Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Leica & Alternative Gear | Join Upload & Sell

  

Archive 2015 · NX1 vs. A7R vs. 5DMk2

  
 
AlanD
Offline
• •
Upload & Sell: Off
p.1 #1 · p.1 #1 · NX1 vs. A7R vs. 5DMk2


I just posted a preliminary comparison of the NX1+16-50/2.0-2.8 against Canon 5DMark II+24-70/2.8L II and Sony A7R+FE5/1.8 at SLR Lounge.

http://www.slrlounge.com/samsung-nx1-vs-full-frame-cmos/

The APS-C NX1 definitely beats the 5D Mark II sensor and holds its own against the A7R up to ISO 3200. It actually preserves colors better than the A7R. At higher ISOs, the A7R wins.

Lens wise, the Samsung 16-50/2-2.8 is very sharp in the center, but it doesn't match the 24-70/2.8L II in the corners.

All in all, an impressive showing from Samsung.



Jan 19, 2015 at 02:15 PM
Mescalamba
Offline
• • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · NX1 vs. A7R vs. 5DMk2


Yep, it was only camera where when I checked RAWs I thought "ok, this really can rival some FF". Its still small sensor tho.. Wouldnt mind if Samsung decided to go big.. (eg. made FF variant of NX1).

Just wish it was more "alt friendly" system..



Jan 19, 2015 at 02:20 PM
AlanD
Offline
• •
Upload & Sell: Off
p.1 #3 · p.1 #3 · NX1 vs. A7R vs. 5DMk2


Yeah, not a great alt-friendly system in that the focus peaking is center frame only (as opposed to Sony's). That's a software issue.

You don't get M mount lenses, but you can run F/FD/EF/R/MD/C-Y which still gets you a lot. The same way we are happy with 35mm instead of medium/large format, it's not horrible to run APS-C *if* Samsung can put out a high-end wide angle lens.

Wide angle and image quality are really the only reasons for FF. Samsung could make a ultra-high res FF sensor for other companies, or do something like a 12MP A7S style sensor in APS-C format and still do well.

The real strength of APS-C is always the price advantage. A FF NX1 would probably cost $3-4k for the body and Samsung's lens line-up isn't good enough to justify that.

Edit: Furthermore, I just tested the 16-50/2-2.8S. It won't cover a 35mm image circle.



Jan 19, 2015 at 03:11 PM
Tariq Gibran
Offline
• • • • • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · NX1 vs. A7R vs. 5DMk2


What I saw when I played around with the NX1 raws from Imaging-resource was that resolution was really great - a testament to both the sensor and the Samsung 16-50 - but I also noticed a dramatic reduction of useable dynamic range compared to pretty much any modern Sony sensor (APS or FF) I have used.

Download this raw for instance and take a look at the white columns. To my eye, even using extreme highlight recovery, the image looks a bit "plastic", 2D and flat with no texture or subtle tones in the highlights at all.

http://www.imaging-resource.com/PRODS/samsung-nx1/NX1FAR2WTW.SRW.HTM

Something about the look of the files seems a little unnatural or fake:

http://www.imaging-resource.com/PRODS/samsung-nx1/YSAM_0797-proto.SRW.HTM

Edited on Jan 19, 2015 at 03:56 PM · View previous versions



Jan 19, 2015 at 03:43 PM
sflxn
Offline
• • •
Upload & Sell: Off
p.1 #5 · p.1 #5 · NX1 vs. A7R vs. 5DMk2


Yes, it looks like a promising video camera. For still photographers, it's very unfortunate they went APS-C. FF prices are crashing, and even without the process advances (like copper), the FF sensors have so much more head room and only get better. Olympus, Panasonic, and Fuji didn't get the memo when Canon, Nikon, and Sony started crashing APS-C prices into the hundreds and FF is getting closer to the $1000 mark every month. Samsung didn't get the memo either. Two years from now, it's going to be a hard sell for a crop sensor camera to sell for mid-1000.

We're on the verge of the next-gen FF sensors. Sony already have ones that do full sensor readouts at ridiculous frame rates and higher dynamic range, and Canon is rumored to finally have a competitive FF sensor.



Jan 19, 2015 at 03:54 PM
AlanD
Offline
• •
Upload & Sell: Off
p.1 #6 · p.1 #6 · NX1 vs. A7R vs. 5DMk2


FF always has more headroom (laws of physics) but APS-C is always cheaper to manufacture (laws of physics).

From the big picture, competition is good -- so even if you stick with Canon or Sony, you'll be in good shape. They'll have to bump their timetable to compete.



Jan 19, 2015 at 05:19 PM
cputeq
Offline
• • • • •
Upload & Sell: Off
p.1 #7 · p.1 #7 · NX1 vs. A7R vs. 5DMk2


Thanks for the comparison. The Samsung looks like a pretty good little system, but there's no way I was going to pay their asking price for the NX1. Plus -- and this is just me -- I'm now very hesitant to buy gear from a "quasi" camera company. I realize Samsung is massive and has tons of resources to put out some good stuff, but sometimes I feel these guys just don't "get" it, or don't allocate the love to their photographic divisions (sorta like Sony).


Also, a few tips in case you ever decide to do another comparison -- ISO testing on a circuit card isn't the best target - you're missing lots of fine detail (except the letters) and it's hard to gauge noise performance because most of the colors are black, dark grey, or dark green and they're constantly interrupting one another -- looks like you had access to a color chart, so I would have just used that instead (with maybe a high-detail 'something' in the middle of it)

It looks like the lens was kept at the same focal length (kudos for adapting the lens), which means the Samsung crops look larger than the A7R's crops. I know opinions vary as to whether a FL-equivalent test should be done or just a straight-up 200mm on each camera test, but it might be a bit unfair to the Samsung, because the noisy areas are "enlarged" compared to the other two cameras.

Still ,thanks for the good write-up and samples!



Jan 19, 2015 at 06:28 PM
sflxn
Offline
• • •
Upload & Sell: Off
p.1 #8 · p.1 #8 · NX1 vs. A7R vs. 5DMk2


AlanD wrote:
FF always has more headroom (laws of physics) but APS-C is always cheaper to manufacture (laws of physics).


That's my point about them not getting the memo. APS-C from Canon, Nikon, and Sony are selling for $800 and less. Only Olympus, Panasonic, Fuji, and Samsung think people will pay 2009 prices for cropped sensors in 2015. People can get a 1st gen A7 for less. Samsung will have a small market for video.

It's a nice camera, and I wouldn't mind trying it, but that price is not a try it price. That's a switch your system price.



Jan 19, 2015 at 07:32 PM
AlanD
Offline
• •
Upload & Sell: Off
p.1 #9 · p.1 #9 · NX1 vs. A7R vs. 5DMk2


cputeq wrote:
There's no way I was going to pay their asking price for the NX1.


Unless you use the 4K, it's definitely pricey. Putting the same sensor in a $800 body without 4k video (and "only 9 fps") is aggressive though. It's hard to know what will happen with Samsung. Sometimes if they are TOO massive, they won't allocate their love to the photographic divisions. As an owner of the Canon 60Da, you can tell that Canon is selling a camera with pretty low margins out of love of the field. It's like a modern-day Nikonos.

Also, a few tips in case you ever decide to do another comparison -- ISO testing on a circuit card isn't the best target - you're missing lots of fine detail (except the letters) and it's hard to gauge noise performance because most of the colors are black, dark grey, or dark green and they're constantly interrupting one another -- looks like you had access to a color chart, so I would have just used that instead (with maybe a high-detail 'something' in the middle of it)

Agree. The test scene is actually much more interesting, but it was hard to cut everything down to the format @ SLR Lounge and not overwhelm with too much data. And I was a guest writer -- they have a "real" NX1 review coming up.

