Olympus OM-D (EM-5) comparison samples are now.. (continued)

Started Mar 17, 2012 | Discussions
Gianluca101 Forum Member • Posts: 63
Re: Andy...Big Ga's comments is spot on it seems....Re: No, no, no

Thomas_A wrote:

gianlucaficociello wrote:

Nex7 have + 1/3 ev attested in your reviews

E-m5 have -1/3 ev attested in this preview

Side by side those camera's may have same aperture and shutter speed but not same iso

Is it the same to say that both cameras will expose slightly incorrect using the same ISO (under or over-expose by 1/3 EV), but still both are within the ISO specifications of + - 1/3 EV?

I only know that with nex7 I can get iso 3200 - f5,6 - 1/200s when maybe with e-m5 the same exposure it's reached at iso 5000 - f5,6 - 1/200s, and this, according for the only mesurement here at dpreview.

Simon Cowell Senior Member • Posts: 2,543
Re: Olympus OM-D (EM-5) comparison samples are now.. (continued)

Andy Westlake wrote:

panos_m wrote:

So if I understand it correctly the only two variables that you try to keep constant between cameras tested in your studio setup is the rendered middle gray value of the OOC jpeg at standard settings (factory jpeg settings out of the box) and the manufacturer stated iso.

Or stated a little differently you try to align the exposures between cameras based on the rendered middle gray value of the OOC jpeg at standard settings (factory jpeg settings out of the box) at the manufacturer stated iso.

Am I right?

Yes, in terms of exposure.

Sorry to reenact this thread and I really appreciate Andy's effort to clarify the issue.

However, I feel I need to ask this:

Based on the answer above, how valid is it to present the comparison results in terms of the manufacturers's ISO settings when it is known that the ISO 12232:2006 standard allows the manufacturers to set the ISO values arbitrarily and is applicable to sRGB output (not raw) only?

Note, I'm not arguing here that what you do is invalid for a particular manufacturer/camera model, I'm arguing is not valid to compare/present noise results across camera models for a particular ISO value.

Thomas_A Regular Member • Posts: 382
Re: Andy...Big Ga's comments is spot on it seems....Re: No, no, no

So this means both cameras are within specifications according to the ISO standard, but at the very end of the limits of + - 1/3 EV.

This makes a difference of 2/3 E in the dpreview comparometer between the OM-D and Nex-7 (and also Nikon D7000).

Compared to e.g. Panasonic G3 or Canon 600D which are "spot on" there is a difference of about 1/3 EV.

Comparing the detail in RAW between OM-D ISO 1600 and NEX-7 ISO 800 (1 EV difference, I see that the OM-D holds up quite well although there are some moiré. This may be lens-related, but anyway, impressive since the final output is what counts in a system.

T

Andrew Westlake Senior Member • Posts: 2,928
Re: Olympus OM-D (EM-5) comparison samples are now.. (continued)

Simon Cowell wrote:

Based on the answer above, how valid is it to present the comparison results in terms of the manufacturers's ISO settings when it is known that the ISO 12232:2006 standard allows the manufacturers to set the ISO values arbitrarily

That's something of a misapprehension. While the REI definition within ISO12232:2006 sort-of says that, our testing is essentially based on using the Standard Output Specification definition. That's what we report on, and that's how all of our tests are conducted. Most cameras conform to it pretty well in practice.

REI is actually a method of dealing with the vagaries of multi-pattern metering, but our testing doesn't rely on the camera's meter. It also lets manufacturers use context-sensitive tonality-adjustment processing for their JPEGs that can adjust the brightness after the event (called variously Auto Lighting Optimizer, Active D-Lighting, Dynamic Range Optimizer, Auto Gradation, iDynamic, etc.); needless to say we turn all of these features off.

and is applicable to sRGB output (not raw) only?

How does it make any sense to suggest a camera has a different ISO setting dependent upon whether you're shooting RAW or JPEG? The camera doesn't change its metering or anything when you change file format or color space.

Note, I'm not arguing here that what you do is invalid for a particular manufacturer/camera model, I'm arguing is not valid to compare/present noise results across camera models for a particular ISO value.

We test ISO, essentially according to the SOS method, and present the results in every review. All of our further tests are implicitly based upon that, because they're exposed so particular reference grey patches are rendered at a standard brightness. You do need to pay attention to those ISO test results when comparing cameras .
--
Andy Westlake
dpreview.com

 Andrew Westlake's gear list:Andrew Westlake's gear list
Fujifilm FinePix F30 Zoom Olympus Stylus Verve S Olympus XZ-2 iHS Canon EOS 5D Panasonic Lumix DMC-GF1 +11 more
Just Having Fun Veteran Member • Posts: 3,869
Sony and Oly are the same in my test

Thomas_A wrote:

This makes a difference of 2/3 E in the dpreview comparometer between the OM-D and Nex-7 (and also Nikon D7000).

Do you understand that the lighting was different? DPR said over and over they changed the lighting.

I did a quick test with my Sony SLT and an Olympus E-PM1 in the same scene with the same lighting. They were exactly the same in every test. If the Sony was F/5.6, 1/125 and ISO1600, the Olympus was exactly the same. If some one wants to give me an EM5 I will gladly repeat the test.

Louis_Dobson
Louis_Dobson Forum Pro • Posts: 27,494
Re: Amazing. (from a Nikon D7000 owner)

I fine the 7-14 Zuiko, 7-14 Panny, and 14-24 Nikkor, all of which I owned at the same time, to be near indistinguishable. All excellent lenses.

The 12mm f2.0 is very good too, though not quite such a star.

RicksAstro wrote:

I find the wides and ultra wides of m43 are in the same league if not better than the big boys. The Oly 12 f2 and Panasonic 7-14mm are fantastic. Not Nikon 14-24 league quite, but other than that they are up there.

Steve Bingham wrote:

After downloading the raw files, I am amazed. It LOOKS like this micro 3/4 is performing slightly better than my Nikon D7000 (from a resolution and noise at base ISO stand point). Gasp! It looks like itty-biddy cameras are here to stay!!! Now if the micro 3/4 people would develop some exceedingly sharp wides and ultra wides I would be tempted. Although the 50mm f2 used in this test is EXCELLENT, it isn't what I would normally use.

Man, the digital field is really moving!!! I love the D800 but canceled my VERY early pre-order (within minutes of being available from B&H) after considering the weight of the glass and camera. My age and physical condition dictate that I travel light.
--
Steve Bingham
http://www.dustylens.com
http://www.ghost-town-photography.com

 Louis_Dobson's gear list:Louis_Dobson's gear list
Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Fisheye 8mm F3.5 Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +7 more
Brian Mosley Forum Pro • Posts: 20,709
Andy, I hope you're not the E-M5 reviewer...

I'd hate to think the review could be delayed by you giving your time in here

However, it is much appreciated.

Cheers

Brian
--
Join our free worldwide support network here :
http://www.ukphotosafari.org/join-the-ukpsg/
UK, Peak District Local Olympus Safari Group : http://snipurl.com/bqtd7-ukpsg
Keep up with me here : http://twitter.com/alert_bri

 Brian Mosley's gear list:Brian Mosley's gear list
Sony RX1 Sony RX100 IV Olympus E-M1 Fujifilm FinePix X100 Olympus OM-D E-M5 +1 more
Paco 316
Paco 316 Senior Member • Posts: 1,151
So... what is the purpose of this thread?

Maybe someone wants a job at DPreview to show them how it's done?
--

http://www.pueblostudio.com

 Paco 316's gear list:Paco 316's gear list
Canon EOS 20D Canon EOS 40D Olympus E-3 Panasonic Lumix DMC-GH2 Olympus OM-D E-M5 +12 more
John Carson Veteran Member • Posts: 4,253
Confused reasoning...

Bilgy_no1 wrote:

Raist3d wrote:

Continued to answer Andy Westlake:

Raist3d wrote:
Andy Westlake wrote:

A> How does the lighting intensity affect the high ISO noise?

R> If the camera compensates for the exposure with different shutter speed it sure can! A capture of the same final intensity under a longer exposure accumulates more noise.

So, the E-M5 is actually at a disadvantage. Less sensitivity means longer exposure means more noise? How does this prove that Olympus are cheating?

The passage you are quoting refers to variations in lighting, not variations in the reporting of ISO values. Here is what is being said

  1. Holding aperture and ISO constant, lower light means longer exposures are necessary to compensate for the lower light, and these longer exposures (marginally) increases noise.

  2. Holding aperture, reported ISO and lighting constant, overstated ISOs will mean longer exposures are necessary to compensate for the lack of gain being applied to the signal, and these longer exposures mean more light is being received by the sensor, which reduces noise (or, more precisely, boosts the signal to noise ratio).

