The paper "A Pixel is Not a Little Square" is a great read for anyone looking for more insight into how that sort of data can be processed, and why some of those things work.
I've never understood how dedicated interpolation software works. Sharpening, contrast, saturation and the rest I get, and the role they play in enhancing the perception of "sharpness", among other things. Does it add pixels based on the proximity of similar pixels to effectively make a larger file and offer more information?
Edit: I took the opportunity to look it up. It seem bicubic interpolation, which interpolates the nearest 16 pixels and is considered the best, degrades the image between 120 and 150% enlargement, which doesn't seem like a lot.
It's often said that the M4/3 format is the digital inheritor of the 35mm film tradition. There are too many variables to make a direct comparison, including the evolution of lenses in the last twenty years and viewing technology, so I thought I'd stick to maximum print size that retains full resolution, close up.
This was triggered by a debate elsewhere where people claim to have made very large prints from micro four-thirds cameras, including a 7ft 4in (225 x 150 cm) image from a 16mp M4/3 file, on YouTube. It isn't news that small formats can produce extremely large printed images, exhibitions routinely contain prints several feet across, sometimes themselves reproduced from a small print source. Given sufficiently viewing distance there are no limits to print size, as billboard advertisements illustrate.
However in a domestic environment, or a gallery where framed images invite the viewer to look ever closer, sharpness and resolution have limitations that depend, among other factors, on the format used. There is a subjective element in determining how sharp is sharp, and sharpness isn't the only factor in determining what people think of as "sharpness", which is a mix of factors. Nevertheless I thought I'd do my own tests using optimum apertures and focal lengths, and especially by downloading files from lenses thought to be exemplary and resolving the highest possible detail on test charts.
My conclusion was that if you require absolute detail in a context like landscape, where the ability to discern and describe individual pieces of foliage and stones at distance is a way in to the subject matter, 16 x 12" is a useful working limit. Which happens to coincide with my experience of 35mm film over the years.
This may seem conservative, and there's no doubt many would be satisfied with much larger prints. Nor does it speak to the aesthetic qualities of the image, or the visceral effect size might have on the viewer, or the technical parameters of print technology. It's simply my appraisal of the potential of the micro four-thirds format for nose-to-the-glass examination some subjects and certain situations invite.
Any thoughts?
Improvements in this realm are coming at a rapid pace. Bicubic interpolation is a "dumb" method in that it only predicts missing pixels based on nearby pixels. Thanks to AI, it's actually possible to blow up pictures to huge scales with minimal loss in detail. Computers these days are able to not just see details on the pixel level, but understand what an object is and infer missing pixels based on how the object is supposed to look, and not just a pixel's neighbors. And the longer these programs are around, the more images they'll learn from, and the more accurate they'll become.I've never understood how dedicated interpolation software works. Sharpening, contrast, saturation and the rest I get, and the role they play in enhancing the perception of "sharpness", among other things. Does it add pixels based on the proximity of similar pixels to effectively make a larger file and offer more information?
Edit: I took the opportunity to look it up. It seem bicubic interpolation, which interpolates the nearest 16 pixels and is considered the best, degrades the image between 120 and 150% enlargement, which doesn't seem like a lot.
Out of curiosity, why does this format suck on 1 to 1? Just wondering what the parameters are that you're using?This format simply sucks on 1 to 1. But it gives small cameras and lenses. And ok prices.
If you print on 8x10, letter size it will be just fine.
My problem, I like to open it 1:1. 35 film sucks, 35 sensor in my Leica M-E is perfect. I mean, addictive for 1 to 1 viewing
Out of curiosity, why does this format suck on 1 to 1? Just wondering what the parameters are that you're using?
Context is important, too. One of the traditions of landscape photography, is that the photograph reveals more than the casual glance. In other words more detail can be rendered in a photograph than can be considered by the eye at one moment, because the photo freezes time giving the viewer a longer period to take it all in. This sells lots of expensive cameras to landscape photographers, but it isn't the only tradition. A landscape photo can also be reductive, translating a breadth of detail and tonality into symbolic values and graphic tonality. This foregrounds the photographic process in a more obvious way, but doesn't make for worse pictures. Jem Southam I referred to earlier, says he shoots landscapes on a digital camera before sunrise and after sunset. This requires embracing the digital artefacts such an approach makes unavoidable, and he says he has to lose half the shots because the noise is too much for his taste.This format simply sucks on 1 to 1. But it gives small cameras and lenses. And ok prices.
If you print on 8x10, letter size it will be just fine.
My problem, I like to open it 1:1. 35 film sucks, 35 sensor in my Leica M-E is perfect. I mean, addictive for 1 to 1 viewing
I've googled both your terms. I can't get a hit at all for "gobbles" using gobbles m43 or gobbles digital photograph as search terms. Is it spelled right? What I find for pixelation looks more like 80's video game graphics than any photo I've seen. Is there any chance you could provide me with an example? Only reason I ask is I've recently purchased a m43 camera and I'm quite happy with the results but I'd like to see where it's limits lie. I'm going to try printing the same photo at various sizes to see the maximum print size I can get from a well taken image at full resolution for my camera (My own subjective opinions and not one calculated through math). This pixelation and gobbles thing though, I haven't experienced yet and I'd like to know what conditions might be more apt to produce it so I can take steps to compensate. We aren't talking about the result of noise are we? When using high iso? I can see stippling/noise or maybe it's the pixelation you're talking about when I use an iso above 1600.Very simple parameters. No pixelation, no gobbles. mft often has it at 1:1.
I don't see pixilation at 1:1. Noise perhaps, but no pixels. A low resolution screen can cause jaggedness that people confuse with pixilation.When I pixel peep my current low iso images by zooming in to 1:1 I can't see that pixelation.
I've googled both your terms. I can't get a hit at all for "gobbles" using gobbles m43 or gobbles digital photograph as search terms. Is it spelled right? What I find for pixelation looks more like 80's video game graphics than any photo I've seen. Is there any chance you could provide me with an example? Only reason I ask is I've recently purchased a m43 camera and I'm quite happy with the results but I'd like to see where it's limits lie. I'm going to try printing the same photo at various sizes to see the maximum print size I can get from a well taken image at full resolution for my camera (My own subjective opinions and not one calculated through math). This pixelation and gobbles thing though, I haven't experienced yet and I'd like to know what conditions might be more apt to produce it so I can take steps to compensate. We aren't talking about the result of noise are we? When using high iso? I can see stippling/noise or maybe it's the pixelation you're talking about when I use an iso above 1600.
When I pixel peep my current low iso images by zooming in to 1:1 I can't see that pixelation.
Kind of new at this digital photography thing. Been shooting film for the past 7 years. Lots to learn.
Thanks for your help.
I don't see pixilation at 1:1. Noise perhaps, but no pixels. A low resolution screen can cause jaggedness that people confuse with pixilation.
I think the question is is it enough, not is it too much. I don't see a downside to too much. Given the same image size, is 42MP worse than 24MP? Is FF worse than m43? If so, how?In a discussion of the aesthetic limits of M4/3, we're forced to ask how much is too much?
How ever did people create those superb historic photographs when 400 ASA was fast? In film and digital I rarely move from base ISO of 200. It's easy to see how good high ISO performance is useful for a professional working in a dimly lit church, for normal photography I find 200 good enough, or I introduce off-camera flash. High ISO is a product of processing, darkness remains dark.ISO1600 is considered low iso these days. Blotching, gobbles and pixalisatoin more often appears with high ISO.
How ever did people create those superb historic photographs when 400 ASA was fast? In film and digital I rarely move from base ISO of 200. It's easy to see how good high ISO performance is useful for a professional working in a dimly lit church, for normal photography I find 200 good enough, or I introduce off-camera flash. High ISO is a product of processing, darkness remains dark.
Early on, I did some reportage. I would occasionally push to ASA 1600 to avoid flash. I found ASA 3200 virtually unusable.During film only era it was common to push film.I push my 400 bw @3200. I'm just not into the landscapes or else static as most likely you are.
Early on,, I did some reportage, I would push to ASA 1600 to avoid flash. I found ASA 3200 virtually unusable.[/QUOTE]During film only era it was common to push film.I push my 400 bw @3200. I'm just not into the landscapes or else static as most likely you are.
I'm not sure what your point is? Mine was whether noise is too compromising in a large print from a M4/3 source, or does the strength of the image allow us to ignore the technical shortcomings. Regarding the 42 vs 24 mp question, it's widely known that high megapixel cameras perform worse than a lower MP equivalent at high ISOs, but allow for bigger enlargement before pixilation.I think the question is is it enough, not is it too much. I don't see a downside to too much. Given the same image size, is 42MP worse than 24MP? Is FF worse than m43? If so, how?
Indeed, but the image won't look the same pushed. It will have large grain (silver halide clumping in fact), contrast will be much higher with a 3-stop push, and highlights and shadow detail will be less. You may desire that aesthetic, but it's generally a compromise people are willing to make to get the shot, in film or digitalI push my 400 bw @3200
Landscape is an occasional activity. I'm far from static though, and M4/3 allows me to get extra depth of field for an equivalent aperture, so I can keep the shutter speed up. The main reason I keep an M4/3 camera as well as my Fujis, is it's the only high quality zoom lens camera that fits in a pocket. The fact it can shoot sharp, well saturated photographs at 16 x 12" is remarkable, and the excellent video capability makes for a compact digital imaging tool. The trend towards large, sophisticated professional micro four-thirds cameras is not for me, I'd use a larger format, but M4/3 has been adopted by wildlife shooters in a big way and they require the heft and waterproofing of a pro specified camera.I'm just not into the landscapes or else static as most likely you are
Why are you shooting 3200 outside in daylight?
Landscape is an occasional activity. I'm far from static though, and M4/3 allows me to get extra depth of field for an equivalent aperture, so I can keep the shutter speed up. The main reason I keep an M4/3 camera as well as my Fujis, is it's the only high quality zoom lens camera that fits in a pocket. ..
I've never seen a print so large I see pixelization, Loss of resolution (mush), yes; pixelization, no. Maybe billboards.Regarding the 42 vs 24 mp question, it's widely known that high megapixel cameras perform worse than a lower MP equivalent at high ISOs, but allow for bigger enlargement before pixilation.
At the 24mm to 50mm equivalent focal length I usually work, everything is in focus at f5.6 on M4/3. 200 ISO gives 1/500 in daylight, and between 1/000 and 1/2000 in sunshine. Crushing the blacks means 1/4000 and beyond. f4 offers lots of depth of field at 1/250 in dull weather, so no need for IS. Micro four-thirds is useable for action!But for ISO200 and zooms mft makes a lot of sense. Because you really need good stabilization
Actually, that was one of the reasons I decided to go for micro 4/3. Having that extra depth of field is more important to me than having more bokeh, even if it's only a small amount.At the 24mm to 50mm equivalent focal length I usually work, everything is in focus at f5.6 on M4/3. 200 ISO gives 1/500 in daylight, and between 1/000 and 1/2000 in sunshine. Crushing the blacks means 1/4000 and beyond. f4 offers lots of depth of field at 1/250 in dull weather, so no need for IS. Micro four-thirds is useable for action!
Yes, me too.Having that extra depth of field is more important to me than having more bokeh
There also the matter of tonality. British box cameras favoured the 6 x 9 format, and mostly had meniscus lenses. Not very sharp, nor great resolution but with masses of smooth tone or saturation for the rare colour film that was used.Sharpness is largely a measure of how much contrast there is along edges in an image. Resolution is largely a measure of how small the smallest discernible detail is in an image.
I agree with certain reservations. Viewing distance is one, and I also think there's a maximum size per format before sharpening detracts from the balance of a photograph. For example if you enlarge an image in editing software and concentrate on a fine detail like grass, it will sharpen up nicely but the image as a whole can appear over-sharpened in smoother areas, with what people have described as a visual mosquito whine. These things are subjective - some people can look at a heavily HDR'd photograph and think it's an oil painting!In my experience, you can shoot pretty much any format you want and print pretty much any size you want (within reason) and as long you intelligently apply reasonable selective sharpening to the image once it’s at its final output resolution, it will register as a great looking image for most subject matter to most people.
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?