Re: Just when you started believe the low noise myth

DD wrote:
In article <YIpng.15645$ap3.10796@xxxxxxxxxxxxxxxxxxxxxxxxxx>, Dmac@xxxxxxxxxx says...

<snip>

Douglas, what exactly are you trying to accomplish with this latest post? What possible motives could any human being have to want to persist with such an inane interchange?

Who cares who's right or wrong? You might die tomorrow, or Bret might win the lottery. This will remain unresolved forever. You can't change that.

This sort of stuff is just not important. The group was designed to exchange ideas/information in a positive way. This crap doesn't do anything but waste your time and my (and others) bandwidth.

Maybe you missed something here Dallas.
The difference between ISO 100 and 400 is close enough to 2 stops. This is about the maximum latitude in a scene a Canon CMOS digital sensor can cope with before resorting to software enhancement to compensate for noise.

No doubt you are aware I have been a long time critic of the bull*** and poor quality quality control Canon are guilty of. I've bought 4, 20Ds since the day they were released. Only one focused properly and it had a lockup problem when you changed lenses. All the others produce noise in shadows at ISO 100 which 5Ds don't.

The image I posted was under exposed 2 stops but at ISO 100 - the same exposure as ISO 400. If you did this with a film camera, the print would be a darker to be sure but not have any more grain than a correctly exposed one.

The "average" exposure of this scene was actually the "shadow" average of a 20D shot at ISO 100. Forget my barb at Bret (He leaves himself open to them with his attitude) and concentrate on the content of the post.

This shot shows that 20D/30D cameras are indeed noisy and you don't need ISO 3200 to light up the noise. It's there in any scene exposed for the highlights that has more than 2 stops of variation in the shadows.

Someone recently posted in rec digital about the "noise issue" some journalist has all of a sudden discovered in 30Ds - basically a 20D Mk II. Strange, I've known about it in 20Ds for 2 years. Maybe his Canon advertising budget has stopped.

The 5D is much more advanced in noise control than the 20D. It is possible to recover 3 stop under exposed shots with relative ease. Not so with a 20D OR 30D. If it was just noise it wouldn't be so bad but the image picks up contrast and shifts colour too.

Probably because of the pixel density of the sensors. It's not all that far different to the noise issue these Canon journalists went on about with Olympus E300s. It only matters when it matters. The rest of the time it has no impact.

One of the real problems I though this post would highlight, is that reports from sources closely aligned with camera makers - read that as "don't bite the hand that feeds them" - are so distorted as to only present the very best possible report of a product.

Helen Silverburg suggested DPreview as some sort of reference for camera reviews. Their glowing review of 20Ds when Canon showed them favor and gave them pre-release cameras to "test" is but one example of how suspect their reports actually are.

Anyone owning a 20D or a 30D who consistently "exposes for highlights" will sooner or later get pictures like this. 18 months ago when I offered up the suggestion a Panasonic ZLR was better in these situations than a 20D, I got posts suggesting everything from hyperfocal errors to rigged pictures was the cause when it is the design of the sensor which is the real problem, as the FF sensor of the 5D clearly demonstrates.

--
From Douglas...
My photographic site: http://www.douglasjames.com.au
Canvas Archival and Metallic Prints: http://www.canvasphotos.com.au
.

Loading