Reading mode: Light Dark

C2000Z versus C2020Z Noise

  • Thread starter Thread starter Barry Carter
  • Start date Start date
Go to last post
Jack A. Zucker said:
Member said:
Let's try the link anain, don't know why it didn't work the first time!
Think it's case sensitivity on Unix.
Lin



Click to expand...
I'm still having problems seeing Noise.jpg and Filtered.jpg seems to
still have a significant amount of noise present.

What settings did you use for the lowercase excerpts ?

Jaz
Click to expand...
Jack,

I'm not sure that what you are seeing is all noise. There are artifacts of compression in the shot due to my high compression, but if you will copy both the Filter.jpg and your original into a single sub directory and view them with QimagePro's Comparitor feature, you can pull up the pictures side by side and compare apples and apples. Look at the red areas of the rose itself on both sides under magnification and there is an astounding difference in the amount of noise. I can't be sure about the consistency of the green backdrop and whether or not the somewhat differentiated areas between dark and light are caused by digital noise or normal variations which would not be seen except under very high magnification. Without examining the backdrop itself, it's difficult to know exactly, but by using the comparitor you can see the extensive noise reduction. We have examined side by side identical pictures from the D1 and C2500L which is at least as noisy or moreso than your C2020Z. The post filtered C2500L has less noise than the original D1 which has less noise than anything else we've tested. You might ask if Mike still has some of the original side by side shots. It's not perfect, but it's many times better than the original. Print them out and see the differences there. I'm looking at them on a 21" color corrected and calibrated monitor and the differences are astounding. Hope you can see the blow up (noise.jpg) and see the differences.

Lin
 
Lin Evans said:
Barry, Jack, Noel, et al,

The Shadow Noise and Dark Current filters work perfectly.
Click to expand...
Whoa, Lin, that's a tough one to swallow whole...

The filters indeed do remove noise. They do a marvelous job of it. They also remove a significant amount of image detail.

I have found at least 4 techniques so far to remove image noise, and I own software for three of them (only Quantum Mechanic remains out of my arsenal). The three I DO have are Qimage, DigiCam Pal (under development), and Photoshop. In EVERY case, the filters do their job, but also degrade the image in other ways. Actually, I like Photoshop's noise filter the best.

The image degradation introduced by these smart noise filters, to my eye at least, shows up as blotchiness where the eye expects to see detail, kind of as if a fine mist of water were sprayed on an inkjet printout, blurring some spots. It makes the image look like one from one of the older digital cameras to me.

As an example, apply the Qimage filter to a picture of a cat with fine fur. The fur will get blotchy. In another case, I applied Qimage to an image of a small model car on a formica desktop surface. It not only destroys the noise, but makes the desktop surface look curiously smooth with occasional divots.

You'd think from the above rants, noise bothers me terribly. In fact, generally speaking, it doesn't. Frankly, when compared with film, the modern 2+ MPixel cameras do an absolutely MARVELOUS job, and I would probably never have noticed the noise much at all but for the hype I've been reading.

Noel
 
Noel Carboni said:
You'd think from the above rants, noise bothers me terribly. In fact,
generally speaking, it doesn't. Frankly, when compared with film, the
modern 2+ MPixel cameras do an absolutely MARVELOUS job, and I would
probably never have noticed the noise much at all but for the hype I've
been reading.

Noel
Click to expand...
Nor does it me. I agree the cameras are marvelous. My problem is trying to decide if I have received a defective (or below average) one.

Thanks to all that have reported tests on their cameras. I hope others will do the same.

Gerry
 
I tried the same test using my C2020z. My results are that it certainly has a few pixels that are visible but nothing like the C2020 comparison photo. I can count the visible pixels on one hand. Using PSP 3, the histogram looks similar to the one on your C2000 photo but there is no data as your version has. I guess there is some variation in these cameras.
 
Noel Carboni said:
Lin Evans said:
Barry, Jack, Noel, et al,

The Shadow Noise and Dark Current filters work perfectly.
Click to expand...
Whoa, Lin, that's a tough one to swallow whole...

The filters indeed do remove noise. They do a marvelous job of it. They
also remove a significant amount of image detail.

I have found at least 4 techniques so far to remove image noise, and I
own software for three of them (only Quantum Mechanic remains out of my
arsenal). The three I DO have are Qimage, DigiCam Pal (under
development), and Photoshop. In EVERY case, the filters do their job,
but also degrade the image in other ways. Actually, I like Photoshop's
noise filter the best.

The image degradation introduced by these smart noise filters, to my eye
at least, shows up as blotchiness where the eye expects to see detail,
kind of as if a fine mist of water were sprayed on an inkjet printout,
blurring some spots. It makes the image look like one from one of the
older digital cameras to me.

As an example, apply the Qimage filter to a picture of a cat with fine
fur. The fur will get blotchy. In another case, I applied Qimage to an
image of a small model car on a formica desktop surface. It not only
destroys the noise, but makes the desktop surface look curiously smooth
with occasional divots.

You'd think from the above rants, noise bothers me terribly. In fact,
generally speaking, it doesn't. Frankly, when compared with film, the
modern 2+ MPixel cameras do an absolutely MARVELOUS job, and I would
probably never have noticed the noise much at all but for the hype I've
been reading.

Noel
Click to expand...
Ah Noel,

I see you have discovered the Achilles heel in my enthusiastic endorsement of QimagePro. Of course nothing works "perfectly" in either software or hardware, but to my eyes, the QimagePro noise filter still works better than anything else I've seen or tried. Unfortunately, I haven't had the same experience as you with PhotoShop (ver 5.0.2) and my noise filtering experiments in PhotoShop have been met with serious image quality degredation. I too usually am not bothered by the noise levels, except where they disrupt an otherwise clear blue sky. I sometimes mask the sky and save the mask as a separate file. Then I run the mask through Qimage and re-apply it to the original with the noise from the sky gone and everything else left in its pristine condition. I'm a little surprised that with your original lack of concern for noise that you have investigated three separate programs to remedy the pesky non-problem? I don't find the amount of image detail lost to noise filtering in Qimage a serious problem unless one were to examine it under intense magnification. In my printouts at high resolution, the images do look better after filtering, but then "better" is certainly a subjective impression. I think my original post was more prompted by the apparent intensity of noise level concerns which I too feel are highly over-rated. It's kind of "let's get over the noise concerns", and get on with more important issues like advancements in optics and improvements in electronics to make consumer and prosumer level cameras more like the D1 and large Kodaks. I do think that others on the forum who aren't as fortunate as you and me to own PhotoShop or other expensive digital darkroom processing software should be made aware that tools like QimagePro are available for a very nominal fee and will allow them to make beautiful prints, filtered to their personal tastes. I suppose that's why I take the opportunity, whenever it's presented, to promote such products. Sometimes those of us who frequent the forums forget that there are many new visitors who have yet to hear that they don't necessarily have to spend more for software than they did for that new digital toy which their wives or husbands will be reminding them about for the next few years. :-))

Lin
 
Lin, I can't find fault with any of your points. Well said. I, too, am impressed that inexpensive software is available to do the job.
Member said:
I'm a little
surprised that with your original lack of concern for noise that you have
investigated three separate programs to remedy the pesky non-problem
Click to expand...
It's the curious engineer in me. I've actually corrected noise in only several of the many hundreds of shots I've taken, but I have a keen interest in the digital aspects of the hobby and I've really tried to just see what's out there. If I had a real need, I probably would have bought Quantum Mechanic.

I don't know whether I'm getting more used to my camera, or it's breaking in, or what, but I just took about 40 pictures of my son playing basketball, and all but one (where a boy ran in front of me right when the shutter opened) are keepers! Noise appears to be at an all time low (the shots were outdoors on a bright day).

Noel
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top