My X-E1 vs D7100 High ISO comps

Started Jun 29, 2013 | Discussions
Horshack
Veteran MemberPosts: 4,447
Like?
My X-E1 vs D7100 High ISO comps
Jun 29, 2013

I couldn't find many High ISO comparisons online between these cameras and so decided to do my own.

Both were shot raw under 4800K LED lighting, using the same Voigtlander Ultron 40mm lens at f/5.6 (X-E1 used this adapter). Manually focused, tripod, timer delay, processed in LR4 using all defaults except white balance adjustment. The X-E1 produced a lower brightness image at every ISO even though the exposure and lens were identical; to account for this and make the comparison fair I made the following LR4 exposure adjustments to bring the X-E1 brightness up to the D7100's levels (matched histograms):

ISO 3200 +0.60EV
ISO 6400 +0.55EV
ISO 12800 +1.50EV (X-E1 shot max ISO 6400, +1EV base adj in LR4)
ISO 25600 +2.46EV (X-E1 shot max ISO 6400, +2EV base adj in LR4)

Below are 2 side-by-side comparisons at each ISO, one to demonstrate color+detail in midtones and the other to demonstrate shadow noise. Both were downsampled to 8MP to normalize for resolution differences and make this a sensor-area equivalent comparison. I can make the raw files available if there is any interest. X-E1 is on the left, D7100 on the right.

ISO 3200 Comp #1

ISO 3200 Comp #2

ISO 6400 Comp #1

ISO 6400 Comp #2

ISO 12,800 Comp #1

ISO 12,800 Comp #2

ISO 25,600 Comp #1

ISO 25,600 Comp #2

Fujifilm X-E1 Nikon D7100
If you believe there are incorrect tags, please send us this post using our feedback form.
ChicagoRob
Regular MemberPosts: 142
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jun 29, 2013

Very interesting comparison. I was surprised to see the X-E1's color rendering virtually match the D7100.

Rob

Reply   Reply with quote   Complain
Astrophotographer 10
Senior MemberPosts: 4,728Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jun 29, 2013

Great work.

The Nikon is slightly better here. Its also interesting in that you have quantified just how much Fuji overstates their ISO. It matches my experience in the field with the XE1.

It also shows you just how good Nikon is at getting low light performance out of their sensors. A 24mp sensor (I am assuming here this is a Sony Exmor sensor, perhaps the same as used in the Nex 7) with such low noise in APSc size is amazing.

Greg.

Reply   Reply with quote   Complain
Felix11
Regular MemberPosts: 113Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jun 29, 2013

Nice job! Thanks.

Clearly the D7100 is a little bit ahead in these tests but not by enough to concern me. The sensor in the D7100/D5200 currently king of the hill at DXOmark.

In fact I would imagine that before you downsized the D7100 images the apparent noise levels were the same or even a bit in the favour of the X-E1?

However, the actual size is, of course, important because nobody is going to start printing at 9x8 instead of 8x6 just because they bought a camera with a larger sensor! (or to put it another way your iPad will still have a 10" screen not a 12.5", and my monitor will still be 24" not 29")

 Felix11's gear list:Felix11's gear list
Ricoh GR
Reply   Reply with quote   Complain
Felix11
Regular MemberPosts: 113Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Felix11, Jun 29, 2013

I understand that the X-E1 is overstating the ISO (which makes it look good on review sites).

However, what result do you get if you set the shutter speed and aperture and let the camera chose whatever ISO it wants?

Does the X-E1 give you an accurate rendition of the scene or is the image incorrectly exposed?

If the test is done this way then we can ignore whatever the camera chooses to say the ISO level is and just look at colour, noise, exposure to see if the are pleasing.

 Felix11's gear list:Felix11's gear list
Ricoh GR
Reply   Reply with quote   Complain
Astrophotographer 10
Senior MemberPosts: 4,728Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Felix11, Jun 29, 2013

D7100 and D5200 king of the APSc sized sensors per DXO Mark which has no review of the Xtrans that I am aware of. The full frame D800E though is the highest scoring sensor of all at 96 points.

Greg

Reply   Reply with quote   Complain
Felix11
Regular MemberPosts: 113Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Astrophotographer 10, Jun 29, 2013

You are correct; I was being succinct

 Felix11's gear list:Felix11's gear list
Ricoh GR
Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Felix11, Jun 29, 2013

Felix11 wrote:

I understand that the X-E1 is overstating the ISO (which makes it look good on review sites).

However, what result do you get if you set the shutter speed and aperture and let the camera chose whatever ISO it wants?

Does the X-E1 give you an accurate rendition of the scene or is the image incorrectly exposed?

If the test is done this way then we can ignore whatever the camera chooses to say the ISO level is and just look at colour, noise, exposure to see if the are pleasing.

At equivalent f/stop, shutter speed and ISO the X-E1's meter will show 2/3 stop underexposed vs "properly exposed" on the D7100's meter. This means the X-E1 wont underexpose when you follow the camera's meter (as is typical in normal use) but it does mean the X-E1 will require either 2/3 stop slower shutter or 2/3 stop higher ISO to produce the same output brightness as the D7100. When comparing two sensors it's necessary to use the same absolute exposure (f/stop and shutter) so that each sensor receives the same amount of light. The exposure on X-E1 then must be lifted to match the output brightness to make subjective IQ evaluation possible.

Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to ChicagoRob, Jun 29, 2013

ChicagoRob wrote:

Very interesting comparison. I was surprised to see the X-E1's color rendering virtually match the D7100.

Rob

It's to be expected since both were processed using Adobe's color profile, which normalizes spectral differences against a known target.

Reply   Reply with quote   Complain
Astrophotographer 10
Senior MemberPosts: 4,728Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jun 29, 2013

I wonder what accounts for the wrong ISO on the XE1. Either deliberately overstated for marketing reasons or is it a complication from the Xtrans sensor? Or a loophole in the way ISO is meant to be measured by the manufacturer. For example is it measured before the colour matrix is fitted. Perhaps the Xtrans colour array saps a bit more light but the naked sensor is rated the same as say the same naked sensor in the Sony Nex 5/6?

I mean the Sony Exmor 16.3mp sensor is used in several cameras, Sony Nex 5/6, Pentax K, Nikon D7000.

ISO refers to amplifier gain of the signal. So either the Xtrans colour array and resulting demosaicing robs light and signal or the amp gain on the X series is less than it should be for that ISO to meet the ISO standard (deliberately understating it).

Greg.

Reply   Reply with quote   Complain
Felix11
Regular MemberPosts: 113Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jun 29, 2013

Yes, that makes sense.

My question is whether, if you set the aperture and shutter speed manually but leave the X-E1 set to auto ISO, the camera produces an image of comparable brightness to the D7100 (albeit with a different nominal ISO value) which does not need the exposure manually adjusted in LR before you can then do a noise comparison.

And, if the above is true, whether the resulting noise comparison would be any more valid than the one you have presented here? (Due to there being less human intervention in the process)

Also, does the D7100 allow you to create 16mp images directly (once again removing human intervention)?

Thanks

 Felix11's gear list:Felix11's gear list
Ricoh GR
Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Felix11, Jun 29, 2013

Felix11 wrote:

Yes, that makes sense.

My question is whether, if you set the aperture and shutter speed manually but leave the X-E1 set to auto ISO, the camera produces an image of comparable brightness to the D7100 (albeit with a different nominal ISO value) which does not need the exposure manually adjusted in LR before you can then do a noise comparison.

And, if the above is true, whether the resulting noise comparison would be any more valid than the one you have presented here? (Due to there being less human intervention in the process)

Also, does the D7100 allow you to create 16mp images directly (once again removing human intervention)?

Thanks

Provided the Auto ISO follows the meter as it should then the X-E1 would be expected to bump it up by 2/3 stops. I considered using a higher nominal ISO vs PP-adjustment but at these High ISOs nearly all contemporary sensors max out their analog noise gain advantage somewhere short of ISO 3200, so the difference between higher nominal ISOs vs PP-digital adjustment in terms of noise should be imperceptible.

The D7100 has a 1.3x crop mode which produces approx 15MP images but that shouldn't be used to compare noise performance with other APS-C sensors since it only uses a portion of the sensor.

Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Astrophotographer 10, Jun 29, 2013

Astrophotographer 10 wrote:

I wonder what accounts for the wrong ISO on the XE1. Either deliberately overstated for marketing reasons or is it a complication from the Xtrans sensor? Or a loophole in the way ISO is meant to be measured by the manufacturer. For example is it measured before the colour matrix is fitted. Perhaps the Xtrans colour array saps a bit more light but the naked sensor is rated the same as say the same naked sensor in the Sony Nex 5/6?

I mean the Sony Exmor 16.3mp sensor is used in several cameras, Sony Nex 5/6, Pentax K, Nikon D7000.

ISO refers to amplifier gain of the signal. So either the Xtrans colour array and resulting demosaicing robs light and signal or the amp gain on the X series is less than it should be for that ISO to meet the ISO standard (deliberately understating it).

Greg.

There's been a lot of religious debates about the ISO spec and whether the output brightness from post-gamma JPEGs should be used as the standard (which I believe it is) vs raws. I didn't pay close attention to all the arguments so for me at least it's academic. I used to think Canon gamed their ISOs since they went down significantly during the 7D/5DM2 generation but then they increased them back up for the 5DM3 so I'm not sure if it's marketing or engineering driving the change.

Reply   Reply with quote   Complain
Red5TX
Senior MemberPosts: 1,586
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Astrophotographer 10, Jun 30, 2013

Astrophotographer 10 wrote:

It also shows you just how good Nikon is at getting low light performance out of their sensors. A 24mp sensor (I am assuming here this is a Sony Exmor sensor, perhaps the same as used in the Nex 7) with such low noise in APSc size is amazing.

I believe the D7100 uses a Toshiba sensor, not a Sony.

Reply   Reply with quote   Complain
forpetessake
Senior MemberPosts: 3,893
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Astrophotographer 10, Jun 30, 2013

Astrophotographer 10 wrote:

I wonder what accounts for the wrong ISO on the XE1. Either deliberately overstated for marketing reasons or is it a complication from the Xtrans sensor? Or a loophole in the way ISO is meant to be measured by the manufacturer.

My guess, it's the result of Fuji cooking the raw file to improve the noise profile. I noticed that ISO discrepancy isn't a constant but depends on the scene. I took X-E1 and NEX-5N and shot the same subjects in low light keeping the exposures the same (same f-stop, same shutter) and the difference in auto ISO was from 0 to 2 stops, with 1 stop being an average. I also noticed that the tonal compression was often higher on Fuji, but the colors were stronger coming from Sony. When you try to bring images to look the same in PP the level of noise becomes similar from both cameras, thought the structure of noise is different.

Reply   Reply with quote   Complain
Randy Benter
Senior MemberPosts: 2,093Gear list
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jun 30, 2013

Thanks for this comparison. This is much better than the comparison tool here on DPR, which is often misused and leads to incorrect conclusions.

I agree that Fuji overstates ISO. My test showed 2/3 stop at ISO 3200 and 1 stop at ISO 6400 in raw files, which is similar to your results.

I think the D7100 sensor is the best APS-C sensor available today. When processed properly, it can deliver more resolution and have slightly less noise at high ISO (compared to X-Trans). By increasing to 24MP, moire is less of an issue, so there is no need for an OLPF or special CFA.

 Randy Benter's gear list:Randy Benter's gear list
Leica X2 Fujifilm X100S Leica X Vario Leica M Typ 240 Fujifilm X-E2 +27 more
Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Base ISO DR Check
In reply to Horshack, Jun 30, 2013

Here's a quick test of base ISO DR. I took three exposures; the first at Base ISO (200), the second at Base ISO underexposed by 4 stops and pushed 4 stops in LR4, and the last at ISO 3200. The IQ delta between the second and third demonstrate the base ISO DR (shadow noise) and color fidelity retention, which depicts DR and ISOless'ness of the sensor.

Here are three 100% crops of an 8MP downsampled image from each of the 3.

Crop #1

Crop #2

Crop #3

Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Re: Base ISO DR Check - D7100 equivalents
In reply to Horshack, Jun 30, 2013

These D7100 images were pushed 5 stops (ISO 100 -> 3200) vs 4 stops on the X-E1 (ISO 200 -> 3200)

Crop #1

Crop #2

Crop #3

Reply   Reply with quote   Complain
Alessandro63
Contributing MemberPosts: 787
Like?
Re: My X-E1 vs D7100 High ISO comps
In reply to Horshack, Jul 2, 2013

Oh, wow! Those fuji files look quite "cooked"; I mean color noise reduction: given you have left the LR values at default, could it be the raw itself? Loss of color and bleeding reds and blues... Those blacks/dark greys cannot be as clean as they appear without color noise reduction. Indeed, Nikon has always been the best at that.

I'd really like to have a look at (one of) those raws to see if it is unrecoverably cooked, or if it's LR to apply what it is instructed to apply - by the raw I mean. Easy task: open the file with a "naked" converter in the like of RPP, dcraw or such.

Would you mind uploading one of your files somewhere? Just out of my own curiosity, if it's a problem please don't mind, no problem.

Alessandro

Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,447
Like?
Raw files and raw vs jpeg comparison
In reply to Alessandro63, Jul 2, 2013

Sure, here's a link to the ISO 6400 raw files for both the D7100 and X-E1. This link is only good for 4 more days. If you'd like to see others just let me know.

Also, here's a comparison of an X-E1 ISO 25,600 OOC JPEG vs raw. Since the max raw ISO is 6,400 on the X-E1 the raw was shot two-stops underexposed and pushed in LR4:

X-E1 ISO 25,600 JPEG (left) vs ISO 25,600 equivalent raw processed in LR4 (right)

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads