A77 vs A35 RAW comparison

Started Aug 25, 2011 | Discussions
zackiedawg
zackiedawg Forum Pro • Posts: 29,951
Re: Impressions of different cameras with true raw

It is intriguing how good the 5N stuff looks - I went all the way to ISO6400 and it was still looking very good - better not only than the NEX3 & NEX5 and A33, but also better than the A580 & A55 and D7000 by a touch. For JPG, they really seemed to have tuned that 5N wonderfully.

What it makes me wonder is whether the same JPG processing algorithms might be expected on the A65, A77, NEX7, etc in the final production versions. While I have no complaints with the JPGs from my NEX3 & A550, I would welcome the NEX5N's improvements.

Certainly the A77's downsizing ability from the large 24MP will help...if they can also apply a little 5N-style JPG processing magic, they might get even more out of it.

-- hide signature --
 zackiedawg's gear list:zackiedawg's gear list
Sony a6300 Sony DT 18-250mm F3.5-6.3 Sony E 18-55mm F3.5-5.6 OSS Sony E 55-210mm F4.5-6.3 OSS Sony E 16mm F2.8 Pancake +22 more
OP Amateur Sony Shooter Veteran Member • Posts: 5,433
Re: Impressions of different cameras with true raw

Are these images from processed RAW files? If so, very impressive. Looks like a draw to my eyes, so we can take that in comfort knowing A77 is as good as A55/580/NEX-5. Yes there is no breakthrough in terms of IQ but A77 offer whole bunch of usability enhancement, it will be considered as one of top APSC bodies.

dr jim wrote:

A77 ISO 1600 (downsized to match)

Nex 5N ISO 1600

 Amateur Sony Shooter's gear list:Amateur Sony Shooter's gear list
Sony RX100 Sony a77 II Sony 24-70mm F2.8 ZA SSM Carl Zeiss Vario-Sonnar T* Sony 70-200mm F2.8 G Kenko Teleplus Pro 300 AF 1.4x +6 more
dr jim Veteran Member • Posts: 6,284
Re: Impressions of different cameras with true raw

These are just straight out of the Imaging Resource compar-o-meter. I just downsized the A77 image and cropped as close as I could
--
Just for fun!

Jim

 dr jim's gear list:dr jim's gear list
Sony RX100 III Sony RX10 IV Olympus OM-D E-M10 III Olympus E-M1 II Sony a6500 +15 more
rhlpetrus Forum Pro • Posts: 25,685
Thanks, but there's another IQ parameter to check

I also find the G3 has amazing high ISO performance.

At base ISO, well, I think the 16MP APS-C gives better detail, but I based my tests on DPR's RAW samples and there are focus and lenses variations there, so it's hard to decide. Likely same for your tests (which RAW samples did you use? IR's? Lots of lens, aperture and focus variations there as well, can't be used for detail/resolution analysis).

Now, I think you are missing one big point: DR. The Sony sensors, last generation, have better DR at base ISO and likely better shadow recovery abilities than the Panny sensors. Let's see how this one performs. My guess is that it'll be as good as the 16MP one.
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)

 rhlpetrus's gear list:rhlpetrus's gear list
Leica D-Lux (Typ 109) Nikon 1 V1 Nikon D7000 Nikon 1 Nikkor 10mm f/2.8 Nikon AF Nikkor 35mm f/2D +4 more
rhlpetrus Forum Pro • Posts: 25,685
Re: Still unadressed... Dynamic Range?

Not directly related as some may think, depends on how gain for high ISO is done.

I expect DR at base ISO as good as present 16 MP sensor's, close to 14EV, why would it be different?

bendispo wrote:

It seems as if the noise performance is largely comparable, but I'm also still concerned about dynamic range. I do the vast majority of my shooting in outdoor lighting, and dynamic range is just as large of an issue as noise for me.

Would dynamic range roughly be congruent with noise performance?

