Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > Digital Photography > low end digital / low light and macro ?

Reply
Thread Tools

low end digital / low light and macro ?

 
 
clandestin_écureuil
Guest
Posts: n/a
 
      06-30-2008
Mark Thomas wrote:
> John Sheehy wrote:
>> John Sheehy <(E-Mail Removed)> wrote in
>> news:Xns9ACCD188A5FA0jpsnokomm@130.81.64.196:
>>> I'm just stating the facts as I find them.

>
> Indeed. Like asaar, forgive me if I defer to other sources. Just for a
> change from the usual dpreview pages, here are two Cameralabs pages, one
> for the F30 and one for the FZ50. Look carefully at the ISO 1600 shots,
> taken in controlled, but real-life conditions:
>
> http://www.cameralabs.com/reviews/Fu.../page4ca.shtml
> http://www.cameralabs.com/reviews/Pa.../page4ca.shtml
>
> Neither are brilliant, but imo there is more usable detail in the Fuji
> images. The pure noise tests on the previous pages are even more
> damning, but less 'real'. Similarly, every review I can find for the
> FZ50 is critical of its low-light/noise performance, and its
> over-zealous NR. Nice camera otherwise, don't get me wrong.
>
> Eg, here are some reviewers comments:
> (IR) - Fuji has been wise to emphasize the F30's range of excellent
> low-light options...
> (CL) - (The F30's) low light capabilities still lead the market
> (DPR) - F30 is far and away the best low light compact camera on the
> market today... F30's sensor gives you at least a two-stop advantage
>
> (IR) - Panasonic FZ50 really struggles when shooting at higher ISO
> levels, producing images that are marred by excessive chroma noise...
> (CL) - FZ50 sadly runs into trouble at 200 ISO... detail is smeared-out
> by aggressive noise reduction... with its compromised image quality
> above 100 ISO is arguably a step-backwards
> (DPR) - fantastic camera with a less than stellar sensor... smearing of
> detail.. limits the FZ50 to low ISO settings.. only for those users who
> can live without anything over ISO 200...
>
> And there are numerous other examples - are all these people on the Fuji
> payroll? It seems you may be justifying your own purchase decision,
> rather than doing controlled tests. If you wish to show two identically
> lit and exposed images, and a similarly lit and (under)exposed
> unprocessed shot at 3200 from the FZ50... feel free to come back and
> continue the discussion.
>
> I'm intrigued by your claim that FZ50 'collects more photons' - do you
> have a source?
>
> Anyway, this is rather offtopic as no *compact* camera is highly suited
> to low-light photography, and the FZ50 is neither compact or worthy of
> any consideration for someone who is focussed on low-light imaging...
>
> Do you have alternative suggestions for the best *compact* camera, and
> supporting links?



As an addendum to this, there are some very distinct differences between
early FZ50s and those with current firmware. I initially used a friend's
FZ50 and was really impressed but I bought one I was immediately
unimpressed. The performance both in noise levels and focusing was not as
good as the one I borrowed. I subsequently learned, through Google, that
there had been a firmware update and my new camera did not have the current
firmware. I found a thread on the issue and looked at some sample photos
with the new firmware (see: http://tinyurl.com/5kevnv) and immediately
returned the camera. They offered me another - brand new stock with the
new firmware - which I accepted. It was and is quite brilliant. The
difference really was from mediocre to brilliant. The new firmware makes it
faster and gives it better battery life as well less noise and more precise
focusing.

I still find it to be more versatile at low ISO settings than my DSLRs. No
bulky lenses to carry and I can instantly shoot anything from a macro to
430mm zoom with that lovely Leica lens (or with the telephoto conversion at
more than 850mm).

If only it had decent low light capability I would want no other camera.
That is why I was posting about TTL flash units the other day. I have
solved that problem now, at huge cost unfortunately.

Secret Squirrel

--

Ingrid Rose

clandestin.ecureuil(insert missing symbol here)gmail.com
 
Reply With Quote
 
 
 
 
Mark Thomas
Guest
Posts: n/a
 
      06-30-2008
clandestin_écureuil wrote:
> Mark Thomas wrote:
>> If you wish to show two
>> identically lit and exposed images, and a similarly lit and
>> (under)exposed unprocessed shot at 3200 from the FZ50... feel free to
>> come back and continue the discussion.


And I can answer that request myself, having just browsed over the links
asaar provided. At the bottom of this page:
http://www.dpreview.com/articles/com...hiso/page4.asp
there is a direct comparison of the F30, the FZ50 and a DSLR at ISO3200.
Oh dear. Perhaps that is a little unfair, as the Pana is in a reduced
res mode and they didn't include the FZ50 in the lower ISO comparisons.
But it is very clear throughout that test that the F30 is simply in a
class all by itself when it comes to compact high ISO.. Of course the
camera itself wasn't so outstanding in some other aspects but in that
respect, ie low-light sensor performance, it was unchallenged.

Sadly, it's been downhill since then due to the suicidal desire of
camera companies (inc. Fuji) to simply pack more pixels onto sensors, so
finding one of these little gems may be difficult. It does make one
wonder what sort of cameras we *should* be seeing now if the improvement
in sensor/processing quality had continued. Or even if they just stuck
a similar sensor into a design like the FZ50 so it could benefit from
that lovely piece of glass.. Oh well.

> As an addendum to this, there are some very distinct differences between
> early FZ50s and those with current firmware.


That's interesting to know, especially for anyone considering one
secondhand... I wasn't aware of a firmware release addressing noise.

> The difference really was from mediocre to brilliant. The new
> firmware makes it faster and gives it better battery life as well less
> noise and more precise focusing.


I would imagine the 'less noise' is only relevant up to about 2-400 ISO,
judging by your comments. The images I've seen above that from the FZ50
were clearly beyond any sort of NR, or firmware fix.. (O:

> I still find it to be more versatile at low ISO settings than my DSLRs.
> No bulky lenses to carry and I can instantly shoot anything from a macro
> to 430mm zoom with that lovely Leica lens (or with the telephoto
> conversion at more than 850mm).


No argument with that - I use a similar camera and love the
'one-thing-to-carry' convenience. I borrow a DSLR every now and then,
and one day I might win the lottery and get one of those I think are
'worthy' (which at the moment would probably only be the D3...(O

> If only it had decent low light capability I would want no other camera.
> That is why I was posting about TTL flash units the other day. I have
> solved that problem now, at huge cost unfortunately.


I saw the thread but missed that posting - I'll go check it again..
Metz? (no need to answer, I'm just being lazy..)


 
Reply With Quote
 
 
 
 
David J Taylor
Guest
Posts: n/a
 
      06-30-2008
Mark Thomas wrote:
[]
> And I can answer that request myself, having just browsed over the
> links asaar provided. At the bottom of this page:
> http://www.dpreview.com/articles/com...hiso/page4.asp
> there is a direct comparison of the F30, the FZ50 and a DSLR at
> ISO3200. Oh dear.

[]

Not necessarily. To me, there is more graduation in the reflected
highlights on the Casio and Panasonic than on the Fuji, and it might be
that the 2.5MP image would be more than good enough for smaller prints
(say up to 7 x 5 inches). (I've seen good A4-size prints [297 x 210mm)
from a 3.3MP camera.)

So in low-light conditions, reducing the spatial bandwidth by binning may
well be a better comprise then having a sharp, but noisier image. It may
also depend on the image, and the viewer. It is one case where
pixel-peeping at 1:1 zoom just doesn't tell the whole story.

Cheers,
David


 
Reply With Quote
 
ransley
Guest
Posts: n/a
 
      06-30-2008
On Jun 27, 3:10*pm, (E-Mail Removed) wrote:
> I have a Fuji P&S... nice zoom... and, it does great daytime
> pictures... but I have two problems with it.
> This camera will not function once the Sun goes down... without
> flash, it chokes even on a heavilly overcast day.
>
> So_ next, I'll be looking for a digital with two capabilities...
> it needs in reduced lighting situations... and... it needs to
> work totally without autofocus, or, maybe I should say, have
> a manual focus mode.
>
> What kind of digital... inexpensive please... will have these
> capabilities ? *Or... has digital completely forgotten about
> manual focus and low light ??
>
> thanks for any suggestions. *Im sure these same problems
> are perplexing a lot of other 35mm/slr people too.


I have a cheap digital, i just use a tripod, actualy many fujis have
had better low light performance then others by a stop or 2. P&S are
most all a bit slow. but it depends on your model. Dont expect
anything without sun or a tripod but blur.
 
Reply With Quote
 
ASAAR
Guest
Posts: n/a
 
      06-30-2008
On Mon, 30 Jun 2008 10:30:20 GMT, David J Taylor wrote:

>> there is a direct comparison of the F30, the FZ50 and a DSLR at
>> ISO3200. Oh dear.

> []
>
> Not necessarily. To me, there is more graduation in the reflected
> highlights on the Casio and Panasonic than on the Fuji, and it might be
> that the 2.5MP image would be more than good enough for smaller prints
> (say up to 7 x 5 inches). (I've seen good A4-size prints [297 x 210mm)
> from a 3.3MP camera.)
>
> So in low-light conditions, reducing the spatial bandwidth by binning may
> well be a better comprise then having a sharp, but noisier image. It may
> also depend on the image, and the viewer. It is one case where
> pixel-peeping at 1:1 zoom just doesn't tell the whole story.


Someone here has eyes but cannot see. Really, anyone only
needs a quick look at the ISO 3200 and they'll easily see detail
produced by the F30 that would show up in a print. But that same
detail is completely absent from the Casio and Panasonic images, so
it wouldn't be visible in prints of any size. See all of the dots
(rivets?) in the robot's face? They're very clear in the EOS 30D
and F30 images, but completely absent from the Casio and Panasonic
images. It's impossible to compare the text seen in the 30D and F30
images because it's cropped out of the Casio and Panasonic images.
But if you examine the full images, you'll see that the "Tin Light"
that appears on the robot's chest is produced very legibly by the
30D, is a bit noisy but legible (F30),

Now if you're trying to make an A4 print from the FZ50's full
image, the result may be satisfying, but that satisfaction would
quickly turn to dismay if an A4 print from the F30 was placed next
to the one from the FZ50. From uncropped (other than to match the
print aspect ratio), where *no* pixel peeping is involved, the
FZ50's ISO 3200 image just doesn't compare. You may need good
eyesight or a loupe to read the "Tin Light" in the smaller of the
F30's prints. But the same detail from FZ50's image, whether
printed at 4"x6", 5"x7", 8"x10" or A4, is an illegible smudge. You
may be able to make out that it's text, and that's a shame,
considering the detail that the FZ50's Leica lens can produce.

Speaking of "Leica", look at the camera in the test shot. The
F30's image clearly shows Leitz Elmar 1:3.5 F=50mm, as well as the
aperture and other markings. The FZ50 only shows smudges.

One other point - the color produced by the FZ50 may be pleasing,
but it isn't accurate, at least on my monitor. The F30's image of
the "Kodak Gray Scale" is gray, and the supposedly white background
appears quite white. The FZ50's image on the other hand has a
pronounced greenish cast. Since DPReview said about the FZ50 "Vivid
but realistic color", maybe the color balance wasn't closely
monitored during the test. In any case, that's a non-issue for FZ50
RAW shooters. Or for FZ50 owners that wear rose colored glasses.

 
Reply With Quote
 
David J Taylor
Guest
Posts: n/a
 
      06-30-2008
ASAAR wrote:
> On Mon, 30 Jun 2008 10:30:20 GMT, David J Taylor wrote:
>> []
>> So in low-light conditions, reducing the spatial bandwidth by
>> binning may well be a better comprise then having a sharp, but
>> noisier image. It may also depend on the image, and the viewer. It
>> is one case where pixel-peeping at 1:1 zoom just doesn't tell the
>> whole story.

>
> Someone here has eyes but cannot see.

[]

I'm not pixel peeping, simply suggesting that for smaller prints, have a
smaller number (2.5MP) of binned pixels may be preferable to a larger
number (10MP) of poorer quality ones. I don't have the time to try this
test for myself right now, but it would be interesting to know what actual
prints - and not pixel peeping - shows. And I mean viewing the prints at
a normal distance, not with a magnifying glass. It's a test for anyone to
try - I am not pre-judging the result.

Cheers,
David


 
Reply With Quote
 
ASAAR
Guest
Posts: n/a
 
      06-30-2008
On Mon, 30 Jun 2008 20:36:56 GMT, David J Taylor wrote:

>> Someone here has eyes but cannot see.

> []
>
> I'm not pixel peeping, simply suggesting that for smaller prints, have a
> smaller number (2.5MP) of binned pixels may be preferable to a larger
> number (10MP) of poorer quality ones. I don't have the time to try this
> test for myself right now, but it would be interesting to know what actual
> prints - and not pixel peeping - shows. And I mean viewing the prints at
> a normal distance, not with a magnifying glass. It's a test for anyone to
> try - I am not pre-judging the result.


I completely understood that you weren't pixel peeping and in some
cases binning might be preferable. From what I could see, though,
the binning might be more of an advantage for the pixel peepers,
since it seems to really be good at hiding, or smoothing out noise -
unfortunately losing a tremendous amount of detail in the process.

Did you actually look at the images? The one from the FZ50 at ISO
3200 that I described (at the link provided by Mark Thomas) *was*
binned. At least it didn't turn out as bad as the really atrocious,
similarly binned photo from the Casio EX-Z1000. If you want to try
your test, almost everything is supplied. Save the ISO 3200 images
for the FZ50 and F30, add some spice from your favorite editor if
you wish, bake until done and print. As I indicated, if you print
the full sized image the difference may be slight. But if it's a
cropped portion that you print, the difference should be obvious.
Loupe not needed!

 
Reply With Quote
 
David J Taylor
Guest
Posts: n/a
 
      07-01-2008
ASAAR wrote:
[]
> I completely understood that you weren't pixel peeping and in some
> cases binning might be preferable. From what I could see, though,
> the binning might be more of an advantage for the pixel peepers,
> since it seems to really be good at hiding, or smoothing out noise -
> unfortunately losing a tremendous amount of detail in the process.


That I don't agree with - it seemed to me that there was more colour
detail in the binned images, and that operating the small sensor at such
ISOs with noise cleaning lost too much.

> Did you actually look at the images? The one from the FZ50 at ISO
> 3200 that I described (at the link provided by Mark Thomas) *was*
> binned. At least it didn't turn out as bad as the really atrocious,
> similarly binned photo from the Casio EX-Z1000. If you want to try
> your test, almost everything is supplied. Save the ISO 3200 images
> for the FZ50 and F30, add some spice from your favorite editor if
> you wish, bake until done and print. As I indicated, if you print
> the full sized image the difference may be slight. But if it's a
> cropped portion that you print, the difference should be obvious.
> Loupe not needed!


Yes, it was inspection of the images which made me wonder what a full-size
print mught be like. I was thinking of full-size images, not cropped,
i.e. 2.5MP for a 6x4-inch or 7x5-inch print. I don't the time at the
moment to check this myself. Yes, the Casio looked to be binning too many
pixels!

Of course, no matter how you try to fix-up the small-sensor images, they
will not be anything like as good as those from a full-frame or crop-frame
35mm camera (DSLR or otherwise), so as soon as someone says "low-light",
the DSLR would spring to mind as the obvious solution.

Cheers,
David


 
Reply With Quote
 
John Sheehy
Guest
Posts: n/a
 
      07-10-2008
ASAAR <(E-Mail Removed)> wrote in
news:(E-Mail Removed):

> On Mon, 30 Jun 2008 00:35:47 GMT, John Sheehy wrote:
>
>> The day I bought the FZ50, I almost bought the 6MP Fuji (6500, I
>> believe) which had RAW (I went with the FZ50 because of the legendary
>> lens).

>
> OK so far.
>
>> It turns out they both have similar pixel read noise, but the FZ50
>> collects more photons in each of its 10M pixels than the 6500 does
>> in each of its 6M.


> How so? The FZ50 has a 1/1.8" sensor and the S6000/S6500 has a
> larger 1/1.7" sensor. If the larger sensor has fewer pixels, then
> those pixels/sensels must be much larger than those in the FZ50's
> sensor. Unless each of the Fuji's sensels is really tiny, and each
> one is surrounded by a lot of unused, wasted space.


Something like that must be going on. The sensor in the 6500 saturates
at about 3200 photons at ISO 100. 4800 for the FZ50.

>> This is the same 6MP sensor in the "legendary" F31.


> Legendary it may be, but it's not the camera that DPReview
> compared in its "legendary" high ISO/low noise comparison. That one
> was the F30, but close enough. At least for the moment you've
> abandoned your familiar "cartoon-like noise reduction" rant, which
> you've used before when you didn't want to admit that the F30 could
> possibly have low noise at high ISO as well as more detail than all
> but the DSLR in DPReview's test.


I never said that the Fuji NR didn't leave a lot of edge detail. That's
exactly what it *does* do; maintains high-contrast edges, while smoothing
away fine detail and low-contrast detail. I can do the same for any
camera with a sigma filter. The edges that are maintained can be quite
wrong and jagged, of course, at high ISOs.

> Now if you're really on to
> something with your claim that the FZ50's pixels/sensels capture
> more photons than the Fuji's, it should outperform the Fuji. Let's
> see if DPReview found this to be true (selected quote snippets) :


How in the world would DPReview be able to tell this? No one on the
DPReview staff knows anything about RAW data; how to look at it or
analyze it. They are JPEG and converter people.

> Oh well, at least your FZ50 has low read noise. Unfortunately,
> when it comes to high ISO/low noise with high detail, it doesn't
> come close to the performance of the S6000/S6500, which in turn is
> bested by the palm-sized F30. If your FZ50 could produce the same
> "cartoon-like noise reduction" you'd be ecstatic!


No, I wouldn't; I despise that look. It's for optically naive people who
think that anything they don't see colored speckles in is good.

> For those that never saw DPReview's Compact Camera/High ISO test
> (and possibly for you, who apparently wants to forget it),


It's not a matter of forgetting; it's a matter of not accepting anyone in
the DPReview Staff as an authority on RAW imaging.

> On second thought, it'll probably be a waste of your time unless
> you've changed (become more open minded) since you posted this
> comment about DPReview's comparison article nearly a year ago :


>> "I read it a few weeks ago when you first linked to it, and didn't
>> learn anything."


> "A closed mind is a terrible thing to waste." -- anon.


My mind is wide open, and very elastic, and that's why my understanding
goes a lot deeper, and I don't by into the myths that the DPR staff do.

--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <(E-Mail Removed)>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

 
Reply With Quote
 
Chris Malcolm
Guest
Posts: n/a
 
      07-11-2008
John Sheehy <(E-Mail Removed)> wrote:
> ASAAR <(E-Mail Removed)> wrote in
> news:(E-Mail Removed):


>> Now if you're really on to
>> something with your claim that the FZ50's pixels/sensels capture
>> more photons than the Fuji's, it should outperform the Fuji. Let's
>> see if DPReview found this to be true (selected quote snippets) :


> How in the world would DPReview be able to tell this? No one on the
> DPReview staff knows anything about RAW data; how to look at it or
> analyze it. They are JPEG and converter people.


Really? If so, that explains something I've been struggling with in
their reviws for some time. They sometimes make detailed comments
about camera sensors and lens performance without making at all clear
how they avoided having those judgments polluted by well-known
differences in camera or editor jpeg conversions. I've sometimes spent
a long time scanning their reviews for clues as to how they avoided
those problems without finding any clear statements.

I can see there is a specific utility in comparing how one package
combination of camera body, lens, and in-camera jpeg conversion stack
up against another, but if that's all that dpreview are doing then
they really should make that clearer in their reviews.

--
Chris Malcolm http://www.velocityreviews.com/forums/(E-Mail Removed) DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]

 
Reply With Quote
 
 
 
Reply

Thread Tools

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 On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Low carb diabetes diet. Low fat high carb diet. The low carb diet.Low carb diet pregnancy. zyraco C++ 0 11-10-2009 01:44 PM
Low carb calorie diet. No low carb diet. Free low carb diet. Low carbdiet meal plan. zyraco C++ 0 11-10-2009 01:44 PM
low light movie works better than low light still photos why? Brian Digital Photography 31 06-14-2009 04:08 AM
best low-light macro/zoom versatile digital camera? Scott Speck Digital Photography 8 03-09-2005 07:18 PM
Re: Are there any decent indoor light (low light) digital cameras out there? ishtarbgl Digital Photography 0 04-01-2004 07:10 PM



Advertisments