A week plus with the SD1 (long.....very long)

Started Jul 7, 2011 | Discussions thread
Erik Magnuson Forum Pro • Posts: 12,247
Re: I believe this image is actually much better

NarrBL wrote:

  • browser color management. That's a good page, and my c-m browsers check out fine - Firefox 5.0 and Safari 5.0.5. Actually the problem is with Pbase - for any but original size, it clips the profile.

True, but most of us look at original size as the forum tradition dictates. Most of us know that you certainly should never make any judgements about an image from a pbase (or flickr or dpreview) resize. You jumped to the conclusion that since the OP saw something you did not, he must have been looking at it improperly. I pointed out that is not necessarily the case. (Did you ask if he was looking at original size and uses a profiled/managed system?)

  • mold vs. noise. You sound pretty abrupt on all this, and I am not interested in that form of back and forth, but let me answer what I see anyway. Actually, again, color tinting doesn't show until one has artificially raised the saturation to the point where the whole image is beginning to bloom colors.

The color tinting makes it blatent. As with any artifact, you either see or you don't (or you learn how to see it.) I'm always amazed at how quickly some of the participants here can spot "nutella" a mile away in a resized mosaic image

And guess what: these color patterns do not match the lumpy spots of what I am still thinking is cold wet climate outdoor molds.

I agree they do not. The question is which artifact did the OP notice/comment on? If he saw what I 'm referring to, then the mold is a red herring. The mold did not register to me as looking anything like noise so I paid it no attention.

Lots of words, I know. But I am pretty sure you will find the same thing if you are a bit quiet and careful.

We shall agree to disagree then.

  • now, what is this about SD1 assumed to be perfect under extreme abuse of its images?

Where does extreme abuse enter into it? Both the OP and I noticed something in the image as presented (even correctly profiled for my case). Since you did not see it, I attempted to highlight it for you. If you look, you should be able to see the noise under typical pixel-peeping conditions of simple 100% view. If you say it's too mild to be relevant, then that's your subjective opinion on it's pictorial impact. My point was there is something non-mold to be seen. Whether it's a issue of significance or not is a completely different subject.

Erik, you of all people have some sense of the physics at the root of this and other sensors. That physics says that all things being equal, smaller pixels will have more individual noise.

In the real world, smaller pixels and newer sensors almost always have lower read noise than larger pixels. More, but smaller, pixels can detect both finer real detail and finer shot noise. But the SD1 pixel pitch is not that small where you would expect to see an impact like this and it would not occur in just those areas.

With present smaller pixels, firmware or software image processing becomes more important,

I've often stated that it's always been important - perhaps more so than pixel design.

In other words, again as you well know, there is not Platonic perfection in any digital photography;

Exactly. That is why I found your dismissal of the issue overhasty and attempted to point out what I suspected the OP was talking about. It certainly seems that SPP still needs some tuning to get the right balance of noise/detail/sharpening. That doesn't make the SD1 a bad camera - just one that currently needs a bit more care in use and processing.

-- hide signature --


Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow