PDA

View Full Version : High ISO E-5 comparison with E-3 and E-PL1


Ian
5th October 2010, 03:36 PM
To kick off the new E-5 forum, here is a, I expect the first of several, high ISO comparison of the new Olympus E-5, the old E-3, and an E-PL1 Pen. All three were taken using a Zuiko Digital 50mm f/2 Macro lens, @f/6.3, 1/8th second, ISO 2000, tripod mounted with IS switched off. These are from camera JPEGs, Large Fine setting, with noise filter set to low.

http://fourthirds-user.com/files/forum/E-5_ISO2000resized.jpg

E-5 complete frame.

http://fourthirds-user.com/files/forum/E-5_ISO2000crop.jpg

E-5 100% crop.

http://fourthirds-user.com/files/forum/E-3_ISO2000resized.jpg

E-3 complete frame.

http://fourthirds-user.com/files/forum/E-3_ISO2000crop.jpg

E-3 100% crop.

http://fourthirds-user.com/files/forum/E-PL1_ISO2000resized.jpg

E-PL1 complete frame.

http://fourthirds-user.com/files/forum/E-PL1_ISO2000crop.jpg

E-PL1 100% crop.

Perhaps the most obvious thing I noticed was how desaturated the E-3 sample looks compared to the others. Reducing saturation is one trick for dealing with chroma (colour) noise.

Comments?

Ian

edmund473
5th October 2010, 08:18 PM
What a difference in the resolution between the E3 and the E5, olympus have certainly reduced the amount of noise thanks for the comparison regards JohnR

wanderer
5th October 2010, 08:36 PM
The brown edge shows it up more.
There is the question of how your pictures interpolate on my screen.
It would be interesting to print them and observe.
Which also brings up the question of how many pictures get printed.

But I'm no technical expert.
I think I may be talking in everdecreasing circles.:rolleyes:

Duncan

iMac
5th October 2010, 09:30 PM
The JPEG's from the E-5 does look good, wonder how the RAW files compare as this would eliminate I suspect all noise removal processing the camera is doing. Ian can you post a 100% crop of the base of that wonderful looking meter? I'd like to see what the dark areas look like for noise.

Thanks for postings these, looking forward to seeing more.

Nick Temple-Fry
5th October 2010, 11:54 PM
What a mixed message these comparisons are.

Looking at the 100% crops of the clockface

Clock/Face - numbers around the edge - well I'd rank them E-5, E-PL1, E-3

Wood - extending up above 0.6/0.7, now thats harder but E-PL1, E-3, E-5, I rate the E-3 higher because it's unsaturated image controls the reds well, whereas the E-5 has more red noise slightly obscuring the detail (follow the three textures of wood up). But the E-PL1 has less red noise obscuring the 3 different textures of wood.

I Suspect the slightly lower 'resolution' and the unsaturated nature of the the E-3 shots favours them.

If I was to draw a conclusion then it would be that the E-PL1 and E-5 are very close, with the E-3 losing out primarily on its resolution of the figures at the edge of the clockface.

Actually I dislike peeking this way, now if Ian can arrange pictures where the foreground needs high'ish iso and the background is indistinct/dull. We'll see better random noise artifacts and any banding.

Nick

Crippledsandwich
6th October 2010, 12:18 AM
There is a very natural feel to the amount of noise on the E5. If only i had the money lol.

Ian
6th October 2010, 09:14 AM
In general, professional cameras apply less sharpening than consumer cameras, so the observation that the E-PL1 has more sharpening applied than the E-5 seems logical. The E-3 result is less saturated, which I feel is part of the Olympus strategy three years ago to control chroma noise. If you turn up the saturation to return the image to the levels of the E-5 and E-PL1, I'm sure you will see more chroma noise. Olympus has improved its edge-sharpening algorithms since the E-3, and even over the E-PL1.

As requested, I will add the venerable E-1 to the mix in some tests later today :)

Ian