A7S image comaparison review

[OzRay]

Imaging Resource compares the Sony A7S to a number of other cameras, including the GH4 here: http://www.imaging-resource.com/PRODS/sony-a7s/sony-a7s-image-quality.htm. Whilst there's no doubt that the Sony is a great camera, some of the results just seem completely odd, especially considering the quality of images that I can get from the E-M1 (which I understand has the same sensor as the GH4).

Here is a shot of a bear taken at ISO 25600, with a full crop included, and it's no where near as poor quality as the ISO 1600 shot taken with the GH4 (though the subject matter is different):

Subscribe to see EXIF info for this image (if available)

Subscribe to see EXIF info for this image (if available)

 

[MAubrey]

Imaging Resource's comparison photos are all in-camera jpegs, according to the moderator in the comments of that article.

 

[OzRay]

Even so, when I looked at the base ISO, full sized, images from the A7S and GH4 (the cloth especially), the GH 4 images looked out of focus on my monitor. While they shot JPG, the results just don't look right, there's virtually no texture in the shots taken by the other cameras, the Fuji being the worst and the Nikon being the only exception. Looking at those examples, you'd be led to believe that you shouldn't touch any of the others with a barge pole, if the base ISO results are the best that you can expect. It's a shame really, as camera reviewers should be trying to get the best out of every camera they test, not use settings that potentially produce the worst results, depending on how the manufacturer has set the default settings.

 

[Ulfric M Douglas]

... camera reviewers should be trying to get the best out of every camera they test, ...

For whatever resons they appear not to be doing what they should be doing.
It seems to happen a lot.

 

[OzRay]

This has me intrigued, so I did some testing on a somewhat similar product, but with anything but perfect lighting. The following was shot with an E-M1 and 14-35mm f2 lens at f2, illuminated with a 25W bedside lamp about 2m from the pillow, in an otherwise darkened bedroom. The photos show the full shot and a 100% crop. The only post-processing was to convert the ORF into a JPG with LR5 and no adjustments. The images were then cropped and/or reduced in size in CS4.

Imaging Resource's testing procedures are clearly selling short the capabilities of the cameras that they test.

ISO 6400:

Subscribe to see EXIF info for this image (if available)

Subscribe to see EXIF info for this image (if available)

ISO 25600:

Subscribe to see EXIF info for this image (if available)

Subscribe to see EXIF info for this image (if available)

 

[hankbaskett]

It seems that they're shooting to jpeg with a bunch of noise reduction on?

I'm not sure why they feel like that's appropriate, but it might explain the discrepancy in their tests and yours.

 

[fortwodriver]

I thought I was going mad.

My E-M1 doesn't look as bad as they made it out to be...
I shoot RAW (and noise-reduce, like seasoning one's food, to taste).

I'm very impressed at how usable the high-ISO images it makes can be!

 

[tomO2013]

Blanket ISO6400 to ISO6400 comparisons across sensor sizes need to be taken at face value - they fail to take into account real world usage, the impact of IBIS for keeping ISO's lower for longer and also that the 'shallow' depth of field argument that often gets thrown about the place between larger formats and m43 is a double edged sword with regard to high ISO on larger formats. Larger formats typically require you to stop down to keep a subject in focus, negating the theoretical high iso advantage....
I'm just hoping that lots of GH4 owners saw this comparison and now realize that their GH4's are worthless and should be sold off. I'm prepared to help those poor souls out and take a few off their hands for them...

 

[Jonathan F/2]

Blanket ISO6400 to ISO6400 comparisons across sensor sizes need to be taken at face value - they fail to take into account real world usage, the impact of IBIS for keeping ISO's lower for longer and also that the 'shallow' depth of field argument that often gets thrown about the place between larger formats and m43 is a double edged sword with regard to high ISO on larger formats. Larger formats typically require you to stop down to keep a subject in focus, negating the theoretical high iso advantage....
I'm just hoping that lots of GH4 owners saw this comparison and now realize that their GH4's are worthless and should be sold off. I'm prepared to help those poor souls out and take a few off their hands for them...

I have a Nikon D3 with an older 12mp sensor, though it's still very good especially with high ISO AF tracking and good glass. It also blows away my M43 kit when I want to get the FF look especially when paired to the razor sharp Sigma 50mm 1.4 Art.

Yet my E-M10 is my go to camera for street shooting especially in low light. The smaller body, 3 axis IBIS, more DOF while shooting with faster glass and CDAF single shot accuracy are aspects of real world shooting that never get mentioned. I'm sure the A7S sensor is awesome, but I'm kind of cheap and like getting the most bang for the buck while meeting a certain criteria of performance.