Olympus cheating with inflated iso numbers?

Started Mar 16, 2012 | Discussions
DonSantos
Senior MemberPosts: 1,118Gear list
Like?
Olympus cheating with inflated iso numbers?
Mar 16, 2012

gf1 iso 1600 has greater real iso than ep3 3200 iso

 DonSantos's gear list:DonSantos's gear list
Nikon D600 Sony Alpha 7 Voigtlander 40mm F1.4 Nokton Classic Nikon AF-S Nikkor 85mm f/1.8G Sigma 35mm F1.4 DG HSM +1 more
Olympus PEN E-P3 Panasonic Lumix DMC-GF1
If you believe there are incorrect tags, please send us this post using our feedback form.
Rriley
Forum ProPosts: 21,844Gear list
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to DonSantos, Mar 16, 2012

rubbish
nobody cheated

the difference is b/se DxO use a non standard measurement that they call ISO pretty much for their own purposes measuring from the white point instead of 18% grey like everyone else does

Olympus like every other manufacturer has to comply with an agreed standard which applies a different definition to ISO. Olympus like every other manufacturer therefore cant apply to DxOs system of measurement.

please read this from Andy Westlake, indeed the whole thread
http://forums.dpreview.com/forums/read.asp?forum=1022&message=30481253

The 'multiples of a standard' we are dealing with here are not between different manufacturers, who actually all appear to use the same standard, but between them and DxO who use a different one .

Fact is that if you want ISO to mean a value which when combined with shutter speed and aperture gives you the correct exposure, all of the manufacturers ISOs will work, but DxO's won't. Ultimately this is because DxO are using a definition which allows a specific approach to comparing RAW sensor data but tells you nothing about how to operate the camera (as they themselves make quite clear on their site) .

The bottom line is that if you take an E-3, Sony A900 and Nikon D300, point them at a grey card and let them expose it according to their own devices, they'll give the same result despite having different sensor ISOs according to DxOs definition. This isn't DxO being wrong, it's just them choosing a different definition of ISO which makes more sense to them when profiling a camera for DxO Optics Pro

-- hide signature --

Riley

any similarity to persons living or dead is coincidental and unintended
support 1022 Sunday Scapes'

 Rriley's gear list:Rriley's gear list
Canon EOS 5D Olympus E-3 Olympus E-5 Panasonic Lumix DMC-GH2 Olympus E-M1
Reply   Reply with quote   Complain
DylanHoen
Regular MemberPosts: 372
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to DonSantos, Mar 16, 2012

I think, in their previous cameras, ISO 200 was ISO 100 with a different curve applied to the image. This fixed the problem with highlight clipping that the e-510 had.

If you look specifically at the 18% gray point, I think the iso's should match, but outside of that point, there is a curve applied, which makes it hard to compare between camera manufacturers.
--
Victoria, BC, Canada
Olympus E510, 14-42, 40-150, 70-300, 50mm Macro, FL-50R
Canon PowerShot S1 IS, Raynox DCR250 Macro Lens
Canon PowerShot S70
http://www.dylan.hoen.ca/programming/java/depthOfField/

Reply   Reply with quote   Complain
John Carson
Veteran MemberPosts: 4,033Gear list
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to DonSantos, Mar 16, 2012

This subject has been done to death. A brief summary:

  1. There are different ways of measuring ISO. DXOMark and DPR use a different method and DXOMark's method tends to give larger discrepancies.

  2. Olympus does tend to overstate ISO and Panasonic understated it in its early models. However, Panasonic stopped this with (if I remember correctly) the G3.

DPR estimated that Olympus over-stated ISO by 1/6 stop

http://www.dpreview.com/reviews/olympusep3/13

and that the GF1 and GF2 were understated by 1/3 stop,

http://www.dpreview.com/reviews/PanasonicDMCGF2/9

giving a 1/2 stop discrepancy in total.

DonSantos wrote:

gf1 iso 1600 has greater real iso than ep3 3200 iso

-- 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
Reply   Reply with quote   Complain
MAubrey
Senior MemberPosts: 1,479Gear list
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to DonSantos, Mar 16, 2012

Here we go again...
--
--Mike

 MAubrey's gear list:MAubrey's gear list
Olympus OM-D E-M5 Sony Alpha 7 Olympus M.Zuiko Digital ED 9-18mm 1:4.0-5.6 Voigtlander 58mm F1.4 Nokton SL II Voigtlander 35mm F1.2 Nokton +7 more
Reply   Reply with quote   Complain
benwood
Regular MemberPosts: 211Gear list
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to John Carson, Mar 16, 2012

I wouldn't worry at all about 1/6 stop, so long as the exposures are accurate. I shoot my GF1 down 1/3 stop all the time; I guess that partly explains why.

-- hide signature --

-- Ben

 benwood's gear list:benwood's gear list
Sony Alpha DSLR-A700 Panasonic Lumix DMC-GF1 Panasonic Lumix DMC-GX1 Panasonic Lumix G 20mm F1.7 ASPH Panasonic Leica D Summilux Asph 25mm F1.4 +2 more
Reply   Reply with quote   Complain
John Carson
Veteran MemberPosts: 4,033Gear list
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to benwood, Mar 16, 2012

benwood wrote:

I wouldn't worry at all about 1/6 stop, so long as the exposures are accurate. I shoot my GF1 down 1/3 stop all the time; I guess that partly explains why.

No, it doesn't. The reporting of ISO values and the auto exposure algorithm of a camera are independent of each other.

The only relevance of the under- or over-reporting of ISO values is that it means that noise/detail comparisons at the same ISO value can be misleading.

-- 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
Reply   Reply with quote   Complain
amalric
Forum ProPosts: 10,584
Like?
Silly thread...
In reply to DonSantos, Mar 16, 2012

How many should newbies open, before they discover it's rude and that they have no notion of the facts of photographic life?

Suffice to say that the DxO scale is totally arbitrary, that it doesn't help to compare different brand cameras.

That the only way to do it is set two cameras at the same settings, let them shoot the same greycard, and then observe the noise. Is it so difficult to understand?

Am.
--
Photostream: http://www.flickr.com/photos/amalric

Reply   Reply with quote   Complain
DonSantos
Senior MemberPosts: 1,118Gear list
Like?
Re: Silly thread...
In reply to amalric, Mar 16, 2012

It's important to understand because many comparisons aren't done with the same shuter speed and fstop. They mainly compare noise at a certain iso which is the only arbitray setting in a camera.

amalric wrote:

How many should newbies open, before they discover it's rude and that they have no notion of the facts of photographic life?

Suffice to say that the DxO scale is totally arbitrary, that it doesn't help to compare different brand cameras.

That the only way to do it is set two cameras at the same settings, let them shoot the same greycard, and then observe the noise. Is it so difficult to understand?

Am.
--
Photostream: http://www.flickcom/photos/amalric

 DonSantos's gear list:DonSantos's gear list
Nikon D600 Sony Alpha 7 Voigtlander 40mm F1.4 Nokton Classic Nikon AF-S Nikkor 85mm f/1.8G Sigma 35mm F1.4 DG HSM +1 more
Reply   Reply with quote   Complain
cmorse
Regular MemberPosts: 493Gear list
Like?
Re: Silly thread...
In reply to DonSantos, Mar 16, 2012

DonSantos wrote:

It's important to understand because many comparisons aren't done with the same shuter speed and fstop. They mainly compare noise at a certain iso which is the only arbitray setting in a camera.

amalric wrote:

How many should newbies open, before they discover it's rude and that they have no notion of the facts of photographic life?

Suffice to say that the DxO scale is totally arbitrary, that it doesn't help to compare different brand cameras.

That the only way to do it is set two cameras at the same settings, let them shoot the same greycard, and then observe the noise. Is it so difficult to understand?

Am.
--
Photostream: http://www.flickcom/photos/amalric

If you take a hand held light meter and use it to determine what aperture and shutter speed you want at a certain ISO you'll find that you'll get much better results using the camera's ISO rating rather than DxO's. While there are a few people who do some rather specific things with RAW files who might get more use out of the DxO rating, for most peoples' use its little more than an indicator as to whether the camera is more likely to do better in shadows or in highlights.

 cmorse's gear list:cmorse's gear list
Olympus OM-D E-M5 Olympus Zuiko Digital ED 70-300mm 1:4.0-5.6 Olympus Zuiko Digital 14-54mm 1:2.8-3.5 Sigma 105mm F2.8 EX DG Macro Panasonic Leica Summilux DG 25mm F1.4 +3 more
Reply   Reply with quote   Complain
amalric
Forum ProPosts: 10,584
Like?
Yes but it doesn't work that way...
In reply to DonSantos, Mar 16, 2012

Go read the DxO's technical papers about their choice of ISO, among 4 different ones.

