Getting it right in camera

[arthurking83]

is a bit of a misnomer really.. especially if you shoot in the raw :p
If you shoot jpg, then the software used to display the image should, ideally, make no difference.

I've notice this same effect regardless of the thirdparty software used... having used PS, and hence Adobe's ACR directly, and then LR3(whether it uses a version of ACR seamlessly I don't know), and finally Bibble5(trial).
The difference in the way raw images are rendered in non manufacturer software is almost always going to be different.
I only use Nikon software for my raw images, and the images are rendered exactly as per the camera does, and hence, the foremost important aspect of getting it right in camera is to have your software render the image as you captured in it camera.
Remember this is only for raw captures.
Jpgs should be rendered the same, in camera and also via any software.
Raw images are not pictures as such, but a collection of data, that needs to be interpreted or decoded by software to produce an image.
I can't imagine that any thirdparty software company will be able to decode the raw data as well as the manufacturer's own will be... ever!
While thirdparty software can be set to emulate a particular style for your camera, my opinion is that the manufacturer knows a lot more about their own file formats than any generic software maker will.
I've seen this in many trials of other software, but never thought to actually catalogue the differences, and I'm annoyed with myself for not doing so when I had LR3B2 installed as a trial.
Trial now over, and LR3 exorcised from my PC(but not fully I expect! :rolleyes:), and with no intention of ever installing Adobe software on my PC again, I can only explain the recent (distant :D)memories of my frustrations with LR3.
Bibble was in some ways horrendously inaccurate in rendering the raw files. I say inaccurate, and that's only compared to the camera's idea of the file, the basic manufacturer's software available(Nikon ViewNX), and their more elaborate software.. CaptureNX2.
We all know that embedded into the raw data is a jpg file, so we know we have a reference image file to judge the exposure by, but for some reason thirdparty manufacturers cannot reproduce the same image from the raw file(as per the jpg image).
FastStone's FSViewer renders the raw file exactly as the raw file should be displayed(judging by the jpg file and the manufacturers software), but in converting the raw file, it produces the same terrible results as other non manufacturer software does. It only displays the raw file as the camera does due to the fact that it displays the embedded jpg file by default, but can be set to display the raw file.
(I've never used it that way, as it apparently slows the program down, and I like it because it's fast :th3:)

I have no issues with other software, per se!
For editing and working on non raw images, there are literally millions of non manufacturer software available that may be better in many ways than the manufacturers software .. so this thread is limited to raw file rendering/editing/processing.
eg. as an image manipulation processing software, Nikon's CaptureNX is basically terrible. With a very limited feature/tool set when compared even to free programs like the GIMP, or FSViewer. It can't merge images, or HDR, or stitch them together!... nothing but the basics that even MS Paint can probably do! :p

But the one thing you should be able to rely on... 100%, it's that the manufacturers software should render/display and convert the raw file exactly as you captured in on the camera.
In a recent discussion on another topic, an ex member explained how the manufacturers only want you to see a nicely rendered image, and that the image is actually not properly exposed!!! :Doh:
I'm seriously thinking this ex member is totally deluded in this method of reasoning!
That thirdparty software manufacturers know more about the inner workings of another manufacturers proprietary files!!.. was this person serious? Or were they simply looking for an argument where no argument exists?

So the nett results of this line of thinking is:

Camera displays a raw file in a certain manner.
Camera manufacturer's computer software displays the same raw file in the same manner as the camera does.
Thirdparty software vendor displays the same raw file differently ....

therefore the manufacturer must be wrong! :lol:

:umm: ... OK... whatever you think buddy. It's your life... you make the mistakes you need too!

My recent experience with Bibble5, while a great program in itself(fast and well featured), the rendering of raw images, was horrific!
So much so that if I were totally reliant on it for my image processing needs, I'd either shoot in jpg mode (hahaha!... :Doh:) or give it all away, and take up croquet or knitting teacozys, or something easier.

here are some screen shots of how badly thirdparty can make your raw images look:

First up a direct display of the raw file in ViewNX, which is 100% identical as I captured it on the camera.
http://www.ausphotography.net.au/staff/AK83/raw_converter/Nikons_NX%20raw-or-jpg%20display_01.JPG
The image itself looks a littel different compared to the image in the camera screen, and that's perfectly normal and expected, as the camera's screen won't show the same level of detail, as a calibrated PC screen. BUT!... the histogram is identical.. to the minutest bump, as per the camera's histogram as per the review screen.

