You're not wrong. I think the importance of that issue has been over-exaggerated.

@Pete_Murrell

People who don't care for process RAWs can just use Olympus's JPEGs in SHQ (very low compression, I think it's like something about 1:1.6 compression ratio) and turn the NR to the lowest setting that particular model allows (some model is low, some is none).

Another route is just batch process all the RAW files into Olympus Viewer (or Olympus Master or Olympus Studio). That is, just automatically save all RAWs into JPEGs but zero post-processing (I personally program my batch process so it auto corrects geometric distortion, turns NR to none, but nothing else. Yes, Olympus Viewer can do automated batch process). That way you preserve the beautiful Oly color and IQ of RAWs with just a few clicks. If you prefer less sharpness, you can program that too.

@SunLyte

You're not wrong about your analysis, but it's the same argument why Olympus says 12-bit is enough. The reason is (*drum rolls please*) a typical LCD monitor people use are 8-bit only. Most people, won't notice any IQ difference between RAW and JPEGs (with very low compression) unless they use a professional monitor, which specifically states it has 10-bit process engine and wide-gamut (97% sRGB and up..., yes, even pro models may not have 100% sRGB gamut, let alone AdobeRGB gamut). I rarely see any monitor that has 12-bit process nor 14-bit ones (those can cost $1,000+).

Most people usually capture their photos in 72dpi, which is fine enough for LCD monitor, but not print quality 300dpi. I really think this subject is really moot point for most casual photographers.

Louis, I thought of you when I read on another site that the M5's base ISO of 200 is in reality 120. I recall that you take issue with the upward-creeping ISO situation, given that you value low ISO pic quality for landscapes (as do I).

@ Antony John:

DXO Mark results:

Canon 50mm 1.4 : 1.6TStops
Nikon 50mm 1.4: 1.6TStops
Zeiss 50mm 1.4: 1.6TStops
Sigma 50mm 1.4: 1.7 TStops
Sony 85mm 1.4: 1.6 TStops
Sigma 85mm 1.4: 1.6 TStops
Nikon 35mm 1.4: 1.7TStops

So where is that huge difference?

The difference between real and nominal aperture is much significant, that difference is transmission. Look at the lens patents. The real aperture of an 1.4 lens is about 1.35 to 1.45.

@ ogo:

There is no connection between histogram and a cropped image where we try to find details and define the noise.

Download two images, load them into an image editor software, and measure the pixels at the same area. They are very similar, but exposure values show 2/3 EV difference.

@ Manmachine, quick look, Panny 45-175 5.2 TStop.
And it doesn't have "15 to 20 elements"

manmachine242,

More white on image means more incoming light and thus less exposure time, this is what I say. You can't say ISO is cheated if you don't expose the same quantity of light.

Also look at the G1X (Canon) samples which is 4:3 ratio. They have the same exposure value than the E-M5 at 6400 ISO. Did Canon also cheat ?

"More white on image means more incoming light"

:D

http://en.wikipedia.org/wiki/Light_meter

@ Antony John:

Do you really know what that word "aperture" means?

Panasonic 45-175 is an 1:4.0 - 5.6 lens
The TStops: 4.3 - 5.7, the 5.2 is an average value.

Click on "measurements" and click "transmission" after.
Look at the bottom diagram called "T-Stop difference to manufacturer"
All that you see is the effect of entrance pupil/focal length ratio, not the effect of reflection on surfaces.

And now, do the same thing with a prime lens.

The only thing I can do is repeat myself: Your examples show that you simply do not understand the simplest basics of optics, like aperture etc.

manmachine242, I give up with you if this can help your ego. Have better things to do :)

PS: I know what is aperture, light, light meter because I use them to make photos :) I also know that if I shoot a white panel, the exposure time will be less that if I shoot a grey one (perhaps even more with a black one LOL). You do not seem to agree, you must be right :)

@ Antony John

You could not show any example from DXO where difference in transmission more than 0.1 EV with similar lenses, but you talk about 0.7EV effect.

So which statement was wrong?

@ ogo

"I give up with you"

It is a best thing from anybody who continously mix up the ttl and the external metering.

@ Antony John:

Quote:
"@ Manmachine, quick look, Panny 45-175 5.2 TStop.
And it doesn't have "15 to 20 elements""

So what we are talking about?

This sentence shows that you simply do not know what stops and aperture means.

