Samsung NX1 - FF level quality in an APS-C?? Imaging Resource samples...

Neutral said:
dtaylor said:
jrista said:
That's my point. Statistically, the NX1 has lower noise, which leads to richer color and contrast (hence the reason the parts of the NX1 image that are supposed to be dark look dark! :P )

It does have very slightly lower noise, but that is not the reason it's a darker image which is what leads to the difference in our perception of color and contrast.

Again I'll note that with color NR the 7D II ends up looking a tiny bit cleaner, i.e. lower luminance noise. But in the end neither requires a different work flow or more work. Neither has "better data."

My perception is a bit different and I tend to agree with Jrista.
My comparison of RAW samples (no NR at all - no Chroma no Luminance NR) at DP comparison tool which I posted a bit earlier shows that at ISO3200 NX1 RAW samples visually look better/cleaner/crisper than ones from 7Dm2 and as I also noted NX1 noise pattern is visually better and more pleasant than noise pattern of 7Dm2. NX1 is less blotchy , has a bit more higher noise frequencies and a bit more regular - so should be easier to clean up compared to 7Dm2 . Blotchy noise pattern from many Canon Cameras is something that was always irritating me as well as low performance at low contrast details in red channel - just smearing them away. 1Dx fortunately is better in this respect though also suffering a bit.
Also on this samples NX1 is not downsampled/normalized to 7Dm2 resolution which would also add more difference in favor of NX1.

I was interested to see if NX1 could compete with A7S - but so far A7S is far ahead of all the competitors.
Just see here as an example one of test picture done handheld by A7S in extremely low light conditions at ISO20000.
One small screen snapshot and also full image exported from LR


Thanks for the analysis and image samples. I think you nailed it on the head, and it's the same stuff I've been saying for a while now. Noise characteristic matters. It isn't simply the amount of noise, it's how the noise presents. I am right there with you...I really dislike the blotchy color noise of Canon cameras. Give me clean, random, primarily luminance noise, and even if it is slightly higher, I'll be much, much happier. Gaussian luminance noise distribution is so much easier to clean up, and it required minimal application of NR.


One thing regarding the A7s...it is a 12mp camera. I would be very curious to see how the NX1 performs relative to it when the image sizes are normalized. I don't think the NX1 will be better, it is APS-C afterall, and the total light gathering capacity is still limited, despite the BSI design (which, now that more information is out, may actually NOT be an ISOCELL design...not sure why, guess they figured the pixels were big enough to not suffer enough from color crosstalk to warrant the ISOCELL?) I do think, though, that when normalized down to 12mp, the NX1 will approach the A7s's IQ more closely.
 
Upvote 0
Can't really start a poll within the thread, but which of the two images below do people think looks better? Both are crops of ISO6400 shots that were downsampled to 8MP and with minor chroma noise reduction but nothing else. You'll need to click on the image, or even better view it in a new tab, to see the full resolution.

2nZS6Q1.jpg
.
 
Upvote 0
As Jrista mentioned in his previous post it would be interesting to see comparison of normalized to 8mb images from NX1, 7DM2, 5Dm3 and A7S. Attached are ones at ISO6400, ISO12800 and ISO 25600.
Normalized images comparison just illustrates what was said before about NX1 noise characteristics which are better than 7Dm2 and on normalized images comparison it is even more obvious that NX1 is much less noisy at high ISO than 7Dm2 and not very far behind of 5Dm3. To my opinion for general photography NX1 is much more advanced camera and has more potential than 7Dm2 and I definitely I would recommend NX1 in favor of 7Dm2 if someone asks my advice. Also NX1 AF tracking capabilities looks much more advanced to me than 7Dm2 AF system adopted from 1DX. NX1 AF tracking system looks like something borrowed from military object tracking systems used in optical weapon object recognition and targeting system on missiles and military jet fighters. Maybe Samsung does something this area and they used that experience this in NX1
I have feeling that now when semiconductors giants like Samsung and Sony are on the market with the latest imaging sensor technologies and their resources in R&D and manufacturing capabilities and the rapidly increasing speed of new technologies development and implementation Canon just cannot compete any more on the sensor market (which is also semiconductor technology). Even if Canon would develop something in their R&D LAB they still need to manufacture that and this is where is their big weakness (manufacturing processes/technologies etc.) which is very difficult to overcome and they cannot run long in semiconductors technology race (as this is not their primary business), unless they ask some other big player (e.g. Samsung) to manufacture their chips, like Apple was doing for their iPhones and iPads.
I would be interesting to see (just my wish) Samsung coming on MF sensor market and possibly build up their own affordable ML MF camera. They just can buy one of the MF makers (e.g. Hasselblad) and enter market easily with top-notch sensor technologies.
On comparisons shots is also could be seen that 1Dx and A7S are almost on par with each other, maybe A7S a bit better. But practically (from my experience) A7S images are easier to clean and they have more margin for post processing (could be pushed more) before exhibiting any problems.
Also there are couple of comparisons with P645Z.
Images from P645Z almost looks one stop better that 1Dx or A7S at ISO 12800 and ISO25600.
So far nothing can compete with latest MF sensor from Sony for low light performance.
Hopefully we will see soon ML MF camera from Sony.
 

