NikonGear'23
Gear Talk => Camera Talk => Topic started by: Michael Erlewine on June 16, 2016, 01:59:58
-
This shot of a spider on a Gerbera flower was taken with the Pentax K1 and the Sigma 35mm f/1.4 ART lens. This lens is not a macro lens and, as you can see from the full-frame, this is just a normal shot. The crop shows a kind of detail I can’t get with my Nikon D810. I would like to see how some of the Zeiss Otus lenses do on the Pentax for close-ups, with perhaps a bit of extension.
The purity of the color, without any real processing, is remarkably clear IMO.
Sorry for the poor photo, but this is just a snapshot to see what kind of detail the Pixel-Shift option picks up. Answer: quite a bit. Imagine if we really devote some time to this kind of work.
-
Colours look great. However, there are a disturbing number of hot pixels in this crop - did you use a long shutter time?
-
Colours look great. However, there are a disturbing number of hot pixels in this crop - did you use a long shutter time?
Not on purpose, however, the ISO was 100 and it was f/10. Now, in Pixel-Shift Mode, which I am playing with, the camera takes four successive photos, with a bit of a pause in between. This always seems long for me. It then combines them into a single photo. Could this have produced the effect you are noticing. It is a lousy photo to begin with.
This camera is new to me, very frustrating compared to a Nikon interface, and missing crucial amenities that I have grown accustomed to. Still, that being said, this concept of recording each photosite (RGBG) via four photos is a different approach to detail than the Bayer interpolation.
I keep thinking I don't need to learn another camera, but then... I am such a gear-head that I keep seeing noteworthy progress, something I am waiting to hear more of from Nikon.
Here, at last, perhaps, is a powerful camera at a price almost anyone can afford. Unless I find something really horrible about it, we have a better mousetrap, even if only for those of us who like still life.
-
I don't know Michael, I feel like I've seen better quality images from your other cameras and I don't mean from a composition, DOF/stacking, lighting stand point. In the 100% crop the flower pedals have that long exposure noise reduction look to them, some of the tips of the flowers appear to be doubled and other edges have a grid like pattern going on, and the back leg of the spider looks like an RGB grid of some sort. All artifacts of combining 4 photos into one I would suspect.
-
I don't know Michael, I feel like I've seen better quality images from your other cameras and I don't mean from a composition, DOF/stacking, lighting stand point. In the 100% crop the flower pedals have that long exposure noise reduction look to them, some of the tips of the flowers appear to be doubled and other edges have a grid like pattern going on, and the back leg of the spider looks like an RGB grid of some sort. All artifacts of combining 4 photos into one I would suspect.
Yes, you can look at it that way, but I was trying to show folks the spider and not the flower. What is important, IMO, is this new way of approaching detail by purifying color, which may (or may not be) apparent in these small files.
-
I like the experimental attitude. This information can only be to the benefit for the greater audience in the longer run, whether the camera fulfils expectations or not.
-
I've always been fascinated by the color rendition of your images, Michael, pure or otherwise. So, no complaint or criticism about that.
That said, I also see some double images of the petals (the one on the farthest right is the easiest example). I suspect that the ever-so-slight camera shake caused the double images.
-
I do (I think) notice that the highlights with the Pentax K1 are more difficult to handle than with Nikon, IMO. Of course, it does not have the ISO 64 of the D810, but actually this pixel-shift mode is a whole different kind of color, as I see it, kind of increased resolution by more accurate color. I have yet to learn to handle it. I am still just fooling around, gathering lenses that will work, fiddling with adapters, and making mistakes, of course.
I can't just send this camera back as I have so many others. It would go against my grain, at least at this point. I love color. I am also learning the Sony A7rII, which is another beast entirely, and much more similar to my Nikon D810. The Pentax really is a maverick.
-
I've always been fascinated by the color rendition of your images, Michael, pure or otherwise. So, no complaint or criticism about that.
That said, I also see some double images of the petals (the one on the farthest right is the easiest example). I suspect that the ever-so-slight camera shake caused the double images.
Again, that is not the point here, but I have no doubt you are correct. Does anyone see the sharpness of the spider, which is enhanced not by acutance or resolution, but by purifying color? That is the advantage of the Pentax K1.
-
This technique undoubtedly puts extraordinary strain on the colour performance of the lens used. I suspect many lenses will not be able to stand up to the requirements.
-
This technique undoubtedly puts extraordinary strain on the colour performance of the lens used. I suspect many lenses will not be able to stand up to the requirements.
Exactly. I now have an adapter for Nikon lenses to fit on the Pentax K1. I knocked out the cheap lens that allows you to go to infinity, which is worthless when you get there, because the image is so poor. I can now mount the Otus lenses directly. I have to decide whether I want to use rails to support the weight of the Zeiss lenses. We will see. The Sigma 35mm (used below) is a pretty nice lens, but not at the Otus level.
-
Again, that is not the point here, but I have no doubt you are correct. Does anyone see the sharpness of the spider, which is enhanced not by acutance or resolution, but by purifying color? That is the advantage of the Pentax K1.
