E-M5 is still noticeably better at high ISO particularly in dark areas where noise shows up the worst. Too bad. I had hoped Panasonic would push the level of competition up a notch. Nice camera with a good feature set, but waiting to see what the GH3 has in store now in terms of improving image quality.

I am only looking at RAW files. I'll leave it to all the other million sites to look at the in-camera JPEGs .

I think it's in pushing shadows where the E-M5 will prove a bit better. Without pushing, I'm hard pressed to see much difference. I'll show some comparisons when time permits.

__________________
Amin
Mu-43.com Webmaster (Site FAQ | Help Forum | My Disclosures | My Flickriver | My G+ Profile)

Note: Links in this post may be to our affiliates. We get a small referral fee when you buy after clicking our link. Your price is unaffected.

I think what Yohan is saying (and I agree) is that noise pattern in the E-M5 files is more controlled (for lack of a better word). I agree with getting detail out of shadows though.

BTW, I have to respectfully disagree with your assertion (in another thread) that all sensors today excibit the same noise patterns in RAW, differing only on the ISO settings depended noise. I have recent experience only with Canon (APS-C) and Olympus (12Mp and 16Mp) and I have to say that the Sony sensor in the OM-D gives much better looking noise patterns at the same ISO settings. Regretfully I cannot readily provide evidence; it would mean comparing quite similar photos and it's -to be honest- a lot of work to find and process them (crops must be set-up for sensor crop factor, etc).

My assumption -which I have no way to evaluate or prove except empirically- is that different processors implement noise patterns differently, even in RAW. I believe there are noise manipulation routines in RAW processing, dependent on the particular ISO setting. Let's not forget that "ISO" in digital cameras is not a chemical sensitivity index; it's a measure (to put it very simply, I know) of signal amplification.

__________________
http://500px.com/Eyes_Unclouded_Photography
http://eyesunclouded-photography.blogspot.gr/
http://www.facebook.com/eyesunclouded

Okay, disagreement acknowledged . For clarification, my statement specifically was regarding Bayer sensors. The patterns from the Leica M monochrome and Sigma Foveon sensors look different to me. Other notable exceptions include cameras with on-chip smoothing, like the Nikon 1 System cameras, Panasonic LX2, or Pentax K-5 at high ISO. However I've been doing these comparisons for a long time with a lot of cameras, and when I take an "honest" RAW converter like dcraw or Raw Developer and look at processed RAWs from the vast majority of Bayer sensors, the noise patters look the same to me.

Here's an example using the 1st gen Panasonic MFT sensor and the Sony sensor from the Nikon D700: Panasonic G1 - Nikon D700 ISO Shootout

__________________
Amin
Mu-43.com Webmaster (Site FAQ | Help Forum | My Disclosures | My Flickriver | My G+ Profile)

Note: Links in this post may be to our affiliates. We get a small referral fee when you buy after clicking our link. Your price is unaffected.

Thank you for the link!

So, do you think that, apart from on-chip smoothing (which it was my assumption that, more or less, all processors do to a degree), choice of the RAW converter plays a significant part? Meaning "less honest" (sic!) RAW converters?

__________________
http://500px.com/Eyes_Unclouded_Photography
http://eyesunclouded-photography.blogspot.gr/
http://www.facebook.com/eyesunclouded

I think that only a minority of cameras apply on-chip smoothing in a way that affects how "grain like" the noise appears, and some of the ones which do so only filter the color noise. The choice of RAW converter, on the other hand, makes an enormous difference.

Here's an example of LR vs Viewer: Olympus E-P3 Lightroom-Converted RAW and In-Camera JPEG Comparison at High ISO

Lightroom 2 vs LR 3 default NR: http://seriouscompacts.blogspot.com/...reduction.html

An old comparison of LR 1.2 vs C1 v4 Beta and Olympus E-410 high ISO RAW: http://farm2.staticflickr.com/1004/1...cafc614a_o.jpg

Four different developers at their minimum NR settings for the same D700 file: http://farm4.staticflickr.com/3018/2...cf91ac6a_o.jpg

DP1 crop in ACR vs Sigms Photo Pro: http://farm4.static.flickr.com/3066/...68125eb3_o.png

Probably the best example of all: http://forum.getdpi.com/forum/5168-post2.html

__________________
Amin
Mu-43.com Webmaster (Site FAQ | Help Forum | My Disclosures | My Flickriver | My G+ Profile)

Note: Links in this post may be to our affiliates. We get a small referral fee when you buy after clicking our link. Your price is unaffected.

If the raw performance differences are so minimal, then why did't the GX1 get better reviews? I am puzzled.

__________________
Pax,
Richard
--------------
I hate cameras. They are so much more sure than I am about everything. ~John Steinbeck

I am debating getting this camera, but how are the files to work with? At normal iso is it easier to pull shadow detail? I am still using a gf1 and I find it really hard to work with the files when I screw up the exposure.

__________________
Photography isn't about the gear, but the ability to capture a moment

Micro Four Thirds Street Photography Group

Or conversely why was the em5 so acclaimed? Maybe more the convergence of design, sealing, ibis plus mostly comparable IQ?

Sent from my iPad using Mu-43 App

Thanks Amin.

I meant to say I have already looked at the JPEGs (@ISO 1600) and it looks like G5 kept finer details(higher resolution) as can be seen in the map, etc).

I was trying to ask if you can see if G5 has a higher resolution advantage in RAW.

I appologise for not framing the question clearly.