[iCrop]

OK. The files are now available. Ed has been busy and hasn't been able to give me details on the files, so what follows here is my best guess from the EXIF data in the files. Ed can clarify things as he gets time. It appears, from the EXIF data, that Ed shot in manual exposure mode, and bracketed 3 exposures at 0, -2 and +2 EV at ISO100.

This is a huge amount of data, 33 files total, so I've tried to break it down so it's relatively easy to get what you are looking for. If there are any requests beyond the organization presented here, just post and I'll do my best to make things available. I have zipped the files into archives by lens type. Those can be found here:
A35 12 files, 671,279KB,
DFA35.zip 12 files, 682,904KB,
DA28-45.zip 9 files, 508,980KB

Individual files, by lens, can be downloaded here:
DA28-45
IMGP5622.PEF, IMGP5623.PEF, IMGP5624.PEF, IMGP5625.PEF, IMGP5626.PEF, IMGP5627.PEF, IMGP5631.PEF, IMGP5632.PEF, IMGP5633.PEF

A35
IMGP5634.PEF, IMGP5635.PEF, IMGP5636.PEF, IMGP5637.PEF, IMGP5638.PEF, IMGP5639.PEF, IMGP5640.PEF, IMGP5641.PEF, IMGP5642.PEF, IMGP5643.PEF, IMGP5644.PEF, IMGP5645.PEF

DFA35
IMGP5652.PEF, IMGP5653.PEF, IMGP5654.PEF, IMGP5655.PEF, IMGP5656.PEF, IMGP5657.PEF, IMGP5658.PEF, IMGP5659.PEF, IMGP5660.PEF, IMGP5661.PEF, IMGP5662.PEF, IMGP5663.PEF

For those curious to see the scene Ed shot without downloading the files, here's a taste; DA28-45, f16, 1/60s, ISO 100:


Given that I don't know how many will be downloading these images, I may be testing the 'unlimited bandwidth' aspect of the agreement of my web site host, inmotion hosting. If there's any dialog about this, I'll give as much notice as I can before taking any 'action' they may deem appropriate.

Enjoy!

[TheDocAUS]

I was there when Ed took the photos. We were on a slight rise about the footpath. Ed wanted something in the foreground (fence & lamp post) and something in the background (Opera House). He took a sequence of photos with each lens at different apertures. Focused using live view.

DA28-45 was set to 35mm for photos.

Cannot be 100% sure, but I think Ed focused on the fence. EDIT on the Opera House.

There were some photos taken into the sun, not sure if they are posted yet.

[mikeSF]

I downloaded them and did an initial scan(at 200%), ranking blindly with stars as i do in Adobe Bridge when i first review my own images from a shoot.
In this first pass,
the A35 looks terrible at shallow apertures(3.5-4.5) and quite good at f/8, falling off noticeably at f/16.
The DA28-45 is pretty average all the way, never getting very sharp by f/8
The DFA35 is decent at shallow apertures, looks GREAT by f/8, and falls off a little by f/16.

I will study contrast and colors next pass, but these comparison shots are valuable.

Thanks!

[SeattleDucks]

Thanks Ed and iCrop for providing the files. Took a look at the new DFA 35 vs. the A 35, both at f/8, and the results are interesting. The old manual A lens is clearly superior in the lower right quadrant, while the new 35 is better in the middle right quadrant. Perhaps due to differences in field curvature, slight decentering, or minute focus differences. Here are the screen caps (viewed at 200% in Lightroom with all sharpening and processing disabled, and using Huelight standard color profile):

Attached Images

       

---

Last edited by SeattleDucks; 04-04-2016 at 07:31 PM.

[Ed Hurst]

My sincere thanks to Mike (iCrop) for his hard work in organising and posting the files, supported very badly by my tardiness!

And thanks again to TheDocAUS for the use of his DFA 35mm lens.

Here is a rundown of the file details...
For all three lenses, i took shots at f4.5, f8 and f16. For the two primes, I also included f3.5 (could not do so on the zoom, of course).
Each shot was bracketed -2, 0 and +2. You will find that some of the -2 files are identical to the 0 files, since there was no faster shutter speed available to make the exposure any darker, so these files are redundant.

To be more exact, the files are as follows:
DA 28-45mm
f4.5 -2 5622
f4.5 0 5623
f4.5 +2 5624
f8 -2 5625
f8 0 5626
f8 +2 5627
f16 -2 5631
f16 0 5632
f16 +2 5633

A 35mm
f3.5 -2 5634
f3.5 0 5635
f3.5 +2 5636
f4.5 -2 5637
f4.5 0 5638
f4.5 +2 5639
f8 -2 5640
f8 0 5641
f8 +2 5642
f16 -2 5643
f16 0 5644
f16 +2 5645

DFA 35mm
f3.5 -2 5652
f3.5 0 5653
f3.5 +2 5654
f4.5 -2 5655
f4.5 0 5656
f4.5 +2 5657
f8 -2 5658
f8 0 5659
f8 +2 5660
f16 -2 5661
f16 0 5662
f16 +2 5663

The pictures were focused on the Opera House sails manually using live view zoomed in to the maximum extent. The ambient conditions were quite bright on the screen, so it was not the easiest task to get right, but I took my time and was as precise as I could have been. I wonder if the differences SeattleDucks found were simply due to focusing (i.e. that the DFA was focused slightly further away than the A lens). We'll have to examine other near/far parts of the files to answer that...

The camera was mounted on a heavy Gitzo tripod with an Arco Swiss Cube head, but the head itself is in need of servicing so there is a tiny amount of play in the gears - which means there are tiny framing differences between the shots taken with the different lenses as the lens changes move the camera slightly. Not to an extent that matters, I think, but please don't assume that they are aligned to a pixel level - they aren't.

The flare shots that TheDocAUS mentions may not be worth posting, as the sun was moving quite quickly between frames - I will look at the more closely again.

[SeattleDucks]

My sincere thanks to Mike (iCrop) for his hard work in organising and posting the files, supported very badly by my tardiness!

And thanks again to TheDocAUS for the use of his DFA 35mm lens.

Here is a rundown of the file details...

Ed, thanks for the details, and thank you again for shooting these and providing them to the forum, and thanks again to iCrop for the work in getting them uploaded.

These tests have been helpful to me. First, as I have been suspecting for some time, processing software such as Lightroom has become so good, that after I made proper adjustments to the files I was able to get the manual A 35mm lens files to look as good as the new/modern DFA35 and 28-45. That's not a slam on the new glass. I am quite sure the new optics offer better flare resistance for those who shoot toward bright light, and better performance at wider apertures. But in global contrast, microcontrast, color and sharpness, I was able to get the old A lens looking just as snappy and beautiful as the other two lenses, by adding only a few more points of clarity/contrast (I viewed at 200% on a Retina screen to confirm).

Additionally, these results were further confirmed today when a DFA35 arrived at my home and I tested it at f/8/11/16 at infinity and again at mid distance against my own copy of the A35. At wider apertures the new DFA is better and it also shows the better flare resistance of modern coatings when a bright light source was included in the frame. But at f/16 where I live for my landscape work, my old A copy is actually a bit sharper throughout the frame, has slightly less CA, and less corner vignetting at infinity. (By the way, those concerned about diffraction at small apertures will find that deconvolution sharpening recovers most of the detail. I just printed a 30x40 landscape shot at f/22 with the 67 105mm and it looks oustanding in detail and clarity even with my nose up to the print)

The DFA35 is of course a very fine lens and was a great deal at B&H in 'used' 10 condition at $1350 (looks brand new, and I'm wondering if this is the copy B&H loaned to Diglloyd for his tests), but I can't justify keeping it when my A35 copy is already so sweet, so back it goes. Someone wanting the DFA35 at a discount should keep an eye on B&H's used listings for this sample to come back to them.

Cheers,
Ross

[mikeSF]

I won't be doing my own comparison since I have already sold and shipped my A35 to its new owner. I never found that lens to be lacking in stopped down landscape rendering; in fact, it was one of the best performers in my rig.
I primarily went for the DFA35 to update my rig and that is my most used focal length - i will cross fingers that it fares as well as your tested copy!

[2351HD]

I won't be doing my own comparison since I have already sold and shipped my A35 to its new owner. I never found that lens to be lacking in stopped down landscape rendering; in fact, it was one of the best performers in my rig.
I primarily went for the DFA35 to update my rig and that is my most used focal length - i will cross fingers that it fares as well as your tested copy!

I will do my own tests Mike, as I have the 28-45 and by next Monday will have the DFA35. I have a feeling that I will most likely sell the 28-45, not because its not great, it is, but because when I need wider than 35mm (28mm eq) I really want to go to 18mm or 15mm. I really wish that Pentax would have kept the DA25 or at least a redesign to work around whatever the problem was why they discontinued it. I think you will be pleased with the DFA version. I sold my A35 so that I could use the lens more wide open if required, plus having AF is handy at times.

Also, I don't take much stock in the test photos supplied in this thread. While they serve a purpose and some effort went into them, if the results were any indication of what the 645z and these lenses could produce, well you wouldn't see my buying into the 645Z system at all. I feel personally that something went awry in the test images supplied, they just aren't of the same quality as I would expect from the Z.