As I mentioned in another thread, the sharpness feels a bit harsh to me. Even though the sharpness of this level is achieved by the full color pixels thanks to the sensor shift, the monitor images are still rendered by the combined RGB pixels instead of full color ones.
The fine details you mention look like the iridescence to me, sorry to say...
-
As I mentioned in another thread, the sharpness feels a bit harsh to me. Even though the sharpness of this level is achieved by the full color pixels thanks to the sensor shift, the monitor images are still rendered by the combined RGB pixels instead of full color ones.
The fine details you mention look like the iridescence to me, sorry to say...
Ah, the night is still young.
-
Michael, I'm not trying to disappoint you. But the sensor (pixel) shift technology seems to be abandoned in the world of MF digital back. There could be some critical reason for that.
-
Michael, I'm not trying to disappoint you. But the sensor (pixel) shift technology seems to be abandoned in the world of MF digital back. There could be some critical reason for that.
Yes, like motion. But I do still life, so I may find this camera useful.
-
Ah, the night is still young.
:)
-
Talking about colour, in the studio or in the outdoors there is more than a technical difference.
-
Here we are again. I am not trying to show you the flower or the spider, but just the color and resolution, etc. of the spider. This is the Pentax K1 with the Voigtlander 90mm APO with close-up attachment. This is f/16, which obviously has some diffraction showing. More important, as far as my old eyes can see, the sensor/pixel/shift is out-resolving the lens here. This is a problem with the K1, not being able to get really great lenses that challenge the sensor and the pixel'shift. This for me is a serious problem.
I do have an adapter that I knocked the lousy glass out of that I can mount Nikon-mount lenses, although only good for close-up. That may help somewhat, but will have to wait for another day.
Any thoughts on this particular point, and not (please) about the photo, the flower, the spider, etc. Anyone else out there working with the K1 on this kind of question?
-
The 100% crop does show a peculiar form of uneven 'grittiness' (for lack of a better word) distributed over parts of the image. Could this result from the pixel shifting technology?
I assume that studio flash lighting is not a viable solution for the k1 due to the short exposure time from these light sources. Thus some kind of continuous illumination probably is required and that introduces potential issues with flickering. Is this assumption correct?
-
The 100% crop does show a peculiar form of uneven 'grittiness' (for lack of a better word) distributed over parts of the image. Could this result from the pixel shifting technology?
I assume that studio flash lighting is not a viable solution for the k1 due to the short exposure time from these light sources. Thus some kind of continuous illumination probably is required and that introduces potential issues with flickering. Is this assumption correct?
This is natural light from outside. Yes, there are a lot of variables that have to be sorted out. To my eyes, the K1 has harsher highlights than I am used to, but I have to learn to tame it, which I will try. I believe, as I mentioned, that the most serious obstacle is the lack of very fine lenses and the difficulty adapting them to the not just the mount, but to the camera controls. I have lots of things to do, and so keep looking for reasons to dismiss this camera, but it keeps getting saved by the bell, so to speak.
-
This is the Pentax K1 with the Voigtlander 90mm APO with close-up attachment.
Did you use the supplied close-up attachment?
-
The second sample looks pleasingly smooth and not harsh at all. Perhaps it is more about the lighting (the first image was a bit backlit, I suppose?). I observe no iridescent-like anomaly.
-
The K1 should not have overly harsh highlights. The recent Ks have had good dynamic range, right up there within 1/2-1 stops of the good Nikon bodies. So how much headroom are you finding with the K1? The D810 has so much headroom that I could imagine being used to shooting it and then overshooting with the K1 during first tests.
Pentax colours can be a little contrasty & harsh, but I have not looked yet at any K1 raws and do not know if that still holds. However I do see some blockiness as Bjørn mentioned in some of the tones in the crop. The K1 may be a few bits shy of the colour depth of the Nikons, but it should not do that. Is the K1 set to 14-bit (or whatever it has for that setting)?
I wonder if you are encountering some conversion woes?
Temporarily turn off the colour and look at the tone roll-offs in desaturated (B&W) mode. Is there till some blockiness? Set the black and white points appropriately. Set highlights/shadows appropriately with a curve. And then turn the colour back on. What happens? What hits the wall, if anything? Do artifacts appear with the colour on that are not there with the colour off?
(I really apopreciate all your tests with the K1. I still would like to have this camera. And I am really looking forward to the a7R2 tests!!)
-
Here is a Pentax mount adapter which supposedly permits mounting of Nikon lenses with proper flange focal distance. However, no information yet as to whether it works on the K1. Do be sure to read the product warnings page. "-)
http://adaptist.weebly.com/
http://adaptist.weebly.com/product-warnings.html
Kindly ignore this if you already know about it.
-
Here is a Pentax mount adapter which supposedly permits mounting of Nikon lenses with proper flange focal distance. However, no information yet as to whether it works on the K1. Do be sure to read the product warnings page. "-)
http://adaptist.weebly.com/
http://adaptist.weebly.com/product-warnings.html
Kindly ignore this if you already know about it.
LLoyd Chambers posted about this. Let's wait and see. If it does what it says, it is inexpensive enough. Not sure what it does to the way the Pentax K1 does not like uncoupled lenses. It is interesting.