And read what Andy Weslake, editor here, did say about that. DxO is completely off the mark with m4/3, possibly with Nikon too.

There are two other threads going on about it, and each ends in total disaster because people are unable to understand that ISO setting is totally conventional, and MFTR defined.

All you can compare between cameras is SNR at MFTR stated ISO. A camera responds more or less well (noise) than another, to the same amount of light. End of the problem.

Am.

PS It has been shown that Oly cameras expose exactly the same of other cameras for the same stated ISO. It's only with some RAW developers that they might seem underexposing.

Thus you relight, and get more noise. But not so if you have hard/soft denoising, like the E-M5, which seems to have very little noise.

Therefore the only thing that matters is SNR at stated ISO. QED.

-- hide signature --
Reply   Reply with quote   Complain
Banger
Contributing MemberPosts: 579Gear list
Like?
Re: Olympus cheating with inflated iso numbers?
In reply to DonSantos, Mar 16, 2012

DXO measurments are totally rubbish (IMO), and difficult to relate to the real world of photography and image quality. Whilst I have some minor misgivings about the DPR approach to image quality and camera (inc. sensor) measurements, they are well ahead of the meaningless drivel that DXO produce.

Ignore them.....Rgds, rob

-- hide signature --

Give a wise man instruction and he is yet wiser !

Olympus E500, E520, E30, 14-45mm, 40-150mm, 11-22mm, 14-54mm, 70-300mm, 25mm 'Pancake' , FL36R

 Banger's gear list:Banger's gear list
Olympus E-30 Olympus PEN E-P3 Olympus E-M1 Olympus 12-40mm F2.8
Reply   Reply with quote   Complain
DonSantos
Senior MemberPosts: 1,118Gear list
Like?
Re: Yes but it doesn't work that way...
In reply to amalric, Mar 16, 2012

Am I crazy? No.

Here is the simple plain truth no matter what dxomark says. exifs don't lie. Both shot with the same lens and lighting setup. These are raw conversions. I linked the jpg to only show exif. Both have the same ev shutter and f-stop. See the difference in iso???

e-p3: 1/1,000 sec, f/6.3, ISO 3200

http://regex.info/exif.cgi?imgurl=http%3A%2F%2Fwww.dpreview.com%2Freviews_data%2Foly_ep3%2Fboxshot%2Fp7191068.acr.jpg

gf2: 1/1,000 sec, f/6.3, ISO 1600

http://regex.info/exif.cgi?imgurl=http%3A%2F%2Fwww.dpreview.com%2Freviews_data%2Fpanasonic_dmcgf2%2Fboxshot%2Fp1010242.acr.jpg

 DonSantos's gear list:DonSantos's gear list
Nikon D600 Sony Alpha 7 Voigtlander 40mm F1.4 Nokton Classic Nikon AF-S Nikkor 85mm f/1.8G Sigma 35mm F1.4 DG HSM +1 more
Reply   Reply with quote   Complain
John Carson
Veteran MemberPosts: 4,033Gear list
Like?
Re: Yes but it doesn't work that way...
In reply to amalric, Mar 16, 2012

amalric wrote:

And read what Andy Weslake, editor here, did say about that. DxO is completely off the mark with m4/3, possibly with Nikon too.

DPR review reaches the same conclusions about understated and overstated ISO values, albeit by different amounts, so it is hard to justify the claim that DXOMark is "completely off the mark with m4/3".

All you can compare between cameras is SNR at MFTR stated ISO. A camera responds more or less well (noise) than another, to the same amount of light. End of the problem.

Which entirely misses the point that if you shoot at the same ISO, achieve a similarly exposed final image, and ISO is misreported, then the two cameras will not be getting the same amount of light.

PS It has been shown that Oly cameras expose exactly the same of other cameras for the same stated ISO. It's only with some RAW developers that they might seem underexposing.

Source?