It looks like the lens was kept at the same focal length (kudos for adapting the lens), which means the Samsung crops look larger than the A7R's crops. I know opinions vary as to whether a FL-equivalent test should be done or just a straight-up 200mm on each camera test, but it might be a bit unfair to the Samsung, because the noisy areas are "enlarged" compared to the other two cameras.

It's a combo. I tested "platform" by trying to do focal length equivalent. Realistically, you need autofocus if you're going to do 15 fps and if the samsung native lenses were bad, it wouldn't be worth it.

For the sensor tests, i used the same lens. The 70-200/2.8IS II isn't my sharpest lens, but it was the sharpest lens i had with a tripod collar. I felt that would give more consistent results. Ideally, a Leica 180/3.4 would have been the best lens to use and adapt to all 3 platforms.

The argument for equal focal length is that no single lens is perfect at all focus distances. So if I used a zoom lens and zoomed out, it'd be a test of the lens not the sensor. if I used the same lens, but moved further away, it would affect the focus + depth of field. therefore, I tested the sensor in true crop mode. That 1:1 is the fairest test in a way because it lets you really decide for yourself. FF is awesome when you want wide angle. APS-C is better when dealing with tlephoto. I tried downsampling everything to 16MP (1Dx equiv) but TBH, the overall appearance is similar.

I would probably say that you're better off grabbing full res images off another site and downsampling it yourself.

The only thing I didn't test was the 24-70/2.8II *on* the Samsung. That would be interesting.

Still ,thanks for the good write-up and samples!

Thanks. Whether you're a Samsung fan or not, hopefully this lights a fire behind Canon to really invest in sensor technology again. As you can see from that last pic in the article -- I've historically shot Canon over everything else.



Jan 19, 2015 at 08:59 PM
Mescalamba
Offline
• • • •
Upload & Sell: Off
p.1 #10 · p.1 #10 · NX1 vs. A7R vs. 5DMk2


Condsidered for who it is, its probably intended price..

Tho I kinda doubt anyone will switch to pure Samsung. Maybe if we get some Metabones speed boosters first?



Jan 19, 2015 at 09:09 PM
justruss
Offline
• • • • •
Upload & Sell: Off
p.1 #11 · p.1 #11 · NX1 vs. A7R vs. 5DMk2


AlanD wrote:
FF always has more headroom (laws of physics) but APS-C is always cheaper to manufacture (laws of physics).


Not really true... if you understand physics.

At a certain point the smaller overall size of the APS-C sensor (cheaper) is no longer the primary determinant of total cost in the face of serious physical limits when it comes to smaller process/node sizes (more expensive). Take a stroll down CPU/GPU lane for a view of what happens re leakage on tiny circuits and the approaching quantum mechanics scale funk.

It's no surprise to me that the NX1 is around the same price as the A7(r,s,II). APS-C only stays cheaper in the long run if it doesn't attempt to match FF resolution (or as long as it stays below the spot on the curve where the smaller process increases costs more than the smaller overall size decreases costs).

In some ways it is analogous to APS-C lenses needing to resolve detail to a much higher level than their FF or MF siblings. Up to a certain point the materials keep things cheap, but beyond a certain point the complexity/tolerances/physical limits become the driving force behind costs.



Jan 20, 2015 at 04:43 AM
rscheffler
Offline
• • • • • •
Upload & Sell: Off
p.1 #12 · p.1 #12 · NX1 vs. A7R vs. 5DMk2


IMO the primary 'target' camera of the NX1 is the 7DII and perhaps to a lesser extent, the a77II. It seems to be between the two in terms of price...


Jan 20, 2015 at 05:55 AM
SoulNibbler
Offline
• •
Upload & Sell: Off
p.1 #13 · p.1 #13 · NX1 vs. A7R vs. 5DMk2


Neato, you've been slashdotted.
rscheffler, I agree about the target being the 7Dii, It feels that way in the hand.
I kinda hope that Samsung will release another firmware update soon that makes the AF comparisons (as far as control goes) fit that target a little better.

I'm frankly surprised to see the iso 3200 performance so high as compared to the A7R, this has not been my experience at all, however as I mentioned earlier much of that may be down to using LR as opposed to C1.

Nice review though! I hope to see more and a DXO test and proper C1 support soon.



Jan 20, 2015 at 07:13 AM
Steve Spencer
Offline
• • • • • •
Upload & Sell: On
p.1 #14 · p.1 #14 · NX1 vs. A7R vs. 5DMk2


AlanD wrote:
I just posted a preliminary comparison of the NX1+16-50/2.0-2.8 against Canon 5DMark II+24-70/2.8L II and Sony A7R+FE5/1.8 at SLR Lounge.

http://www.slrlounge.com/samsung-nx1-vs-full-frame-cmos/

The APS-C NX1 definitely beats the 5D Mark II sensor and holds its own against the A7R up to ISO 3200. It actually preserves colors better than the A7R. At higher ISOs, the A7R wins.

Lens wise, the Samsung 16-50/2-2.8 is very sharp in the center, but it doesn't match the 24-70/2.8L II in the corners.

All in all, an impressive showing from Samsung.


I am less impressed by the Samsung here. If you look at the table that is provided earlier in the review and look at the ratio of gain (signal) to noise, you will see that rounded to 2 decimal points at ISO 1600, the A7r has a signal to noise ratio of .37 in all three channels, whereas the Samsung has a signal to noise ratio of .21. That is almost a stop better for the Sony full frame. And if you look at the shots, which aren't equalized for size, which is a major problem, for not only noise but also for detail you will see the same basic issue reappears. The Samsung has low noise, but also low detail, whereas the Sony has about the same level of noise (maybe a bit higher) but a lot more detail. When you compare images at high ISO you also have to consider not just noise, but in effect signal to noise ratio or said more photographically detail to noise ratio. If you just want low noise, you can always blast an image with noise reduction, but personally I think that looks terrible. A high ISO image with the same noise and higher detail is a better high ISO image and you can process it so that either you retain the detail with the same noise or you can match detail and have lower noise. I have seen a number of comparisons lately between APS-C and FF that seem to have missed this point. They try to say look they have equal noise, and therefore equal high ISO performance. Yeah, the FF has more detail, but it isn't any better for noise. Well, it could be if you just processed it differently and matched detail, then the FF would have a lot less noise.

What does this say about the Samsung. It actually is a good showing for such a high pixel density APS-C sensor. If I have the math right I think you would expect the FF to have at least double the signal to noise ratio of APS-C (and I think that would represent a 1 stop advantage). The numbers suggest a little less than a stop, so a good showing for the Samsung, but we shouldn't try to pretend that the FF Sony isn't quite a bit better. This analysis also assumes that both Sony and Samsung cheat the same amount on reported ISO (almost all manufacturer's ISO rating don't hold up when they are measured), if they don't then the difference could be larger or smaller.



Jan 20, 2015 at 11:58 AM
AlanD
Offline
• •
Upload & Sell: Off
p.1 #15 · p.1 #15 · NX1 vs. A7R vs. 5DMk2




justruss wrote:
Not really true... if you understand physics.

At a certain point the smaller overall size of the APS-C sensor (cheaper) is no longer the primary determinant of total cost in the face of serious physical limits when it comes to smaller process/node sizes (more expensive). Take a stroll down CPU/GPU lane for a view of what happens re leakage on tiny circuits and the approaching quantum mechanics scale funk.

It's no surprise to me that the NX1 is around the same price as the A7(r,s,II). APS-C only stays cheaper in the long run if it doesn't attempt to match FF
...Show more

okay. Always is the wrong word.

But your die size is still a major factor for sensor costs at least for the next 5 years.



Jan 20, 2015 at 04:03 PM
AlanD
Offline
• •
Upload & Sell: Off
p.1 #16 · p.1 #16 · NX1 vs. A7R vs. 5DMk2




SoulNibbler wrote:
Neato, you've been slashdotted.
...
I'm frankly surprised to see the iso 3200 performance so high as compared to the A7R, this has not been my experience at all, however as I mentioned earlier much of that may be down to using LR as opposed to C1..


heh, I have lost count of the number of times I've been slashdotted. Today's servers are far more robust.

Make sure you use 5.7.1 with the NX1. It has the chromatic aberration fix.

I was surprised too. Using the exact same lens helped me see the diff. Also, it is chroma detail more than luminance.



Jan 20, 2015 at 04:06 PM
Jman13
Online
• • • • • •
Upload & Sell: On
p.1 #17 · p.1 #17 · NX1 vs. A7R vs. 5DMk2


I'm excited to test out the NX1. I've got one coming on Monday with the 16-50 for review.


Jan 22, 2015 at 09:22 AM
curious80
Offline
• • • •
Upload & Sell: Off
p.1 #18 · p.1 #18 · NX1 vs. A7R vs. 5DMk2


Tariq Gibran wrote:
What I saw when I played around with the NX1 raws from Imaging-resource was that resolution was really great - a testament to both the sensor and the Samsung 16-50 - but I also noticed a dramatic reduction of useable dynamic range compared to pretty much any modern Sony sensor (APS or FF) I have used.

Download this raw for instance and take a look at the white columns. To my eye, even using extreme highlight recovery, the image looks a bit "plastic", 2D and flat with no texture or subtle tones in the highlights at all.

http://www.imaging-resource.com/PRODS/samsung-nx1/NX1FAR2WTW.SRW.HTM
......


This does not really indicate a limited DR, its basically a metering + tone curve issue. If the highlights have been blown then there is no way any sensor can recover that regardless of how much DR it has. The cameras which are better at highlight recovery are the ones which use a tone curve with a gentle roll-off at the highlight end. Essentially the tone curves of those cameras allocates a larger portion of overall available DR to the highlight region, allowing you to be less careful about metering / exposure. But there are other cameras which despite having quite a bit of DR, do not allocate enough of that to the highlight region. With those cameras you just have to be more careful with your metering and exposures.



Jan 23, 2015 at 04:39 PM
Mescalamba
Offline
• • • •
Upload & Sell: Off
p.1 #19 · p.1 #19 · NX1 vs. A7R vs. 5DMk2


I have hunch that you might like it..

..thats if you get over size and weight.



Jan 23, 2015 at 04:42 PM
Tariq Gibran
Offline
• • • • • •
Upload & Sell: Off
p.1 #20 · p.1 #20 · NX1 vs. A7R vs. 5DMk2


curious80 wrote:
This does not really indicate a limited DR, its basically a metering + tone curve issue. If the highlights have been blown then there is no way any sensor can recover that regardless of how much DR it has. The cameras which are better at highlight recovery are the ones which use a tone curve with a gentle roll-off at the highlight end. Essentially the tone curves of those cameras allocates a larger portion of overall available DR to the highlight region, allowing you to be less careful about metering / exposure. But there are other cameras which despite having
...Show more

Possibly. With most cameras based on Sony sensors (regardless of brand and including APS format) when a histogram looks as decent as this one does from the example I linked (showing a fairly normal exposure and no clipping), it's fairly easy to recover the highlights. If you go digging into the shadows, noise shows up fairly quickly as well (with the NX1). Thus the issue I initially saw in the highlights does not seem to translate into more useable DR in the shadows in this instance (which would indicate lower DR I believe). I'm very curious to see what DxO finds if they measure the sensor. Working with these NX1 raws feels somewhat to me like working with raws from a smaller sensor than APS (lower DR, more noise). Of course, what I'm seeing could also be an Adobe raw issue with Samsung in particular...



Jan 23, 2015 at 04:57 PM





FM Forums | Leica & Alternative Gear | Join Upload & Sell

    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.