GH3 studio shots available

Started Mar 11, 2013 | Discussions thread
Anders W
Forum ProPosts: 18,722Gear list
Like?
Re: GH3 versus EM-5 RAW IQ: EM-5 clearly ahead!
In reply to texinwien, Mar 13, 2013

texinwien wrote:

Anders W wrote:

amtberg wrote:

Anders W wrote:

amtberg wrote:

Anders W wrote:

amtberg wrote:

Anders W wrote:

amtberg wrote:

dr_elis wrote:

Anders W wrote:

dr_elis wrote:

In the EM-5 review you can directly select the GH3 for comparison. The EM-5 RAW image looks at base ISO clearly better, sharper, punchier, more contrast, more detail. Could be the lens though. Was the same lens used to test both cams?

Yes. Same lens and same f-stop (6.3). For MFT bodies, they always use the FT Zuiko 50/2 macro for the studio scene.

Ouch, that´s a rather disappointing result for the GH3 then. Looks indeed as if Panasonic uses a more heavy AA filter in the GH3 compared to the EM-5.

Yes, I think the GH3 is using a heavier AA filter. Not a big deal, IMO, as it just requires a notch more sharpening. The most surprising thing to me is that the GH3 seems to show significantly better DR in shadows, although it could be a result of lighting/exposure. Check the thread spools inside the box to see what I mean.

Are you talking about the RAWs? I see no difference at all in the noisiness of things inside that box that wouldn't easily be explained by the fact that the E-M5 samples are slightly more contrasty and slightly more saturated.

No, not the noise, but the contrast ... dynamic range.

I don't follow you here. Contrast is not the same as dynamic range. The contrast of the E-M5 shot is generally higher. But how do you determine DR by looking at contrast within the box with thread spools?

Provided the highlights are comparable, shadow noise is the way you can judge DR. That's why I mentioned it. According to the standard engineering definition

DR = M/R

where M is max signal and R is read noise. Differences in read noise shows up most clearly in the shadows where it dominates over photon noise.

But I do think the GH3 seems to do better with shadow noise, while the EM5 does better with noise in better exposed areas.

In the shadows, I'd rate them as equal with regard to noise except for what might be expected on the grounds I already mentioned. Where in the better exposed areas do you think the E-M5 looks better?

You can also judge DR if a camera goes to black in the shadows while another is resolving some detail, which is what it looks like in the camparo gizmo.

No. Judging DR by such means doesn't work. The shadows go to black quicker with the E-M5 because of the more contrasty tone curve chosen (probably a matter of what Adobe has found appropriate, for one reason or another, to use per default). It has nothing at all to do with DR as conventionally understood and defined and is unrelated to any property of the sensor.

I'm sure you're right, from a technical viewpoint, but from a practical perspective it certainly does give you less DR if the camera manufacturer has set up the curve in such a way as to reduce the amount of shadow information available to you.

I wonder whether amtberg thinks the data is lost in the RAW file rather than just not included in the JPEG output due to the tone curve that was applied to the RAW in order to generate the JPEG.

Could be. Let's hear what he says.

The data is still there - it's not lost, at least not in the RAW file, and that's what matters.

Precisely. And if something is actually lost in RAW, it's because the noise level is too high to distinguish it, not because of the choice of tone curve or contrast level since there is no such thing in RAW, just a (normally) linear/linearized response to differences in light levels.

Yes. But the DPR samples don't really tell you that, do they? On the one hand, the tone curve used in these samples is just one of many potential examples. The E-M5 allows you to control the shape of that curve not only via the usual contrast setting but also by other means such as the newly introduced highlight/shadows control. On the other hand, I personally don't care at all about these OOC jpeg options. I haven't even for a moment considered using them since a) pretty much any RAW converter can do it much better and b) I don't want to take time and concentration away from more important things when I am shooting.

What I care about is not whether and how the camera allows me to manipulate the tone curve but whether the sensor is good enough for sufficient manipulation of that kind at later stages, with other tools, without, and that's the important point, exceeding reasonable noise limits. This is not something that the base-ISO DPR samples really let us judge well enough. You can learn a bit more by downloading the RAWs and do a bit of shadow pushing but I wouldn't say the samples are really ideal for that purpose either.

But I haven't downloaded the files and it may well be possible to boost the OM-D's shadows and bring out the same level of detail. Of course, if you can bring up the detail you will also bring up the noise level....

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
NoNew
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow