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!

Unscientific but interesting: AR7, RX1, D800, M240

fmueller

Active member
I am a bit taken aback by how poorly the Sigma lens is suited for a landscape photo. I presume that I am seeing a field curvature simply not present in the other two lenses. The other three combination are vastly preferable from a focus point of view.

That is my non expert opinion from a pixel peeping session with the downloaded jpegs...
 

CharlesK

New member
Very interesting test! IMO, I like the RX1 the best closely followed by the M240 and A7r with the 35 Lux FLE. There is more detail with the A7r, but the 35 Lux FLE is very useable. Maybe the difference would be less if the 35 Cron Aph was used:) I am surprised by the results with the D800/Sigma.
 

cunim

Well-known member
Thanks, Tim. Ideal shots for evaluation. Now I am really confused.

I want something to take pictures that approach MF quality, but in a smaller image. For the past year I have been carrying an APSC camera in my briefcase - just for quickies. I have found the small format, even with good lenses, ... difficult and disappointing. My fault, I'm sure.

Aha, methinks! Get a full frame A7r, put a nice 50 on it and I may actually like a few of the briefcase shots.

The A7r is coming this week, and I will probably just stick the Sony/Zeiss 55mm on it. It will be portable and that is its primary application. However, looking at your shots and the others springing up on the web, I am probably not going to be much happier with the PQ than I was before. Sigh.

Peter
 

Taylor Sherman

New member
Just looking at the web-sized shots, I actually liked the A7 shot the best, thinking it was probably the RX-1! But, it's obviously super subjective. The D800 is just washed out, and the M shot is too dark.

One thing, of course, is clear: the M has more birds!
 

Guy Mancuso

Administrator, Instructor
There really should be no difference in a small change of color temp to a file if you go in post with a WB card to correct. Noticed I said small, big changes like shooting tungsten in daylight or vise versa can have a big impact trying to correct in post to a WB card. Reason is some separate color data in the file is just to far a stretch in lost color to correct. Shooting tungsten for instance is heavily weighed on the red channel which can have a impact on your blue channel for instance. Rule of thumb never use AWB as the camera will always be influenced by the scene color itself. In daylight shoot to a daylight preset than do your adjustments in post. Back to the small change comment you simply do not want big swings in color data. Also in a test like this its best to include a WB card in the shot as to get the color correct in post for evaluation. ACR changing the color profile in post can have a big effect. I'm not a fan of ACR or LR never was because you can't use a ICC profile. When I got the D800 out of the gate I had a really hard time getting color correct in these programs and Nikons own software. Finally when C1 supported it than life got a lot better as C1 uses ICC profiles. This thing I won't buy it until C1 supports it. That's me though

BTW Tim did say unscientific and it truly is without a WB card in the scene you have no idea what correct really is when it comes to these kinds of evaluations. Color will be all over the place with raw converters. You can take that Nikon file in four converters and have 4 radical changes in each without a standard to go by, than you comparing raw converters itself.
 

fotografz

Well-known member
Thanks Tim, I'm evaluating a Demo M240 right now and these are helpful (even though I do not shoot many landscapes).

It confirms what I found in a controlled "quasi-scientific" sort of test using all constant elements. The M240 exhibits some IR contamination (somewhere between the M8 and M9 as far as I can determine so far) ... see attached test with M240 in Tungsten light with, and without, an IR filter on the lens). In discussions on the LUF I have come to understand why a weaker IR filter was used on the M240. However, it is there.

It is also apparent in your real world images in the sky, (see crops from your samples of the M240 sky, which reveals the magenta cast when placed next to the A7R sky, which is relatively neutral. This may also account for the slightly more vibrant fall colors (only confirmed IF the shot was done with and without an IR filter) ... maybe okay for landscapes, perhaps not so good for other applications.

I agree with Taylor that the A7R has been over sharpened ... someone suggested rather than LR RC5.3, it may be better to use the latest DNG converter and then process A7R RAW files in LR4.4. Since my computer is older, I can't install LR5, so I used the DNG converter for the A7R files I shot, and can say I didn't witness such brutal sharpening artifacts. Not sure that proves anything ... it would require A RAW file be processed both ways to really tell.

- Marc
 

tashley

Subscriber Member
Just looking at the web-sized shots, I actually liked the A7 shot the best, thinking it was probably the RX-1! But, it's obviously super subjective. The D800 is just washed out, and the M shot is too dark.

One thing, of course, is clear: the M has more birds!
D800 files always look washed out unless you give them some curves... They are really raw raw. Sharpening had to be the same to be fair to all but the rc5.3 doesn't 'do' 7r files well, whatever the sharpening!
 

tashley

Subscriber Member
Thanks Tim, I'm evaluating a Demo M240 right now and these are helpful (even though I do not shoot many landscapes).

It confirms what I found in a controlled "quasi-scientific" sort of test using all constant elements. The M240 exhibits some IR contamination (somewhere between the M8 and M9 as far as I can determine so far) ... see attached test with M240 in Tungsten light with, and without, an IR filter on the lens). In discussions on the LUF I have come to understand why a weaker IR filter was used on the M240. However, it is there.

It is also apparent in your real world images in the sky, (see crops from your samples of the M240 sky, which reveals the magenta cast when placed next to the A7R sky, which is relatively neutral. This may also account for the slightly more vibrant fall colors (only confirmed IF the shot was done with and without an IR filter) ... maybe okay for landscapes, perhaps not so good for other applications.

I agree with Taylor that the A7R has been over sharpened ... someone suggested rather than LR RC5.3, it may be better to use the latest DNG converter and then process A7R RAW files in LR4.4. Since my computer is older, I can't install LR5, so I used the DNG converter for the A7R files I shot, and can say I didn't witness such brutal sharpening artifacts. Not sure that proves anything ... it would require A RAW file be processed both ways to really tell.

- Marc
Marc, I honestly think that if you want to use mainly M glass, the M240 is the better choice though I noticed,when I reviewed it, the same IR contamination. Trouble is, with wider apertures, the 35fle and other Leica wides I have tried on the A7r, the colour shading is the worse of the two evils...
 

tashley

Subscriber Member
Thanks, Tim. Ideal shots for evaluation. Now I am really confused.

I want something to take pictures that approach MF quality, but in a smaller image. For the past year I have been carrying an APSC camera in my briefcase - just for quickies. I have found the small format, even with good lenses, ... difficult and disappointing. My fault, I'm sure.

Aha, methinks! Get a full frame A7r, put a nice 50 on it and I may actually like a few of the briefcase shots.

The A7r is coming this week, and I will probably just stick the Sony/Zeiss 55mm on it. It will be portable and that is its primary application. However, looking at your shots and the others springing up on the web, I am probably not going to be much happier with the PQ than I was before. Sigh.

Peter
Peter, don't judge it until you see it shot with a native FE lens and a better raw developer... If you want close to MF quality in a briefcase, the a7r and the Rx1 are the only games in town IMHO.
 

tashley

Subscriber Member
Right. After yesterday's discussion about WB, I decided to run a quick test.

My historical feeling and therefore practice has been that for cameras with colour shading corrections turned off, it does not matter what WB you shoot if you are shooting RAW because the camera captures all the spectra available to it and the balancing of those spectra is a decision as to how that captured data is later marshalled and presented. In other words, you could leave your camera on Fluorescent for a day at the beach as long as you corrected the WB in your raw developer.

Others felt that that the setting of WB at capture stage, even for a RAW file, would affect what was captured such that later attempts at balance would be stymied. This had never occurred to me so I thought I would test it.

The linked shots are on an a7R with 35 Lux FLE at F8. I shot them with Fluorescent (setting '0'), Incandescent, Cloudy, Shade and Daylight setting on camera, then shot a reference frame with a WB card and then in LR RC5.3 I applied to BW from that frame to all the RAW files. here they come, first the 'as shots' in the order above, and then all files corrected to custom WB as captured. Of course, if this proves anything it relates only to one camera, does not take account of what might happen if shading corrections are turned on with a relevant lens, and doesn't test for whether metering and (therefore exposure) might differ with different on-camera WB settings. It also doesn't test for what might happen if I did the same thing with a strongly biased light source such as a lightbulb. All are done with adobe RGB throughout the process.









 

Floyd Davidson

New member
There really should be no difference in a small change of color temp to a file if you go in post with a WB card to correct. Noticed I said small, big changes like shooting tungsten in daylight or vise versa can have a big impact trying to correct in post to a WB card.
When shooting in RAW it makes absolutely no difference at all. None. Zilch. Zip.

The raw sensor data is exactly the same regardless of how the White Balance is set. And when processed the parameters for converter configuration can be set to exactly the same as would have been set in camera for each or any of the different possible camera White Balance settings.

As for the images, I don't see any differences that cannot be equalized with proper configuration of the RAW converter. There is no reason to assume that the camera defaults and the converter defaults are either optimal or that they should be the same. Each camera has a different built in tone curve, and each converter has different curves for each camera. The defaults are merely what either engineering or marketing decided would suit their market best. That is not necessariy a match for the audience here! And note too that major camera manufacturers use different tone curves for different markets, and hence the same model sold in Germany will produced a different default image than one sold in Japan.

The whole point of a test should be to see what the "best" is that can be done with each camera/converter, not a comparison of the defaults. For example, using the same amount of sharpening on a D800 with 36MP as is used on a camera with say 14MP is a gross error. Incidentally, using the same sharpening on images from a D800E and a D800 would also be invalid for comparative purposes.
 

tashley

Subscriber Member
Now the same images re-WB in Post to the setting determined by a WB reference shot:











Many eyes are better than one. What do people think? Again, a link to larger processed files is here.
 