Then I open the raw file in Bibble5..
http://www.ausphotography.net.au/staff/AK83/raw_converter/Bibble%20raw%20display_01.JPG ..

and whoa! what happened to my colour!? While the in camera shot is slightly over exposed, that was a deliberate decision, so as to not lose any shadow detail in the pelican's black areas. Turns out that it wasn't required, but as this is one of three exposures in succession, if it didn't work, delete it.. no harm done. Check for sharpness on the review screen(D300's screen is more than capable for that purpose, if it looks OK.. keep it and review it later on the PC. But not via thirdparty software!!

I did notice LR3 was similar, but not to the same extent as Bibble5. Where B5 probably needs -1.3Ev compensation to bring it back into line with what the camera saw, from memory I think LR3 needed at least -0.3Ev and maybe -0.7Ev as a general guess... so call it even at -0.5Ev compensation.
But if I shoot with the view to compensate by -0.3 to -0.7 at the time of exposure by slightly blowing the highlights(which are more than easily and accurately recoverable!!), what of these thirdparty software? :confused013
LR3 did render all my D300 images with almost zero contrast, so it appeared as though it was rendering more detail in the shadows.. but then again so did B5 too!

OK, I have to admit that I never loaded any camera presets for the D300 with LR3. Didn't know they were there to be honest! But that's another point.. software is supposed to be easy, not difficult and cryptic and involve more work.
I just want to see what the camera did, if I did it wrong, review quickly and delete any stupid mistakes.. before everyone realises that I make billions of mistakes :Doh:

So we know that these thirdparty software vendors are more than capable because even B5 can render the image exactly as per the camera as long as the image file is a jpg(and I suspect a tiff as well)...
http://www.ausphotography.net.au/staff/AK83/raw_converter/Bibble%20jpg%20display_01.JPG
While not 100% the same histogram, the differences in each colour channel is absolutely insignificant and for all intents and purposes considered to be 'the same'.

Why bother with non manufacturer software to do your raw file to <image format> conversions then?

Beats me... I never would.

If I were(the type of image manipulator) to process an image via a thirdparty software, I would open the raw file in my manufacturers free software, make any basic adjustments in there.. colour, exposure, contrast.. ie. only basic edits .. and then convert to 16bit uncompressed tiff, for any further enhancements in my thirdparty software.


OK, that's it.

all comments, experiences and screenshots of differences anyone else has noted are welcome too.

[Kym]

I use Lightroom - it works. It is visually the same as SilkyPix (Pentax bundled software).
I've also used (no longer due to 64bit Win7) the Pentax raw codec http://www.pentax.jp/english/support/digital/rawcodec_vista.html which looks like LR.

Also: Pentax can put either PEF or DNG format raw files. No difference in images.

[joele]

I liked Bibble 4 a LOT (bought it) but have to say I get worse results from Bibble 5, though it does have some nice features..

LR works well with my Canon...

[junqbox]

If I had a lifetime to wait for CaptureNX to render anything, you might be right. Until then I'll keep using Lightroom, and getting it right in camera for that piece of software until then.
It's really not all that different than exposing for a particular type of film, processing or paper, once you know the parameters you work within.

[elyshiamadison]

I had the same thoughts with Lightroom - I know they are way different to the shot I viewed on camera. So much so I went back to using the Canon software (DPP) for a few weeks.

I have since decided, speed and easy processing is more to my liking, and have adjusted lightroom to suit my needs.... (my camera overexposes reds a huge amount! So desaturating them now is part of my import process) I'll persevere with Lightroom, and hopefully I can eventually sort out all of the settings so that what I shoot is what I see on the screen... the one that bugs me the most is seeing a severely over exposed image - that I know wasn't on my camera's shot histogram!

[MattC]

Jeez Arthur - you don't half hold a grudge do you! ;)

Anyway- I think what you are finding that Bibble is automatically adjusting the image to what it thinks it should look like.
Picasa does the same thing, and I agree it is annoying. How you turn this off (if you can), I don't know.
It is easy to test this - take a series of bracketed shots, and you will see that the under exposed images, are all pushed to look like the 'correctly' exposed image, usually with an increase in noise.

[Dylfish]

I went from lightroom back to PS5/Bridge Combo. seemed to have more control over the images.

[Speedway]

I don't know about all this. I take photos using general settings to suit the conditions and lens. Process them to my liking with elements5, then go and take more photos. If I got caught up in all this over tech stuff I wouldn't have time to take photos so what would be the point.
Keith.

[Analog6]

I went from lightroom back to PS5/Bridge Combo. seemed to have more control over the images.

I agree. I also use Capture One (version 5 basic) and it does a wonderful job with conversions, much more control than anything else I've tried.

[arthurking83]

.... If I got caught up in all this over tech stuff I wouldn't have time to take photos so what would be the point.
Keith.

Keith, I think you've missed the point.

The point being, that you constantly hear the term .. 'getting it right in camera' which is clearly a state of confusion if you work with the two different image editors, as I had(in this case).

I get it right in camera.. according to the camera, I get the image home, onto the PC and find that Bibble5 is trying to tell me something I don't want to hear(that I've over exposed the image), but on the camera the image is not over exposed.

Who do I believe?.. who should I believe?

The initial post was more of a retort to an ex members point of view, that the manufacturer of the camera know or understand less about their own formats than do the likes of Adobe, Apple, Bibble Labs, etc.
His comment in another thread(long closed) was that the manufacturers software only want you to see a nice image on screen, because they want you to believe that you're taking nice images(or some other garbage to that effect), and that Apple's Aperture(his choice of program) was much better at reproducing a raw file more accurately than the original manufacturer of the gear that produced that file!

I remember seeing similar exposure differences when comparing the LR3 rendition of an image compared to either of the Nikon software's version of that same file. Not as badly as Bibble proved to be, but a lot more flat in contrast and slightly over blown.

I guess ultimately the point is that I prefer to rely on what the camera tells me I want to see. I desperately need to see and use this info, as I like to 'get it right in camera' in the clean and technically correct version of the term.. not get it right in camera, but conditional on some thirdparty application's approval process

Imagine a situation where, you shoot 90K images give or take 10K, and you open each one with this third party program and find that they're over exposed and not recoverable, so you delete them all, or many of them.
Imagine later on that had you used the software from the manufacturer of the device that can read the raw data in a more accurate manner and find that the ones that were supposedly over exposed in the other application were not!!
I'd be pissed off violently with the manufacturer of that third party program.. maybe even to the point where I'd be looking to sue them!(unless they have a clause in their license agreement where they explicitly state that accurate reproduction of raw data is not guaranteed ... or other words to that same effect).

Based on Bibble's idea of that image(being the same image in both programs) ... I'd have deleted it(not immediately, but ultimately at years end) (ugly)Blown highlights are not something I tend to tolerate too well, the pelican lost way too much detail in the Bibble rendering. It was approximately 1.3Ev over exposed relative to the ViewNX rendering.. for the same image!.. not a similarly captured image, the same file.

If Capture have a trial of their software, I may try in a while. I'm also interested in trialling DxO's software too.

[[ kane ]]

Only way to get it right in camera is to shoot film, even the different labs will produce different results.

Work with what you know.

Thanks
Kane

[yogestee]

Being a Canon shooter and shooting in RAW or sRAW1, I only use Canon's Digital Photo Professional to aquire, tweak and convert to Jpeg.. If needed, jpegs are further imaged in CS4.. For me this work flow works.. Also,, I find what I see on my camera's LCD screen is quite accurate to what I see when loaded into DPP..

In reality, I do very little post processing.. I have always been taught (a long time ago) to get everything right in the camera,, exposure, composition etc.. Getting it right in the camera saves me time,, I'm not a big fan of sitting in front of a computer screen post processing.. I'd rather be out shooting..

[crumplex]

I would say Lightroom as well