[SeattleDucks]

I won't be doing my own comparison since I have already sold and shipped my A35 to its new owner. I never found that lens to be lacking in stopped down landscape rendering; in fact, it was one of the best performers in my rig.
I primarily went for the DFA35 to update my rig and that is my most used focal length - i will cross fingers that it fares as well as your tested copy!

The DFA is a sweet modern upgrade to the old A and most people will love it. Better coatings, better f/8 performance (much much better than my A at f/8), autofocus, nice petal hood. I'd be keeping it and my A35 if money was no object.

[TheDocAUS]

Also, I don't take much stock in the test photos supplied in this thread. While they serve a purpose and some effort went into them, if the results were any indication of what the 645z and these lenses could produce, well you wouldn't see my buying into the 645Z system at all. I feel personally that something went awry in the test images supplied, they just aren't of the same quality as I would expect from the Z.

A decision was made to minimise any artistic content in the images, so people focused on lens performance, not the view. A 35mm lens was the wrong lens for that location and placing a post and fence smack in the bottom third of the photo does not help. The posted images are all about testing the lens, nothing more. Some will find them helpful, others will not.

[SeattleDucks]

I will do my own tests Mike, as I have the 28-45 and by next Monday will have the DFA35. I have a feeling that I will most likely sell the 28-45, not because its not great, it is, but because when I need wider than 35mm (28mm eq) I really want to go to 18mm or 15mm. I really wish that Pentax would have kept the DA25 or at least a redesign to work around whatever the problem was why they discontinued it.

Agreed, this system really needs a wider prime added back in. Ricoh has been slow to release the promised ugraded zooms, and other lenses could use the upragde also, and now with them jumping into full frame I think 645 lenses will be even slower in coming. Their lens roadmap for the K-1 is not encouraging either, putting them at 2017 and beyond.

[mikeSF]

i do have and use the DFA25mm but would like to see a 21mm prime. Heck it could even be an f/5.6 - fine w me!

[2351HD]

i do have and use the DFA25mm but would like to see a 21mm prime. Heck it could even be an f/5.6 - fine w me!

Yesss, a 21 or 22mm lens would be great, and year f/5.6 to keep the size manageable.

---------- Post added 04-07-16 at 01:39 PM ----------

A decision was made to minimise any artistic content in the images, so people focused on lens performance, not the view. A 35mm lens was the wrong lens for that location and placing a post and fence smack in the bottom third of the photo does not help. The posted images are all about testing the lens, nothing more. Some will find them helpful, others will not.

I understand completely. My comment was more in relation to testing methodology and it not being the way I would have done it. For example, if you focus at infinity then all parts of the scene in front of infinity that are really close, ie the footpath details, are worthless to compare because no effort was applied to attempt to get them in focus.

Maybe if one lens had substantial forward field curvature towards the edges then it may pop up here, but its a long shot.

I would have taken a shot that had detail within what would be an acceptable DOF range of a 35mm lens, for example the foreground detail starting at around 3-5m and going to infinity. This would have been able to be focussed at f13 and get mostly the whole scene in focus, showing each lens' strengths. Then maybe a shot focussed at middle ground in the same scene, that way the wider apertures would show field curvature if there was foreground/background detail in focus.

Its a lot of work to do correctly and I understand if it was out of the question, I certainly don't have the time right now to do that. I guess thats why Lloyd Chambers charges for access to his site.

Scott

[Ed Hurst]

You make good points and I am sure there are many ways in which a test could be done better. I am certainly no lens tester by profession. However, this sequence of shots does represent broadly how I would use the lens in most cases (even if the shot composition does not), so any differences that show up here would have some relevance to landscape shooters. Some differences have shown up here which I think are instructive.

For example, I am interested in the lenses' absolute resolution when used at around f8 and focused in the distance (not quite infinity). I am interested, when used that way, how the edges and corners look, etc etc.. For that purpose, these shots, I think, have some utility.

Nevertheless I would love to see some fully controlled tests done on these lenses (and the FA35).

Hope they are of use to some people as they are.

[TonyW]

Just a thank you from me for the work you have put into producing these. As a 645z newbie I am finding my way around and of course lens testing and comparisons very valuable particularly with a real subject - although I still like to see test charts as well .

Selecting a few of the files I see something a little strange when opening in ACR i.e. they are inverted, something I have never seen before (see attached). Well of course from each of our perspectives (You Australia and me UK) we are upside down and surprisingly neither of us has fallen of the world yet , so this is probably down to that and of course water circulating round the plug hole in opposite directions .

Obviously not a real problem but honestly this is how the images appears for me.