Jump to content

DigLloyd reviews M246


jrp

Recommended Posts

  • Replies 44
  • Created
  • Last Reply
Guest JonathanP

Its odd that someone who purports to be so methodical with their testing would write "I strongly suspect the damage is baked into the raw files I shot" without bothering to open the file in RawDigger to see.

 

Its less odd that someone who runs a subscription review site needs to find an alleged problem with a new camera to draw in new subscribers. Usual Lloyd click bate I'm afraid. It would have only taken a few seconds with RawDigger to confirm it, but that wouldn't have generated the drama queen headlines, would it?

Link to post
Share on other sites

I used to subscribe to Sean's reviews but no-flash really killed it for me as the best place to read his articles (for me) was on the way to work and back while on the train, which means iPad. I would renew in a heartbeat if he provided no-flash version

Link to post
Share on other sites

More samples on there now and you can see the problem.  I've not seem anything like it on any of my shots.  

Sometimes I have had a line of pixels (left jumper sleeve) when I've pushed the shadows up high but nothing like this.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

"UPDATE: I have confirmed with Brian Griffith of Iridient Digital that the spots are seen when converting with Iridient Developer. So unfortunately the black spots are “baked in” to the raw file."

Or "baked in" to the raw developer?

Link to post
Share on other sites

I can see a bit of "shooting the messenger" here. Something weird is going on, and I hope somebody finds out soon what is going on. BTW, has Leica given any comments on the 12 bits thing? Would be interesting to know the full story.

Link to post
Share on other sites

4:1 crops ........

 

can't find any issues with the rendering of patterned fabric or anything similar to what Lloyd posted ......

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

 

there is banding with extreme shadow recovery ...... but no more than you would expect 

Link to post
Share on other sites

I can see a bit of "shooting the messenger" here. Something weird is going on, and I hope somebody finds out soon what is going on. BTW, has Leica given any comments on the 12 bits thing? Would be interesting to know the full story.

It is indeed all too easy to shoot a reviewer with a not-spotless record of technical excellence, but in this case his examples are convincing.

Still, it  looks like a software error , whether caused by camera firmware or by raw

developer(s) remains unclear.

What is strange is that it seems that not all users can reproduce it.

That would point in another direction, a manifestation of a Moiré-like effect. 

 

The last would be easy to check: Does the effect disappear with slight defocus?

Link to post
Share on other sites

This is an unprocessed original exposed for the highlights .......

 

..... I ran out of steam on the other thread I posted and didn't show the shadow recovery properties of the camera .......

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

this is what you get if you increase exposure and +100 shadow recovery in LR ...... (this is a screenshot)

 

...... there is a rather wide horizontal banding ......

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

Guest JonathanP

Would be interesting to see if Lloyd tries again with uncompressed DNG. The way lossless DNG compression works is optimised for a 2x2 Bayer array, by stacking rows to make an image twice as wide and half as tall, in order to improve adjacent (top+bottom) pixel likeness for the compression. I could imagine that such an algorithm could expose a buggy implementation (either encode or decode). Several raw converters I have tried fail to decode M246 compressed DNGs (the image appears split almost into quadrants) but are fine on uncompressed.

 

I still think he's being over-dramatic as usual - less haste and more rigorous investigation of his findings before publishing would make his claims much more credible.

Link to post
Share on other sites

at 1:1 this doesn't seem so obvious ...... but vertical streaks are apparent ......

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

 

...... but this original is 12500 iso and the shadow areas almost black ....... no sane person would attempt this degree of shadow recovery from an already highly amplified sensor output and expect results with no compromising of the image quality ...... 

 

.... and again it is difficult to know just how much of this would appear in a print ....... 

Link to post
Share on other sites

thighslapper, you are right. near blacks boosted substantially are always ugly. What I would say is that when you must process in such a way that slight banding does appear, Nik Dfine does a good job of sorting it out: http://thephotofundamentalist.com/?p=1655

 

I have had to use this with my Canon files quite a bit, when photographing urban spaces at night, due to very deep shadows and a need to hold highlights.

Link to post
Share on other sites

The challenge with people like digilloyd is with the substantial number of pieces of equipment they revieiw and the time it takes to assess write, publish and so on. It leaves little time for photography, so why do it ? Well it's about add revenue right ? So you end up with what is for me a fundamental problem.  Limited time, poor alignment between reviewer and potential customer and uninteresting descriptive images. If you can download full size raw images and make your own judgement that is a much better option. I also value seeing lots of published images from the camera and whilst most love Mr Reid, I have subscribed twice and regretted it, interesting once, but there is only so many shots of overwieght people at summer fates mid America I can view.

 

Sorry for being grumpy, it must be Monday...

Link to post
Share on other sites

The latest Irridient Raw Developer seems to have some M246-related fixes.  Not sure whether the dots are one of them.

 

Meanwhile, Digilloyd declares B&Ws from the 50Mpx Canon to better the M246. http://diglloyd.com/blog/2015/20150620_1118-Canon5DSR-comments.html

 

(I'd have thought that something like the D810, which has the highest dynamic range sensor of full frame cameras would make the best b&w camera, since b&w is largely about tonality.)

Link to post
Share on other sites

The latest Irridient Raw Developer seems to have some M246-related fixes. Not sure whether the dots are one of them.

 

Meanwhile, Digilloyd declares B&Ws from the 50Mpx Canon to better the M246. http://diglloyd.com/blog/2015/20150620_1118-Canon5DSR-comments.html

 

(I'd have thought that something like the D810, which has the highest dynamic range sensor of full frame cameras would make the best b&w camera, since b&w is largely about tonality.)

At 950g without a lens he is welcome to it

Remember that Lloyd get clicks by declaring something controversial every few months.

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...