To state the second point more simply, if you are overstating ISOs, then you are "really" shooting at a lower ISO than claimed and this, as everyone knows, gives less noisy images.

-- hide signature --

john carson

 John Carson's gear list:John Carson's gear list
Fujifilm X-E2 Fujifilm XF 18-55mm F2.8-4 R LM OIS Fujifilm XF 55-200mm F3.5-4.8 R LM OIS
Gregm61 Forum Pro • Posts: 15,664
Re: Jeez..who'd be a DPR tester...

papillon_65 wrote:

it's almost as bad as being a football referee and the people making the most noise on these tests are people who probably won't even buy the camera anyway!

The Pentax SLR forum is really slow, and posts on the Olympus DSLR forum are now staying on page one for up to three days.

This one was inevitable......

 Gregm61's gear list:Gregm61's gear list
Leica M9 Olympus E-M1 Leica M (Typ 262) Olympus E-M1 II Olympus OM-D E-M1X +30 more
OP Raist3d Forum Pro • Posts: 38,615
Re: Olympus OM-D (EM-5) comparison samples are now.. (continued)

Andy Westlake wrote:

panos_m wrote:

[]

Shutter speeds have absolutely minimal impact on noise.

To quip a bit- this depends at what shutter speeds you are talking about. I will agree with you at 1/640, 1/1600, 1/3200 it would be minimal. I don't buy once you are hitting 1/30 vs 1/15, and below that's true and dark noise subtraction of the camera won't help you in a range until you get to 1-4 seconds shutter speeds.

But fair, it's not what is going on here.

If you shoot the same scene at ISO100 1/125sec F16 and ISO100 1/2000sec F4 you really won't see any difference in the noise. To me this is so self evident from everyday shooting I'm actually quite surprised to find myself typing it here. Then again, I did spend several years of my life testing lenses, so I guess have a bit more experience of this than most.

DxOMark's ISO measurements are fine in their own way - they provide a logical framework for DxOmark's own particular method of RAW data comparison. They just bear no relation to our tests based on SOS ISO - the two simply don't intersect. You can't say our ISO tests are wrong because DxO's are different, and you can't reinterpret our data based on DxOMark's measurements. The two testing regimes just use entirely different methodologies.

But there is one problem with the test presented here that DXo's methodology does catch - this particular test the cameras performance seems better than real world low light scenarios. If the intent is to show how camera ISO and noise reduction work, I find that DXo's analysis highlights better how sensors data is really mapped to the dynamic range as to get crushed shadows or better highlights.

So I think in terms of showing the JPEG engine, it's better. In terms of showing what you will be able to do ISO wise with each camera in a real life lower light situation (one of the situations where you would like to crank that ISO), it will show cameras doing a bit more equal along each other than the real world will bear.

This because of the light intensity and I don't mean because of the difference that was there between the EM5 shot and the other cameras. I mean it in general for both.

-- hide signature --

Andy Westlake
dpreview.com

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

papillon_65
papillon_65 Forum Pro • Posts: 27,030
Re: Jeez..who'd be a DPR tester...

Gregm61 wrote:

papillon_65 wrote:

it's almost as bad as being a football referee and the people making the most noise on these tests are people who probably won't even buy the camera anyway!

The Pentax SLR forum is really slow, and posts on the Olympus DSLR forum are now staying on page one for up to three days.

This one was inevitable......

Yep, looks like it. It's funny, I don't remember the same issues being raised when the Pentax K-5 was reviewed/tested, who'd have thought it.
--
It's a known fact that where there's tea there's hope.
Tony
http://the-random-photographer.blogspot.com/

 papillon_65's gear list:papillon_65's gear list
Fujifilm FinePix X100 Sigma DP1 Merrill Sigma DP2 Merrill Sigma DP3 Merrill Sony Cyber-shot DSC-HX50V +7 more
Bilgy_no1
Bilgy_no1 Veteran Member • Posts: 4,880
Re: Confused reasoning...

John Carson wrote:

Bilgy_no1 wrote:

Raist3d wrote:

Continued to answer Andy Westlake:

Raist3d wrote:
Andy Westlake wrote:

A> How does the lighting intensity affect the high ISO noise?

R> If the camera compensates for the exposure with different shutter speed it sure can! A capture of the same final intensity under a longer exposure accumulates more noise.

So, the E-M5 is actually at a disadvantage. Less sensitivity means longer exposure means more noise? How does this prove that Olympus are cheating?

The passage you are quoting refers to variations in lighting, not variations in the reporting of ISO values. Here is what is being said

  1. Holding aperture and ISO constant, lower light means longer exposures are necessary to compensate for the lower light, and these longer exposures (marginally) increases noise.

  2. Holding aperture, reported ISO and lighting constant, overstated ISOs will mean longer exposures are necessary to compensate for the lack of gain being applied to the signal, and these longer exposures mean more light is being received by the sensor, which reduces noise (or, more precisely, boosts the signal to noise ratio).

To state the second point more simply, if you are overstating ISOs, then you are "really" shooting at a lower ISO than claimed and this, as everyone knows, gives less noisy images.

The reasoning was as follows:

  • Olympus overstates ISO

  • As a result, it gives cleaner pictures at a stated ISO

  • This is cheating.

DPR debunk this allegation of cheating, by saying that lighting has been changed in the studio set-up, and that the lower light intensity being compensated by longer exposure shouldn't influence the noise test.

Instead of just accepting this, Raisted than says that longer exposure does influence noise; it gives more noise. I say, if that's the case, then instead of cheating, the Olympus is at a disadvantage compared to the others.

So, what the H3ll are we talking about? Maybe I am confused about who argues for what. But I am not confused about the logics of reasoning, unlike some others. I'm simply astonished at the amount of flack that DPR are getting on the m4/3 forum for publishing test shots of a m4/3 camera that make it look great.

OP Raist3d Forum Pro • Posts: 38,615
No son….

papillon_65 wrote:

it's almost as bad as being a football referee and the people making the most noise on these tests are people who probably won't even buy the camera anyway!

I think the EM5 is fantastic- I have said that by now many times. I just love the concept of reality and using your brain. When I mentioned to dpreview that the E-3 does have fantastic writing card speeds with a Lexar Pro card instead of the San Disk they used in which they found "pedestrian writing speeds" I didn't see you making fun of me

-- hide signature --

It's a known fact that where there's tea there's hope.
Tony
http://the-random-photographer.blogspot.com/

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

OP Raist3d Forum Pro • Posts: 38,615
Because….

papillon_65 wrote:

Gregm61 wrote:

papillon_65 wrote:

it's almost as bad as being a football referee and the people making the most noise on these tests are people who probably won't even buy the camera anyway!

The Pentax SLR forum is really slow, and posts on the Olympus DSLR forum are now staying on page one for up to three days.

This one was inevitable......

Yep, looks like it. It's funny, I don't remember the same issues being raised when the Pentax K-5 was reviewed/tested, who'd have thought it.

The Pentax and the other cameras had the same shutter speed and aperture in the comparison for the most part?

Are you really this dumb?

-- hide signature --

It's a known fact that where there's tea there's hope.
Tony
http://the-random-photographer.blogspot.com/

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

OP Raist3d Forum Pro • Posts: 38,615
Re: No, no, no

Andy Westlake wrote:

Big Ga wrote:

[]

Adobe's philosophy is to match the default brightness of its conversions to the out-of-camera JPEGs, if necessary we may tweak slightly to ensure a perfect match. This is the correct thing to do based on the SOS definition of ISO. I don't understand why you'd want a raw converter to arbitrarily set a different brightness as its start point, as that would mean allowing it to re-rate the camera's ISO all the time, in essence ignoring its metering.

Actually I would because it would allow best use of dynamic range by allowing to shift the data. Remember the E-620 for example uses ISO 200 as an under exposed ISO 100 and the values are brought back up. This has been happening in every Olympus camera since the E-30 all the way to the latest Pen and looks like the EM5 is doing similar. While this fullfills the definition of exposure and ISO, it also means that the dynamic range window shifted up from the E-3 to catch more highlights at the expense of shadows.

This is fantastic for a wedding shot in full daylight, horrible for the reception at night. This is also why the "unexplicable" one stop "DR loss" on the E-30 from ISO 200 to ISO 100- they are one and the same- the DR is the same but it's shifted.

Canon, Pentax, Sony allow us to shift the range in-cameras. This affects how the RAW file will be written as the exposure picked by the camera will vary. It would be great if Olympus offered the option also (they need them more than the others) and a RAW converter that straightforwardly honors the manufacturer's pick doesn't allow for a proper shifting of this.

it seems to me (although there is so much confusion here now that I might be wrong!), that you would need to check the DxO graphs if you were comparing different cameras, to try and glean what ISO values would actually result in specific shutter speeds if you were shooting RAW and using an external lightmeter for example.

No. DxOMarks ISOs will not necessarily agree with an external lightmeter (although with some cameras they might), instead the camera's metering is calibrated to the JPEG ISO (of course). Because of this I don't personally think it makes any sense to refer to a different RAW ISO.

Sorry ... I'm obviously being a bit dim here - can you clarify this bit. What exactly am I needing to note?

When you look at the test images at different ISOs, and compare between cameras, you need to bear in mind the ISO Accuracy tests for each camera. Then you know how they sit relative to each other, based on the ISO definition that we use to set the exposure for these tests.

But what DXo does is give us an accurate idea of the sensor performance, and this will bear in the RAW data as it is pushed and pulled when necessary. This assumes a rather optimal raw converter for that particular camera of course.

-- hide signature --

Andy Westlake
dpreview.com

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

OP Raist3d Forum Pro • Posts: 38,615
No son….

It was clarifying the different exposures that we saw from the files.

You see, Andy clarified the light intensity they used vary as they changed their equipment setup. That's cool. And I was answering to something he replied with.

Without that information, the test would surely have been wrong.

However, as many here, it seems you seem drawn to threads that you find useless for some reason.

Paco 316 wrote:

Maybe someone wants a job at DPreview to show them how it's done?
--

http://www.pueblostudio.com

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

OP Raist3d Forum Pro • Posts: 38,615
Which would make...

a difference in image quality assuming this is at the RAW level.

gianlucaficociello wrote:

Nex7 have + 1/3 ev attested in your reviews

E-m5 have -1/3 ev attested in this preview

Side by side those camera's may have same aperture and shutter speed but not same iso

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

OP Raist3d Forum Pro • Posts: 38,615
Re: Olympus OM-D (EM-5) comparison samples are now.. (continued)

Andy Westlake wrote:

Raist3d wrote:

I don't agree with you we need to get to longer than a second or two to see the difference particularly at high ISO and bad light, but I will agree with your point that at very fast shutter speeds like the ones shown it should not matter. Asking the general question, I would say it can certainly affect it.

So, if I'm reading this correctly, you agree that it doesn't matter if we adjust shutter speeds to compensate for any changes in the intensity of the lighting in this test.

For the fast shutter speeds being used in this test , yes, I agree. I do think the test as presented can be read as wrong because of the issue in discussion- without knowing that the light intensity changed, one may think that one camera sensor is running with a way different sensitivity.

This of course may happen anyway but in this test as presented when seeing the exif, it's a one full stop behind, making the point of comparing the shots at the ISO's pointless - but this is not what you guys did, it just that it would suggest it so.

I will say though and this goes in general for the majority of reviews I am seeing out there- shooting high ISO where the camera needs to close the aperture (high F) and fast shutter speed is hardly a stress on the high ISO ability of the camera. This means that in good light, all cameras tend to do better than the conditions of say F2.8 1/30th and ISO 1600/3200 which are some of the conditions that I find in the real world as a photographer where i want to use those ISO's.

Probably partially because of this: http://forums.dpreview.com/forums/read.asp?forum=1041&message=40936605 (assuming that is correct, at the raw level)

Thank you for taking the time to read and reply, it is appreciated.

-- hide signature --

Andy Westlake
dpreview.com

-- hide signature --

Raist3d/Ricardo (Photographer, software dev.)- "You are taking life too seriously if it bugs you in some way that a guy quotes himself in the .sig quote" - Ricardo

exdeejjjaaaa
exdeejjjaaaa Veteran Member • Posts: 8,263
Re: So... what is the purpose of this thread?

Paco 316 wrote:

Maybe someone wants a job at DPreview to show them how it's done?

if you really want to know how to evaluate sensors, you do not read dpreview - you read people like http://www.dpreview.com/members/9263714680 or http://www.dpreview.com/members/8931023692 but not Westlake or anybody who dares to use Adobe's raw converters to illustrate anything

 exdeejjjaaaa's gear list:exdeejjjaaaa's gear list
Sony a7R II Sony FE 55mm F1.8 Phase One Capture One Pro +25 more
Keyboard shortcuts:
FForum MMy threads