You thought Olympus is noisy? huh!!!

shg

Senior Member
Messages
1,376
Reaction score
33
Location
US
Any comments? does that mean badly exposed photo at high iso is
noisy no matter what camera you have? Or noisy sensor is noisy
whether good or bad exposure at high iso?
I think some probably have an unrealistic view of what ISO 3200 should be, no matter the model or brand. For ISO 3200 and blown up to the size I'm looking at on my screen, I'd say that looks pretty darn good. Thinking back to my days of processing my own TMAX 3200 black & white film, it's absolutely outstanding. ISO 3200 isn't going to be completely grainless on any camera.

I've been in the process of looking at several things as I'm about to expand my system and, to be honest, I've been also looking at models like the Nikon D200 and the Canon 30D. The D200 is no noise king above ISO 640 or 800. Go to this review of the D200 and scroll down to the book case and check out the shots from ISO 800 and above...

http://photo.net/equipment/nikon/D200/

The Canon 30D is quite a piece of work, but to be honest, I see nothing in the Canon or (especially) Nikon lens lines I like better than what I have right now (7-14, 14-54 and 50-200 Zuikos) so tomorrow when I go online and place my order it'll be for an Olympus E330. From what I've seen, I'll be perfectly happy processing E330 files in Adobe Camera RAW and, using the software within both ACR and the smart sharpen filter of Photoshop, should be able to produce perfectly nice images at even the higher ISO ratings like I did here with my E-1 at ISO 1600:

http://gmchappell.smugmug.com/gallery/2455125#128748041

Who knows what that guy did with those photos on PBase. What sharpening settings did he use, how did he process the images? They don't look to be underexposed to me, something that would have made the grain worse. To be honest, we all should hope for performance like that at ISO 3200.
 
That's a top of the line camera for 2004 but still, 1D Mark II is 3 year old. At that time ISO 3200 was a big thing, and even now it is, except for some Canon, Fuji and Nikon top of the line cameras.

Any other camera (except for top models from N/C/F), even today, would miss that shot with an F2.8 lens. The picture looks noisy, true, but it's still pretty sharp and usable for most (with little PP).

The built-in AS/IS of K10D or A100 wouldn't help and for any Oly with a similar lens, you'd just get blur. Canon is unbeatable in high action area, it's just that they are too expensive and for people with limited budget looking for value, they are not suitable.

For an Oly you could probably use a faster lens and get the same photo for less $$ (even today Mark II body costs $3600) , but not sure if Oly has such a lens.
Here you go (read the camera), watch the noise and boy's hair who
is holding the ball

http://www.pbase.com/slamothe/image/74368773

Any comments? does that mean badly exposed photo at high iso is
noisy no matter what camera you have? Or noisy sensor is noisy
whether good or bad exposure at high iso?

--
http://www.pbase.com/shg2

 
sports is the intention, then a 1 series is probably the best bet.

Of course, if you are not interested in indoor sports (you'd have to pay me to go) then it isn't really relevant :-)
 
Any comments? does that mean badly exposed photo at high iso is
noisy no matter what camera you have?
Correct
Or noisy sensor is noisy
whether good or bad exposure at high iso?
Well ... sort of. But bad exposure just makes it worse and good exposure makes it better. However your comments seem to suggest you have fallen into the trap of thinking that some other cameras are somehow 'noise free'.

You should be asking the question: "are all sensors noisy at high ISO"
and the answer is YES

But then the real question is "so what's the difference in real life between camera models"

And this then becomes a lot more tricky because it depends on lots of things like absolute noise, visual unpleasantness of the noise and resolution so when you downsample, the noise gets reduced in final output.

I asked this question a few weeks ago in the pro forum. I pretty much knew the general ballpark answer (grin) but I wanted to see if people could prove what they thought.

http://forums.dpreview.com/forums/read.asp?forum=1014&message=21794833

Unfortunately, nobody really answer my question as to its effect in normal size prints, so maybe I'll have to do the print tests myself. Enjoy the read. And for anyone thinking of jumping systems thinking the noise is suddenly going to go away. You might be in for an unpleasant reality check (or at least a realisation that you've spent a whole bucketful of cash for a very small improvement alongside a whole boatload of disadvantages).
Swings and roundabouts ... swings and roundabouts ....

G.
 
That 3200 looks like my 1600. But that is OLD sensro technology. If Oly were to make a 10MP camera with that ISO 3200 today, they would be lauded.
--



--
Zach Bellino
'I prefer my lo-mein of the veggie variety.'
--ZJB
'There are three kinds of lies: lies, damned lies, and statistics.'
Benjamin Disraeli (1804-1881)
 
IMO, my E-1 is next to useless at ISO3200, and even 1600 if you don't overexpose a bit. Which is the same as cheating and going to less than 1600, I guess :)

I know folks here do a good job of processing these sensitivities, and I take nothing away from them.
 
Here you go (read the camera), watch the noise and boy's hair who
is holding the ball

http://www.pbase.com/slamothe/image/74368773

Any comments? does that mean badly exposed photo at high iso is
noisy no matter what camera you have? Or noisy sensor is noisy
whether good or bad exposure at high iso?
Yes, an under-exposed shot could look bad and noisy even at low ISO.

This is ISO 3200 - probably as good as ISO 800 on my E-1.

Gábor
 
. . .ISO 1600, so that means Olympus is making progress. Oh, wait, so hasn't Canon since the 1D. Snap.
--



--
Zach Bellino
'I prefer my lo-mein of the veggie variety.'
--ZJB
'There are three kinds of lies: lies, damned lies, and statistics.'
Benjamin Disraeli (1804-1881)
 
The 1D is as noisy as the E-1 and worse at lower ISO's if underexposed. I have worked with files from that camera and was really shocked that it was so noisy.

IF you need 8fps than you can't beat the 1D. For color and noise, it is not top notch. I was ready to sell my E-1 and buy into Canon or Nikon until I started actually working with images from Canon/Nikon cameras. Then I realized how much I like my E-1

I shoot at 1600 with no problems as long as it is well exposed . The new model Canon's (20/30D/5D) will tromp on the E series camera noise wise. You do have to like very smooth skin though, as their process in camera is heavy.

The 1D is a sports machine though. It is also very heavy and the same age as the E-1.

--
See profile for equipment

http://JuliePoole.com
 
Much ado about very little. 10 years from now, noise at 3200 will be history. people will be complaining that there is to much noise at 64000. Anyway, we have some great noise reduction software available today that makes noise almost a non issue. Personally, I do not mind noise if it looks and feels like film grain.
--
http://www.zenfolio.com/digitalphotonut
 
Recently I was shooting at high isos with the E-1:

This is 3200iso, no noise reduction:



This is 1600iso, again no noise reduction:



To my mind these are both quite impressive for "un-useable iso values"... and I've been critical of high iso noise from the E-1 here in the past.

As the fascinating link in Big Ga's post suggests.... so much depends on the light quality, whether high or low contrast, whether daylight or not, and if not whether concert/theatre style, fluorescent or tungsten as well as even lighting, or lots of deep shadow... and of course your exposure too. My 'problem' is the unpredictability of the E-1's low light performance... it isn't as straightforward as good vs bad exposure....although of course that has a bearing. Plenty of good exposures give dreadful results as well as useful results.

With film I shot a great deal of low light and I convinced myself that I had to have a 5D... then baulked at the cost and decided to wait and see what the E-1 replacement can do & the Leica f1.4 25mm on the E-1...

.. and in reading Big Ga's link, it isn't as cut and dried as I thought it was, although I think that if you do shoot a lot of low light, the 5D is probably better than current Oly offerings.

But still, I would like to see some fast glass from Oly... it would give an extra stop or so.. and the 4/3rds format will always be playing catch-up vs larger chips in this respect (short of a significant jump in sensor technology).

Most don't shoot low light, so it isn't an issue for most. It isn't the single defining issue for me either, I'd just like to see Oly doing its upmost to get the best possible performance under all conditions with a combination of iso improvements and glass speed.

Best wishes-
Andy
--
http://www.flickr.com/photos/andymclean5/
 
Thanks for the feedback everyone.

MY point was, that even 1D from canon generates noise and it is not a habit of Oly's cameras, the shot that I posted is not mine and I am not thinking to move to another system, but what I thought is if that photos had in caemra PPed then it does not look that impressing and specialy the hair of the guy who is holding the ball and how it smashed because of the processing, but if that photos shot raw and just resized for web then again, the quality of the photo is not impressing from a camera that i being sold maybe 5 times more than Oly's cameras, I was also interested to know if any Canon camera would have similar noise result like Oly' when shot raw.

--
http://www.pbase.com/shg2

 
Ask or read Dave Eckel's reviews. He often translates IQ regarding noise into maximum print size.

But, to throw a wrench into this discussion, suppose you need to crop an image?
Any comments? does that mean badly exposed photo at high iso is
noisy no matter what camera you have?
Correct
Or noisy sensor is noisy
whether good or bad exposure at high iso?
Well ... sort of. But bad exposure just makes it worse and good
exposure makes it better. However your comments seem to suggest you
have fallen into the trap of thinking that some other cameras are
somehow 'noise free'.

You should be asking the question: "are all sensors noisy at high ISO"
and the answer is YES
But then the real question is "so what's the difference in real
life between camera models"
And this then becomes a lot more tricky because it depends on lots
of things like absolute noise, visual unpleasantness of the noise
and resolution so when you downsample, the noise gets reduced in
final output.
I asked this question a few weeks ago in the pro forum. I pretty
much knew the general ballpark answer (grin) but I wanted to see if
people could prove what they thought.

http://forums.dpreview.com/forums/read.asp?forum=1014&message=21794833

Unfortunately, nobody really answer my question as to its effect in
normal size prints, so maybe I'll have to do the print tests
myself. Enjoy the read. And for anyone thinking of jumping systems
thinking the noise is suddenly going to go away. You might be in
for an unpleasant reality check (or at least a realisation that
you've spent a whole bucketful of cash for a very small improvement
alongside a whole boatload of disadvantages).
Swings and roundabouts ... swings and roundabouts ....

G.
 

Keyboard shortcuts

Back
Top