-- 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
Reply   Reply with quote   Complain
amalric
Forum ProPosts: 10,584
Like?
Re: Yes but it doesn't work that way...
In reply to DonSantos, Mar 16, 2012

DonSantos wrote:

Am I crazy? No.

Here is the simple plain truth no matter what dxomark says. exifs don't lie. Both shot with the same lens and lighting setup. These are raw conversions. I linked the jpg to only show exif. Both have the same ev shutter and f-stop. See the difference in iso???

e-p3: 1/1,000 sec, f/6.3, ISO 3200

http://regex.info/exif.cgi?imgurl=http%3A%2F%2Fwww.dpreview.com%2Freviews_data%2Foly_ep3%2Fboxshot%2Fp7191068.acr.jpg

gf2: 1/1,000 sec, f/6.3, ISO 1600

http://regex.info/exif.cgi?imgurl=http%3A%2F%2Fwww.dpreview.com%2Freviews_data%2Fpanasonic_dmcgf2%2Fboxshot%2Fp1010242.acr.jpg

So what? Do they have the same noise or not? My PL3 goes up to 12800. Do you think it's a real ISO?

LOL. Wake up.

Am.

-- hide signature --
Reply   Reply with quote   Complain
amalric
Forum ProPosts: 10,584
Like?
Re: Yes but it doesn't work that way...
In reply to John Carson, Mar 16, 2012

John Carson wrote:

amalric wrote:

And read what Andy Weslake, editor here, did say about that. DxO is completely off the mark with m4/3, possibly with Nikon too.

DPR review reaches the same conclusions about understated and overstated ISO values, albeit by different amounts, so it is hard to justify the claim that DXOMark is "completely off the mark with m4/3".

But it's very easy to prove that a Pen exposes perfectly at a stated ISO. So what matters?

BTW DPR measures DR in a different way of DxO, and are right on the spot for Oly cameras, contrary to DxO, which always underrates it. QED.

All you can compare between cameras is SNR at MFTR stated ISO. A camera responds more or less well (noise) than another, to the same amount of light. End of the problem.

Which entirely misses the point that if you shoot at the same ISO, achieve a similarly exposed final image, and ISO is misreported, then the two cameras will not be getting the same amount of light.

That's entirely your assumption based on circular reasoning - LOL

PS It has been shown that Oly cameras expose exactly the same of other cameras for the same stated ISO. It's only with some RAW developers that they might seem underexposing.

Source?

Just look in the other threads. The proof I have is that when ACR wasn't yet ready for my PL1 or PL3, Raw Photo Processor could open the file but it underexposed it. It wouldn't honour the MFTR's tone curve.

Am.

-- hide signature --
Reply   Reply with quote   Complain
John Carson
Veteran MemberPosts: 4,033Gear list
Like?
Re: Yes but it doesn't work that way...
In reply to amalric, Mar 16, 2012

amalric wrote:

John Carson wrote:

amalric wrote:

And read what Andy Weslake, editor here, did say about that. DxO is completely off the mark with m4/3, possibly with Nikon too.

DPR review reaches the same conclusions about understated and overstated ISO values, albeit by different amounts, so it is hard to justify the claim that DXOMark is "completely off the mark with m4/3".

But it's very easy to prove that a Pen exposes perfectly at a stated ISO. So what matters?

"Easy to prove that a Pen exposes perfectly". You are so full of cr*p.

Perfect exposure or not, the accuracy of stated ISO matters when you compare two cameras at the same stated high ISO and draw conclusions about the relative quality of the two shots.

If one camera's ISO is over-stated relative to the other camera's ISO, then comparing them both at, say, a stated value of ISO 1600 makes the first camera look better than it is. The first camera will have a longer exposure or wider aperture — i.e., it receives more light — and apply less gain, making it an unfair comparison.

BTW DPR measures DR in a different way of DxO, and are right on the spot for Oly cameras, contrary to DxO, which always underrates it. QED.

I have already noted that the two organizations measure ISO differently but that both conclude that Olympus over-states ISO , albeit not by much in the case of DPR. I gave a link in my first post in this thread.

Which entirely misses the point that if you shoot at the same ISO, achieve a similarly exposed final image, and ISO is misreported, then the two cameras will not be getting the same amount of light.

