Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
10-01-2015, 05:45 AM   #16
Pentaxian




Join Date: Nov 2013
Posts: 4,540
QuoteOriginally posted by Kolor-Pikker Quote
The key point with TV dynamic range lays entirely within the black level. A display can have "infinite" contrast ratio simply by having blacks that are pure black, but our eyes don't get strained by an absence of light. For a display to truly show more than 7 stops DR it would not only have to have a very low rated black level, but a very bright backlight as well, even though in practice though most displays are calibrated to a brightness target that's comfortable for extended viewing regardless of what the display is capable of.

You're still being mislead by your visual system. Our eyes don't work like photographic nor motion video devices, they continually scan and update an image that's recorded to visual memory. It's impossible or at least not within our conscious control to truly "focus" on something so as not to allow our brains to fill in the blanks. Our eyes move and update various points in space over 200 times a second even when we think we're looking straight-on, and they have to, since only 1% of our visual field is responsible for nearly all of the detail that we see.

The reason you can look at a sunset and see a lot of the tonal range with such small sensors/optics stuck in your face is due to HDR processing occurring along every stop of the way from your eyes to your imagination. And imagination has a lot to do with things, after all, do you know what the color pink is? The near-visible radiation scale begins with ultraviolet and ends with infrared, so where does pink fit in? It's actually a non-existent color your brain makes up in absence of green, so pink is actually "negative green".

Facts concerning the human visual system get increasingly more freaky from here on out as you ask more and more questions, like how do out eyes deal with cloning out all the blood vessels in our eyes, among other things. When you take every factor into account, our visual system is constantly doing one heck of a Photoshop job, essentially taking a bunch of random data from the visual cortex and reconstructing it in a way we can comprehend.
Man you can say it all you want... If I go to a church, I can see the stained-glass window with every details and colors. I also see the surrounding wall with all details of the stones.

If I ask my camera, if it does the metering on the wall, the windows are pure white and if it does the metering on the windows, the wall is black. I can push the wall but then it is VERY noisy. To mitigate that I need to go into exposure bracketing, HDR and tone mapping. Then some part of the image look unnatural so I may have to merge different rendering and spend quite some time to not get what I can feel when I'am actually in the church.

You could say that the camera is better because in theory the sensor is better, bigger, larger with more dynamic range... But the results are not there. So no, it is not as good.

10-01-2015, 07:14 AM - 1 Like   #17
Veteran Member
Kolor-Pikker's Avatar

Join Date: Dec 2014
Posts: 333
QuoteOriginally posted by Nicolas06 Quote
Man you can say it all you want... If I go to a church, I can see the stained-glass window with every details and colors. I also see the surrounding wall with all details of the stones.

If I ask my camera, if it does the metering on the wall, the windows are pure white and if it does the metering on the windows, the wall is black. I can push the wall but then it is VERY noisy. To mitigate that I need to go into exposure bracketing, HDR and tone mapping. Then some part of the image look unnatural so I may have to merge different rendering and spend quite some time to not get what I can feel when I'am actually in the church.

You could say that the camera is better because in theory the sensor is better, bigger, larger with more dynamic range... But the results are not there. So no, it is not as good.
Ultimately, you just can't cheat physics, photons in = image out. The same rules apply to everything. The only thing we can do better is for our brains to imagine things to look better than they really are. Photos, having the ability of freezing a moment in time, allow us to actually look at all the details and inconsistencies as long as we want, whereas we often don't know what we're actually looking at.

How do you know if your visual system isn't performing HDR if you don't ever actively look for the signs of it? You certainly can't compare the results of what a computer can achieve with that of the human brain, of course it would seem seamless to you, but it's really not. You're always looking at the composite of a wide range of color and tonal data. For one, we actually have separate receptors for tonality and color, while cameras do not. We can actually see something that's simultaneously extremely bright or dark and also have a color at the same time, while in the RGB system you cannot on a fundamental level.

QuoteOriginally posted by Nicolas06 Quote
I can push the wall but then it is VERY noisy.
The wall might seem noisy, but not as much noise as we have in our vision, we just don't notice it, because we can't zoom our vision in 100%, and also because among other things, as mentioned, our brain attempts to cover it up. If you get up in the middle of the night and take a moment to stare at a wall, you'll probably notice more noise than you're comfortable with, you'll also probably notice that your vision is almost entirely monochromatic, since the number of tonal receptors we have far outnumbers the color-sensitive ones, and under moonlight we practically see in black and white.

Our eyes are not better at anything, nature simply came up with a lot of clever ways of maximizing what little we have to work with.

The 20-stop DR figure is simply derived from the fact that, without the aid of ND filters, no optical lens is capable of resolving more than 20~24 stops of dynamic range, so they picked the conservative maximum.

Last edited by Kolor-Pikker; 10-01-2015 at 07:31 AM.
10-01-2015, 12:07 PM   #18
Pentaxian




Join Date: Nov 2013
Posts: 4,540
But think of it again Kolor-Pikker... All of this look great in good company. Yes the eyes work differently but what count is the result.

In the meantime, with my camera if I want to capture the dynamic range I see naturally I need to do an exposure braketing and that's boring.
10-02-2015, 05:52 AM   #19
Veteran Member
Kolor-Pikker's Avatar

Join Date: Dec 2014
Posts: 333
QuoteOriginally posted by Nicolas06 Quote
But think of it again Kolor-Pikker... All of this look great in good company. Yes the eyes work differently but what count is the result.

In the meantime, with my camera if I want to capture the dynamic range I see naturally I need to do an exposure braketing and that's boring.
To be frank, with the 645Z I haven't had to capture an HDR yet, even in cases where I though I needed it, I ended up processing just the middle exposure; having 14+ stops is just crazy.

To really push things, I once tried shooting the setting sun with a lone high-rise building in the frame to prevent reflected light from falling on it's walls, just to see how far DR could be pushed. Needless to say, the lens ran out of contrast with a 5-shot bracket of 2-stop intervals, there was no black left to recover, only full-screen glare/veiling. 8 + 14 stops = ~22, but in practice it must have been a lot less, since I didn't even need to recover shadows to see the glare. I need an HDR lens to make use of HDR now.

10-04-2015, 07:51 AM   #20
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by Nicolas06 Quote
TV are typically advertising 1:1000, 1:2000, 1:5000 that 10EV, 11EV, 12EV.
No more words needed. What I said, plus the comment about black levels above.

Your further comments imply that the eye is more acting like a video camera rather than a stills camers. True, but that's not our topic. Here, we talk about still images or single frames in a stream of images. Where again the human eye cannot compate anymore to say, an A7s. Moreover, the ability of a video camera to quickly swicth aperture between say, F1.4 and F32 outclasses the human eye even more than do its still capabilities. The iris simply isn't this fast and wide gammut to adapt.

The one point where I actually do agree to what you said is the requirement for a camera to create an immersive 3D full dynamic range experience at presentation time. This is actually commercially relevant as the requirement for 32-bit 360 panoramic backdrops for 3D renderings of marketing presentations such as automotives etc. As of today, these cannot be produced from a single shot indeed, they require multiple shots at different angles and exposures. However, the human does actually require time with many "frames" to digest this amount of information too.

Last edited by falconeye; 10-04-2015 at 08:05 AM.
10-05-2015, 12:05 PM   #21
Pentaxian




Join Date: Nov 2013
Posts: 4,540
This link (Clarkvision Photography - Resolution of the Human Eye) shows that from a practical point of view:

Many people are able to see a 1:1000000 contrast in very dim light (looking at stars), that's approximately 20EV. The other link show that this is still valid in bright ligting: How does the dynamic range of the human eye compare to that of digital cameras? - Photography Stack Exchange there lot of interresting responses.

Me I know I see this dynamic range too.

We can go as deep as we want on the theory of why the eyes or the brain or whatever is cheating. But we are not CAMERA and sensor centric. We are HUMAN centric. What count is for the device to satisfy humans, not the reverse. If the devise can't match, what ever the reason that's a device issue, not the human cheating. The goal is to please the human, not to be kind with the device.
10-06-2015, 08:23 AM   #22
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by Nicolas06 Quote
Many people are able to see a 1:1000000 contrast in very dim light (looking at stars), that's approximately 20EV.
But not w/o clipping the highlights (stars).

I don't talk about cheating. Humans have wonderful capabilities. But a camera must be compared to the eye, not entire humans. In the end, it is still the same eye and human who will digest the image.

If you would indeed replace the entire human (say, by an autonomous robot with its own vision system), then the camera capabilities would already exceed that of humans by a large margin. However, the human would still excell at pattern recognition, resulting 3D reconstruction and fast adaption of its optical system to identified points of interest (attention-steering).

It does make sense to make a distinction between the capabilities of a human and his camera aka eye. Therefore, when I compare cameras, I do so against the human eye, not entire humans. Cameras do not replace humans, not normally at least.
10-06-2015, 09:43 AM   #23
Pentaxian




Join Date: Nov 2013
Posts: 4,540
QuoteOriginally posted by falconeye Quote
But not w/o clipping the highlights (stars).

I don't talk about cheating. Humans have wonderful capabilities. But a camera must be compared to the eye, not entire humans. In the end, it is still the same eye and human who will digest the image.

If you would indeed replace the entire human (say, by an autonomous robot with its own vision system), then the camera capabilities would already exceed that of humans by a large margin. However, the human would still excell at pattern recognition, resulting 3D reconstruction and fast adaption of its optical system to identified points of interest (attention-steering).

It does make sense to make a distinction between the capabilities of a human and his camera aka eye. Therefore, when I compare cameras, I do so against the human eye, not entire humans. Cameras do not replace humans, not normally at least.
I agree that I would not expect a camera to thing or even for that purpose that the picture here is cat and not a dog.

But what the purpose of a picture ? What do we make picture?

Some peoples might see it as a scientific tool, like analysing the spectral frequencies and deduce it's chemical composition from it or things like that.

I think that many people are interrested not in the camera, not how it capture the picture but in the actual result. The printed picture in their hand, the vacation book they can make, what they can share with their family and friend and see on a screen.

Choosing by design to change the reality like "improving" a face shape in portraiture can be a very good thing. As changing the colors and so on. Still another aspect is to give an image that the human understand, like and that match the original scene as seen.

White balance is a typical example. Yes our brain change the colors of thing and does it own write balence. But if you choose to ignore it, then most people will not want to buy your camera that doesn't manage white balance.

Sharpness is another example. When our eyes see in reality, they never get printing artifact or pixels except when staring at a print or screen. It reduce immersion. Our eyes have limited resolution but because they can focus to a small specific part, the actual resolution of the scene the brain can reconstruct is enormous. Because the camera take the picture (in general) in 1 single shoot, it has to have a sensor with much more resolution than the eyes. You can ignore this and make a camera with 1-2MP and say this is enough, people will not want it.

Now for the dynamic range, this is another issue. People don't like that they take a church picture and the windows are all white or the wall are all black. There a reason why more and more camera (and even phones) have an HDR mode.

You can say no we should compare camera sensors with eyes only... But that not how it is. Many photographers post process and don't hesitate to remove/change things to make the photos better. This is used at large scale in fashion and for famous people. On the other side, human will not disable their brain anymore to be more fair to your camera than people photographer will suddenly stop to remove all the imperfections of their models. In fact, even less: the photographer can stop doing it... I don't think many people can on purpose change the way they see things to be more fair to camera sensors.


Last edited by Nicolas06; 10-06-2015 at 09:55 AM.
10-07-2015, 07:34 PM   #24
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
@Nicolas06: I can only assume you completely misread what I said. So, let's stop it here.
10-08-2015, 03:29 AM - 1 Like   #25
Site Supporter
sculptormic's Avatar

Join Date: Dec 2012
Location: Amsterdam
Posts: 220
Now what I am interested in; did DxO Labs add The Pentax already to its scope? I am interested to try it out.
10-08-2015, 04:44 AM   #26
Veteran Member
Kolor-Pikker's Avatar

Join Date: Dec 2014
Posts: 333
QuoteOriginally posted by sculptormic Quote
Now what I am interested in; did DxO Labs add The Pentax already to its scope? I am interested to try it out.
Was about to post the same thing. What's the ETA on this information anyway?
10-08-2015, 07:45 AM   #27
Senior Member




Join Date: Mar 2015
Posts: 160
Once confirmed as out, is it worth buying their software?
10-08-2015, 12:04 PM   #28
Site Supporter
sculptormic's Avatar

Join Date: Dec 2012
Location: Amsterdam
Posts: 220
Some time ago, when I had a problem with my Adobe license - talking to people in India for days -, I used a trial for some time. I quiet liked it. It has most bells and whistles of LR in a different lay out, but without the for me terrible file system. That's why I use ACR instead of LR so I can do my own file system.
I think I do one more trial as soon it supports the lenses and decide afterwards.
10-08-2015, 01:06 PM - 2 Likes   #29
Site Supporter
Zygonyx's Avatar

Join Date: Jun 2011
Location: Ile de France
Posts: 2,951
Original Poster
Here we go :

PENTAX 645Z and the 5 Ricoh originating lenses are available for DxO OP 10.5.1 :

HD PENTAX-DA 645 28-45mm F4.5 ED AW SR
SMC PENTAX-D FA 645 55mm F2.8 AL [IF] SDM AW
SMC PENTAX-FA 645 75mm F2.8
HD PENTAX-D FA 645 90mm f/2.8 Macro ED AW SR
SMC PENTAX-FA 645 120mm f/4 Macro




As the corresponding tests were made last august ; the rest of lenses + 645D tested in september, should join within a few weeks.

Last edited by Zygonyx; 10-08-2015 at 01:21 PM.
10-08-2015, 01:39 PM   #30
Site Supporter
sculptormic's Avatar

Join Date: Dec 2012
Location: Amsterdam
Posts: 220
That is good news.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
645, 645d, 645z, camera, dxo, dxo lab, dxomark, f/2.8, f/4, f/4.5, fa, medium format, medium format camera, pentax, pentax news, pentax rumors, pentax-d, pentax-fa, phone, sensor captures images, site, smc, version
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
How DxO Labs tests hot cameras like Canon's latest SLR gawan Non-Pentax Cameras: Canon, Nikon, etc. 3 04-20-2012 07:36 PM
Does it make sense to add a DA* 300mm to my 70-200mm? HenrikDK Pentax SLR Lens Discussion 37 03-28-2012 05:37 AM
Even if $ is no real issue, it's not easy to replicate a Pentax system. pcarfan Pentax DSLR Discussion 86 08-14-2009 09:40 AM



All times are GMT -7. The time now is 03:14 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top