Go Back   Steve's Digicams Forums > Digital Cameras (Point and Shoot) > Konica Minolta

Reply
 
Thread Tools Search this Thread
Old Nov 6, 2002, 9:18 AM   #91
Senior Member
 
Join Date: Jun 2002
Posts: 161
Default

Quote:
Originally Posted by fotograafdigi
:? Why in the world would I care how these (or any other cam's images) look?????............

My equipent works to my satisfaction . I have no need to look into what I dont have or need, let alone in what I dont want to have.

Well, this is my problem. I would love to contribute. But I haven't done comparisons. What I have done is blow up Minolta images to large dimensions, far larger than was ever recommended, print them out and get very good results. Now, in this http://marksquires.com/images/PICT0020_boatparty.jpg photo, shot in RAW and max everything, there is some noise perhaps--esp in the roof. But keep in mind too that I took this image and printed it out in a 16 x 20 poster, a size that often bothered my print camera at times, and it looked darned good. This jpg image had no photo processing---what you see is what I got. Smaller prints have been exceptional.

Given the price, features, results---I'm extremely happy, and I'm extremely glad NHL helped me along the way to understand the features of the camera.
monsieurms is offline   Reply With Quote
Old Nov 6, 2002, 2:40 PM   #92
Senior Member
 
Join Date: Oct 2002
Posts: 163
Default

Based on all I've seen and read so far, here are my provisional conclusions about this issue.

1. Even though the Sony f707/717, Nikon 5700, Olympus E20 and the Minolta 7/7i/7Hi all use some version of the same CCD chip, that doesn't imply that they all have the same amount of noise prior to filtering/processing. Differences in a number of factors, such as lenses, internal IR filters, color filter arrays, power supply, and other electronic componenents do/might affect the amount of noise as well as how noticeable a pattern it forms.

2. Noise is more apparent in the shots taken with the 7x than those taken with its competitors when in similar conditions and with similar camera settings. This appears not to be due to greater noise removal processing in the other cameras, at least not entirely, since this is also reflected in the images shot in raw.

3. The level and pattern of noise in 7x shots is usually not a problem. Generally, good 8x10s and even larger prints can be made without post-processing, though those who have less tolerance for noise generally do use post processing. The noise will be more of a problem for those who (like me) intend to crop heavily as a means to extend the zoom.

(As an aside, cropping a 200 mm zoom 5 MP image to 3 MP extends the zoom to an equivalent of about 260mm. Cropping to 2 MP, which could be done by removing about 37% in both dimensions, extends to about a 315 mm equivalent. At 1 MP, or about 55% off each dimension, you get about a 450 mm equivalent.)

4. The noise can be lessened up front by using less in-camera sharpening and lower contrast, 100 ISO, fill lighting and perhaps other devices. However, such measures don't entirely solve the problem. Post processing can usually remove any remaining problematic noise without decreasing the image quality in other ways, though not always.

5. There is a theory abroad that Minolta is the only camera in its class that offers images without noise processing, and that this is a superior imaging choice. The 7x actually does have in-camera noise reduction available for some conditions. As far as I can tell, its competitors that offer raw file shooting don't do noise processing in raw (though I don't really know for sure), so the choice appears to be available to go without it.

In the Sony, there may be noise removal processing for all files (don't really know about that either). However, whether that represents a defect isn't clear. Some complain of an artificially clean or plasticky look in Sony shots, others just think they're wonderfully sharp, clean and accurate. For those Sony owners who prefer more of a noisy texture, there are filters available (e.g. in Photoshop) that can add it.

6. It has been suggested in various places that Minolta should offer optional in-camera noise reduction, as this could save post processing time at least in some cases. This appears to work well in similar cameras, which have equally high or higher levels of accurate textural detail with less noise right out of the camera. It would still leave the option of no noise reduction.

I have a theory--just an idea--on why Minolta doesn't offer this option (except for long exposures). For in-camera noise removal to be effective without harming overall image quality, the noise must be limited, somewhat predictable and not too mixed up with real details. My theory is that the level and type of noise distribution in the 7x doesn't lend itself to automatic noise reduction software as well as that found in its competitors. It requires more individual attention. Since other cameras apparently have less noise to deal with, and with different characteristics, it might not be so difficult for them to control automatically in-camera if one so chooses. (The reason the 7x does have noise removal for long exposures is that at least one type of noise common in such exposures can be removed automatically to a limited extent without much harm--the trade off works better than no noise reduction in those cases.) Not sure about this theory.

Happy shooting.
Sanpete is offline   Reply With Quote
Old Nov 6, 2002, 3:55 PM   #93
Member
 
Join Date: Oct 2002
Posts: 99
Default

We got clobbered with rain here, if anyone has some good weather and a 7,7ior 7hi please reproduce the test I wanted to do varying the contrast and sharpness settings.
Binji7 is offline   Reply With Quote
Old Nov 6, 2002, 6:47 PM   #94
Member
 
Join Date: Aug 2002
Posts: 32
Default

Quote:
Originally Posted by Sanpete
1. Even though the Sony f707/717, Nikon 5700, Olympus E20 and the Minolta 7/7i/7Hi all use some version of the same CCD chip, that doesn't imply that they all have the same amount of noise prior to filtering/processing. Differences in a number of factors, such as lenses, internal IR filters, color filter arrays, power supply, and other electronic componenents do/might affect the amount of noise as well as how noticeable a pattern it forms.
Lenses - no. Lenses can cause fixed position artifacts, but can not directly cause noise (random fluctuations from pixel to pixel)
Internal IR filter - may accentuate or attenuate noise based on the white balance changes.
Color filter array - these CCDs all have the same color filter array (factory installed), so no effect.
Power supply and electronic components - this is possible. The LCD in the Minolta is very close to the CCD and could increase noise by increasing the temperature of the CCD.

Another possible difference is that Sony manufactures the CCDs - If Sony kept the least noisy CCDs for its own use, then all other camera makers would have higher noise.

Quote:
Originally Posted by Sanpete
2. Noise is more apparent in the shots taken with the 7x than those taken with its competitors when in similar conditions and with similar camera settings. This appears not to be due to greater noise removal processing in the other cameras, at least not entirely, since this is also reflected in the images shot in raw.
More apparent on screen, yes. More apparent in prints, not necessarily.

No one has posted any evidence on either side for the raw file issue. To be meaningful, the raw images would have to be taken of the same picture under the same conditions, and converted with the same software at the same settings - an extremely hard task. (using different conversion software means different algorithms for conversion, and thus different tradeoffs of noise/detail)

A side note: Minolta's RAW converter does NOT do a good job with noise - it actually increases noise quite a bit, due to the contrast increase it does.

The difference in the type of noise between the cameras means that the Nikon noise (blobs of discolored areas) tends to show up on prints more than the Minolta noise (individual discolored pixels).
This changes when prints get larger, but it is much harder to process the Nikon noise further, and it is possible to process the Minolta noise if needed.
I don't know about the Sony - the 707 took pictures with more intensified color (which looks better to a lot of people, but is less accurate) and this increased color makes the pictures look more plastic to me, even without noise involved.

Quote:
Originally Posted by Sanpete
(The reason the 7x does have noise removal for long exposures is that at least one type of noise common in such exposures can be removed automatically to a limited extent without much harm--the trade off works better than no noise reduction in those cases.) Not sure about this theory.
In technical terms the hot pixel removal is not 'noise' as such - the effect of hot pixels is predictable, and can be removed quite easily by the dark frame subtraction method. The dark frame subtraction does potentially increase the random noise, since you are now taking 2 exposures (each with noise) and subtracting them from each other. Since hot pixels are usually several orders of magnitude more problematical on long exposures, this is usually a very good tradeoff.
abredon is offline   Reply With Quote
Old Nov 6, 2002, 8:24 PM   #95
Senior Member
 
Join Date: Oct 2002
Posts: 163
Default

abredon

Thanks for the comments. This is a long thread, so you probably didn't read all that went before. Some of my comments will be repeating what was said earlier.

Quote:
Lenses - no. Lenses can cause fixed position artifacts, but can not directly cause noise (random fluctuations from pixel to pixel)
Since the lenses have different coatings and also affect white balance and IR transmission in various ways (IR being a source of heat as well), they too might have some effect on noise.

Quote:
Color filter array - these CCDs all have the same color filter array (factory installed), so no effect.
Sony offers two different CFAs on this chip, a primary color one and a complementary color one. Minolta, Sony and Olympus use the primary color one, Nikon the complementary.

Quote:
Another possible difference is that Sony manufactures the CCDs - If Sony kept the least noisy CCDs for its own use, then all other camera makers would have higher noise.
Yeah, I left that one off the list--I don't think it's likely, but it's possible, I suppose.

Quote:
More apparent on screen, yes. More apparent in prints, not necessarily.
According to Minolta and other camera owners who have made comparisons, the noise is less apparent in prints, as is the noise from other cameras. However, the prints made from 7x files are still noisier when printed than similar shots from other cameras, according to the reports I've read.

Quote:
No one has posted any evidence on either side for the raw file issue. To be meaningful, the raw images would have to be taken of the same picture under the same conditions, and converted with the same software at the same settings - an extremely hard task. (using different conversion software means different algorithms for conversion, and thus different tradeoffs of noise/detail)
Quite so. My primary assertion is that those who claim that all the cameras produce the same amount of noise before filtering should be providing some evidence. The fact that none have even claimed that the raw shots from other cameras are as noisy indicates to me that they don't have any direct evidence. I've never read reports from owners of other cameras complaining about such a level of noise in their prints from raw files, but I have read complaints from Minolta owners on this score who have also shot raw with other cameras. So I base my belief on that. But if anyone is really interested, and has a fast conection (I'm 56K), there are raw files available at imaging-resource.com that are suitable for comparisons (far field test--not necessarily the best subject for a noise comparison, but OK, I suppose).

Quote:
A side note: Minolta's RAW converter does NOT do a good job with noise - it actually increases noise quite a bit, due to the contrast increase it does.
Can't that be avoided by selecting the lowest contrast setting?

Quote:
The difference in the type of noise between the cameras means that the Nikon noise (blobs of discolored areas) tends to show up on prints more than the Minolta noise (individual discolored pixels). This changes when prints get larger, but it is much harder to process the Nikon noise further
That's the first time I've heard either of those claims. What do you base them on?

Quote:
I don't know about the Sony - the 707 took pictures with more intensified color (which looks better to a lot of people, but is less accurate) and this increased color makes the pictures look more plastic to me, even without noise involved.
My impression from Phil Askey's measurements was that the Sony 707 had a slight bluish or grey tint as well as a little extra saturation. The 717 colors, however, were as close to accurate as any, on the whole. Most people who have complained that the Sony shots look plastic have attributed it to noise reduction.
Sanpete is offline   Reply With Quote
Old Nov 6, 2002, 9:05 PM   #96
Senior Member
 
KCan's Avatar
 
Join Date: Jul 2002
Posts: 1,625
Default

Well, ...
What a long but very informative thread from all of you, my brain have an overdose now :shock: ...

As fotograafdigi and monsieurms (hic :mrgreen: ) , I bought the cam without trying all the brands and printing, checking each one's output , decision was features and +versus price of course ( if not , sure I just bought the top Nikon DSLR as many I guess)

Nevertheless, me too, I am very happy with my choice, despite some noisy low light shots. Thanks again NHL for all the tips and information's we got from you and your support b4 and after my acquisition ( and from other chatter as well ) to make us appreciate and learn more about our D7xx camera.

But do you know guys, there is also another not mentioned yet source of noise , I hear myself ***ing when I miss a shot , and no sir, even photoshop ver 10 (to come) won't be able to cure that noise :mrgreen:

BTW, I try some tips in photoshop mentionned by NHL, and just installed NeatImage to see how I could be more happy about pics I take, that's all that matters.

Keep cool everybody and Cheers !
hic! ( that's to monsieurms )
KCan is offline   Reply With Quote
Old Nov 7, 2002, 6:29 PM   #97
Member
 
Join Date: Oct 2002
Posts: 99
Default

Here is the link to my test shots, the recommendations for noise reduction had a HUGE effect on the outcome of the shots!!


http://www.pbase.com/binji7/dimage_7i_noise_testing
Binji7 is offline   Reply With Quote
Old Nov 7, 2002, 8:21 PM   #98
Senior Member
 
Join Date: Oct 2002
Posts: 163
Default

Well done, Binji7. Very helpful.
Sanpete is offline   Reply With Quote
Old Nov 8, 2002, 5:56 PM   #99
Senior Member
 
Join Date: Jun 2002
Posts: 161
Default

Kcan, glad to see you're having some wine...

Re: the posted photos, This was a test regarding the issue of noise on D7s. Frankly, all the shots look perfectly acceptable to me--I actually like the first one best, the extra contrast being better, and the relative supposed addition of noise insignificant.
monsieurms is offline   Reply With Quote
Old Nov 8, 2002, 7:26 PM   #100
Member
 
Join Date: Oct 2002
Posts: 99
Default

If u zoom in 200% or so, you can see the difference in noise from the various settings. Although when u 'look' at a pic, u don't usually look at 400% :lol: But for those wanting to make larger than 8x10" prints you may have an issue.
Binji7 is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 3:39 PM.