@Manmachine,
SIgh, please tell me which of those lenses you listed were used for the Oly E-M5 tests please.
And
if neither
Then what the TStop (or EV) difference was there with the lens used to test the E-M5 and those you listed.
Finally, the John Cleese "Hungarian/English phrase book" and accompanying "Optics as I see it" should not be used as reference material. ;-)

After you wrote technical supidities and you could not demostrate any of them the only weapon that resmains for you is scoffing and indulging in personalities.

"Why can't people just admit that a 4/3 sensor can be good at high ISO, and instead of that just say that Olympus cheated on high iso ?

Now think about this :
- R. Butler clearly stated that you can't use those studio samples to judge ISO scale"

The first point you pointed out, is enough to question the performance shown. I think that's sort of the point.

The IBIS in the E-P1 works very well indeed and the E-M5 is supposed to be even better. People who have actually used the camera confirm this.

As for the slight variation in "ISO" ratings from manufacturer to manufacturer, so what? The E-M5 is clearly doing very well regardless - add a stop and it's still good, way better than Olympus have done in the past and better than any would expect from that sized sensor in 2012.

@coroander
But the price and size/weight of such 0.95 lenses kill all advantages of the m4/3 system. And do not forget about AF ;).

@ pomovile
I have E-PEN1. And I state (with the hand on the Bible), that its in-body stabilization is practically useless. It barely gives 1 Stop advantage, most probably 0.5 Stop if any.
Even if they say that EM-5 has brand new IS, I really, really doubt that it would give more than 1.5 Stops advantage.

I aggree about the skewed ISO rating of the Oly E-M5 camera.
For example, when checking DPRs test pictures at ISO1600.
Sony NEX5N shoots at f8 and 1/400.
Oly E-M5 shoots at f6.3 and 1/400.

"Useless" in-body stabiization? One can bicker the issue in the case of still photos, but hand-held video without stabilization is useless, period. The OM-D's 5-axis IBS may be over-taughted, but let's see some tests. It might be quite nice.

I have to jump in here about the IBIS. I'll admit I don't know much about the finer points of measuring ISO, but if Olympus has exagerrated, my attitude is "So what?" Just choose the appropriate ISOs in the studio comparison tool. This camera still looks good to me.

As far as IBIS, I currently shoot a Panasonic G2 and an Olympus E-510. I love the G2, it has better dynamic range than the E-510 and is better for manual focus, but I REALLY miss the IBIS of the E-510. It makes a huge, noticeable difference in any kind of handheld shot in minimal light. So I don't know where this post is coming from. Everything suggests that this camera has better IBIS than ever before, and it is a big part of my decision to make what for me will be a major investment.

Read the reviews of the Panasonic 25mm 1.4. Tack sharp wide open. It may be as sharp wide open (or nearly so) as at any other aperture. That lens + this camera gives you as much capability in low light as anything else out there.

@bobbarber

The IBIS might give you a "huge" advantage on your E-510, but it is practically useless on my E-P3. Different models, I guess. I leave it off all the time, as it adds blur to high-speed shots.

@John Koch

IBIS does not work when shooting video on PEN cameras. The PEN cameras automatically turn off IBIS and use digital IS for video.

The two-stop IBIS result was in a test of the M5 which I accessed from the m43 rumours website.

@jorginho

That's interesting. Thanks for making a constructive comment.

@ NZ Scott

You may be right about camera models. I have read complaints on these forums about IBIS on the E-620, which was worthless in some people's estimation. It sounds like you don't like it on the E-P3 either. I guess we'll have to wait for the verdict on this system. I'll repeat though, IBIS is excellent on the E-510, which I have shot with since 2006. So my opinion is not based on 100 shots or anything like that. If this system gives me what the E-510 does, even if all the hype about 5-axis is a lot of hooey and it doesn't do /more/ than what the E-510 does, it is still a feature worth having.

bobbarber is correct; the Panasonic Leica Summilux f1.4 is very sharp wide open. Also m43rds offers larger DOF with large apertures, and for most low-light work this is likely to be advantageous -- certainly more will be in focus. There's a potential 2 stop advantage to m43rds in low light work, plus another potential 2 or 3 or more stop advantage with IS. This camera still promises to be outstanding (if not the best solution) for a lot of low-light work.

digitalkamera.de, as far as I can tell, seem to only judge by out-of-cam JPEG, which clearly has a very steep tone curve applied, rendering shadows even darker, which may account for them to assume a 2/3 EV ISO difference.

ISO is defined by out-of-cam JPEG, so they're right to do so.

About IBIS

I can't shoot handheld without IS around 1/60", but totally fine with IBIS turned on on my E-620. It's not useless.

If you use a telephoto lens, please make sure to turn the focal length section at the bottom of the screen to correspond to the focal length you're using. That is, if you used 200mm lens, change IBIS to optimize for 200mm.

There are three modes of IS with E-620 (there are more with the E-M5). IS.1 is for both horizontal and vertical shakes, which is most useful to me, as my hands shake a lot. IS.2 and IS.3 are designed for minimize noise caused by IS, so they only reduce vertical shake only (IS.2) or horizontal shake only (IS.3). To me those two modes are useless.

In high speed shots, IS does very little. IS is most useful around borderline shutter speed, that is 1/30" to 1/250". 1/250" is usually considered safe handheld shutter speed for anything under 100mm focal length.

And please don't call E-M5 "M5." That name usually refers to Leica M5.

So as 5N

To everybody arguing that the IBIS will be useless, the tests are out there showing otherwise. Quickest one I could find:
http://www.pekkapotka.com/journal/2012/2/24/olympus-e-m5-5-axis-image-stabilization.html
There's another with a graph hanging around, but I couldn't locate it.

Your conjecture is unsupported, so your confidence in your assertions is unjustified.

@Zanton: I actually mostly agree with you about EP-1 IS (disappointing, though not entirely useless as you claim) but you're entirely wrong about the E-M5.

@joes49

Most likely those who don't want to carry big cameras and don't have SHG lenses. Basically, to those E-3/5 are overkill to begin with.

Yes it is true, really don't know why Olympus did this, and why they need to cheat a bit, on the past they had always the correct ISO

Refresh the page in the browser. If still nothing, try to clear the cache and load the page again (or try a different browser).

Doesn't work in FireFox, does in IE8 tho.

I got it to work on Firefox after pressing refresh a couple of times and it is working okay now.

That's exactly right. These camera companies are the football/soccer teams.

LOL @backayonder, i guess you're not the one that believe the three lions going to win euro 2012 or every euro & WC for that matter :D

@radix1,

No those you mention are the ones who spend big bucks on their cameras, lenses and tickets hoping to photoraph the winning goal. When it doesn't come they then berate their equipment :-)

Do you really think that Olympus would change to an APCs sized sensor?

Not actually the same - micro 4/3 sensors are smaller in area than the APS-C in the Sony NEX5N -, but some suspect the E-M5's sensor has been made by a manufacturer other than Panasonic. If that's right (Olympus play their cards close to the chest), Sony is a possibility. (But so are Fujifilm and Samsung...)

No, just a variation or design similar to that, but with a different size.

On Amazon's page it clearly says it's LiveMOS censor. I highly doubt it's Sony who makes the sensor (unless they have the plant to make it..., and given they have trouble pushing the NEX-7 out, I doubt they have the capacity). If Panny is not making it, I guess it may be a third-party outsourcer who Panny has given the permission (and given the technology to) to make it.

yes got the same impression, particularly in the RAW samples that the output looks closer to what the Sony sensors do (yes I realise one is APS-C and the other M 4/3)

All depends on what's done with the analogue signal from the sensor. It's been said that Nikon can get better low light performance from the Sony cameras than Sony can. The only difference being the 'pipeline' after the sensor - The ADC and any other 'manipulations' to the signal that might occur.

@ coroander
Good luck focusing where your eyes cannot see.

@Pelasdf: that exposure is for about EV-2 (ISO 100) or about the same as a full moonlight night with the moon overhead, plenty bright enough to see. One great thing about this camera is the availability of IS with very fast prime lenses.

Oly 45/1.8 is closer to a 50/1.4 than the 50/2.0 Macro.

@Prognathous

Then you have to crop the image to give the same coverage, and 45mm and 50mm give different angles of view.

Though using 50mm f/2 macro for 4/3 on m4/3 camera is a little weird. On one hand, 4/3 lenses tend to be sharper, on the other, when you use a non-native lens with an adapter, IQ will suffer a little because of the light bouncing inside the camera body (refraction noise) because the lens is designed for a different flange distance, therefore, the rear element isn't optimized for m4/3.

I have both Oly 50mm f2 and Nikon 50mm 1.4. There's no contest. The Oly 50mm does a better job resolving and creating those beautiful Oly straight out of camera jpeg colors. After looking at the raw files at 3200, especially the shadows, I am glad I went with a D7000 instead of going m43rds. This camera is a vast improvement for Oly and that's good for competition. Please keep the tech wars going for when I have to replace both my Oly and Nikon DSLRs!

