[orangepeel]

I'm starting this new thread coz I don't wanna hijack the other one where this originally came up...

Originally Posted by orangepeel

note that the plugins will process your images differently. the nikon one appears to use your camera settings (saturation, contrast, sharpening etc) wheras the adobe one doesn't. this means that the images you load into PS will look different depending on which plugin you use.

Originally Posted by skipper954

You can set Adobe Camera Raw so that it opens with your camera settings

Originally Posted by skipper954

see my post in tutorials. Hope it helps.

Just had a go with your tut and while it's super useful, it doesn't seem to do as advertised



This is a the same shot opened using the nikon PS plugin, ACR with my settings and ACR using skips tut.

This is the nikon plugin with the defaults (only option of WB and EV level)


This is Adobe Camera Raw as I used to have it.


This is using skips settings.



Now the nikon one is pretty much how the images appear on the camera LCD. None of them above have the corrected white balance. It was set to auto on the cam and should have been tungsten.



http://www.orangepeels.nildram.co.uk...6_WB_Nikon.jpg
http://www.orangepeels.nildram.co.uk...636_WB_ACR.jpg
http://www.orangepeels.nildram.co.uk...36_WB_Skip.jpg


So wotcha think of them apples? Ya think this is anything to do with the encrypted white balance nonsense that nikon forced on us?
Incidently, ACDSee Pro has a third different look again although it's a lot closer to the nikon plugin.

[skipper954]

I am not sure that I am going to explain myself well, but here goes.

My settings were not the ones you should use or pay attention to. They were the settings for my camera on the particular image that I was working on.

WB is one of the major flaws in digital cameras. Auto works well some of the time, but not always. You can generally trick your cam into a good one by forcing it into a warmer or cooler tone depending on the image. If you set your cam to 5000K or cloudy day/shade mode, it will force an image with more yellow and red and therefore warmer, and visa versa.

On some cameras you can take a closeup image of a neutral gray with the lighting you are using and ask the camera to set a custom WB accordingly. I know it can be done with the Nikon and probably with Canon.

However, one of the advantages of shooting in RAW is that you can post process the WB to where you want it, so that the correct WB in camera is not THAT critical.

In ACR, if you have a series of images in the same setting, open them in CR (or PS), set the WB of the first image the way you want it, then select all the images and synchronize them. Actually, you can manipulate the other parameters (exposure, shadows, brightness, contrast and saturation) on the first image along with the WB, then synchronize all the settings at once.

So bottom line here is that WB isn't automaticly perfect because the cameras settings are not always perfect. Adjustments have to be made on the 'puter but ACR makes it fast and easy.

Hope this is clear and helpful.

skip

[orangepeel]

Is the point not that ACR and the Nikon RAW interperate the files differently?

[skipper954]

When you open the file yes. But at least in ACR, you can and should adjust. Probably in Nikon RAW as well, but I don't use it so I don't know.

All of the programs use different algorithms.

skip

[orangepeel]

Heh. Kinda makes a mockery of "RAW" then eh?

p.s. The Nikon plugin only lets you set WB and EV levels. This is one of the reasons I say it sucks - the other being it doesn't transfer EXIF information.

[skipper954]

Originally Posted by orangepeel

Heh. Kinda makes a mockery of "RAW" then eh?

No, I don't think so. The raw image comes out of the camera uncompressed, the way the camera "sees" it. If your exposure or WB or other settings are off, that is the way the camera sees it and when you open an image in ACR, those incorrect settings are shown. A raw setting is not supposed to produce a perfect image.

Various processors interpret the data slightly differently, and the secret is that you tweak the settings to your liking.

The real value of raw is, as I said, there is no loss of data, and global changes can be made without losing any pixels. In other words, you can ALWAYS go back to the original shot out of the camera.

skip

[j sotelo]

You should try an Expodisc white balance I just bought one and it makes a HUGE difference in the white balance....it saves you a ton of color correcting....

[orangepeel]

Originally Posted by skipper954

No, I don't think so. The raw image comes out of the camera uncompressed, the way the camera "sees" it. If your exposure or WB or other settings are off, that is the way the camera sees it and when you open an image in ACR, those incorrect settings are shown. A raw setting is not supposed to produce a perfect image.

I still think you're missing the point. I'm not fussed about right/wrong WB settings. I'm bothered by the fact that I know of 3 different RAW processing routines for my D50 NEF files that produce different output. (ACDSee Pro, ACR, Nikon NEF Plugin.8Bi)

I'm hoping that another person with a non-nikon or a nikon that doesn't use these encrypted NEF files can also load their raw with different plugins and get different results. If that's the case then we're all in the same boat and misery loves company If it's not the case, then it's an effect of Nikon stupidly encrypting the WB settings and/or compressing the D50 raw files.

Originally Posted by skipper954

Various processors interpret the data slightly differently, and the secret is that you tweak the settings to your liking.

Then as I said, the data isn't really RAW. Whole point of digital information is a "one's a one for aw that". You can't interpret a one as a five and still say the data is the same.

At the end of the day, I want what my camera saw, not some softwares interpretation of what my camera saw. Because with the interpretation it's not really raw.

Take an example where I loose a generated PSD/TIFF/JPG. 15 years down the line I want to reproduce these files again from the original NEF. I can't do it unless I have the EXACT (RAW) software I used initially because to use different software produces different results. The only way round it is to backup the RAW plugin and all the software used at the time of the image creation. Oh and hope the version of windows at that time still runs 15 year old software.

[skipper954]

I see what your point is. It is probably moot, because each company's software should be backward compatible, but I don't know the answer. Perhaps someone else on the site does. In addition, I will google

skip

[orangepeel]

Originally Posted by skipper954

It is probably moot, because each company's software should be backward compatible

Perhaps you mean new versions of software should open the old data files - well if 3 versions of current software produce different results, it seems likely to me, future versions will suffer the same problem too.

As for old software running on new platforms, that's a major problem. I've got a cupboard full of old software that I can't run anymore and that's on a 7 year old OS. Sometimes it's for hardware reasons, no serial port for dongles or lack of floppy drive, other times it's because the underlying API has changed so much that the software simply can't run.

Adobe Lightroom is a case in point. You can't run it on Win2K. It illustrates the fact that it's in the developers interest for their software not to run on older systems. Gives them a lever to force users to upgrade, thereby squeezing more money out of them.


Sorry, I'm not trying to argue with you I'm just trying to work out why people refer to RAW files as the "raw sensor data" when patently it's not the case due to software interpreting the data differently... for my cam at any rate.

[skipper954]

I think that we are both saying the same thing.

The 0's and 1's are exactly the same. The software algorithms are different, so that the image appears a bit different on each program.

No way around it except to use the same program for everything, therefore getting consistant results. I prefer ACR because there are so many things that you can do with it.

skip