Jump to content

The Sony A7 thread [Merged]


dmclalla

Recommended Posts

Advertisement (gone after registration)

That's how the M9 uses it, as quoted from to the manual.

 

pg 160 M9 manual

Working With DNG RAW DATA

 

If you have selected the standardized and futureproof

dNG (digital Negative) format, you require highly

specialized software to convert the saved raw data

into optimum quality, for example the professional

Photoshop® Lightroom® raw data converters from

Adobe®. It provides quality optimized algorithms for

digital color processing, delivering exceptionally low

noise photographs with incredible resolution.

during editing, you have the option of adjusting

parameters such as white balance, noise reduction,

gradation, sharpness etc. to achieve an optimum

image quality.

Link to post
Share on other sites

x
  • Replies 4.9k
  • Created
  • Last Reply

My recollection (if you do a bit of searching on "coding" in the M9 threads), is that the Leica firmware applies a lot of correction to the raw file for wides - colour shift, vignetting and distortion. That's the Leica approach - apply as much correction in camera as necessary to provide a relatively uniform raw file, regardless which Leica lens you use.

 

That's rather the point of the A7 cameras - no correction in camera - just raw (with the exception of white balance). That's been the challenge, and also why we're getting odd results with different Leica lenses. The fatal issue for me will not be pixel peeping the edges, but distracting smearing in normal use.

 

So far, with my lenses,it hasn't been a problem, but to be honest the year end has been hectic and I have hardly taken a photo in weeks (Monochrom and Noct carried from pillar to post, unused).

 

So far, the camera has performed okay. I was considering ditching it, but felt I had not really explored it properly. I don't have an M(240) and have no intention of getting one; this was always a small addition to the cameras I have. So, I will play some more, and I will perhaps try one of the new Zeiss lenses next year - a ZE mount Otus would be appealing, or an Otus style 28?

 

The idea, however, of a fine sensor, producing pure raw files, for correction in post, is interesting (and I suspect, the future, as it provides a level, stable platform for any lens). The difficulty seems to be that the CMOSIS sensor handles the edges better. This surprises me, as I had thought that Sony would do a better job.

 

My task is to see which of my lenses are unworkable - the Distagon 2.8/15 is the leading candidate, at this stage.

 

Cheers

John

 

Sent from my iPad using Tapatalk HD

Link to post
Share on other sites

I made the point that your example didn't make the point very well if at all. Certainly not to me. Perhaps the downsized jpegs don't allow me to see the problem.

 

If the Summilux is all you are setting out to deal with there are still plenty of problems with the A7r sensor that your point doesn't cover. The extensive smearing and lower resolution in the corners.

 

If you were interested in that point and the given example, you would have seen the hint directly under the image saying that you can click in the image to see other resolutions (including full size).

 

The question adressed there has nothing to do with the sensor although it is obvious that here are several people whishing so.

 

I recall examples with the voigtlander 21/1.8 which showed similar curvature, which it apparently doesn't show on film, so it's a ' feature' of the camera?

 

Quite strong field curvature of the Ultron 21/1.8 was also reported by Leica M users.

Link to post
Share on other sites

Advertisement (gone after registration)

Thanks John.

 

Isn't one difference between the approaches of the two companies which problem they consider to be worse?

Corner smearing versus residual infrared contamination?

Apparently one cannot avoid both completely for all M lenses.

 

BTW I don't mind at all being corrected if I state something that isn't correct.

Actually I thrive on correct factual feedback.

Thanks.

Link to post
Share on other sites

Ummm, I think so...

 

I wouldn't want any camera's "processor" be it Nikon, Sony, or Leica, to apply anything to a RAW file. It's not really RAW at that point. I want to take it out of the camera and look at it. Not to be blatantly argumentative, but I don't want an effing "smile sensor" either.

 

My Nikon's NEF's aren't corrected by the camera, but they do have the EXIF lens info inserted into the file when I shoot with an AF lens or an unchipped Ai or Ai-S lens. Programming an adapted lens choice into the menu is something Sony cameras don't have the ability to do. They SHOULD be able to do at least that, with the earlier mentioned, semi worthless, lens compensation APP. Snip....

 

Are you sure? I have always assumed that the NEF files I get are corrected for chromatic abberation and light fall off at least. I don't believe the motley collection of nikon fit lenses I have produces such marvellous results any other way.

 

Gerry

 

Sent from my GT-I9505 using Tapatalk

Link to post
Share on other sites

Well my understanding is that ever since the M8 leica have had to manipulate the RAW/DNG files in some way otherwise why would we have coded lenses which give info to the camera so that corrections can be added, also why all the firmware upgrades to try and improve the corrections for various lens issues. Surely we all take it for granted

Link to post
Share on other sites

Are you sure? I have always assumed that the NEF files I get are corrected for chromatic abberation and light fall off at least. I don't believe the motley collection of nikon fit lenses I have produces such marvellous results any other way.

 

Gerry

 

Sent from my GT-I9505 using Tapatalk

Nikon is well known for applying (very good) noise reduction to their raw files.
Link to post
Share on other sites

Thanks John.

 

Isn't one difference between the approaches of the two companies which problem they consider to be worse?

Corner smearing versus residual infrared contamination?

Apparently one cannot avoid both completely for all M lenses.

 

BTW I don't mind at all being corrected if I state something that isn't correct.

Actually I thrive on correct factual feedback.

Thanks.

 

True. I don't really think m lens can be improved over wha the M9, Monochrom and M(240) can achieve. The interesting thing will be Zeiss ZE lenses on the A7r. My plan is only to get one, so I want to make sure it's a good one.

 

 

Sent from my iPad using Tapatalk HD

Link to post
Share on other sites

None of which gives any hint on how much processing has been done in the camera.

 

Err…no.

 

Nikon is well known for applying (very good) noise reduction to their raw files.

 

 

who's talking about noise reduction?

 

Go on...

Link to post
Share on other sites


To be quite clear:

The DNG format allows cameras to store the raw - unprocessed - image data, along with instructions to any downstream software on how to enhance the images. The same format also allows the camera to store processed image data. It may be worth knowing that the DNG format is merely an application (or "extension") of the well known TIFF.

This does not imply that any cameras will actually store raw data. The format just allows the cameras to do so.

Leica state nowhere that I'm aware of that the DNG files written by any of the digital M cameras contain raw image data. In fact, they seem to carefully avoid doing so. Several discussions about imaging problems are worded as if the image corrections were applied in-camera (as opposed to "hinted at" in some instructions to the downstream processing entities).

One possible reason for doing the processing in-camera and not downstream is, of course and strictly IMO, the ease of reproducing the possibly tricky post-processing by third parties if it's specified in plain view within each image file.

It seems that one of the earlier digital Ms (some copies of the M8, IIRC) even boasted an option to save the image in a RAW format, in addition to the then provided DNG format. I can't recall the statistics, but they were briefly discussed in one of the subforums here.

Hence: There's no evidence which supports the notion that Leica's DNG format as written by the digital M cameras contains raw image data; to the contrary, there's every reason to assume that the image data in the DNG files is subject to processing within the camera.

I don't think that continuing this debate is fruitful without citing any new sources which directly support one or the other side of the debate. Merely pointing out capabilities of a very flexible file format will not help, I think.
Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...