Attachments

  • ISO6400  NX1-7Dm2-1Dx.-A7S.JPG
    ISO6400 NX1-7Dm2-1Dx.-A7S.JPG
    126.5 KB · Views: 302
  • ISO6400  NX1-7Dm2-5Dm3-1Dx.JPG
    ISO6400 NX1-7Dm2-5Dm3-1Dx.JPG
    128.6 KB · Views: 292
  • ISO6400  NX1-7Dm2-5Dm3-A7S.JPG
    ISO6400 NX1-7Dm2-5Dm3-A7S.JPG
    130 KB · Views: 251
  • ISO12800  NX1-7Dm2-1Dx.-A7S.JPG
    ISO12800 NX1-7Dm2-1Dx.-A7S.JPG
    138.2 KB · Views: 256
  • ISO12800  NX1-7Dm2-5Dm3-A7S.JPG
    ISO12800 NX1-7Dm2-5Dm3-A7S.JPG
    143.6 KB · Views: 318
  • ISO12800  NX1-7Dm2-A7S -P645Z.JPG
    ISO12800 NX1-7Dm2-A7S -P645Z.JPG
    140.1 KB · Views: 291
  • ISO25600  NX1-7Dm2-1Dx-A7S.JPG
    ISO25600 NX1-7Dm2-1Dx-A7S.JPG
    157.1 KB · Views: 283
  • ISO25600  NX1-7Dm2-A7S-P645Z.JPG
    ISO25600 NX1-7Dm2-A7S-P645Z.JPG
    152.6 KB · Views: 295
Upvote 0
ajfotofilmagem said:
For my personal taste, the images on the left are better in color, contrast and has more organic grains.

+1

although I don't really care for either of the 2 pics. On my moniitor, both look underexposed by at least 2/3 of a stop. Colors ... too reddish on the left [skin tones], too yellowish on the right.
 
Upvote 0
Neutral said:
My comparison of RAW samples (no NR at all - no Chroma no Luminance NR) at DP comparison tool which I posted a bit earlier shows that at ISO3200 NX1 RAW samples visually look better/cleaner/crisper than ones from 7Dm2 and as I also noted NX1 noise pattern is visually better and more pleasant than noise pattern of 7Dm2.

In the tool the NX1 image is larger and therefore gives a better impression as to sharpness and detail. Psychologically you will always pick the larger image even if the smaller is actually a little better, and even if one image is displayed at two different sizes and you are told they are different cameras. This does not happen when the images are viewed at the same size, even when scaling the 7D2 up.

As to the noise, same as in the IR scene. With no NR the Canon is slightly worse (more color noise). With color NR...which completely eliminates the color noise in ACR btw...the NX1 is sightly worse (more luminance noise, and that's harder to clean up). Never the less, the differences are far too tiny to matter in print.

I would bring up mountains and mole hills but this is more like mountains out of ant hills.
 
Upvote 0
jrista said:
I am right there with you...I really dislike the blotchy color noise of Canon cameras.

Color noise in the 7D2 samples cleans up completely, with no loss of detail, using CNR in ACR. This is true for the NX1 as well.

Cleaning up luminance noise, no matter how random/fine, is always a trade off involving noise and detail.
 
Upvote 0
dtaylor said:
jrista said:
I am right there with you...I really dislike the blotchy color noise of Canon cameras.

Color noise in the 7D2 samples cleans up completely, with no loss of detail, using CNR in ACR. This is true for the NX1 as well.

Cleaning up luminance noise, no matter how random/fine, is always a trade off involving noise and detail.


I have exactly the opposite experience. Per-pixel color noise isn't difficult...the real difficulty with color noise is the blotches...the stuff that spans 50-100 pixel areas. I believe Neutral knows what I'm talking about. It shows up in shadow areas and at high ISO, and it is nearly impossible to clean up without obliterating detail. I've tried using TGVDenoise as well as several multiscale noise reduction routines on it within PixInsight, and it is just NOT easy to clean up, and impossible to clean up without a visible cost to detail (due to it's scale...you have to factor in large pixel areas, so of course it's going to affect detail.)


I never even saw any color noise in my A7r images (I was actually able to reduce the slider from 25 to 10 in LR before any of it, which was WAY deep in the shadows, even began to reveal itself), and cleaning up Lum required shifting the Luminance noise slider in LR to +10. That was it. I have to remove color noise AND luminance noise with some fairly significant edits with all of my Canon images.


Neutrals screenshots clearly demonstrate the lower color noise of the NX1 relative to the 7D II. It does not appear to be as clean as the 5D III or A7s. It's somewhere inbetween the 7D II and 5D III though.
 
Upvote 0
jrista said:
I have exactly the opposite experience. Per-pixel color noise isn't difficult...the real difficulty with color noise is the blotches...the stuff that spans 50-100 pixel areas. I believe Neutral knows what I'm talking about. It shows up in shadow areas and at high ISO, and it is nearly impossible to clean up without obliterating detail. I've tried using TGVDenoise as well as several multiscale noise reduction routines on it within PixInsight, and it is just NOT easy to clean up, and impossible to clean up without a visible cost to detail (due to it's scale...you have to factor in large pixel areas, so of course it's going to affect detail.)

This is what surprised me about the comparison I posted above, the 7D2 is on the left and the NX1 is on the right. From the DPReview low light samples what I'm seeing is that at 6400 (and 12.8K although I didn't post it since the NX1 seems to fall apart more significantly over 6400) the 7D2 definitely looks worse before chroma noise reduction but then the 7D2 looks considerably better after a small amount of chroma noise reduction.
 
Upvote 0
raptor3x said:
jrista said:
I have exactly the opposite experience. Per-pixel color noise isn't difficult...the real difficulty with color noise is the blotches...the stuff that spans 50-100 pixel areas. I believe Neutral knows what I'm talking about. It shows up in shadow areas and at high ISO, and it is nearly impossible to clean up without obliterating detail. I've tried using TGVDenoise as well as several multiscale noise reduction routines on it within PixInsight, and it is just NOT easy to clean up, and impossible to clean up without a visible cost to detail (due to it's scale...you have to factor in large pixel areas, so of course it's going to affect detail.)

This is what surprised me about the comparison I posted above, the 7D2 is on the left and the NX1 is on the right. From the DPReview low light samples what I'm seeing is that at 6400 (and 12.8K although I didn't post it since the NX1 seems to fall apart more significantly over 6400) the 7D2 definitely looks worse before chroma noise reduction but then the 7D2 looks considerably better after a small amount of chroma noise reduction.


Are you reducing the noise on Neutrals screen captures? If so, the NX1 has been downsampled. Reduce noise at full size then downsample, and no matter what you do to the 7D II, the NX1 still look better.
 
Upvote 0
jrista said:
raptor3x said:
jrista said:
I have exactly the opposite experience. Per-pixel color noise isn't difficult...the real difficulty with color noise is the blotches...the stuff that spans 50-100 pixel areas. I believe Neutral knows what I'm talking about. It shows up in shadow areas and at high ISO, and it is nearly impossible to clean up without obliterating detail. I've tried using TGVDenoise as well as several multiscale noise reduction routines on it within PixInsight, and it is just NOT easy to clean up, and impossible to clean up without a visible cost to detail (due to it's scale...you have to factor in large pixel areas, so of course it's going to affect detail.)

This is what surprised me about the comparison I posted above, the 7D2 is on the left and the NX1 is on the right. From the DPReview low light samples what I'm seeing is that at 6400 (and 12.8K although I didn't post it since the NX1 seems to fall apart more significantly over 6400) the 7D2 definitely looks worse before chroma noise reduction but then the 7D2 looks considerably better after a small amount of chroma noise reduction.


Are you reducing the noise on Neutrals screen captures? If so, the NX1 has been downsampled. Reduce noise at full size then downsample, and no matter what you do to the 7D II, the NX1 still look better.

I applied the standard chroma noise reduction to the RAW files and then downsampled to JPGs. The 7D2 definitely looks better at 6400 and 12800; the NX1 has a kind of funky quality to the noise.
 
Upvote 0
raptor3x said:
jrista said:
raptor3x said:
jrista said:
I have exactly the opposite experience. Per-pixel color noise isn't difficult...the real difficulty with color noise is the blotches...the stuff that spans 50-100 pixel areas. I believe Neutral knows what I'm talking about. It shows up in shadow areas and at high ISO, and it is nearly impossible to clean up without obliterating detail. I've tried using TGVDenoise as well as several multiscale noise reduction routines on it within PixInsight, and it is just NOT easy to clean up, and impossible to clean up without a visible cost to detail (due to it's scale...you have to factor in large pixel areas, so of course it's going to affect detail.)

This is what surprised me about the comparison I posted above, the 7D2 is on the left and the NX1 is on the right. From the DPReview low light samples what I'm seeing is that at 6400 (and 12.8K although I didn't post it since the NX1 seems to fall apart more significantly over 6400) the 7D2 definitely looks worse before chroma noise reduction but then the 7D2 looks considerably better after a small amount of chroma noise reduction.


Are you reducing the noise on Neutrals screen captures? If so, the NX1 has been downsampled. Reduce noise at full size then downsample, and no matter what you do to the 7D II, the NX1 still look better.

I applied the standard chroma noise reduction to the RAW files and then downsampled to JPGs. The 7D2 definitely looks better at 6400 and 12800; the NX1 has a kind of funky quality to the noise.


I noticed a slight amount of something when I denoised the 6400 and 12800 images. Not sure what that is. Hopefully it is simply very early versions of the demosaicing support for the NX1 files in ACR/LR.


I have been testing out CaptureOne Pro. CaptureOne does seem to render a better quality noise with Canon files than LR does. I'm not going to say it's groundbreaking, but it doesn't seem to exhibit that horrid blotchy color noise that I loath so much in my Canon files with Lightroom. Sadly, C1 does not seem to have any support for Samsung cameras... :(


I am beginning to think now that Adobe's algorithms in Lightroom are indeed becoming rather dated. They don't render the data in the RAW files as well as they could be. I don't remember who asserted that in the past, but I think they may be right. Here's to hoping Lightroom 6.x gets a much-needed rewrite of the rendering pipeline, one that eliminates the color blotch with Canon files.


I did notice that LensRentals.com now has the NX1 on preorder:


http://www.lensrentals.com/rent/compact/samsung/samsung-nx1


Woo! Not sure that I'll have a chance to try it out before the end of the year, but I'm glad they have it. They have a number of lenses as well. I think I'd give the NX1 and the 16-50 a try at the very least...but since this is something I'm curious about as an alternative to a 7D II for birds, I may hold off until Samsung releases their 300mm f/2.8 lens first. One thing I did learn recently is there do not seem to be teleconverters for the NX line mount yet. That might be an issue for me...if I picked up the 300mm f/2.8 lens, I would want both 1.4x and 2x teleconverters, and not having them would probably be a deal breaker. Maybe if Canon lenses AF well on the NX1 with an adapter...but I suspect that wouldn't be the case for a while. :'(
 
Upvote 0
jrista said:
I noticed a slight amount of something when I denoised the 6400 and 12800 images. Not sure what that is. Hopefully it is simply very early versions of the demosaicing support for the NX1 files in ACR/LR.


I have been testing out CaptureOne Pro. CaptureOne does seem to render a better quality noise with Canon files than LR does. I'm not going to say it's groundbreaking, but it doesn't seem to exhibit that horrid blotchy color noise that I loath so much in my Canon files with Lightroom. Sadly, C1 does not seem to have any support for Samsung cameras... :(


I am beginning to think now that Adobe's algorithms in Lightroom are indeed becoming rather dated. They don't render the data in the RAW files as well as they could be. I don't remember who asserted that in the past, but I think they may be right. Here's to hoping Lightroom 6.x gets a much-needed rewrite of the rendering pipeline, one that eliminates the color blotch with Canon files.


I did notice that LensRentals.com now has the NX1 on preorder:


http://www.lensrentals.com/rent/compact/samsung/samsung-nx1


Woo! Not sure that I'll have a chance to try it out before the end of the year, but I'm glad they have it. They have a number of lenses as well. I think I'd give the NX1 and the 16-50 a try at the very least...but since this is something I'm curious about as an alternative to a 7D II for birds, I may hold off until Samsung releases their 300mm f/2.8 lens first. One thing I did learn recently is there do not seem to be teleconverters for the NX line mount yet. That might be an issue for me...if I picked up the 300mm f/2.8 lens, I would want both 1.4x and 2x teleconverters, and not having them would probably be a deal breaker. Maybe if Canon lenses AF well on the NX1 with an adapter...but I suspect that wouldn't be the case for a while. :'(

I know they've had the NX1 up for pre-order for at least a month or two as I almost tried to order one for testing before an event I shot over the thanksgiving weekend. As for the weird NX1 noise, I can definitely believe that it could have something to do with the LR conversion although I'm not quite sure how DPReview arrived at their DNG files. The craziest part of those results to me is that before color noise reduction the NX1 definitely looks better but after you remove the color noise the 7D2 looks better, at least to my eye. I don't think I've ever seen that happen before. I've seen cameras become essentially equal after you remove the color noise but never swap positions like that.
 
Upvote 0
raptor3x said:
jrista said:
I noticed a slight amount of something when I denoised the 6400 and 12800 images. Not sure what that is. Hopefully it is simply very early versions of the demosaicing support for the NX1 files in ACR/LR.


I have been testing out CaptureOne Pro. CaptureOne does seem to render a better quality noise with Canon files than LR does. I'm not going to say it's groundbreaking, but it doesn't seem to exhibit that horrid blotchy color noise that I loath so much in my Canon files with Lightroom. Sadly, C1 does not seem to have any support for Samsung cameras... :(


I am beginning to think now that Adobe's algorithms in Lightroom are indeed becoming rather dated. They don't render the data in the RAW files as well as they could be. I don't remember who asserted that in the past, but I think they may be right. Here's to hoping Lightroom 6.x gets a much-needed rewrite of the rendering pipeline, one that eliminates the color blotch with Canon files.


I did notice that LensRentals.com now has the NX1 on preorder:


http://www.lensrentals.com/rent/compact/samsung/samsung-nx1


Woo! Not sure that I'll have a chance to try it out before the end of the year, but I'm glad they have it. They have a number of lenses as well. I think I'd give the NX1 and the 16-50 a try at the very least...but since this is something I'm curious about as an alternative to a 7D II for birds, I may hold off until Samsung releases their 300mm f/2.8 lens first. One thing I did learn recently is there do not seem to be teleconverters for the NX line mount yet. That might be an issue for me...if I picked up the 300mm f/2.8 lens, I would want both 1.4x and 2x teleconverters, and not having them would probably be a deal breaker. Maybe if Canon lenses AF well on the NX1 with an adapter...but I suspect that wouldn't be the case for a while. :'(

I know they've had the NX1 up for pre-order for at least a month or two as I almost tried to order one for testing before an event I shot over the thanksgiving weekend.


Hmm, odd. I only saw the Samsung entry in their list of brands show up recently.


raptor3x said:

As for the weird NX1 noise, I can definitely believe that it could have something to do with the LR conversion although I'm not quite sure how DPReview arrived at their DNG files. The craziest part of those results to me is that before color noise reduction the NX1 definitely looks better but after you remove the color noise the 7D2 looks better, at least to my eye. I don't think I've ever seen that happen before. I've seen cameras become essentially equal after you remove the color noise but never swap positions like that.


The thing about the 7D II is that after color noise removal, as in the more speckled finer grained color noise, you have that blotchy color left behind. Again, though...maybe that really is just poor demosaicing algorithms. C1 doesn't seem to do that. Kind of funny, to think that Adobe's RAW engine is now old and potentially less effective than it could be today...but I think that may be the case.


Personally, the large blotchy color left behind after your regular color noise reduction is what bugs me the most. There is very little that can clean that up nicely. If Adobe could fix their RAW engine to NOT produce that in the first place, then one of my biggest complaints about Canon RAW images would be gone. They still wouldn't have the dynamic range, but, at least the data would be cleaner. I don't really want to spend the couple hundred bucks on C1 Pro, as it's workflow doesn't seem as nice to me as Lightrooms, and it has a limited range of DSLR compatibility...but I may jut do that for the IQ.
 
Upvote 0
jrista said:
Personally, the large blotchy color left behind after your regular color noise reduction is what bugs me the most. There is very little that can clean that up nicely. If Adobe could fix their RAW engine to NOT produce that in the first place, then one of my biggest complaints about Canon RAW images would be gone. They still wouldn't have the dynamic range, but, at least the data would be cleaner. I don't really want to spend the couple hundred bucks on C1 Pro, as it's workflow doesn't seem as nice to me as Lightrooms, and it has a limited range of DSLR compatibility...but I may jut do that for the IQ.

I know exactly what you mean, this was the issue i had with the A7 at high ISO. I'm not sure I saw anything like that in the 7D2 samples I played with but I'd need to check again. Do you see this on the 5D3 as well or just 7D2 samples?
 
Upvote 0
try a different raw converter

I've downloaded the new beta 3 of Irridient Developer to try on my older Fuji files and found the new v3 demosaic also to be quite competent; I'm seeing increased detail and accuity from old XE1 files that rival what I get from my XT1.
You might want to compare what it does to deBayer vs DxO, C1 and ACR modules.

mac only tho, I think

http://iridientdigital.com/products/iridientdeveloper_download_v3beta.html
 
Upvote 0
raptor3x said:
jrista said:
Personally, the large blotchy color left behind after your regular color noise reduction is what bugs me the most. There is very little that can clean that up nicely. If Adobe could fix their RAW engine to NOT produce that in the first place, then one of my biggest complaints about Canon RAW images would be gone. They still wouldn't have the dynamic range, but, at least the data would be cleaner. I don't really want to spend the couple hundred bucks on C1 Pro, as it's workflow doesn't seem as nice to me as Lightrooms, and it has a limited range of DSLR compatibility...but I may jut do that for the IQ.

I know exactly what you mean, this was the issue i had with the A7 at high ISO. I'm not sure I saw anything like that in the 7D2 samples I played with but I'd need to check again. Do you see this on the 5D3 as well or just 7D2 samples?


I see it in every Canon file once I start lifting the shadows enough. It usually doesn't take much, a stop and a half. It's pretty bad with the 5D III, it seems milder with the 7D II. I think Canon may have moved to their newer fabs for the 7D II sensor. If Roger Clark is right about the dark current, and if the Q.E. really is 59%, then this is the first sensor from Canon in a long while that is starting to rival Exmor as far as dark current levels go. If Roger is right, it may even be a little better in terms of dark current than an Exmor.


If the color blotch problem is a consequence of the RAW engine, then Canon has certainly made some strides. They eliminated vertical banding and gained a little bit of horizontal banding (but it is soft, so, not nearly as intrusive as what the 5D III has), lowered dark current, and increased Q.E. Read noise is introduced by the readout pipeline, probably primarily by the ADC units Canon uses. So, that is probably something they could fix (basically, anything that reduces ADC frequency should help.)


I am pretty amazed at how clean the NX1 files are though. Very clean, very neutral random noise, much lower than Canon's. I am hoping Chipworks tears apart both sensors and gives us a detailed look at the designs. I'd love to see what's changed at a low level in Canon's sensor, and what Samsung has done with theirs.
 
Upvote 0