Using CaptureNX(2) for Nikon files

[arthurking83]

hey peeps, I had to do my CaptureNX tute allover again, as I had made up one way back on '06 and it got vaporised when the server had a crash a while back. No mind.. It needed a bit of an update anyhow.
I've been using CNX since it first came out, and immediately upgraded from Nikon Capture.
I first had PSP7(or 6) to begin with at the very start, but was given a copy of PSv something by a family member that had upgraded to a newer version. So I started using both Nikon Capture(totally different to CNX software) and PS at about the same time.
AND! no prior background with any image editing software, before I had to start a journey on image editing.

First things first.. I found PS to be a convoluted POS! trying to do anything of importance.. like cropping may be easy once you learn how to do it.. but up til that point it's about making too many mistakes before hitting the downhill run. PSP(paint shop pro!!) and Nikon Capture were miles ahead in that respect. And that respect was from a noobs perspective!
PS may be a technical tour de force, but you needed to have the force with you, to make any use of it(to begin with). I get frustrated easily and PS sat idle for many years at a time, only occasionally coming out for cloning/healing work.

So with that small bit of background I make no apologies for my dislike of PS and how much easier and quicker it is for me to edit photos using CaptureNX.

Of course CNX does have a few serious limitations, but at the price difference it's really kind of acceptable too based on price.
(All comments about serious limitation in CNX will be highlighted in red, like that)
At 1/5 of the price you get far more(for photo editing!!) that 1/5th of the product.
have I mentioned how PS is a piece of...!
So LR cost close to 3x the price as a download. I got my boxed upgrade for approx $100.
Had I known that Nikon were to give away v1 of CaptureNX with every D300/D3 back then, I wouldn't have bought my copy! .. oh well!
The moral of the story is that CaptureNX is dirt cheap compared to other software.

Other software of importance is ViewNX.
Now as I only deal with Nikon NEF files, any other branded raw file users can just sit back and watch(only!!)

I'm going to start with ViewNX which can do very basic global processing work.
If you use LR for your images simply to do global exposure compensations and highlight/shadow recovery and that's it.. you can also do that with ViewNX.
In a very few way(maybe only 1), ViewNX's batch conversion tool is slightly better than CaptureNX's one. I'll get to that point later. One thing that really annoys me tho is that you can't edit or convert any images that have been processed with CaptureNX, using ViewNX. You know this by the use of a small icon in the ViewNX thumbnail of an image. It has the Nik software Colour Control Point Icon/logo meaning edited by Capture. It'd have been nice if you could at least batch convert the NEF files created by CaptureNX, into jpgs or tiffs using ViewNX.
Sometimes I find that I have a series of images where many don't require the use of CNX, but if one does, I have to convert/resize any NEF's using CNX only. So that means having to stuff about with two programs to do the one job.

Start with ViewNX:


As a basic NEF editor it's ideal.
I like the interface. It's not confusing(like LR is!).
I just don't get the concept of exporting.. sorry! when all other programs use the concept of saving files, adobe need to confuse you with extortion!.... oops 'exportion' or whatever the new trendy name ends up being in the next incarnation (ok! I've stopped ranting again). Save file, save it to where I want. I can upload with a proper upload tool when I want too.. with security measures that I know are safe!.. ok.. ranting again

In VNX you see your thumbs. I have two screens and in saving the screen shot, I didn't realise that it'd save both screens. So I see this, the main thumbnail screen and my loupe on my secondary monitor. If you have this setup, with an image active(clicked), wherever the mouse is on that loupe view, if you hit the Shift key it gives you instant 100% view. Good for deciding which image is sharp or not. As you quickly navigate though your thumbnails, hit the numerical keys to give the image a rating. The rating Nikon use can be made to coincide with Adobe ratings.. but I hate Adobe Hitler, and Adobe should be tortured into submission... to Nikon's much better system! ... Oops, I know I'm ranting again.
My system is that I rate up to 7(look closely and you see a pinkish dot under each thumbnail, that's the numbered system). 7 is more of a purple colour, and stands out more. I use that as my base keeper rating. 99.99% of images rated less than 7 are trashed, eventually. I do have a few 6's but they are so bad they'll never be displayed. They may be images of my kids or something of intrinsic value. Maybe 1 or 2 images over 75K images.. I can afford that. I'll also have a series of of similar/same images, like a birder does when they 'spray and pray'. I then use Ctrl + 1-5 on the number keys to rate them even more detailed. So an 8 means upload to AP and other places, and the star ratings are based only on how each image rates against the others in the series. things like exposure/sharpness/moment of capture/etc... all contribute to the star rating.

I'm not going to go into VNX too much other than; up top is a link to CaptureNX.
easy way to open an image in CNX from there(green arrow).
You can configure VNX to add any other application to it's list of "Open with" dialogue, but you can't add a quicklink to the toolbar(silly!.. but not important).. Only CNX exists.

Other Icon of note is the Geo Tag icon. If you want to share geo information about the image, this takes you to a new page where you get to Google Maps, and you select the location of the image from the maps. This simple feature has delayed the purchase of a GPS for me to tag my NEF's with. I once tried using a tried and tested program to add geo info into an NEF and the NEF got corrupted!.. never again will I do that, on any important images.

..... /2 Starting CaptureNX.

[arthurking83]

Now be warned this tute wil be long and heavy with large-ish images!

