The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Will it ever end?

jlm

Workshop Member
my biggest gripe is the used market gets out of whack. a two year old digital back should not depreciate 50% or more but it does do to perceived "obsolescence".

what will happen to those perfectly good A7R's when a new one is released this spring?

an old story, but exacerbated by the new product madness.

if sony comes out with a 50mpx A7r, those will be tiny pixels
 

ErikKaffehr

Well-known member
50 MP 135 FF release imminent

Hi,

It seems that the release Canon 5Ds/5Ds R at 50.6 MP is imminent.

The pixel density on full frame 645 would correspond to around 135 MP.

With some probability, Sony and Nikon will also release 50 MP digital cameras within a few days.

Best regards
Erik
 
Even if resolution seems be increasing at an alarming rate, it's not that big of a difference as it seems, because we need exponentially more pixels to increase the amount of real detail as far as the final print goes.

We need 4x as many pixels to increase linear resolution (amount of pixels between any two points) by two, so going up from say 40mp to 50mp is a mere 12.5% increase in visually discernible detail. I think +50% is a better value to go by as a reasonable detail increase, since you're getting 9 pixels where you used to have 4 - enough to far more accurately determine the correct color and luminance for any point.

Needless to say, digital processing technology needs to be able to keep up too... where computers were previously well off using a single graphics adapter or even an integrated solution, 4K and 5K displays need not just the fastest hardware, but sometimes two or more cards just to run properly, as the workload has quadrupled even compared to 2560x1440 displays, and this of course applies to the electronics used in cameras as well.
 

ErikKaffehr

Well-known member
Hi,

The first Canon with CMOS sensor was introduced in late 2000, that camera had 3.1 MP resolution. That was 14 years ago. My first DSLR dates back to December 2004 and had a 6 MP CCD-sensor. December 2008 I bought my Sony Alpha 900 sporting a 24 MP CMOS-sensor, that was 6 years ago.

So my interpretation of history differs from yours.

Best regards
Erik



Btw, it would be amusing to be having this discussion with the current state of the art tech maybe even 5 - 7 years ago. I remember mortgaging the first born for a 2.74mp Nikon D1 which was the state of the art everywhere outside of NASA.

Talk of 30mp+ is so pedestrian today that it's kind of funny when you think about it.
 

Quentin_Bargate

Well-known member
We can already see what overkill resolution looks like through stitching. 200mp is also already available with the H5D-200ms, although limited to static objects, of course. Its good but honestly, a bit boring, because we already have enough to meet most photgraphic situations. I hope for better, not more, so Lets move beyond the compromised Bayer designs and look for true three layer RGB technology, similar to Foveon but hopefully with better high iSO performance.

Give me a 50mp true RGB sensor, and I will be happy.
 

Paul2660

Well-known member
Btw, it would be amusing to be having this discussion with the current state of the art tech maybe even 5 - 7 years ago. I remember mortgaging the first born for a 2.74mp Nikon D1 which was the state of the art everywhere outside of NASA.

Talk of 30mp+ is so pedestrian today that it's kind of funny when you think about it.
Graham, this brought back some memories. I can remember the same way, in 2000, when the D1 was announced. I ran out to my local camera store (back then they were still around in Little Rock), and put my name on the list. The only folks in front of me were some local newspaper photographers.

Before the D1, I had been working with the Sony 505, (still wish I had it as it was a bit unique) and then the Nikon 990, both non DSLR and non 1.5 crop styles.

The D1 was a huge step, and I can remember all the test shots I took, wondering if digital would handle water (longer exposures) the same way as film did.

Then within 1.5 years camea the D1x at 6MP. I did think for a while that that would be the end. I shot the D1x for quite a while and tried briefly the 660 and 560 (Kodak chip in a Canon body). Wish I still had the 660, as I did love the look of those files.

Many have forgotten the D1x, with the strange layout of the CCD (and I believe that was a Sony CCD). That chip was a strange layout and had some really harsh interpolation errors. Then Qimage and Bibble came out with the 12MP output from the D1x, anyone remember those days?

So my first move to CMOS was the Canon 1ds, in early 2003. Stayed in their camp until April 2014 and won't be going back anytime soon. The race is pretty much over for me. But it's fun to look back.

Paul
 
Last edited:

stngoldberg

Well-known member
Personally the most exciting aspects of photography are the creativity and the composition. While for a long time, I bought the latest technology because I wanted more and more detail in my images, the current state of availability allow me to tell my story with 50 megapixels and technical camera lenses.
Here is some food for thought; in Palm Beach, Florida on the most exclusive street is a photography gallery called the Holden Luntz Gallery.
He sells old masters along with emerging terrific masters of composition at prices usually exceeding $10,000.
Many of the pieces he sells will not pass muster with the community on this forum because they are grainy and lack the contrast and sharpness that we all crave for.
But the "ART" flies out of his gallery to well educated and obviously wealthy customers who see photography as a valuable form of expression- not as a technical achievement.
I had dinner this week with a dear friend who was excited to read a rumor about a new Sony camera which will include some technology from cannon along with a 55 megapixel chip.
I told him that I may have purchased my last camera....I am done chasing technology.....I stay up nights chasing great lighting, great subject matter, more creative ways of telling a pictorial story
Stanley
 

jerome_m

Member
In the days of film 8"x10" was medium of ultimate resolution and image quality. And it has withstood the test of time. A photographer could go from medium format to 4x5, 5x7 but 8x10 was basically it. Ultra-large film formats eg. 48"x48" or even 16"x20" did not become main stream and 8x10 remained the choice of photographers seeking that quality.
That is an interesting remark.

I think that the reason why 8"x10" was the biggest widely used format is the maximum size of prints. Fine art prints are rarely bigger than 1,5 meter, maybe 2 meters. The reason is that this is also the average size of humans, so any larger print cannot be hung so as to be observed close. Billboards are larger, but one cannot get close.

Fine art papers (digital or analog), printers (epson or canon inkjets) or processors (lambda or lightjet) top at these sizes as well.

To print at these sizes, a 8"x10" negative is sufficient.
 

Wayne Fox

Workshop Member
I glanced through the discussion, maybe I'll read it closer sometime, but over the last few years I've sort of adopted my own philosophy, be it right or wrong.

1. How much resolution is enough is an old debate, and their is no correct "truth". What's true for one isn't true for another. Easy to prove, yet people that don't need the resolution think there is something wrong with those that do. I have hundreds of images taken with less capable sensors that cannot hold up to the print sizes I desire. I have images with the same sensors that could be cropped to an interesting image, but then certainly won't hold up. When I am on a shoot, I rarely know what I'm going to get and whether any image will be worth printing large, but my goal is always to never limit my options later ... so I'll always use the highest resolving system I can, and more often than not stitch to take it further.

2. The discussion of sensors out resolving lenses has been going on since the 1Ds Mark 2, a 16.7mp sensor. The idea is once you out resolve the lens more resolution isn't helpful. but resolution of detail using a bayer sensor array is pretty complex, and considering that detail is made up of blur circles, you have to over sample the data.(I think Eric demonstrated some of this in his replies). And most lenses are better than you would think at this. to me the goal is to resolve whatever image the lens forms on the sensor with enough data that you can reconstruct the image without artifacts or other issues at the size you want to display it. So even if there is some "softness", it's hard to capture that softness accurately. Ctein has some interesting articles about this here and here. So until we can get to a point that a sensor can completely out resolve the best lens meaning adding more pixels can't produce an observable difference there isn't a reason not to keep improving as technology allows.

3. I really can't criticize camera makers for improving their products, after all without at least some new product cycles they soon may struggle to stay in business. It's not like you have to buy the new model, but someday the current one will wear out or die, and why not have something that's better. And this isn't something new to digital ... film cameras saw constant improvements to stimulate sales as well.

4. As far as the whole idea that only photographers look at an image up close and you only need to print based on the viewing distance ... not buying that one. While it is true that if the viewing distance is controlled and limited then that can be factored in (although consider that viewing distance maybe close ... I have large panos on the walls of 4 and 5 foot hallways), the idea only photographers view an image up close just isn't true. sure the majority don't. But bottom line, most large images getting printed now don't get observed close because they look bad up close. So why keep getting closer if the image has nothing to offer. Many images won't get looked at up close because the image just isn't appealing enough to the observer to look at period. But a great image may intrigue some to get pulled in, examining small areas to enjoy wonderful textures or shapes, the interplay of colors and details in small things ... maybe the character of a face in a portrait. As a photographer I try hard to make images that appeal to me personally, and while many may think they are rather mundane or non exciting, I don't see the world that way. I see the landscape as beautiful, peaceful, calming, relaxing. And I don't want to limit what an observer of my image that appreciates that quality to be limited in how they can enjoy the image. I don't think photography is different in this regard than any other art ... some pieces will pull some viewers in to examine things closer, the brush strokes, the use of technique and color to create detail ... and sometimes it's a "how in the world did they do that" kind of a thing. Some viewers also look at some photographs with similar thoughts.

So for me resolution is important and I hope makers keep trying to offer better capturing systems. I don't mean ignore other aspects, it's all important, and I"m sure there are tradeoffs, so please improve all the aspects.

I know many also mention the idea that a better camera won't make me a better photographer, and for the most part I completely agree with that, but to be honest I've always thought for skilled photographers to say that on a forum which is visited mostly by skilled photographers is sort of a weird thing to say ... sort of a "duh" statement. I print hundreds of images each week by photographers in my area, and some of them are very very good, and some of their images are fantastic ... and most of them can't make a decent 30x40 print, let alone 40x60 or 90" pano. Ok, maybe the don't "want" to ... or maybe they just say that because they can't. And I know it's often not their fault, I understand budget constraints. but I see most of these arguments as excuses for these limitations, and not a valid reason to continue improving systems. To offer these kind of tools to talented photographers seems only fair. And hopefully some day I can get the quality I'm after from a system half the size of what I haul around now.
 

dchew

Well-known member
4. As far as the whole idea that only photographers look at an image up close and you only need to print based on the viewing distance ... not buying that one. While it is true that if the viewing distance is controlled and limited then that can be factored in (although consider that viewing distance maybe close ... I have large panos on the walls of 4 and 5 foot hallways), the idea only photographers view an image up close just isn't true. sure the majority don't. But bottom line, most large images getting printed now don't get observed close because they look bad up close. So why keep getting closer if the image has nothing to offer. Many images won't get looked at up close because the image just isn't appealing enough to the observer to look at period. But a great image may intrigue some to get pulled in, examining small areas to enjoy wonderful textures or shapes, the interplay of colors and details in small things ... maybe the character of a face in a portrait. As a photographer I try hard to make images that appeal to me personally, and while many may think they are rather mundane or non exciting, I don't see the world that way. I see the landscape as beautiful, peaceful, calming, relaxing. And I don't want to limit what an observer of my image that appreciates that quality to be limited in how they can enjoy the image. I don't think photography is different in this regard than any other art ... some pieces will pull some viewers in to examine things closer, the brush strokes, the use of technique and color to create detail ... and sometimes it's a "how in the world did they do that" kind of a thing. Some viewers also look at some photographs with similar thoughts.
Wayne,
This is a good point. I saw it play out with my most recent show where I included some six foot prints. Three out of 40 images were that big; each was a 2-image stitch from the IQ180, so native 240ppi at that size. I had the chance on a few different days to observe people. They would get as close or closer to those prints than the others which ranged from 18x24 up to 22x30. It is obviously content-dependent, but for landscapes with detail I think you are spot-on. People get drawn in, or "absorbed" in the detail; it becomes part of the viewing experience. For certain subjects (which happen to be the subjects many of us like to shoot), viewers will get as close as as the detail allows.

Dave
 

ErikKaffehr

Well-known member
Re: Will it ever end? (Some examples)

Hi,

I made a recent experiments with some stuff I happen to have:

First, I shot a scene here in the town I live with a 100/3.5 lens on my Hasselblad using my P45+ and Adox 120 CMS ultra fine grain film. I also shot the same image with my Sony Alpha 77 SLT at 100 mm.

If we look at resolution it is something like this:

P45+ -> 6.8 micron pixels -> 73 lp/mm
SLT 77 -> 3.8 micron pixels -> 131 lp/mm
Adox CMS -> 150 lp/mm

I made a tiny crop that show some signs and rescaled it to 290 pixels width

The P45+ image with it's largish pixels looked like this:



The SLT was shot at the same focal length, so it shows what the P45+ would look like if it had 125 MP (instead of 39 MP)


The small pixel image is much cleaner and legible. This indicates that the 3.8 micron sensor on the SLT 77 is a good match for the 100 mm lens used on that camera.

The Adox film on the Hasselblad resolved like the SLT 77, indicating that the 100/3.5 lens on the Hasselblad performs well down to something like 130 lp/mm. But, the film image is noisy. This was shot on the SLT 77 with a Photar 25 and about 20 cm of extension.



I also scanned the Adix 120 CMS image on my Minolta Dimage Scan Multi Pro at 4800 PPI.



So, what are my conclusions from this:

  • It seems that 3-4 micron sensor pitch is a good match for both 100 mm lenses.
  • It also seems that 6.8 microns is to large to properly render the image from the 100/3.5
  • Film clearly outresolves the P45+, but the image is noisy

The film could probably have more exposure which would result in a cleaner image.

Best regards
Erik
 

Paul2660

Well-known member
Hi Erik

Actually to my eyes, the top shot looks a bit better overall as there is just enough noise to make it look the way I like to see such a shot. The lower image is much smoother, which implies a bit more noise reduction.

You can see the aliasing from the P45+ in the top shot, and I wonder if you had moved to F16 if some of that might have been removed?

P45+ is still a great performer in the base iso ranges. Again yours has always seemed much better than mine was. The details in the shadows in the top shot are very close to the 2nd shot. Again to my eyes.

Paul
 

ErikKaffehr

Well-known member
Hi Paul,

My demo here is rather about demonstrating that there are benefits of going to smaller pixels.

Regarding the colour aliasing, or rather demosaic errors it is my experience that it disappears at f/16. But that comes with loss of sharpness.

It seems that much of what is lost because of stopping down can be regained with aggressive sharpening.

My take is that reducing pixel size to say 3-4 microns is benefitial and also that even pretty old lenses can deliver some contrast at those pixel densities. I did shoot a resolution pattern with the 80/2.8, and it reached a bit over 150 lp/mm on the Adox film.

Now, there are clearly advantages to larger pixels, like being less sensitive to movements and beam angles.

Regarding the P45+, I am quite happy with it. I am enjoying shooting with it and it is often the camera I reach for first. The resolution advantage is obviously there, even if I would say that the 24 MP of the Sony A99 is quite enough for my standard print size A2. Shooting a classical camera is nice. I even have 6 lenses for the Hasselblad, guess it is a bit crazy.

The way I see it having both small pixel and large pixel cameras is a good thing. Small pixels give good resolution and good reproduction.

Of all the cameras I have the P45+ has the largest pixels and the SLT 77 has the smallest ones. So it is sort of a natural comparison. With the Hasselblad I can also use film. The testing I did here was mostly to see how much detail I can get with film. The answer seems to be "plenty but with a lot of noise". Additionally it seems that the Adox film needs to be scanned at something like 8000 PPI to extract all detail.

The image below is the P45+ image without rescaling


Thanks for feedback, always nice to hear from you!

Best regards
Erik



Hi Erik

Actually to my eyes, the top shot looks a bit better overall as there is just enough noise to make it look the way I like to see such a shot. The lower image is much smoother, which implies a bit more noise reduction.

You can see the aliasing from the P45+ in the top shot, and I wonder if you had moved to F16 if some of that might have been removed?

P45+ is still a great performer in the base iso ranges. Again yours has always seemed much better than mine was. The details in the shadows in the top shot are very close to the 2nd shot. Again to my eyes.

Paul
 

darr

Well-known member
LOL.....Happiness is not giving a flying V because I now make all of my income off of shooting and printing black and white MF and LF film....
That is so cool! :clap:
Can you point me to your website or gallery where your work is being displayed for sale, as I'd love to see your portfolio.

Kind regards,
Darr
 
Top