would like to do is take a lens I have a lot of duplicates of and test them all to demonstrate the variability of individual lens.
(I have 5 x M50_1.7 and 2x A50_1.7 which together would be a good comparison.)
Actually your 55mm at 5.6 is very close to the Macro tak.
Thanks Baz! Your way of showing is much better. Now to see if I can figure out how to do the same in LightZone - the freebie editor I use

BTW - always enjoy your shots in the 'Reach' and other threads (300?)

I am currently working on thinning out my four duplicate M42 28mm - a SMCT 28mm F3,5, A Vivitar Komine 28mm 2.8, a Kino Precision Soligor 28mm 2,8, and an Accura 28mm 2.8. So far the Accura is clearly the worst - likely to be expected. Soligor running third...

However, what I seem to be finding out is: What lens works best on a K-01 at 1.5x crop is not necessarily the one that works best on the 5.6x Q crop - and vice versa. (More on this later hopefully). I've been running the original 15" x 22" APS-C chart with the K-01, and my 'new' revised 9" by 12" chart for the Q. I do the APS-C one first, as I think it tells me more about falloff at the edges. Then the Q which tells me more about CA and resolution. At this point, the SMCT, seems better on the Q than the Viv, as its resolution (at F5.6) is a tad better and CA seems less. While on APS-C the Viv seems better because the SMCT falls off much worse at the edges there... But the Viv seems to have a really bad OOF CA problem wide open, making it unusable except at 5.6 even on the K-01. Never is simple.. Except for the Accura - its a clear dog...

Last edited by cahudson42; 04-07-2014 at 04:52 PM.
04-07-2014, 04:13 PM   #11
Veteran Member
crewl1's Avatar

Join Date: Jul 2010
Location: Los Angeles, CA
Photos: Gallery | Albums
Posts: 9,795
Wow Hans the 06 looks really good. Is it possible the other lenses would do better at lower f stop since they may be subject to diffraction?

Also, for everyone, if you are doing this testing on a lens that you have also posted to the adapted lenses index, you may want to post the link there to these results.
04-07-2014, 04:52 PM   #12
Pentaxian
hnikesch's Avatar

Join Date: May 2009
Location: Michigan, USA
Photos: Albums
Posts: 1,246
Yes I filled the LCD frame with the 12X9 field with the 4X6 printed chart at center then cropped the image to only the above part of the chart. I didn't put charts at the corners so I don't have corner unfo on the 06, maybe later.

Larry when I posted the lenses in the adapted thread I chose the best aperture to post and used it here, so I don't think a larger aperture would do better but I will check. As for linking this info to the adapted thread I think it would be better to edit the orig. post and add the info
04-07-2014, 05:05 PM   #13
GUB
Loyal Site Supporter
GUB's Avatar

Join Date: Aug 2012
Location: Wanganui
Photos: Gallery | Albums
Posts: 2,280
QuoteOriginally posted by cahudson42 Quote
However, what I seem to be finding out is: What lens works best on a K-01 at 1.5x crop is not necessarily the one that works best on the 5.6x Q crop - and vice versa. (More on this later hopefully
I totally agree. My 55-300 is a very mediocre lens on my K30 but seem to have real potential on the Q. Only subjective observations so far though.

---------- Post added 04-08-14 at 12:08 PM ----------

QuoteOriginally posted by hnikesch Quote
Just for the hell of it I PP'ed the above shots in my normal Q process in LR5
I think that is a valid thing to do. Your A501.7 appears to outdo your DA50mm and the PPing clarifies that fact.

---------- Post added 04-08-14 at 12:11 PM ----------

QuoteOriginally posted by cahudson42 Quote
Now to see if I can figure out how to do the same in LightZone - the freebie editor I use
I use gimp--another freebie.
04-07-2014, 06:00 PM   #14
Inactive Account




Join Date: Oct 2011
Posts: 147
Original Poster
Why?

Well, I was asked: 'Why waste so much effort on tests? Just get out and shoot!'

Its simply because 'the right tool for the job' adds fantastically to the pleasure of using it. And the results.

I am retired, and another of my hobbies is Woodworking. There are no better power tools than Festool, and no better hand tools than LeeValley and Lie Nielsen.

Zero-ing in on the best price/performance for lenses is the same as my Woodworking approach. Best bang-for-the-buck, gives me the greatest pleasure - and results.

Now it looks I need to retire my 50mm Taks on the Q for the 06. So be it!

But I REALLY would like to see some edge 06 shots first!

Chris
04-07-2014, 06:54 PM   #15
Site Supporter




Join Date: Sep 2006
Location: midwest, United States
Photos: Gallery
Posts: 3,225
Great tests and info everyone. The Q is a great test for sharpness and pf. There are many other distortions to be considered. Especially on larger format sensors.

A Leica lens designer once commented that it took over 200 measurements for them to know if a lens was good or not. Wow. That is a lot to consider.

I have lenses that swap performance on different formats. My Olympus OM 350mm f2.8 has greater sharpness than my Pentax DA*300, on 4/3 or APS-c. Edges of items are more crisp. It is hard to match th resolving power of a 5" front element. I was excited to put the Q on this lens, and then horrified to see the pf at f2.8. On the Q the Pentax DA*300 handily beats the Olympus. a total reversal of performance.

thanks
barondla
On 28mm lens comparisons for FF or APS-c, I would also want to know how rectilinear each lens is. Straight lines that curve can ruin some shots. It can be fixed in post, but there is resolution loss.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, canon, crop, f1.7, f1.8, f5.6, frame, freebie, length, lens, lenses, mirrorless, pentax, pentax q, pentax q10, pentax q7, pp, q10, q7, resolution, sensor, smc, test
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Adapted lenses tested on the Q : the reference thread crewl1 Pentax Q 1054 11-06-2016 09:22 PM
Do M39 adapted lenses focus to infinity on the Q? GibbyTheMole Pentax Q 14 10-16-2013 07:49 AM
Help with adapted lens on my Q stormtech Pentax Q 39 05-10-2013 09:25 AM
Autofocus on adapted Pentax lenses that have an SDM motor colonel00 Pentax Q 8 02-19-2013 05:20 PM
Comparing bokeh on 50's manual focus lenses. Voe Pentax SLR Lens Discussion 13 07-12-2008 06:13 AM



All times are GMT -7. The time now is 12:29 PM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top