Ron Pfister

Member
Comparing the low-res images on my un-calibrated, low-gamut 11" MBA display, I'm tempted to say they all look strikingly similar. ;)
 

Floyd Davidson

New member
My historical feeling and therefore practice has been that for cameras with colour shading corrections turned off, it does not matter what WB you shoot if you are shooting RAW because the camera captures all the spectra available to it and the balancing of those spectra is a decision as to how that captured data is later marshalled and presented. In other words, you could leave your camera on Fluorescent for a day at the beach as long as you corrected the WB in your raw developer.

Others felt that that the setting of WB at capture stage, even for a RAW file, would affect what was captured such that later attempts at balance would be stymied. This had never occurred to me so I thought I would test it.
I can provide a proof positive that the RAW file is not affected by the WB setting on the camera at the time the image is shot.

Dave Coffin is the author of DCRAW, a basic raw converter which he updates regularly to include virtually every new model as it is produced. DCRAW is Open Source Software, and is the basis for many other raw converters. It happens that I use UFRAW, written by Udi Fuchs, which is based on DCRAW. The source code for both programs is freely available.

The significance is that the WB adjustments in the UFRAW menu specifically agree with the WB adjustments available in the camera. The author specifically will list only the exact same configurations as the camera has! And the exact same RGB channel multipliers are used too. That is, if you have a Nikon D800 and set the in camera White Balance to "Daylight" the channel multipliers used to generate a JPEG for previewing will be recorded in the Exif metadata. If the NEF file is then processed by UFRAW with the WB set to "Daylight" it will use exactly the same channel multipliers as the camera does. (I've personally recorded and forwarded the appropriate data on multipliers to the author of UFRAW for several Nikon models.) It is also very easy to generate a "preset" with the camera and then add the same values to UFRAW and recompile it to have that preset on the menu.

It is very clear, from the perspective of writing software for RAW conversions, that the on camera WB setting affects only the JPEG engine and has exactly zero effect on the RAW data that is recorded.
 

jonoslack

Active member
Hi Tim
They seem to me to be the same - but actually that isn't quite what I meant. I would expect them to be the same under those circumstances - because you're applying a white balance setting you've established at the scene.

My point is that if you shoot away with AWB . . . and then apply the LR daylight setting in post processing - not the one you've established to be 'correct' for the scene.

What happens if you just apply the LR standard 'daylight' setting to all of those different shots? - or indeed any of the other LR standards - do they still all come out the same?

But even if they do, that STILL isn't my point - which is that if you have a level playing field when you shoot (by picking a standard WB) then it's much easier to deal with the colour (if needed) in post processing than it is if the WB is all over the place because you've set AWB when shooting.


all the best
Jono
 

tashley

Subscriber Member
They all look the same to me too. I think we were talking at cross purposes, though I am still therefore slightly unsure as to what Guy means above. I have now converted them all the LR Daylight preset and uploaded them to the same gallery - let me know if they seem the same to you, they do to me.

So at least in something like normal daylight and on this camera, it is as I suspected, that the WB you shoot them with makes no difference though it may give you a quicker start in post if you get it close to right at capture, all the data you need is in the file whatever WB setting you shoot with.
One thing though: the file that I shot with a Daylight preset, when developed with As Shot, looks a little different to how it looks if I apply LR's Daylight profile. This seems to be a difference of opinion as to what is the 'right' daylight and given that it is a matter of opinion, because daylight is a moveable feast, may be neither right nor wrong depending on circumstance but in this case, the LR idea of Daylight is warmer than the in-camera idea and is also warmer than the calibrated shot. In this case it would be easy to know what the camera's default daylight setting is and merely apply it to all files as relevant...

I am going to shoot a lightbulb scene with all WB settings and then correct them all in post to both incandescent and to a Wibal calibration shot and see what happens. This is interesting!

Hi Tim
They seem to me to be the same - but actually that isn't quite what I meant. I would expect them to be the same under those circumstances - because you're applying a white balance setting you've established at the scene.

My point is that if you shoot away with AWB . . . and then apply the LR daylight setting in post processing - not the one you've established to be 'correct' for the scene.

What happens if you just apply the LR standard 'daylight' setting to all of those different shots? - or indeed any of the other LR standards - do they still all come out the same?

But even if they do, that STILL isn't my point - which is that if you have a level playing field when you shoot (by picking a standard WB) then it's much easier to deal with the colour (if needed) in post processing than it is if the WB is all over the place because you've set AWB when shooting.


all the best
Jono
 

Floyd Davidson

New member
But even if they do, that STILL isn't my point - which is that if you have a level playing field when you shoot (by picking a standard WB) then it's much easier to deal with the colour (if needed) in post processing than it is if the WB is all over the place because you've set AWB when shooting.
The raw sensor data that is recorded in a RAW file format is unaffected by the camera's WB setting. As long as you are post prossessing the RAW file (as opposed to a camera generated JPEG or TIFF file) you are starting with a "level playing field".
 
Top