Go Back   Steve's Digicams Forums > Digital Cameras (Point and Shoot) > Panasonic / Leica

Reply
 
Thread Tools Search this Thread
Old Aug 25, 2009, 4:42 AM   #1
Junior Member
 
Join Date: Jul 2008
Posts: 11
Default They call this "noise-free"? (FZ35/38)

I was reading this review which stated:

" The 1/2.33 inch, 12 megapixel sensor used in the Panasonic Lumix DMC-FZ38 produces noise-free images at ISO 80-200..."

Then I took a look at one of the sample images which was apparently taken at 1/125 sec f/4.4 | 486mm | ISO 100.

The noise is horrible, clearly noticeable on the flat colored clothes, and the hair regions are a disaster. Before people start saying that those were compressed JPEG, not RAW, etc. It's clear to see that the noise is those images are not due to JPEG compression artifacts.

Now what I'd like to know are:

1. Are these the standards by which compacts are measure, quality like those considered noise-free?

2. Do all Panasonic compacts produce results like this?

Last edited by splunker; Aug 25, 2009 at 8:09 AM.
splunker is offline   Reply With Quote
Sponsored Links
Old Aug 25, 2009, 1:15 PM   #2
Senior Member
 
Join Date: Jun 2002
Posts: 508
Default

There are all kinds of photos (test & otherwise) from FZ38 in the review for this camera on the "trusted reviews" website. The provided photos look quite good. There's even a detailed analysis of the noise across ISO which uses close cropping to allow close inspection of any noise. In fact, trusted reviews rates the FZ38 IQ=10 !!!

Meanwhile, the link provided in the post dos not work (at least for me - get forbidden) & I certainly wouldnt draw any hard conclusion from one photo especially from an unknown source.

The FZ28 is a very well liked & respected camera with enough user testimonies to substantiate what Im saying. It is unlikely that Panasonic has now suddenly "lost the bubble" with the development of its FZ line with release of FZ35/38.

Last edited by sdromel; Aug 25, 2009 at 1:18 PM.
sdromel is offline   Reply With Quote
Old Aug 26, 2009, 5:41 AM   #3
Junior Member
 
Join Date: Jul 2008
Posts: 11
Default

Why would you say that photographyblog.com an unknown source?
Anyway not sure what's causing the 404 error, but if you clicked the link of the review provided, you would have been able to see the review and link to the sample images page here:

http://www.photographyblog.com/revie...sample_images/

Look at the third sample image (kids wearing blue/black), click it for full size and you will see the noise in the hair especially.
splunker is offline   Reply With Quote
Old Aug 26, 2009, 11:04 AM   #4
Senior Member
 
Tullio's Avatar
 
Join Date: Jan 2006
Posts: 7,370
Default

I agree that some of those sample photos look pretty noisy (horrible, really). The problem I have with sample photos included on the reviews is that in order to compare apples to apples, the reviewers tend to use default settings on the cameras they review. Unfortunately, not all cameras produce their best results at the default settings. I have the FZ28 and if I leave it at 0 everything, the images it produces are not that great. Increase Sharpness, Saturation and Noise reduction to +1 and reduce Contrast to -1 and the results are much much better. Images are sharper and cleaner. The reason these cameras offer the adjustments is because one size won't fit all. Some manufacturers prefer to apply heavier noise reduction at the default settings to make the image look cleaner without any intervention. The down side of it is, images tend to lack resolution and in many cases, users will knock down NR to gain detail (at the cost of more noise). To a certain extent, it's a marketing decision. Knowing that reviewers will most likely keep the settings at their default values when they do their reviews, by producing a clean image at those settings manufacturers can position their cameras above the others on the review results. Real life experience is a different story, though.
__________________

Tullio
Tullio is offline   Reply With Quote
Old Aug 26, 2009, 12:45 PM   #5
Senior Member
 
Join Date: Jun 2002
Posts: 508
Default

Well here's another point of view from an actual owner/user:
"FZ38 vs FZ28: some comparisons"

http://forums.dpreview.com/forums/re...ssage=32801287


Update:
Looks like the results from owners/users are beginning to pile in and FZ38 is looking really good (OIS & WB appear improved & noise not noticeably greater over FZ28):

http://forums.dpreview.com/forums/re...ssage=32808150

(Sorry Steve - always hate when I have to do that)

Looks like FZ35/38 is gonna be a big winner.

And regarding noise, finally always remember that the FZ is marketed as a P&S not a DSLR.

Last edited by sdromel; Aug 27, 2009 at 3:22 AM.
sdromel is offline   Reply With Quote
Old Aug 27, 2009, 11:04 AM   #6
Member
 
Join Date: Jan 2004
Posts: 51
Default

As in most cases with such bad/extreme results, this thread is tending to the conclusion of "user error." Both in the operation of the camera (settings) and in the reviewer's assumption that the sample photos are representative of what the camera can do.

john
jmgiv is offline   Reply With Quote
Old Aug 27, 2009, 9:30 PM   #7
Junior Member
 
Join Date: Aug 2009
Posts: 1
Default Notice the ISO

The photo you are debating on is taken at ISO 400. Everybody knows higher ISOs have more noise. check another photo taken at ISO 80 to judge
mshsoft is offline   Reply With Quote
Old Aug 28, 2009, 10:14 AM   #8
Member
 
Join Date: Jan 2004
Posts: 51
Default

Quote:
Originally Posted by mshsoft View Post
The photo you are debating on is taken at ISO 400. Everybody knows higher ISOs have more noise. check another photo taken at ISO 80 to judge
A ha! Dueling specs. Which is it; 400 of the OP or 100 as above?

john
jmgiv is offline   Reply With Quote
Old Aug 28, 2009, 10:41 AM   #9
Senior Member
 
Tullio's Avatar
 
Join Date: Jan 2006
Posts: 7,370
Default

Quote:
Originally Posted by jmgiv View Post
As in most cases with such bad/extreme results, this thread is tending to the conclusion of "user error." Both in the operation of the camera (settings) and in the reviewer's assumption that the sample photos are representative of what the camera can do.

john
I find review sample photos not very representative of any camera's capabilities. They are usually take at default settings and not all cameras perform their best at those settings. What one can conclude from those samples is that camera A performs better in AUTO mode than camera B. So, if the user is interested in a camera with good AUTO output, then the sample photos should be taken more seriously. The other thing is, some reviewers take the sample photos outdoor, under different lighting conditions and of different subjects> Although this can be more representative in terms of real life examples, one looses terms of comparison. A camera taking pictures under bright sunlight on a clear day will perform much better than under cloudy skies. Just take those samples with a grain of salt.
__________________

Tullio
Tullio is offline   Reply With Quote
Old Oct 28, 2009, 5:42 AM   #10
Junior Member
 
Join Date: Oct 2009
Posts: 7
Default The Skinny on the Lumix Venus Engines

Disclaimer: This comment is not intended for those who don't care about accurate image detail, or don't post-process (and thus don't crop, or look at images at 100%, etc.), and/or smear details anyway (via down-sizing or printing) and want to make "good 'nuff" arguments about image detail.

In short - it is not intended to engender the tired old "pixel-peeper vs. satisfied pragmatist" debates which become dogmatic battles over personal preferences.

The commenter (quoted below) cares about detail - so this comment is addressed to folks who do concern themselves with image detail ...


Quote:
Originally Posted by splunker View Post
I was reading this review which stated:

" The 1/2.33 inch, 12 megapixel sensor used in the Panasonic Lumix DMC-FZ38 produces noise-free images at ISO 80-200..."

Then I took a look at one of the sample images which was apparently taken at 1/125 sec f/4.4 | 486mm | ISO 100.

The noise is horrible, clearly noticeable on the flat colored clothes, and the hair regions are a disaster. Before people start saying that those were compressed JPEG, not RAW, etc. It's clear to see that the noise is those images are not due to JPEG compression artifacts.

Now what I'd like to know are:

1. Are these the standards by which compacts are measure, quality like those considered noise-free?

2. Do all Panasonic compacts produce results like this?

and/or take a look at the foliage on the trees in: http://img.photographyblog.com/revie...mc_fz38_22.jpg

or (even worse) the foliage on the trees in:

http://img.photographyblog.com/revie...mc_fz38_24.jpg

In response to "Splunker's" reasonable questions about image quality and the evolving Lumix line of products:

Both shots are at low ISO, and taken low or moderate focal length. The evident loss of detail in foliage, hair, eyebrows, and eyelashes (in shots where the subject exists within areas that should be in focus) are the "canaries in the coal mine" of what the point-and-shoot market (Lumix included) has devolved to in the age of
"super-stuffing" pixels onto tiny image sensors, then butchering image detail so severely in the quest to vanquish noise that any and all advantages of the superior depth-of-field (and, thus, superior image detail) provided by such smaller sized sensors are completely decimated by extremely aggressive noise-reduction algorithms. What (I) see in the hair (as well as on eyebrows and eyelashes) in the originally linked image is not so much a (chrominance or luminance) "noise" as it is effects of the "trash-mashing" of overly-aggressivein-camera noise-reduction on the optical information itself.

Here is my woeful tale of Lumix (and P&S in general) devolution - though I DO have 600 fine gems out of tens of thousands of snaps with four various Lumix cameras:

In all cases I have (religiously) avoided all the (alleged) "auto-brain" modes on all cameras mentioned, shot at minimum ISO, (wherever possible) minimized the levels of the brutish NR, Sharpening, and Contrast settings, and concentrated on trying to take properly exposed shots with adequate depth-of-field for post-processing that does not include (any) further noise reduction.

My three year old LZ5 (6 Mpix on a 1/2.5" sensor; Venus Plus) and FZ30 (8 Mpix on a 1/1.8" sensor; Venus II) do a far better job in terms of preserving image details (albeit at their minimum ISO settings of 80). This spatial-frequency information, once lost, cannot be artificially regained in any kind of post processing!

Less is more (even with highly regarded post-processing noise reduction programs like NeatImage, etc.) where it comes to smearing image details in order to reduce sensor in dark and/or unfocused areas of the image.

My two year old FZ50 (10 Mpix on a 1/1.8" sensor, Venus III) is an ergonomic dream (with a fine Leica zoom lens system, and priceless *manual* zoom and focus rings), but - the Venus III and it's on-board JPG processing are an absolute unmitigated disaster for image detail. Pretty useless - unless one shoots in RAW format (where the sensor delivers respectable detail up through ISO 200 that needs NO post-NR at ALL, just a bit of mild de-mosaicing in Silkypix DS SE).

My one year old TZ4 (8 Mpix on a 1/2.5" sensor, Venus IV) has better color rendering, and makes a similar level of detail-smearing look (slightly) less "painted-on" than the Venus III engine. However, image detail is similarly devolved. And most vexing (and reported for many Venus IV models), the auto-focus (especially in lower light) is a pitiful disaster that (at least) 3 post-production firmware updates released for the TZ4/TZ5 from Lumix never remedied. One's only hope is using Spot Focus (in all cases), and praying. It is excellent for macro shots of (most) flowers, with a fine depth-of-field of 1 Inch at a mere 6 Inch subject distance. Other than that, you can forget about resoving image details. The best news - it only cost me $200 ...

I've peeped with much interest at the Venus V (which is the same as Venus HD) Engine sample photos taken with the more recent incarnations of (way, way) too many pixels stuffed on (roughly) 1/2.5" image sensors. Foliage at moderate distances (that I know *should* exist within the "depth-of-field" of camera focus) continues to look progressively more smeared. Oh, well.

With the exception of the (expensive and discontinued) LX3 (10 Mpix on a 1/1.63" sensor, Venus IV), each new model degrades image detail quality (viewed at 100%) in a manner nearly INVERSELY proportional to it's number of Mpixels divided by 6 Mpixels. Basically a useless exercise in causing a problem (excessive sensor noise due to pixel density) only to (allegedly) "solve" the same problem with horribly over-aggressive NR, sharpening, and contrast. The result is irretrievable in post-processing, and a devolution of overall quality in favor of playing the "big-Mpixel" game on the same tiny image sensors.

(At least), each new incarnation of the Venus Engine appears to handle on-board color rendering better, anyway. However, I find that the Venus IV in doing this saturates colors to the point where any further saturation in post-processing yields negative and detrimental results. Welcome to the age of over-automatic-ization!

And the LX3's image detail does not look all that impressive, either (especially considering that it cost more than my FZ50). The Venus IV is yet another miss.

What I can see of the Venus V (aka Venus HD) models is that (given the big Mpixel-wars), Panasonic has decided that their largest proportion of customers do not mind highly-overexposed images at excessive contrast levels with NR-decimated detail patched with pretty color saturations and egregious over-sharpening ... I sure do!

And, while I do like the Lumix optical image stabilization (2-3 stops in the LZ5, and 3-4 stops in the FZ30/FZ50), Panasonic (humorously, and I suspect falsely) claims a full two-fold (one stop) improvement with each new incarnation of the Venus Engines ... Let's see, that would imply that the Venus V has an (astronomically good) 5-6 stop mechanical stabilization factor ... something that I sincerely doubt is the actual case. Marketing "sizzle".

Having (foolishly) given my FZ30 away when I bought my FZ50, I now rely on my FZ50 (in RAW format only) or my little LZ5 for everything - other than the macros of (not too detail intensive) flowers that my TZ4 handles well (in high light levels in Spot Focus mode, anyway).

Pardon my puritan ethic, but - if Lumix had simply upgraded the 10 Mpixel FZ50 with a 1/1.63" sensor, and a decent (larger than it's 2.0") diagonal LCD viewfinder, and kept the Venus II Engine with some minor improvements in it's color rendering - they would have so many happy camera owners that few would eternally lust for some new "messiah-model" that only recedes like a mirage farther and farther away, ensuring *maximal* sales of new hardware for Panasonic, and *minimum* satisfaction on the part of users who would like to use their cameras (and their own brains) to take quality shots that can be successfully improved in post-processing (as opposed to vastly over-processed).

Note: I am not evangalizing my personal preferences here. If you like your new little model just fine - more power to you! If you feel like pointing out that I am expecting too much from point-and-shoots, and should go RAW all the time with DSLRs, etc., I've heard it before.

My LZ5 and FZ30/FZ50 (in my opinion) show that P&S and "bridge" cameras under $500 (used to) have real potential - knowingly squandered by the manufacturers, and blithely accepted by the majority of the customer market. Such product value does not maximize profits.

"Thinking" is out of fashion, and "stinking" is where overall image quality has (and will) continue to head. The "auto-everything" camera of the future will simply match the viewed image to a vast resident library of pre-packaged cartoons, and upload a "SIMs-like" facsimile - and precious few may even know or care ...
.
Detail Man 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 10:49 AM.