Samsung 20MP vs Sony 16MP

I often read that people want a better sensor than the current Samsung 20MP sensor. However when I compare RAW images in the Studio Scene Comparisson on this site the sensor seems to do pretty well against the Sony 16MP sensor. Only with regards to DR there’s quite a big difference. Has anyone here used both a camera with the 20MP Samsung sensor and a camera with the 16MP Sony? How do they compare in real life situations?
--
http://www.flickr.com/photos/klaastuin

I have a K5 and a NX200...

I can tell you the difference is quite big between these two!

With the K5, setting an ISO is pretty much optional... There is nearly no difference between a 1600 iso shot and an underexposed 100 iso pushed by 4Ev in post-processing.

The NX200 does not have that much leeway (but that can be just a RAW file limitation)...

The NX200 also have a tendency to drift toward purple in high-iso (same as the K7 I had).

I've also found that there was no great differences between the NX200 and the NX100, despite the added MP, and in some situations the NX100 even fares better (try lifting shadows with the NX200!)... The NX200 has slightly better high iso, but nothing groundbreaking!

dlacoutu said:

I have a K5 and a NX200...

Click to expand...

I did have a D7000 and a D5100. Same Sony sensor.

dlacoutu said:

I can tell you the difference is quite big between these two!

With the K5, setting an ISO is pretty much optional... There is nearly no difference between a 1600 iso shot and an underexposed 100 iso pushed by 4Ev in post-processing.

The NX200 does not have that much leeway (but that can be just a RAW file limitation)...

Click to expand...

I can easily push my NX1000 files 3-4 stops and see no pattern noise. Pretty much the same as my D7000 and significantly better than the Canons I used.

dlacoutu said:

The NX200 also have a tendency to drift toward purple in high-iso (same as the K7 I had).

Click to expand...

This is not my experience with Raw Therapee or Lightroom 4.1.

dlacoutu said:

I've also found that there was no great differences between the NX200 and the NX100, despite the added MP, and in some situations the NX100 even fares better (try lifting shadows with the NX200!)... The NX200 has slightly better high iso, but nothing groundbreaking!

Click to expand...

My NX1000 is considerably better than my NX10. I do a lot of post processing and lifting shadows is no problem at all. Don't think the NX200 is different.

Maybe you are using the wrong software?

--
Regards,

Robert
http://www.sondek.smugmug.com

judging by the Raw samples on dpr, I can see that the sony does quite better (both for dr and noise) at iso higher or equal than 1600. at lower values, there is not much difference, with samsung possibly having a bit more details at base iso.

overall I prefer Sony rendition, since I like shooting in low light situations, but since nex system has poor lenses, I still hope samsung can improve high iso in the next gen.

de_klaas said:

I often read that people want a better sensor than the current Samsung 20MP sensor. However when I compare RAW images in the Studio Scene Comparisson on this site the sensor seems to do pretty well against the Sony 16MP sensor. Only with regards to DR there’s quite a big difference. Has anyone here used both a camera with the 20MP Samsung sensor and a camera with the 16MP Sony? How do they compare in real life situations?
--
http://www.flickr.com/photos/klaastuin

Click to expand...

I have the same experience as Robert. I shoot exclusively RAW (high ISO JPEG from the Samsung has some offensive noise reduction) with both a D7000 and an NX20. I feel comfortable using both in similar situations. The Sony might have a small lead on it, but maybe 1/3 stop. Nothing that stops me from using either one for any particular situation. Both "feel" like APS-C sensors in terms of handling.

The difference between the NX100 and NX200 in RAW is huge, probably 1 stop improvement in SNR. If you shoot JPEG you might not notice much difference.

Here I show the difference in shadow noise between the by boosting the shadows: http://erphotoreview.com/wordpress/?p=2600

And here is a high ISO comparison:
http://erphotoreview.com/wordpress/?p=2725

Sorry, I didn't include the NX10 in the high ISO comparison as I felt that you can see it does worse in the first comparison, so I thought I would throw in one of the intermediate Sony sensors that were good in low light, but not as good as the current batch.

Notice that although the D7000 is a bit cleaner, neither really looks that much better. They are pretty close. These images are normalized for resolution too. Higher MP sensors always look worse at 100% (you are viewing a smaller area of the sensor). More important to photography is what the image looks like at the image level, as this is what you see in a printed or screen viewed photograph.

Eric

--
I never saw an ugly thing in my life: for let the form of an object
be what it may - light, shade, and perspective will always make it
beautiful. - John Constable (quote)

See my Blog at: http://www.erphotoreview.com/ (bi-weekly)
Flickr Photostream: http://www.flickr.com/photos/28177041@N03/ (updated daily)

I just wanted to upload some test shots, but your comparison is much better.

--
Regards,

Robert
http://www.sondek.smugmug.com

I have a NX200 and a NEX 7 (24 MP) and, despite widely varying subjects and exposures, I see little advantage to the more than doubly priced Sony, other than the convenience of the EVF in bright sunlight. DR and noise at higher ISO appear much the same, for which I find Noise Ninja (included in Corel AfterShot Pro) useful. Generally speaking, Samsung's lenses are at least as good as Sony's and there is more choice.

For the NX200, I have both basic zooms and the 16mm and for the Sony I have the 50/f1.8 and both Sigma lenses.

I used both. The most significant difference between the sensors is the Sony's higher DR. The Samsung has a bit more resolution at low ISO, and the Sony is about 1/2 stop better at ISO 3200 & up. If you shoot raw under normal lighting conditions and use the same processing the results are more or less identical. The main differences between Samsung and Sony cameras are in features, ergonomics and lenses, not image quality.
--
http://www.gridenko.com
http://www.flickr.com/photos/alex_virt/

Thanks for all the input! Looks like the sensors are quite close. The side-by-side-comparisson was really helpfull thanks!
--
http://www.klaastuin.nl

these comparison is quite amazingly done! cheers!

I currently have a 16MP a57 and have owned an NX100 and NX200.

There's more to life than alleged DR, of course, but in general I'd say the NX has slightly more vibrant and accurate colors, punchier base ISO JPEG and ever so slightly more room to pull shadows in underexposed low ISO pics.

The a57 has more highlight headroom vs. clipping most notable at higher ISO, cleaner chroma noise even considering the slight added boost needed to clear the 1/3 light lost to the pellicle mirror. The 16MP is simply a champ of high ISO noise management. That said, the very fine and regular chroma pattern in both 20MP and 14MP NX images is very easily mitigated with Lightroom, requiring just a small bump of the slider in most cases.

I had no such issue with my NX100 or 200, either. Purple? Not unless we're talking really high ISO and hugely pushed.

I don't care much for cameras that apply mandatory NR to RAW.

tecnoworld said:

tecnoworld wrote:

overall I prefer Sony rendition, since I like shooting in low light situations, but since nex system has poor lenses, I still hope samsung can improve high iso in the next gen.

Click to expand...

nex-5n with 50mm f1.8 and maybe 35mm f1.8 ...

or manual 50mm f1.4/f1.2 or rangefinder lenses (voigtländer 35mm f1.4)

expensive lenses though.... except 50mm primes like minolta md or canon fd

I use 16 MP sensor with my K5 and 20mp sensor with NX1000.

The is no huge difference between the two specially High dynamic range, excepted the crop factor (but you need a good lens with 20MP to see the advantage on the 16MP). Using 20-50mm is just the limit for good quality but not the best. Comparing with my old NX100, the crop factor was perfect with this lens (14MP). So keep in mind to use a prime lens with the 20MP sensor or you will be disappointed (specially at high iso).

Here 2 good lens for the NX1000:

-samsung 45mm 1.8

-samsung 30mm 2.0

de_klaas said:

de_klaas wrote:

I often read that people want a better sensor than the current Samsung 20MP sensor. However when I compare RAW images in the Studio Scene Comparisson on this site the sensor seems to do pretty well against the Sony 16MP sensor. Only with regards to DR there’s quite a big difference. Has anyone here used both a camera with the 20MP Samsung sensor and a camera with the 16MP Sony? How do they compare in real life situations?
--
http://www.flickr.com/photos/klaastuin

Click to expand...

I have tried a Sony 16MP raw file (Asked one friend to give me one), and the lighting condition was not ideal, so don't really want to comment on that.

I have downloaded some NX1000 raw samples from Photography blog website and tried them with Lightroom 5. Lifting shadow is very difficult, and even a little sliding of the highlights slider shows the DR is not up to work. I have used Sony RX100 raw samples with the same type of processing, and being a smaller sensor camera, it holds up quite well against Lightroom.

Or maybe I am used to the way D800 holds up in lightroom, but surely if RX100 can do better than a crop sensor Samsung, then the DR may not be up to scratch with other crop sensor camera. Judging by the raw file size, I was expecting samsung to do quite well, but it didn't live up to expectations.

Very strange, my findings are totally different than yours and so are those from dxo, that recently tested the nx300. Have a look at my recent night pics:

http://www.dpreview.com/forums/post/51861593

All of them shot at high iso 1600 to 3200 and a couple of them with exposure further pushed in pp.

tecnoworld said:

tecnoworld wrote:

Very strange, my findings are totally different than yours and so are those from dxo, that recently tested the nx300. Have a look at my recent night pics:

http://www.dpreview.com/forums/post/51861593

All of them shot at high iso 1600 to 3200 and a couple of them with exposure further pushed in pp.

Click to expand...

I can't seem to see any exif's on these images, must have got stripped while uploading.

I really wanted to get a NX1000, and considering the price, it would have been a good purchase. Maybe NX300 sensor is better than NX1000?

Robert N said:

Robert N wrote:

Robert N said:

I can tell you the difference is quite big between these two!

With the K5, setting an ISO is pretty much optional... There is nearly no difference between a 1600 iso shot and an underexposed 100 iso pushed by 4Ev in post-processing.

The NX200 does not have that much leeway (but that can be just a RAW file limitation)...

Click to expand...

I can easily push my NX1000 files 3-4 stops and see no pattern noise. Pretty much the same as my D7000 and significantly better than the Canons I used.

Click to expand...

Try a dark underexposed raw in 100ISO. Then in lightroom, push the exposure from 1-5 stops.

You will quickly see the limitations of NX200/210/20/1000 raw files.

Manzur Fahim said:

mfahim27753 wrote:

Manzur Fahim said:

tecnoworld wrote:

Very strange, my findings are totally different than yours and so are those from dxo, that recently tested the nx300. Have a look at my recent night pics:

http://www.dpreview.com/forums/post/51861593

All of them shot at high iso 1600 to 3200 and a couple of them with exposure further pushed in pp.

Click to expand...

I can't seem to see any exif's on these images, must have got stripped while uploading.

I really wanted to get a NX1000, and considering the price, it would have been a good purchase. Maybe NX300 sensor is better than NX1000?

Click to expand...

NX20 and NX300 are maybe better than NX1000, but not a huge amount. One issue is really long exposures will be noisier, or if you use a native lens you might get some color banding from corrections done to the data to correct periphery illumination. You can run into this issue on Sony too from what I understand.

I think to do a fair comparison you have to have both cameras side by side set in manual and do the same exposure with both. If you look at DxOMark the NEX 5R and NX300 are almost identical in terms of RAW sensor performance.

Eric

best idea is try nx300 if have chance.