-- hide signature --

Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)

 rhlpetrus's gear list:rhlpetrus's gear list
Leica D-Lux (Typ 109) Nikon 1 V1 Nikon D7000 Nikon 1 Nikkor 10mm f/2.8 Nikon AF Nikkor 35mm f/2D +4 more
RicksAstro
RicksAstro Veteran Member • Posts: 3,779
Re: Impressions of different cameras with true raw

I did my raw-raw processing to the 5N and they were 100% identical in noise to the C3...no difference whatsoever. Sony did an awesome job with the JPG engine with the 5N apparently, as the results with JPG look fantastic.

zackiedawg wrote:

Try throwing the NEX-5N results in there - because on the comparometer at least in the JPGs, they seem to have taken a massive quantum leap in processing - the 5N compared to the C3 is several orders better at ISO6400...it's not a small difference, but a huge one. I wonder if the RAW shows the same, in which case Sony has really worked on the JPG processing for the 5N - which positively body-slams the A580, NEX-C3, and A77.

-- hide signature --
zackiedawg
zackiedawg Forum Pro • Posts: 29,951
Re: Impressions of different cameras with true raw

Thanks for the effort - that actually is very good and hopeful news for not only current C3 owners, but all Sony APS-C owners, that possibly a firmware update using the new JPG processing algorithms might someday be released. Indeed they seem to have seriously improved the JPG especially in dealing with shadow detail and black/shadow noise removal, as well as fine detail retention.

-- hide signature --
 zackiedawg's gear list:zackiedawg's gear list
Sony a6300 Sony DT 18-250mm F3.5-6.3 Sony E 18-55mm F3.5-5.6 OSS Sony E 55-210mm F4.5-6.3 OSS Sony E 16mm F2.8 Pancake +22 more
Patman888 Regular Member • Posts: 497
Re: Impressions of different cameras with true raw

Sure the G3 had more detail. heh heh

RicksAstro
RicksAstro Veteran Member • Posts: 3,779
images from my playing...

Here's the very raw unprocessed ISO 6400 images (from imaging resource) with the different cameras (all resized down to the nex 5N size):
http://www.ricksastro.com/temp/iso6400rawcomp.jpg

And here are the low ISO images (up upsized to the A77's size):
http://www.ricksastro.com/temp/lowisorawcomp.jpg

No sharpening, no noise reduction, no curves...just demosaic, and linear stretching to normalize each color channel.

The differences are subtle (other than the EP3, which isn't holding up at high ISOs). Basically, go with the camera you like!

RicksAstro wrote:

I did my raw-raw processing to the 5N and they were 100% identical in noise to the C3...no difference whatsoever. Sony did an awesome job with the JPG engine with the 5N apparently, as the results with JPG look fantastic.

zackiedawg wrote:

Try throwing the NEX-5N results in there - because on the comparometer at least in the JPGs, they seem to have taken a massive quantum leap in processing - the 5N compared to the C3 is several orders better at ISO6400...it's not a small difference, but a huge one. I wonder if the RAW shows the same, in which case Sony has really worked on the JPG processing for the 5N - which positively body-slams the A580, NEX-C3, and A77.

-- hide signature --
RicksAstro
RicksAstro Veteran Member • Posts: 3,779
Re: Impressions of different cameras with true raw

Patman888 wrote:

Sure the G3 had more detail. heh heh

Here you go...all images upsized to match the A77's pixel resolution, so it's about as fair as possible. No sharpening or noise reduction whatsoever. I'd say it's about a draw between the A77 and G3 as far as detail resolution, with the G3 having a little more contrast. Probably down to lens selection more than sensor.

See the file names for which camera is which and what ISO each was shot at.

-- hide signature --
Iliah Borg Forum Pro • Posts: 24,569
IR "test" shots

RicksAstro wrote:

I was examining IR's test shots.

Different focus, different effective exposure, different lenses, different light - what to compare?

-- hide signature --
powertoold Contributing Member • Posts: 505
Re: Impressions of different cameras with true raw

Yea, and the 5N has slightly BETTER detail resolution than the A77.

This is laughable. Basically, if you use the A77, you're working with 5N files that are 1.5 times larger and more cumbersome to deal with. You're also sacrificing the frame buffer.

Sony's decision to march up the megapixel path is dumb to say the least.

powertoold Contributing Member • Posts: 505
Re: IR "test" shots

Same lighting, shot at f8 (minimize focusing issues), as close to the same exposure as possible, yes different lenses (but there aren't that many to choose from, they probably go with the most common primes):

You can definitely compare.

Iliah Borg Forum Pro • Posts: 24,569
Re: IR "test" shots

powertoold wrote:

Same lighting

No, it is not - I guess you have not checked.

shot at f8 (minimize focusing issues)

Quite naive.

as close to the same exposure as possible

Different cameras have their exposure meters calibrated differently.