I think all those itsy bitsy nitpicking technical talks are meaningless. The important part is whether the IQ will satisfy you under the perimeter "you" usually shoot with. DPR's review is here to give you an idea how different cameras hold against each other under similar situation under a fairly equal circumstance, so hopefully it'll help you with your buying decision. The really fine differences can only be judged when you do the test yourselves with your very own perimeter. I pixel-peep my photos, but even I wouldn't be so nitpicking about the metering because that's out of control. I would be more concerned about focus accuracy of the test shots and the lenses used than metering differences.

One is 4/3 sensor one is APS-C sensor. They don't capture the same amount of light in the same amount of time to begin with. That's the reason why smaller sensors have more DoF using same aperture, remember? The other part is focal length. Every time you double the focal length, you half the amount of the light captured in a given time, therefore, 1/2 as much DoF for a given aperture (i.e., 50mm f2 will give the same DoF as a 100mm f/2.8). That said, I don't think it's that important. The fact the difference is a consistent 2/3 EV means meters are accurate. Consistency is more important than deviation, because you as a user can control your own exposure without error.

citizenlouie: You have misinterpreted the meaning of exposure value and ISO.

Aparently they changed the lighting for a dimmer one recently. But if you compare to the Panasonic GX1 (same sensor as the G3) you will see same exposure (probably GX1 was shot with the new dimmer light).

ISO 3200:
GX1: 1/800s F/6.3
E-M5: 1/800s F/6.3

@ R Butler: Olympus digital cameras underexpose by 1/3 of a stop is a well-known fact, just like Nikon digital camera over expose by 1 stop. I think it's each brand's preference. (not true for those brand's film cameras). That's the reason you'll find on Olympus's fan website that suggests users to do a +1/3 stop when you receive your Olympus camera (it's programmable in menu, so you don't have to do exposure compensation every time). I personally prefer to slightly under expose because digital sensor is more similar to positive films, so under expose actually improves dynamic range by preserving highlight (if applicable). You can always bring the exposure level back up during post-processing. No need to sacrifice dynamic range which is extremely important for smaller sensor like 4/3.

I am not sure if you understand what I meant by my statement about the exposure... and ISO, but anyways.... Final result is more important than numbers. Everything else is just technicality.

@ citizenlouie

You've made a very interesting comment indeed in your assertion that you prefer to under-expose in order to preserve dynamic range. I remember being advised to under-expose for this very reason, when I shot Nikon DSLR.

Nowadays, I do the exact opposite. I do what professional landscape photographers do.

They use a technique called "exposing to the right", which means over-exposing as much as possible without blowing highlights. The reason for this is that it is actually the highlights that preserve the most detail - not shadows. It is a simple matter to lower the exposure level in post-processing.

The difference in detail is massive - one industry magazine stated that it was a factor of 16 times as much details in highlights as there are in shadows.

@NZ Scott.

This is a perfect example why I mentioned in another post that you can't simply copy photography book's technique and apply to every camera you use.

"Exposed to the right" is the correct terminology, but what you described is for negative film photography. Digital sensors is based on positive film (slide film or color reversal film, all means the same), so everything you do have to be in reverse. You want to preserve "highlight" because once it's gone, it's gone forever whenever you shoot with positive film or digital. To preserve highlight, you expose for shadow to preserve highlight. It's very confusing I know. When you shoot with negative film, you expose for highlight to preserve shadow.

That said, what you said is exactly the opposite of one should do for digital format to save dynamic range. Please refer to Wiki entry on "Zone System" for detail. There are some shortcoming for using this method. Short answer, use histogram. :-D

Less compression, bit more latitude in shadows and highlights. You can take it as very lightweight RAW. During history of digital cameras, there were few that offered either special JPEGs or simply less compressed JPEGs that retained a lot of data and allowed bit more processing than average lvl 8 quality JPEG.

Understood, but a back-lit Stouffer step wedge is painfully absent in the studio setup in order to check the other side side of the coin..

Nevertheless, Brigcam my have a point. Assuming that dpreview tests are all made with the same power of illumination (are they?) some camera manifacturers cheat on the actual iso sensitivity, and the different shutter speed could just be that. Actually, other reviews say that the EM-5 just does that - I hope it's not true

But the practical way that people use these tests is in fact to see the true ISO performance. Sure, it doesn't matter whether you use 1/400 or 1/50 for a stationary scene, but it's of significant important when you have to account for subject motion or camera shake. That's why shutter speed (and aperture) should be constant, rather than ISO.

Bob, maybe Panasonic should rethink their philosphy. Richard, exactly right. People aren't understanding that the exposure of the image is what is most important. Not miniscule differences in shutter speed.

Doesn't matter because haters will hate, and they are out in full force today. Olympus is rattling some cages.