24mm PC-E Episode 2

[tashley]

OK so I got a replacement and had a chance to do a quick run of shots today: I wanted to ensure that it was at least satisfactorily sharp from edge to edge at f5.6 before I got into seeing how it is when shifted. Unlike my first copy, it is indeed acceptably sharp all over. In fact I think pretty darned good. You have to focus it VERY carefully: I think there's some field curvature so you have to make sure that the centre and edge are both in focus in LV within the available DOF (or you have to get lucky!) but get all that right and the signs are pretty promising. Interestingly, though there are mild signs of diffraction when stopping down to F8 or even F11, they are pretty insignificant and you do get a lot more DOF to play with and a touch more corner sharpness at F8.

Many people, including me, have reported problems with this lens on the D800/E but I had heard a couple of people saying it was great. So far, shift tests pending, I am really pleased that I tried another one.

Full sized 91% quality JPEG here:
http://tashley1.zenfolio.com/img/s3/v41/p324326029.jpg

 

[D&A]

Hi Tim,

Of course I"m expressing the obvious when I say the real test is how it performs when lens is shifted to various degrees. Still it's a promissing start and one would like to first observe that unshifted, it appears to be a good sample. Looking forward to you additional tests.

Dave (D&A)

 

[tjv]

Great to hear! There is promise yet!

 

[Shelby Lewis]

having never owned a tech cam... I'd be pretty happy with that as a starting place!

 

[tashley]

OK guys ('n gals)

More tests. I shot with shifts up and down (not showing the downs, they're irrelevant because of DOF and foreground) at up five and ten, and left and right five and ten. I also did my 'parallel floor' test where I tilt the camera down on the cube as far as it goes without the extension, then tilt the lens movement down a little, and use LV to focus in an attempt at getting the whole of the ground from very near to far in perfect focus.

It was a doddle and worked well at both F8 and F11 and I suspect would have been great at wider apertures too.

The shots are all linked here. The shifts were all done at F11 and please start at the second frame, since the first frame was shot yesterday and focussed differently. The shots are in 80% JPEG quality and are at 50% size so as to emulate a 200DPI print and so as to save on huge upload times!

My conclusions: this is a good copy. I would happily use it for five and seven rise and for ten rise if the top of the frame was sky, or only had substantive subject in the centre 2/3rds of the frame. I would also happily use it for shifts of seven left and right; at ten shifts, it shows vignetting and the sharpness isn't great. Could it be better? Yes. Do I still hope to eventually see new PC lenses in Nikon mount that are wider, more flexible in terms of combinations of movements, and optically even better? Yes. But I also want world peace....

uploading now...

 

[D&A]

Tim, from your detailed descriptions, this copy sounds promising. I realize the reasons for not posting at 100%, which is where I would personally (and sure you have) evaluated a lens like this.

I'm not sure with regards to the Current Nikon 24mm PC-E lens, if there is a better performing copy when a near-full shift is attempted, regardless of f-stop. I question if even used on a D700 and defraction limits occured at a smaller f-stop that on the D800 would it result in much of an increase of sharpness at the sides?

Dave (D&A)

 

[tjv]

Thanks Tim, it looks a lot better than your previous copy and give me heart for possibilities in building my own kit, where a 24mm and 45mm PC-E would be my prime go to lenses.

 

[tashley]

And it's also fun: even as a walk around you can get some great effects and can do so without live view with a bit of practice. In this shot I was trying to nail a narrow band of perfect focus running right down the tunnel and I sort of got it!

Full sized version here

 

[danielmoore]

Absolutely lovely image. Your success in finding a good copy inspires me to look again at this lens. In shifting at f/5.6 and f/8 are the corners and edges becoming sleazy? Is f/11 required for optimally balancing sharpness over the full frame? I assumed so but would rather hear it from you. Thanks for these.

 

[Jack]

Tim,

You've given me hope -- I found a used 24PC the owner claimed was a very good copy, had him send me some pics and it looks as though I have a better than average copy arriving as well. If it works as well as I hope, the D800 becomes a *LOT* more versatile for me and MF+tech becomes less attractive to schlep...

 

[tashley]

Absolutely lovely image. Your success in finding a good copy inspires me to look again at this lens. In shifting at f/5.6 and f/8 are the corners and edges becoming sleazy? Is f/11 required for optimally balancing sharpness over the full frame? I assumed so but would rather hear it from you. Thanks for these.

Thanks!

I think the key thing is that this lens is very sharp indeed on centre, but that you really do need to stop down for sharpest edges and corners when shifting. I need a lot more shots under my belt to get the extreme specifics but I suspect that the need to stop down is at least in part a DOF issue: basic trigonometry shows that if you are bang in front of a building and have your sensor exactly parallel to it, then the top part of a shifted frame is a lot further away from the nodal point than is the part of the building directly in front of you. So I am next going to experiment with re-focussing between shifts, which I did not do in the series I posted.

In any event, the resolution of the lens is good enough to withstand a stop down to F11. Sure you lose a bit, it starts to undersample the sensor, but it's still pretty good, and... it IS the only game in town!

 

[tashley]

Tim,

You've given me hope -- I found a used 24PC the owner claimed was a very good copy, had him send me some pics and it looks as though I have a better than average copy arriving as well. If it works as well as I hope, the D800 becomes a *LOT* more versatile for me and MF+tech becomes less attractive to schlep...

I hope yours is a good 'un!

It is SO easy to accidentally leave one of the movements not exactly neutralised and then think you have a sh1t copy...

I am headed the same way as you: this lens on this camera will not match MF but with careful technique and a good copy, it might just be close enough often enough to make some changes... possibly just to what I take where and when, and possibly even to what I own...

 

[Jack]

I it might just be close enough often enough to make some changes...

Yup, that's what I'm beginning to think. And it's certainly more convenient than my other alternatives...

 

[aboudd]

I've had the Hartblei on both the Contax 645 w/ P25 back and the newer digital version on the Pentax 645D. I've owned the 24 PC-E and use it on both on the D3x and the D3s, I can tell you the MF versions of shift lenses are markedly inferior. Until Stefan starts making T/S lenses from Zeiss glass for medium format, the Nikon 24MM PC-E is the only game in town. The 24MM PC-E coupled with the D3x or D800 is an unbeatable combination - until Leica introduces the 30MM Elmar shift for the S2. Given their production record, that should be by 2029.

 

[tashley]

I've had the Hartblei on both the Contax 645 w/ P25 back and the newer digital version on the Pentax 645D. I've owned the 24 PC-E and use it on both on the D3x and the D3s, I can tell you the MF versions of shift lenses are markedly inferior. Until Stefan starts making T/S lenses from Zeiss glass for medium format, the Nikon 24MM PC-E is the only game in town. The 24MM PC-E coupled with the D3x or D800 is an unbeatable combination - until Leica introduces the 30MM Elmar shift for the S2. Given their production record, that should be by 2029.

I am so with you on that: I had the first S2 in the UK and even then, they were blandishing the TS lens as something on the near rather than far horizon!

 

[danielmoore]

One thing I've been curious about is how a moderately shifted 24 PC-E would compare to the 14-24 zoomed out to encompass the same top-to-bottom/left-to-right area when cropped. Where it gets intriguing is when distortion correction is applied to both, especially since the 14-24 begins to noticeably distort afer 24mm and the 24 PC-E doesn't produce rectilinear results, at least by architectural standards, so lens correction is a foregone conclusion for each. Landscape use is a secondary concern for some. Existing resolution tests reveal the 14-24 is sharper towards the shorter focal length so I expect this could potentially be a bit of an eye opener.

 

[D&A]

So I am next going to experiment with re-focussing between shifts, which I did not do in the series I posted.

Tim, this was precisely what I was going to ask you when I made my first post on this thread earlier this morning. Then I deleted and rewrote my post and had forgotten to restate my question.

I recall an earlier (fast and dirty) try of mine where I shifted approx 3 degrees and then refocused and compared to the the version that was not refocused. I repeated this with varying degrees of shift. From what I recall, depending if subject was flat field and in a single plane or not, it did make a surprising difference...but at the same time (depending on aperture), it also resulted in other moderately sharp parts of the image to go soft. Subject dependent and often times there will be some tradeoffs.

Dave (D&A)

 

[D&A]

One thing I've been curious about is how a moderately shifted 24 PC-E would compare to the 14-24 zoomed out to encompass the same top-to-bottom/left-to-right area when cropped. Where it gets intriguing is when distortion correction is applied to both, especially since the 14-24 begins to noticeably distort afer 24mm and the 24 PC-E doesn't produce rectilinear results, at least by architectural standards, so lens correction is a foregone conclusion for each. Landscape use is a secondary concern for some. Existing resolution tests reveal the 14-24 is sharper towards the shorter focal length so I expect this could potentially be a bit of an eye opener.

I too was thinking about this when I had the 14-24...as to whether one could shoot wider than the 24 shift lens and then adjust for distortion and crop. I didn't have the 24mm PC lens at the time, so I couldn't do a direct comparison....but the 14-24mm is generally sharper on the wider end as you stated.

Dave (D&A)

 

[danielmoore]

And in the case of the rented 24PC-E I had, the 14-24 was already notably sharper at 24mm so... ??

 

[Jack]

I too was thinking about this when I had the 14-24...as to whether one could shoot wider than the 24 shift lens and then adjust for distortion and crop. I didn't have the 24mm PC lens at the time, so I couldn't do a direct comparison....but the 14-24mm is generally sharper on the wider end as you stated.

Dave (D&A)

I played with a friend's 14-24 last week -- the center was great, but the corners were not much better on it than my 20 AF-D. I need to see this better 24PC on my D800, but my guess is it's eons better that squeezing and stretching pixels in post.