a7S high ISO claim - - -over hyped or not ?

Started 5 months ago | Discussions thread
ProfHankD
Senior MemberPosts: 2,126Gear list
Like?
5DIII banding problem may be easily fixable
In reply to Rishi Sanyal, 5 months ago

Rishi Sanyal wrote:

andrewD2 wrote:

DXO marks DR results are hopeless, a 5DIII still registers some info at a similar DR to a 6D but the banding of the 5DIII makes the last few stops of DR of very poor quality. Some of the Sony sensors show good detail but the limitation is colour accuracy.

I wouldn't call them hopeless, as they correlate quite well to usability of shadow detail in RAW files. For example, according to their data, base ISO DR goes as follows: a7R > a7S >> 5DIII. Which correlates exactly with my real-world experience. You do bring up a good point about banding, which make the 5DIII shadows even less useful. That's a tough problem to tackle, though.

It's THE problem.

No. It's a problem. And, trust me, not one that is lost on me, given that I use a Canon 5DIII for my people/wedding photography. Sony for everything else.

I don't have a 5DIII, but I have had a lot of webcams that have amazingly bad banding issues -- and I've automatically fixed them in custom software quite effectively, even for webcams where the banding pattern dynamically changed from frame to frame (i.e., was caused by noise from the camera's processor corrupting the analog data). Basically, computationally recognizing and removing a banding pattern imposed on an image is easier than removing truly random noise.

Can you give some examples (ideally including raws) of the 5DIII banding problem?

If this is really a big deal, I might be willing to create free software to fix it as a little research project... sort of like I did by building DeOrbIt to automatically replace Fuji X10 "white orbs" with more credible structures (although in that case, the fixes are completely synthetic, created by recognizing the orbs, synthesizing a credible texture for the missing image data, and then imposing a more appealing highlight). Here's a little sample of before/after using DeOrbIt:

Fuji X10 "white orbs" before and after using DeOrbIt

Not perfect, I know, but certainly more usable... and this is a lot harder than removing banding usually is....

Back on the original topic: in general, "noise" with a distinctive pattern can generally be recognized and removed, so it really shouldn't count against the dynamic range just because Canon (or whoever else provided your firmware/software) was too stupid/lazy to fix it.  

 ProfHankD's gear list:ProfHankD's gear list
Canon PowerShot A640 Canon PowerShot A720 IS Canon PowerShot S70 Canon PowerShot G1 Canon PowerShot G5 +21 more
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow