DP1 vs E-P1 again

[barjohn]

Typically, the pixel density helps most with low light imaging where more photons can be collected for a given site. Yet, I don't think anyone thinks of the DP1 as a low light high ISO camera as above 800 it is terrible (at least from the images I have seen). As the D3x has shown, it is a trade off between resolution (high number of pixels) and high ISO sensitivity (doesn't match D3).

 

[RichA]

I have turned off the noise reduction and the image stabilizer and will do some more tests to see if it has something to do with it.

Check out this thread on Dpreview. The guy is using an E-620 which has the same sensor as the E-P1. His shots look a bit like what you are looking for. Maybe you could ask him what he's doing?

http://forums.dpreview.com/forums/read.asp?forum=1022&message=32502351

 

[Rawfa]

What it looks like you do with the DP-1 files is to expose for the highlights, then bring up the brightness in the other areas while jacking up the contrast. You can do the same with the E-P1. If you can't post 100% image sizes, try cropping the images and posting the crops (of selected areas) at 100%.

That is exactly what I do (note: I always shot jpegs with the DP1). The problem is that when I try to do this with the E-P1 (in RAW, mind you) the files fall apart lacking detail, definition and displaying ugly noise. I don't know, maybe it has to do with what Smokysun said and it's pixel density related...after all when you are bringing out light where there was none if the pixel density is better you shouldn't see it damaged as easily. To be honest, other than the DP1 I've only manage to give this kind of treatment to DSLR files.

 

[tashley]

I've been thinking about this, not that I have a solution for you, but I have been wondering about that shot of the Sagrada from the EP-1 and why the treatment seems so wrong to me. I think it's this: there's so much false light on it that there's effectively no light on it, because you have tried so hard to banish the shadow by bringing up the HDR shadow details.

I spent a while making a Barcelona portfolio so I can appreciate your challenges. The light is hard and contrasty and the shadows deep and that combination is beyond any display medium's dynamic range so either something has to give, or you have to manipulate the files to effectively compress the dynamic range.

But Gaudi knew what he was doing when he designed that facade. If he wanted people to see deep into the shadows he wouldn't have made such a deeply carved surface. He knew how bright and hard the light would be and yet he made a series of surfaces that the light can't penetrate and which will always represent a series of very bright and very dark areas, which may or may not have metaphorical intention.

IMHO the best architectural photographers rarely use any artificial light because (assuming that their subject matter is of high quality) they try to respect the interplay of light and shadow that the architect intended.

Choosing to do the opposite will not only make the the shot look subliminally and disturbingly wrong (as so many HDR shots look) but will also test your equipment beyond its design parameters.

So there's you solution ;-)

Best

Tim

 

[jonoslack]

Hi Rafa - Tim - Rich.
Having been sniffy about HDR, and also having worked with 4/3 files ever since the E1, I have evolved 2 strategies for filling the shadows without actually creating that nasty false light feeling.

The best way is really simple - process the RAW file twice, once with the correct exposure for the sky, and once for the foreground . . . assuming that the exposure will allow it. Then you'll have one file with a completely blown sky - select that, invert the selection and paste it onto the other file as a new layer - obviously it'll need tidying. However, this doesn't give that nasty 'false light' feeling. I've yet to find a program or action that does this well - every shot is different and needs different treatment.

If the exposure varies too much for that, simply take two shots (or use bracketing), one exposed for the sky, the other for the foreground, and layer them again.

These days, I do most of that kind of stuff in Viveza, which works really well.

I think that if you want to get the E-P1 to work for you, then you're going to need to use different strategies to get it to work. Maybe it isn't worth the candle? but I think it probably is.

I've gone through too many cameras, and they always need different tricks to get the best from them

 

[pellicle]

Jono

I've yet to find a program or action that does this well - every shot is different and needs different treatment.

as you say, each is different, but I've been quite pleased with photomatix for doing this ... you just need to get the hang of it for a little bit, but then after that its always only a few quick clicks. I can keep the file as a .EXR file and the XMP to redo it.

samples:

are they 'over the top' to your eye?

 

[jonoslack]

ps. i just went to dpreview to make sure i had the pixel density info correct:

canon 5dII - 2.4

canon 1dsIII - 2.4

leica m8.2 - 2.1

nikon d3x - 2.8

olympus ep1 - 5.1

and the winner is the sigma dp1 - 1.6

HI There - interesting stuff, BUT I think it gives an impression which is definitely incomplete (and rather lopsided). How about:

Canon G10 - 34/cm2 mp

Panasonic Lx3 24 mp/cm2 (with all the razamatazz about larger sensor)

Nikon Coolpix 90 - 43 mp/cm2

Ricoh Cx1 - 33 mp/cm2

Fujifilm S100fs - 19 mp/cm2 , with "one of the largest sensors we've seen in a non DSLR", "Excellent High ISO performance" (quotes from dpreview review).

So, although the E-P1 is twice the pixel density of the D3x, there are cameras which perform well which have 4 times the density of the E-P1.

 

[tashley]

Not over the top at all to me... they subtly bring out shadow detail in a way that you'd not know had been done unless you were told. That, to me, is a well extended dynamic range (or in fact a well compressed one to be strictly accurate!) that tricks the eye rather than blasts it!

Tim

 

[jonoslack]

Jono



as you say, each is different, but I've been quite pleased with photomatix for doing this ... you just need to get the hang of it for a little bit, but then after that its always only a few quick clicks. I can keep the file as a .EXR file and the XMP to redo it.

samples:


are they 'over the top' to your eye?

As Tim says, they look fine . . . but none of them look like shots which would need much treatment - unlike Rafa's picture of the Sagrada.

 

[Rawfa]

The HDR alike picture of the Sagrada Familia taken with the DP1 looks great in my opinion (it's EXACTLY the look I want), the one with the E-P1 doesn't. I have followed the same process and I've tried all the ones I could think of (inclusing different layers with different exposures)...I know, I'm pigheaded aren't I.

 

[pellicle]

ok ... wasn't sure if these would post here ...


certainly this could use help

and a quick flick into this....

hmmm ... you know, comparing the work computer screen to my home laptops screen (despite it being profiled) that's way too punchy in the reds ....

 

[Robert Campbell]

... I have evolved 2 strategies for filling the shadows without actually creating that nasty false light feeling.

The best way is really simple - process the RAW file twice, once with the correct exposure for the sky, and once for the foreground . . . assuming that the exposure will allow it. Then you'll have one file with a completely blown sky - select that, invert the selection and paste it onto the other file as a new layer - obviously it'll need tidying. However, this doesn't give that nasty 'false light' feeling. I've yet to find a program or action that does this well - every shot is different and needs different treatment.

If the exposure varies too much for that, simply take two shots (or use bracketing), one exposed for the sky, the other for the foreground, and layer them again.

...

Take a look at PhotoAcute for this sort of thing.