That's entirely your assumption based on circular reasoning - LOL

Instead of congratulating yourself on your cleverness, you should try a lot harder to be logical. At present, you are failing dismally. My statement is correct and you appear to be clueless.

PS It has been shown that Oly cameras expose exactly the same of other cameras for the same stated ISO. It's only with some RAW developers that they might seem underexposing.

Source?

Just look in the other threads. The proof I have is that when ACR wasn't yet ready for my PL1 or PL3, Raw Photo Processor could open the file but it underexposed it. It wouldn't honour the MFTR's tone curve.

So no evidence at all other than your confused reasoning. Just as I suspected.

-- 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
Reply   Reply with quote   Complain
John Carson
Veteran MemberPosts: 4,033Gear list
Like?
Re: Yes but it doesn't work that way...
In reply to amalric, Mar 16, 2012

amalric wrote:

DonSantos wrote:

Am I crazy? No.

Here is the simple plain truth no matter what dxomark says. exifs don't lie. Both shot with the same lens and lighting setup. These are raw conversions. I linked the jpg to only show exif. Both have the same ev shutter and f-stop. See the difference in iso???

e-p3: 1/1,000 sec, f/6.3, ISO 3200

http://regex.info/exif.cgi?imgurl=http%3A%2F%2Fwww.dpreview.com%2Freviews_data%2Foly_ep3%2Fboxshot%2Fp7191068.acr.jpg

gf2: 1/1,000 sec, f/6.3, ISO 1600

http://regex.info/exif.cgi?imgurl=http%3A%2F%2Fwww.dpreview.com%2Freviews_data%2Fpanasonic_dmcgf2%2Fboxshot%2Fp1010242.acr.jpg

So what?

It has been explained multiple times to you but you just can't or won't understand it.

If the ep-3 at ISO 3200 and the GF2 at ISO 1600 apply similar gain, then for comparisons of the high ISO performance of the two cameras, you should compare the ep-3 at ISO 3200 and the GF2 at ISO 1600. You should not compare them both at ISO 1600 or both at ISO 3200. That is the point.

-- 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
Reply   Reply with quote   Complain
MadsR
Senior MemberPosts: 2,235Gear list
Like?
Being silly is also a skill I guess
In reply to DonSantos, Mar 16, 2012

DonSantos wrote:

gf1 iso 1600 has greater real iso than ep3 3200 iso

What you get here is the raw measuring... What Olympus and Panasonic do different here is the default curve used to make the JPG image. Do read up on the ground rules first... ISO is the specification for correct exposure for the final JPG image, how you get there is difference between manufactorers...

It is well known that Olympus uses a different tone-curve, so they don't have to amplify the signal as much (This is what is proven by this chart, Olympus amplifies less for the same ISO) They use more of the upper values to create the final image enhancing contrast and sharpening to get the result, that should result in less noise than using raw amplification because you get more noise in lower areas.

This is worth noting if you, like me, shoot RAW and don't use Olympus software to process the RAW files, then there are also a number of other things to note, like never use "off" ISO values (only 200, 400, 800, 1600 etc.) and the higher your ISO the more you should ETTR/overexpose to get the result you want. However you should also notice that headroom is not much, because Olympus already uses most of the lighter tones to generate the JPG that is used to show histogram etc. You can get more detail by a sharper default curve in the higher area, but burned out areas will not come back.

 MadsR's gear list:MadsR's gear list
Olympus PEN E-P1 Panasonic Lumix G 20mm F1.7 ASPH +2 more
Reply   Reply with quote   Complain
oluv
Senior MemberPosts: 1,972Gear list
Like?
already known for long
In reply to DonSantos, Mar 16, 2012

this is quite an old issue that came up when G1 was introduced. the G1 had iso-values overstated even compared to the D300 from nikon.

i have a GF2 and E-PL3 and if i shoot the E-PL3 at iso3200 i can shoot the same scene with the GF2 at iso1600 to get the same exposure. noise levels are also comparable, but you should always compare one iso-step lower from the GF2:

 oluv's gear list:oluv's gear list
Panasonic Lumix DMC-GM1 Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH Panasonic Leica Summilux DG 25mm F1.4 Olympus M.Zuiko Digital 45mm 1:1.8 +1 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads