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

Reply
 
Thread Tools Search this Thread
Old Nov 13, 2002, 7:47 AM   #101
Junior Member
 
Join Date: Nov 2002
Posts: 1
Default

This is my first foray into this forum, but I've been reading the dpreview minolta forum for some weeks and this noise issue is clearly a hot topic.
I don't own a digital camera. But I am going to buy a 7hi.

I am aware of the noise issue and, even if I assume the worst - that the Minolta produces more visible noise than the Sony and Nikon competitors without corresponding gains in detail - I am still going to buy it because of the other advantages it has over the competition.

But ...

I would like to get to the bottom of this noise issue.
Until now I have pretty much believed the often reported (in forums) point that since the CCDs are the same, the higher noise levels on the Minolta were due to less post processing in the camera. But as Sanpete points out, where is the evidence of this? There is definite evidence of the fact that there is more noise (yes including evidence from Phil Askey - and many others), but I have seen no evidence of better detail or texture.

Sanpete, I have to say, is the only person (with the exception of a couple of other small on-topic postings) who is making any sense at all on this thread. Kudos to you for staying on topic despite the exasperating distractions.

So let me stipulate:
1. When I say below "this is not the point" I mean that the preceding stipulation is not the point the particular question raised in this, my posting. I do not mean that this is not the point of the camera, of owning a digital camera or a Minolta. Nor do I mean that this is not a relevant issue to others. I just mean "Please don't reply by arguing this point"
2. Sure I'll probably be happy with my camera like other owners - this is not the point.
3. Sure when I print the pictures I'll probably find the noise neglible or invisible - or maybe I won't - this is not the point.
4. Sure the camera is better than the competition - this is not the point.
5. Sure it is a fact of life that small 5mp CCDs produce noise and will never be the same as big CCD DSLRs (and of course there are advantages - like smaller lighter zoom lenses) - this is not the point.
6. Sure noise can be reduced/removed with Photoshop and other tools - this is not the point.

So my question is the same as Sanpete's:

1. Does the Minolta 7xx produce more noise in the image than its 5mp prosumer competitors without a corresponding gain in detail?

2. Put another way: If I were to take the same picture with the Minolta, the Sony and the Nikon (all having the same CCD), under noise producing conditions (eg lots of sky in low light), with all the settings absolutely optimal for this picture for each camera (ISO 100, raw, or whatever), and I were to post-process this picture with the best software available such that the noise level in each image is as close as possible to that of the image which has the least noise to begin with (this probably means reducing noise in the Minolta image to equal the other images - but lets not make assumptions) THEN would the Minolta image have THE SAME, BETTER or WORSE detail/texture than the others.

This is a thought experiment, and would take a lot of setting up in perfect conditions to compare properly. And I'm talking about comparing on the screen - I don't care if the noise or texture is noticable - for the purpose of this test, I want to know if it's there.
But if the so called "noise for texture" argument is fact then the answer should be "THE SAME". If it is indeed a myth, the answer should be "WORSE".
If it is worse, then given that they have the same CCD and are governed by the same laws of physics, Minolta should be able to produce a less noisy camera in line with the competition.

3 And, most importantly, as Sanpete keeps asking, does anyone have ANY EVIDENCE OF THIS one way or the other.


Please don't reply if your response involves one of the stipulations above.
E.g. don't tell me to stop worrying, or that I'll be happy, or that I'll be happier with another camera, or that I should save my money, or that I won't notice it, etc etc etc. Please, just the facts ma'am.
Rhubarb is offline   Reply With Quote
Old Nov 13, 2002, 11:27 AM   #102
Member
 
Join Date: Oct 2002
Posts: 59
Default

You can add the Olympus E-20 to your list of cameras that use this CCD, if you want.

As you said, this would be very difficult to set up to compare properly. I fear that the post processing to equalize the noise levels could skew other image attributes out of proportion to the amount of correction required for noise, invalidating the other comparisons. You would first have to know the post processing software very well. What a research project this would make.
charlibob is offline   Reply With Quote
Old Nov 14, 2002, 4:54 AM   #103
Member
 
Join Date: Jun 2002
Posts: 33
Default

Thanks for the refreshing post!!!

Pretty much changed my view on these things
b.t.w. already own the thing, so know where i'm comming from...

"1. Does the Minolta 7xx produce more noise in the image than its 5mp prosumer competitors without a corresponding gain in detail?"

I dont know. I have a feeling but no hard-evidence. My feeling is: Yes, it produces more noise, but does add detail, but the noise 'gained' is substantially more then 'detail gained'. I think the gained detail is un-noticable in 99,99% of the shots, the noise is noticable in 100%.

So my one word-answer to (1) would be: "YES"

"2. Put another way: If I were to take the same picture with the Minolta, the Sony and the Nikon (all having the same CCD), under noise producing conditions (eg lots of sky in low light), with all the settings absolutely optimal for this picture for each camera (ISO 100, raw, or whatever), and I were to post-process this picture with the best software available such that the noise level in each image is as close as possible to that of the image which has the least noise to begin with (this probably means reducing noise in the Minolta image to equal the other images - but lets not make assumptions) THEN would the Minolta image have THE SAME, BETTER or WORSE detail/texture than the others. "

Okay, first, I think for such a test, RAW is a absolute requirement. The minolta interpolation of the bayer pattern I find definatly worse then that of the competition, especially showing in diagonal sharp lines.

To answer the question: "THE SAME". You have that option, but requires work. There are other options also, that COULD be seen as an advantage, but also as "bothersome".

Again, just my 2cts

But I want to add something why I think the test suggested is inpossible:

The lenses on 2 differnt cameras differ. Each lens (especially zooms) have a 'golden position' (focal length + focussing distance + aperture), at which the lest can render the best detail to the 'film (or in digital case) sensor plane. I dont know (but do approximatly) these settings for 2 lenses of 2 different digital cameras. Anyhow, they most likely will not match. Thus making it impossible to do 'identical shots'. Granted, it wont influence the 'noise' much (however, some people debate that...), but will influence detail, and thereby, the appearance of the final image, with or without post-processing.
fotograafdigi is offline   Reply With Quote
Old Nov 14, 2002, 4:59 AM   #104
Member
 
Join Date: Jun 2002
Posts: 33
Default

For everybody's info:

here's a page I found intresting (especially about using a strong Magenta for fluorescent lighting filter.

http://www.logicaldesigns.com/SRNoise1.htm

Enjoy...
fotograafdigi is offline   Reply With Quote
Old Nov 14, 2002, 9:10 AM   #105
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

Quote:
Okay, first, I think for such a test, RAW is a absolute requirement
Agree here as well, but then one has to find the same software/algorithm to process theses raw files. Rather unlikely since each camera encode their output and embedded their camera parameters differently as well. One need an independent developper outfit that can decode theses raw files using the same and exact algorithm, and not rely on each manufacturer (which are different by design).

The different CFA (RGB vs YCMG) of the CCDs already forced each software to be different in the 1st place! We will never know for sure won't we?
NHL is offline   Reply With Quote
Old Nov 14, 2002, 12:15 PM   #106
Member
 
Join Date: Jul 2002
Posts: 47
Default

for side by side, sony f717, nikon 5700 and minolta 7hi, try this site:

http://www.digital.idv.tw/dc-test/So...17-compare.htm

almost under the same conditon for night shots: s priority, 1 sec, iso 400, tripot used, fine mode.

note: nikon 5700 used noise reduction function.

to me, only the w/b and contracts r different.
zhu001 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 5:36 PM.