You can definitely compare.

Are you a scientist?

-- hide signature --
powertoold Contributing Member • Posts: 505
Re: IR "test" shots

Wow? Yes, all cameras measure ISO differently and have different exposure meters, but if you look at the end result (i.e. the picture), a picture with similar exposure translates into a real world result. Also, if you think f8 isn't adequate to do a sharpness / detail test, you're in lala land. Your standards for a real comparison isn't reachable. Many people use the IR comparometer to compare images, and it has helped them choose the right camera.

Also, IR uses a well-controlled studio to take their comparometer pictures:

http://www.imaging-resource.com/ARTS/TESTS/HMI.HTM

I challenge you to give me an example of where the comparometer has failed at comparing two cameras (i.e. the real world result is better than what the comparometer indicates). Have fun because you won't find an example.

OP Amateur Sony Shooter Veteran Member • Posts: 5,433
Re: Impressions of different cameras with true raw

So there are not RAW files. That doesn't tell much as you scale the size down to much smaller windows (not 100% crop).

dr jim wrote:

These are just straight out of the Imaging Resource compar-o-meter. I just downsized the A77 image and cropped as close as I could
--
Just for fun!

Jim

 Amateur Sony Shooter's gear list:Amateur Sony Shooter's gear list
Sony RX100 Sony a77 II Sony 24-70mm F2.8 ZA SSM Carl Zeiss Vario-Sonnar T* Sony 70-200mm F2.8 G Kenko Teleplus Pro 300 AF 1.4x +6 more
sandy b
sandy b Veteran Member • Posts: 8,954
Re: images from my playing...

Good work. I personally think the A77 is the noisiest of the lot and the nikon the best, but not miles apart, all would make fine images. I think the sensor will be ok when Sony, Pentax and Nikon get done with it. But I think from what I've seen so far, I will stick with d7000, that 16 mp sensor is very good indeed.

 sandy b's gear list:sandy b's gear list
Nikon Coolpix A Nikon 1 J1 Nikon D750 Nikon D7100 Nikon AF-S Nikkor 70-200mm f/4G ED VR +10 more
Roman Stedronsky Regular Member • Posts: 140
Re: images from my playing...

RicksAstro wrote:

Here's the very raw unprocessed ISO 6400 images (from imaging resource) with the different cameras (all resized down to the nex 5N size):
http://www.ricksastro.com/temp/iso6400rawcomp.jpg

And here are the low ISO images (up upsized to the A77's size):
http://www.ricksastro.com/temp/lowisorawcomp.jpg

No sharpening, no noise reduction, no curves...just demosaic, and linear stretching to normalize each color channel.

Please what software have you used? DCRAW? And with what settings?

I would like to convert some RAW samples flying around the internet but I haven't find how to convert files properly (my dcraw output is flat, bright and with poink tone).

 Roman Stedronsky's gear list:Roman Stedronsky's gear list
Minolta DiMAGE A1 Sony SLT-A77 Sony Alpha DSLR-A700 Sigma 50mm F1.4 EX DG HSM Sigma 70-300mm F4-5.6 APO DG Macro +14 more
steelski Senior Member • Posts: 2,555
Re: Still unadressed... Dynamic Range?

rhlpetrus wrote:

Not directly related as some may think, depends on how gain for high ISO is done.

I expect DR at base ISO as good as present 16 MP sensor's, close to 14EV, why would it be different?

Because its a new sensor, with much higher output, (24MP at 12FPS vs 16MP at 10fps). The 16MP sensor is the most flexible sensor released by any manufacturer to date. The Pentax K-5 has a few examples of underexposing the 8 stops, and then being able to recover that detail..... which no other consumer sensor has been able to do to date, (with no banding or color issues). simply by saying "why would it be different" you are ignoring the fact that the previous generation sensors were a HUGE extremely impressive!!!!!

TrojMacReady
TrojMacReady Veteran Member • Posts: 8,729
Re: Still unadressed... Dynamic Range?

Are uou ignoring the obvious fact that those sensors were produced by the same maker? And which sensor in history has performed worse than the one preceeding it?
In other words, there is no solid reason to think it would perform worse.

Keyboard shortcuts:
FForum MMy threads