Starting CNX is a breeze, and I basically never use the TaskBar Icon. It's always from the VNX link.



ps. I also used CNX to create this tute so the spray painting is a little rough around the edges. You can't add text, or symbols or vector graphics or whatever garbage inflates PS to 'bloatware' status! Nope not a rant.. I'd never do that again

First of all well get to some the basics of getting CNX running properly.

if you have multiple hdd's best set to use at least one of them as a temporary and cache disk.
NEF's are tiff files in disguise! They can be huuugely enormous too. 80-90Meg per file is not uncommon for a D5000-D300 at least and I suspect slightly more for a D700 and D3. Apparently 150-200odd Megs for a D3x tiff file. As you edit you're looking at a tiff file representation of the NEF file you 'see'.. so be warned. Many people reckon CaptureNX is slow. Open a similar tiff file in any other program and see how it works too!. A jpg edited in CNX is finished before you even start!

I now have an uber 4core fast-ish system and speed is not an issue any more.
it used to be, but we'll get to that later too.
speed of access to the data is kind of imperative to getting CNX to operate quickly!

Some other configs of note:
In the (toolbar)Edit-> Preferences tab go to:

General and near the bottom is the tickbox to keep all edits active in the edit list. This is the best way to edit, by keeping all edits active all the time, but as the warning says it takes up processor time, so a very fast processor is an advantage. Multiple core is better!(as I now find out ) CNX doens't take advantage of hyperthreading hyperglycosomy or whatever the hyperterm is! But having a fast multiple core processor keeps the system effectively running, otherwise CNX will take full control of the processor time.. your PC will stop responding. I can now watch videos, encode music and edit with CNX all at the same time. CNX's CPU time on my PC is now minimal.
As for RAM, I've never seen CNX use more than 1.5Gig(I have 4G on a 64bit system), so 2G should be ok depending on other requirements on other systems.
Also by keeping the cache(at least) on another drive separate drive and with easy access, I can regularly delete the contents myself. I allowed CNX full access top the space on that drive as the drive is for cache and temp files only. My images are stored on another separate drive too. makes for slightly faster operation again.(something to do with read/write times/queues).

Also in the Preferences tab, you get access to manage settings files(batch edits) and colour management, among other things. Note if you want ProPhoto colourspace, it doesn't come with CNX, You'll have to download a ProPhoto colour space.. usually comes with PS/LR doesn't it? It does work apparently. I just don't use it.

Two things I regularly do now, because I have two screens: is undock the pallettes from the RHS and throw them over to the secondary screen. Gives me even more image real estate. I also like to enlarge the Histogram indicator too, for more detailed info. I edit by the histogram. click the triangle icon and drag a pallette by the serrated edge to wherever.. and click on the anchor icon on the histogram to enlarge it.

The rest is simple. The Quick Fix edit list is always open by default.. which annoys the hell out of me! If anyone knows of a way to stop that annoying behaviour, I would really appreciate it. My main priority in using CNX is usually WB maybe a quick play with a few PictureControl styles.. just to see, but generally for the (ease of using)Colour Control Points.
I barely use the QuickFix section and my immediate reaction is to close it. The I'll reopen it after editing WB or whatever. it kind of seems silly, but that's my workflow.
Actually! makes more sense too because you really need to adjust Wb before making any adjustments to exposure. Once WB is set(away from my default camera setting of Auto) 99.9% of the time the red channel will be slightly blown, in a perfectly exposed image. Standard procedure, as that's how the camera will set WB as part of it's magical routine.
Make the image look well exposed on the back of the camera despite if it looks blue!

No big deal! It's a workflow I actually like and prefer, as setting WB is like adding a warming/cooling/skylight/UV filter except that you can do it in the warmth and comfort of a desk chair with an automatic coffee making kaboodle.. I just wished someone would invent the coffee kaboodle for me and my chair ... to save wasting my time now.

OK, I ranted and raved and got angry with CNX at one point and everyone still knows nothing of the program... except Deb and Andrew of course

we're going to start with adjusting WB(generally easier done in ViewNX, with at least one fewer clicks) but CNX is still good and easy, and gives me 99.999% of the rendition I expect to see.. unlike LR3 that is??(yep! I installed it again.. 64bit stupidity by me this time.. but I'll persevere).


It's easy! OK, I understand that adobe have to backward engineer NEF files for at least WB info, but they've always looked wrong set the same as CNX.. ie. or as I would in camera.
Presest are standard are but with some different naming systems. We get no tungsten preset, I think that's be incandescent in Nikon equivalence. We get Sodium Vapour lamps.. I see nothing like that in LR. Important(ish) if you shoot at night.
(LR folks!! did I miss something?).

anyhow WB is simple and very veryrarely do I have to adjust tint. It'd be nicer to have more finely tuned values like in LR.. but that is only a problem for me if I use my Hoya SuperHMC Pro1 CPL!!
It gives a slight red/magenta cast where +1 tint(ie. green) is too much, too green looking by comparison. The old el cheapo Hoya Lpol has a slightly green cast but perfectly acceptable with the proper WB adjustment.

At this point no WB has been set yet.. I think that's in the next image to come.

Note the small piece of litter in the image!

OK, one more image and much less text:


set WB to Shade, and it looks just about perfect(as I remember) I stopped using the grey card, as I found it was just as easy to do it like this. I have the grey card in my bag near me all the time, it's not just from laziness that I do it this way. Sometimes the grey card is the best way to go tho.. but you can use a grey point in WB as you should be able to with any editor, and if there is some grey in the image use that as a starting point.

this is good at WB = Shade. Sky ended up brighter than I liked it to be, but the next series of exposures were taken with a GND attached, so I made my amens Even with bright sky where the highlights are about 1/3-1/2 blown and therefore still easily managed i didn't end up needing exposure compensation. I tried(tested) exposure compensation but it wasn't needed, and is only done for the purpose of this tute.

...3/ exposure compensation

[arthurking83]

I need to pre apologise for the stupidity of my spray painting of arrows and instructions, but it became addictive and I subsequently couldn't stop!

So still with the same image, many peeps will comment on how crooked the horizon is, without really looking deeper into the image and not seeing the obvious optical illusion! So right click the image and display the grid.



At this resolution you can't clearly see the straightness of the image, even though it looks crooked. I like find obvious optical illusions like this and to capture crooked looking images in an almost perfectly straight manner Yes the rail on the bridge is seriously crooked! Grid helps and the grid is configurabel in the preferences too. I like it fine and to zoom in at 100%(which not as easy to do as it should be).
I'm not referring to a lack of Ctrl-Alt-0, I'm referring to how when you hit Shift in ViewNX it smart enough to zoom in where the mouse cursor is. CaptureNX by comparison is totally dumb!
A small and realistically it's an insignificant issue, as zooming to 100% is not that hard, but when it's easier in their free program.. just doesn't make sense!

So in that image it looks more red because of the grid. no edit to the image. an illusion.

Does it need straightening.. NO! if so there is a straighten tool which is easypeasy to use, and from memory easier than the PS rotation tool, as well as the LR3 toll I think I once tried to use. The need to edit a 0.18° CCW rotation is as easy as following the straightest line with your mouse.

So forget straightening, that's a teade secret I'm not going to reveal anyhow... lets try Exposure Compensation instead.

You can also make tweak adjustments to the levels/curves here protect highlights and shadows are easy to understand, but could be hard to understand for the total beginner. I still don't fully understand what the program does(in a literal sense) to protect highlights. it can be only very subtle, but it;s also different to -ve exposure compensation.
start with Ev adjustments first to a good balance and then use the shadow/highlights protection to fine tune.

NOTE!! if you do end up using both ViewNX and CaptureNX there is a chance that they both render images slightly differently from each other... like mine do now? the difference is very very subtle, but very fine black tones that I can clearly see in CNX are not displayed correctly by ViewNX.. and now I discover by LR3 as well too!(jpg images!) So I have to figure that one out before I can make detailed comments about.. could be colour management. screen calibrations or whatever. But CNX shows me more detailed shadow tones than the other software!
So I made a -0.5Ev compensation edit. you can slide or type. I like to type as it's more accurate and quicker the histogram is now where I like it to be before I edit(if I edit too).
In most of my images I like the security of protected highlights, but not always tho. If I do let them blow or edit them to be so, they have to be appropriate in amount and intensity..

To save you the silly expense!... instead of purchasing Nik's Viveza plug in.. just bloody well get CaptureNX..

LOL! seriously tho, the Colour Control point editing is where it's at. For a doofus non photo, non graphicsy type person like me, editing is basically a chore. make it easy and I'm in. Make it hard... and I'll go play with my Hornsby train set instead
Does anyone have Hornsby train set they can lend me?.. no!? Then lucky for me, and a few other people I know of, also swear by CCP editing.
This may be hard for others to understand, but how you guys that don't use CCP's can toil away without the benefit of CCP's.... is beyond me!

I've tried a few other programs(Bibble, Capture One, LR, etc) and just get myself stuck in a quagmire of .. I have no idea what. It's been so long since I edited an image and then saved it for keeps using another program.. I just can't see past that little circle and the sliders(or number input boxes, which are easier).

CCP's are like using a camera in full Auto scene modes.. point and shoot!

With one teeny 5 second extra step.. slide!

step one is point(or Crtl+Shift+A) then your mousey cursor turns into a dot and you click where you want to darken/lighten/contrast/colour/etc.
There are a few intricate steps to begin with when(if) you start using CNX tho, and that's accessing the advanced editing ability of CCP's. later. for now we want to edit the entire sky by using only CCP's. nothing else. You need to add at least one CCP, and then rightclick and add more for a large area. You can just slide the size of the CCp to make it bigger, but that makes it broader in it's scope, and that's not the point of CCPs.. but it still works.



Note on the RHS, that you have an edit list with the CCP in there, and a single CCP area to manage.


I'm jumping the gun a little here by having added more CCP's already, but as you add them they attach themselves to the same edit step. Important to note that.
In the sky I want to edit an area to see how it looks, so only a single CCP is highlighted in the list of CCP's in that one step. I darken by sliding(or typing a random negative value, experience will help you with speed on that topic). Starting with -25 on the brightness value is a good average, and then you decide from there. it takes 1second to do that, and maybe 5 seconds to decide if that was enough, adjust it more or less and be happy with the result.
As s green post processing noob, I picked up on that after 2 attempts at editing when CaptureNX was first released. That took about 20seconds to learn.

Many people comment on how hard/difficult CaptureNX is to learn!!
I think they're being facetious! .. or just silly, for the purpose of causing friction in forums??

What exactly is so hard to understand about that edit step?
Stupid thing is that it gets even easier(using Selection Control Points), even though it may first appear to be more complicated.. but yet again, I'm jumping the gun, so lets get back to to CCP editing.
Wenh you have only one CCP, the others are independent by default.
EXCEPT!! if you click on the leading title in the edit step, all the editing steps in that title are then active. It's not a hard task to understand either! If the edit is highlighted, it's active. The point is that there are edit steps and then sub edit steps within that main edit step. Easy to understand and use. Just understand that's how it works.
You need to be careful too. If the edit step is highlighted and you like it, AND you decide to add another edit, you can overwrite the original edit. That's actually a good thing, even though there is an element of making errors. Just remember a highlighted step, or series of steps is what you are working on. I think an analogy is like layers in PS. I can never understand why we need layers. My Pops has 150 layers, and the eggs they produce are superb.. totally and 101% free range(gourmet free range too! :th3)

so layers are confusing(to me) and sometimes I forget and keep an edit step active and then overwrite it with another edit of different meaning. CTRL+Z helps there.

So once we learn that you can have a billion CCPs in a single edit step, we can work on one indvidually, or any number of them independently of the others. You highlight each sub CCP step as you like and then adjust one of them only and the others follow suit.
(I hope that makes sense, because unless you do it for yourself, it probably doesn't).


In this screencap, I have many CCPs in the parent edit step, but only one is active(highlighted) over on the RHS, and that's the only one that will change as I adjust something.

..../4 editing multiple steps in one move.

[arthurking83]

Before we get to adjusting all CCP's in one foul swoop, lets look at the highlights being blown out by adjusting the WB earlier. It's probably taken you 10-20mins to read this far, but it took me about 20seconds to do it in real time, so I was jumping the gun and forgetting to do stuff in what I think is a correct order.
Two methods in which it view blown highlights and lost shadows is via:

Shift + H for colour highlights, Shift + S for colour shadows

and down on the histogram is a Double Threshold tickbox too. This shows you black and white points.
You can use that double threshold to set black and white points if you understand what they do!
Up in the tool bar are three eye droppers black grey and white. With double threshold enabled set a black point on one of the black areas indicated in the image now, and set a white point on one of the white areas of the same image. This is the same as levels and curves to a degree.
This other method of checking and adjusting blown highlights is better.



I used Shift + H to show me the blown colour highlights, so you can do one of either two things here. edit the red channel only, as the red channel has clipped or darken the area completely. Blown red channel at this level is not so bad in print, but we're editing for the purpose of tutoring and I'm going to edit it.
One CCP highlighted and I think -25 set to the B brightness channel. helps to saturate a little too, and you can also lower contrast to reduce the blown highlight. Adding contrast will always push the highlight back into clipping point tho.
Adding contrast and Saturation will make the image look over saturated, and probably not very nicely contrasted either. That of course is totally dependent on the particular image and look you want!(and not for me, or anyone else to tell you what to do).

So I edited that single CCP with -25 brightness and then something else, I can;t remember until I also upload the next screencap



Ooops! Apologies it was -40, but once again I jumped the gun, because I used -25 as per usual and it worked, but I then equalised all the CCP's in the sky section of the image and they needed more, so I eventually went down to -40. You should see that screen cap right about now....



because I hate virtual GND's as an alternative to the real thing, I like CCP's because they do a better job. I used a fairly broad size CCP, and the meaning of that is that it works on a broader scale. and so the lamp also got darker too... not by a lot!... but the program is thinking that I want it dark(and fair nuff too), but it's not really adjusting the colour of the lamp post to much either. The sky is a grey/magenta colour and the lamp post is green. it's now easy to bring the lamp post back to a brighter point without affecting the sky too much, but we need a smaller CCP size. And because I'm now not editing the sky any more I've closed the edit step involving the sky only and starting a new edit step for the lamp.
This is because I want any editing of them to be totally independent of each other as we may have to balance one against the other. So it's good to have separate parent steps controlling multiple child edit points.



I zoom in to 100% view here for accuracy, use Ctrl+Alt+0. for the sky the accuracy was not needed, but I have a highlight on the lamp post and I want to test a CCP on it, as wel as avoid it. point and click and slide again. been there done that and now you're all seasoned post processing types

You may find that as you brighten one point the other independent areas will also follow suit, so you bring down those areas too. It's easy and subtle so not always imperative that you do this. Size of CCP makes a difference too! So in many instances many small CCP's may be required instead of fewer larger ones.
Both methods are as easy as each other to adjust as you're probably adjusting the parent edit step and not just a single sub edit step.(but not always the case if the image is very intrciate)

deleting edit steps:


simple. Rightclick any active edit steps, or groups of steps and delete. That's if you don;t want to take you hand off your mouse, other wise highlight the step and press the Delete key on your K/b!

MOST IMPORTANTLY! this screencap shows you the list of available options that you can effect with edits. Copy and paste can be across to other images(like batch processing) or can be just multiplied onto the same image.. or whatever tickles you.

The one aspect that really annoyed me about LR was the fact that I couldn't figure out a way where I could add/subtract/multiply some of my edits in the history list. it was either the last one in succession or all, or none!
For an uber powerful program that's pretty poor. If I missed something,then it's even worse, as it's supposed to have a uber cool workflow system!.. for who? the Eskimos.

The point of having an easy intuitive workflow is for it to be easy and intuitive.. for everybody! Not just the Eskimos(ie. the elite that live in Adobe nirvana

everyone!!... close your eyes.. rant mode is back!!
I really wanted to like LR3 as it has a great noise reduction element. But lack of proper NEF file support and simple things like user friendliness make is .. well it sucks!

I mean.. it took me ages to figure out that zoom in is no longer Ctrl+ + it has to be Ctrl+= I would be inclined to think that equals woullbe more of a equaliser.. ie. make image fit to screen!
But I think it's catering to Apple fanbois that don't have numerical keyboards!?
note! Get a numerical kb dudes.. they've been around for years now! Same with lappies. I won;t ever buy one without.. even if I have to get an external one. Numerical kb's are the bees pants.. or ant's knees.. I'm no racist!

So!.... I've now stooped to equating Adobe and Apple users as spineless insects, with exo-skeletal compositions and soiled undergarments.. and (if they had my knee precondition) probably disabled too!
I can get worse if need be (and I do expect to get some flack back too )

back to the topic at hand.. making Apple toting LR users happy.. give 'em something to smile about!?

..../5 **warning** cloning!!!

(I know you're laughing loudly Andrew and Deb!)

[I @ M]

..../5 **warning** cloning!!!

(I know you're laughing loudly Andrew and Deb!)

Why?

We simply don't need to clone images, we compose them.

Brilliant series, keep it coming.

[arthurking83]

If it all sounded rosy until about now(with very minor obstacles along the way).. now it delves into the impossibly stupid !!

Hey! remember that small piece of litter I mentioned wayyy back in about post number 2!!


cloning and healing out things in CNX is not for the faint hearted.
It's not for the mathematically brilliant super elite computer programmer either!

it's the best example of Chaos Theory ever devised by human kind, and not only proves it exists... but actually started it the annoying trait!

Clone at your peril. And I'm not referring to cloning out massive huge context aware elements. I'm talking about a 2 pixel dust spot. No matter how hard you try, that 2pixel dust spot is either going to:

blend in perfectly with a 4pixel dab of bandaid.

or.... turn into a 4000pixel extraterrestrial network of valleys and extinct watercourses.. and probably red!
They must definitely come from our distant neighbour when they do, because one moment they not there, the next they are.. it's pot luck, and if you get two consistent answers in succession get yourself a lottery ticket before your luck runs out.

I wanted to post a screencap of a clone touchup but it;s just top damned embarrassing sometimes!..

so onto using some external filters..

No! ok. I'll weather the barrage of ridicule in having to put up with clonging in CNX.

it's not called cloning. nup! it's called CLONGING.. with a G. When you apply it you can hear the clonging of some pc parts... probably some RAM transistor banging their virtual heads against the heat sink(my RAM has heat sinks! :th3).. trying to figure out why this instructor(CNX) is asking them to place a shoe next to the girls eye. But there is no shoe there! they all think, and yet the instructor is adamant that the operator wants a shoe next tothe girls eye in this particular image, even though there was only a small black 2pixel bunny there to begin with. And the poor little RAMites just.. band their little heads because its not a full blown BSOD they have to put into effect! No! they're nto allowed even that luxury! they have to toil away like all those poor adobe users and try to figure out a way to stop the operator form ripping out their heats and stomping on them with steelcaps!
(not that I would ever do that...........in my sheepskin mocco's!)

Ok.. here's the embarrass.. I mean image..

\

OK! nuff said! all done and good night!

err.. well. it wasn't a too bad effort, but for a 10pixel wide and maybe 30pixel spread, the first attempt as you can see is pretty woeful.
You would expect for the artificial intelligence part of the tool to figure out some of the local area and apply something appropriate.
To be honest I was expecting more of the girls eye!
Whadday mean what girls eye!?
Isn't it obvious.. the girls eye of the girl I took a photo of back in 1983 with my holga!

That;s how bad it can be. it introduces random elements that have no bearing on the surrounding part of the cloned area!
the pinkish highlight is kind of trying to replicate what the tool actually looks like. a pink worm. There is no selection/sample area like you get in PS/LR to work from. Your only choice is size..... or panadol
From there, Nik does the rest, and you have to wonder what they were smoking!
I reckon it'd have been much simpler to program a totally basic stamping tool, rather than some convoluted intelligent content aware fil tool.

That's what this is supposed to be, except that it wants to fill in parts of an image being edited with CS5 by a anti government operative in Turkmenistan.. and he hasn't even got net connection?

I think Adobe learned a lot from this disastrous attempt at Content Aware Fill!

So this image shows some inappropriate smooth grey mass being used to fill in a rough grey area in this image not an issue, but in many cases it usually is. So I've learnt to clean my camera more thoroughly, or justify dust bunnies as some as artyfarty and yet undiscovered original idea, with some gaussian blur for effect

but sometimes if you persevere a little!



so you're probably thinking that this now doesn't look to bad! I can assure you it's freaking awesome!

but of course I always knew that'd happen.. CNX cloning is so cool....

..../6 some normal type processing

[Debra Faulkner]

Ahhh, excellent, Arthur, I like tutes that make sense Mind you, if we can't compose 'em, we use Paint.net to clone 'em

[arthurking83]

How about some normal processing ability?

Nik Color Efex Pro software is definitely worth some money spent.
I had it ages ago upon recommendation and viewing the results Andrew was getting with it.

I've only really started using it not too long ago for some effects like film emulation, but even more recently using the Tonal Contrast filter..



quick way to get some contrast back into an otherwise flat image.

All my images will look initially flat as I now only use the Standard Picture Control in camera with no enhancements at all. I used to use various methiods of getting some contrast back into the image, all very easy and quick, but this Tonal Contrast is quite ok in some situations. It does produce more grain like pixel structure, so I think exposing the image(more to the right, and hence flatter) with that in mind may be a better way forward if I want to keep on using it.

and then we have the standard barrage of selection tools like:
gradient selction


and radial selection


which I use mainly for vignetting a boring looking image or maybe even for some gaussian blur or whatever.
Note the '-' area selected for using the radial selection tool tho.
You do this one in reverse where you begin from the centre of the image and drag a selection line outwards to the edge of the image.

I've tried the vignette filter in ColorEfex pro, and not too pleased with it.. but I probably need more time to figure out how to work it better??



what you do with a selection tool like as an example is place the starting point at number 1 and drag it towards point 3. Point 2 is kind of like a half way point, but the program always remembers your last use of the tool and starts the 'mid point' there again.
The numbers in edit area in the screen shot correspond to a gradient range indicator up in the toolbar. You can adjust either for the values that you need.
You then apply whatever editing selection you require. It could even be a CCP again. but generally a standard adjustment like brightness/contrast or whatever. I have used this tool when I get stuck with posterisation in blue skies.. and add noise to the selection. Adding a very fine noise to a posterisation affected area removes some of the banding you may see in it!

then you get a few of the normal tools you see in almost al edit software, and these are the normal boring tools I hate using 99% of the time. But sometimes some areas can be easily edited with these tools used in combinations:



as an example of mix'n and match'n selection tools:


I created a lasso selection in the lower half and then applied a reversed gradient selection to it, but I only wanted the start of the gradient from the lower 1/3rd of the image and the graduation range with less opacity towards the edges(as opposed to the standard way it works)
Not that I use that kind of editing all that much, if ever, but it can be handy if ever needed.

another use is to edit strange parts of the image in strange ways. never done this till I did this tute and this screencap. I used the lasso tool to select an area of the bluestone support and then painted a large single brush spot centred just outside the lasso selection.. that's what it looks like.

..../7 more specialised and tedious editing

[arthurking83]

this first one is standard fare using the gradient selection tool again in two spots.



I didn't want to use the radial selection nor the Nik filter as I didn't want or need to darken the upper sky any more. it's already dark. So the gradient selection is handy for some edits steps as half baked as this one usually seems to be

This is another failing of CNX: if you want to go back and redo some of that gradient selection and alter the midpoint range you have to do do it whilst that range selector(the dotted line) is still active.. as it is on the LHS here, but not on the right.
it'd be nice to be able to edit this at will, but once done and a new one is then made the already set gradient is set, and can;t be removed or edited again. To undo the gradient on the RHS corner I have to delete the entire gradient selection and that means the LHS too!
it used to be a problem for me when I wasn't sure of what I wanted to do(with the image), but now it's less of an issue... because I tend to do it less

and what I would do as an example is to darken and contrast the selection..



Now I probably want to attempt to bring back a bit more detail from the middle section of the scene. It was originally a little flat due to there being no light falling in that area, and then it was darkened slightly with the use of the CCP's in the sky. So you could add a few more CCP's again...



and brighten and contrast and saturate that way

and then duplicate them over and over keeping t each one fairly small...



OR!! you could use the Selection Control Points tools

]

These are every bit as easy and cool to use as CCP's are and basically the same thing, but work a little bit more efficiently in getting only an area(or colour) specifcally selected as seen in the image. Note how only the tree areas are selected. The green tint is the selected area about to be edited. whereas with a brush tool you get the selction in th esky and on the grass and just whatever.. as you'd get with a real air brush.. over spray...
as you do here like this


So I used about 4 or 5 of those selection control points along the banks of the Yarra and spruced it up a little bit with :

D-Lighting



Didn't look bad.. but nup! not good either.

so I tried chroma(colour editing) which is a little bit better than the standard Saturation tweak only.
You can use any of the LCH tools individually and independently of each other. If I do LCH, I usually only do Chroma, sometimes Colour Lightness, never really used Master Lightness all that much ..maybe once and never ever hue. I don't even know what it'll do if I did use it, so the fact that I never have used it means that I think my hues look fine.
This application is only going to affect the mid line of the trees grassy banks and parts of the city buildings

....8/ going back through the ages

[arthurking83]

With that last edit step I think I actually kept the D-Lighting step and then copied that main 'parent' edit step and pasted it as a new step, and that initially looks scarily bad, but I wanted to add that LCH edit step to the same areas of the mid line and intricate grass through the detailed balustrading, etc, etc. So instead of creating the same edit steps all over again, I copied and pasted the previous one I was happy with, and then changed the adjustment method from D-Lighting to LCH.

So in a sense it's like creating layers really! But there are no physical layers. the layers are actually edit steps. So unless you're creating complex vector graphics why does PS need layers at all? Some convoluted sneaky manner in which to perform some image manipulation trickery me thinks!


anyhow, here's a screencap of how you go back and re edit an already made edit step that may have been affected by a new edit step, or one that you just weren't happy with after some time.





two screencaps of a copy and paste edit process.

note how the edit step is highlighted in the copy process, but then has to be manually unhighlighted when the paste command is made. If you don't do that, the paste process is made back on top of the original edit step, so you effectively double up on the original edit steps. Of course if that;s what you want to do, then you've done well, but in this case I wanted to create two separate edit types on the same selection.

made easy with this system.
Once again, I could have added any one of the available edit adjustments available.

One thing I could never figure out how to do in PS.. copy and paste parts of the history list.
That manner of doing things like that seems so natural and easy to do.. point and click.

So with edit step listed as #4 in the top section was originally a D-Lighting edit, I went back to that step and changed it to a LCH edit step and the newly 'pasted' step #5 ended up being the D-Lighting edit step.



the priority of placing edit steps also makes a slight difference in some cases.
eg. using NR first, and then adding grain to an image in a subsequent edit step. The grain will still show, as the edit was made after the NR step... etc, etc.

...9/ more to come but I forget what it was

[arthurking83]

Note how in some of the selection based edit steps there is a a small icon to the right of the selection. That is for the purpose of displaying either the selection, or the mask.

Use the show selection it looks like this...



Using the show mask it turns negative where the image turns black and the mask shows up as white areas. any difference?? let me know if you see it

Sharpening is a cinch but tedious and boring!

Jason Odell once recommended a setting base of 55 5 5 for the D300 when the D300 first came out. Jason writes some good how to books on using CaptureNX and gets amazing results with it.. and he basically got me stuck on the program! Google him if you want more info and his CNX ebooks are probably worth the money for some, I like to understand something by feeling my way through it.
I may even be getting to a point where I understand concepts like colour management and the differences in how colour spaces work .... hopefully soon!

Anyway.. I use USM for sharpening again now. I was turned to the dark side of high pass filtering and then I went back to USM I breiefly tried in camera sharpne settings but soon found out that the settings are way to coarse and produce more grainy pattern that even more sharpening via USM. High Pass is still good to tho.

But with USM there is always a few inputs to be made and that always takes three input areas and that means wasting time!... so create a batch job of it.



how to access the sharpening tools

and instead of using that routine over and over again, once saved to a batch job, it;s only tow clicks and no inputting anything again.


Like I have set in my batch edit list there.

So in the list as shown next, you make any number of edits and then you can give yourself the option to save them:



you can save them for any reason or purpose, and I only realy wnat them for using later, so I saved any and all repetitive tasks like my 'save to web'(resize and light sharpen) routine I have in my list. so when you click the save adjustments tab, you get this list...


... and I already used the save dialogue to save on unnecessary screencaps.

In the box on the left is a list of your current edit steps. tick or untick the ones you want to keep. each ticked edit will then go into the saved batch job, and when you hit that batch process from the list of saved settings you'll get instant gratification of creating edit steps faster than ever before!

The only multiple edit saved settings I have are the resize and sharpen settings I regularly use. save to 900pix 1200 pix and now 1024 pixel edits with a very light sharpen step too.
kind'a like save to web. and instant. and when you save the settings(or batch steps) save them to a file where you know you have easier access to them.
because I want to save this kind of data because I know I'm going to use it again, I saved it to the hdd where my photos are stored on my PC. it also gets backed up regularly when I do a drive backup using my RichCopy software too, if I re-install CNX again, I don;t have to go fishing for them in some obscure folder again, they're stored ina folder called CNXBatchJobs or something like that. they're saved as .set files, so no image viewer will ever confuse them for jpg images and apply some bezerk save to web routine by accident


....10/ the end is nigh!

[arthurking83]

cropping is easy as !

One thing that's always caused me grief in PS.



two easy settings and a few small easy to understand variables to adjust if need be, otherwise just use freehand!

NOW! something I do on a regular basis and annoys me to bits.

when I do something like my save to web batch edits, I always forget to unhighlight or deselct the last edit I'd made. So as an example in this screencapture...



where you see the two highlighted edit steps of resize and USM, before I add that batch edit, I'll have the Crop edit highlighted and forget to just blankly click anywhere on the edit step list to de select any edit steps BEFORE I hit the batch step. When I hit the batch job and the last edit step is highlighted, the batch job overwrites the crop edit and the image suddenly is uncropped again!.... at least that's a good visual cue to know I stuffed up.... again!

OH! and don't forget to always save the NEF file too. save any subsequent tiffs of jpgs if you feel is important, but the NEF is much more important.

the reason you used CNX(or even VNX to edit your NEF files is that every one of the edits you made is fully reversible and the NEF can be restored back to it's original unedited state.. or!.. Or that it can also then be edited agan by an even more imporived version of CaptureNX.. like the impending CNX 3

Some folks talk of the possibility that closed format manufacturer specific file formats are not good to use as the manufacturer may cease support for it. Well, after over 11 years Nikon still fully supprt the NEF format and any original 1999 D1x model is stil fully supported by CaptureNX 2, as well as CaptureNX v1, and even still the original Nikon Capture of various versions(I have only version 4.4).

Any and all edits made with earlier version are fully supported by the later versions, but the same in not applicable in reverse.. which is a 'fair enough' situation to have. So CaptureNX 1 cannon see some of the edits that CNX2 can perform.. like the stupid clone tool.
I have CNX2 now and CNX1 is only there for me because I need the serial key to properly install my upgrade version of CNX2. I'm never going to go back to the earlier version. that'd be stupid!.. and yet the later versions still support hardware and software(adjustments) made on files dating back 11years!

I think that at least the NEF format is secure by Nikon for a very long time to come yet.. I can't be sure about other software makers intent tho!
and it's so compact for a tiff based format too.. so it keeps hdd upgrades to as much of a minimum as possible for now.

The end:

[fillum]

Thanks AK, I appreciate the amount of effort you've put into this. I've only had a quick skim through but have already picked up some things I didn't know, even though I've been using CNX2 for a while now. (I'll go through it in detail when I've had less red wine ).

Only thing I didn't see that I use frequently is the "Show Selection" / "Show Mask" options on the Color Control Point and selection tools which show a black-and-white mask for the selected area. I find that this often makes it easier to see the affected area than the light-green overlay.


Cheers.

[arthurking83]

yes! I didn't post a screenshot of that show overlay/mask, as i had way tomany other screencaps to upload too.. had to limit them somehow. But it is mentioned in one of the sections where the green overlay is. I showed the overlay only as that's the default manner in which CNX displays the edit tool.

TBH I've never used the show mask method all that much, and I sometimes use the show overlay method when I've confused myself with invisible editing like NR or sharpen or whatever.

Actually, in making up the tutorial and using that brush tool and colorise edit step to paint the text and stuff, I finally got to see exactly the difference between the use of the different Blending Modes when applying certain selections.

Either on it's own or behind the opacity adjustment dialogue there is a blending mode setting too.

You get things like Normal, Lighten, Screen, etc and they all have different and interesting effects with the selection made.

hard to describe, and I'll get back into the manual to see exactly what differences if any they can have on the final output.

I've only ever really used the Normal mode,and I remember a recommendation to use the High Pass edit step with a Screen blending mode.

[maccaroneski]

Arthur, this is fantastic.

I'm in the same boat as Phil - a quick skim has shown me a few things already, but am looking forward to going through it all in some detail.

Thanks!

[arthurking83]

Start off walking before you try to run!!

I've been using CNX now for 4 years and most of it's abilities are beyond my full comprehension.

All I basically do is CCP's and cropping, to the vast majority of my images.

Only recently I've started to use that Nik filter called Tonal Contrast, which I've had for about a year or more now.
But I remember when using it once before the default 30, 30, 30 settings tuned me off. I got to changing the values to 10, 10, 10 and it looks a little bit nicer(less tone mapped, and HDRish .. which funnily enough rhymes with garish! ).

Despite whether my images look bland or boring, they do end up looking for the most part as I wanted them too.. from how I remember them to look through the viewfinder(when it works!).

Those concepts I've explained up there are things that I have done, but even the edit steps like the vignetting ones both circular and graduated selections are long and tedious for my liking.. even though in reality they're quite quick and easy.

It's really the colour control points and the control point selection tools that work the best, or quick and easiest.

ps. Tony, do a google search on CNX USM settings for the D90.
Jason Odell used to frequent the DPR forums(I don't know if he still does, they're a horridly tedious place!) and offer tidbits of advice about various aspects of CNX for specific cameras.
Even though the D90 shares the D300's sensor, the AA filter may be different.
His recommendations are made with usage experience.

I think Thom may also have advice on that topic too.

You could also try to use the values I now use, and see how they work for ya, and tweak them as needed. the trick is to get the sharpest image with no haloing in the really fine detail like fine twigs and hair and stuff. Always at 100% pixel view, as there may be some haloing at smaller viewing sizes on your screen.

I will add a few more bits of info(with screenshots again) as I get a chance to do them..... but more importantly remember them

If anyone else has tips or hints to add, especially if we can work out how to close down that annoying Quick Fix tab!! ..... feel free to add to this thread(with screenshots).

[maccaroneski]

Arthur I have the Jason Odell e-book - 40/5/5 is the prescription for the D90.

And I love the tonal contrast filter, although I rarely do anything other than either (a) adjust the opacity; or (b) control point or paint the effect in and out.

And I love your idea for a continuation of the "tips with screenshots". If after going through yours in detail I think of anything (although I doubt it) I will post it here. It may even be worthwhile for NX2 users to simply look at something that you've done above, and if they have a more efficient or even different way of doing it (although again I doubt the "more efficient" part) throw that in too.

[HansPeter]

Hello Arthur,
I just want to congratulate you on all the work you pot in to explain the Capture NX 2. That takes a lot of dedication, thank you.
Cheers Hans

[DJT]

nice tut Arthur.

I have an old version of LR (1.4) & also PSCS3.
LR is ok, but very limited in what it does. I don't mind it's cataloging though.
Photoshop can do sooo much more than LR. but if I don't use it for a few days I completely forget how to use it for anything other than the basics. Probably should go to the trouble of writting down a workflow diagram.
I reckon I might download a trial version of Capture & see if I can understand it. When I can't I can just come back & check out this tutorial.

I see it does a nice job on landscapes. Are portraits just as easy should I do a search (too late I already asked).

[maccaroneski]

David,

It sure is.

The attached: tweaked the white balance as the default was a littel cool, a couple of control points to brighten up the what were greyish areas to get the pure white background, adjusted the contrast, gave it a curve bump, applied a skin softening filter (part of Nik Colour Efex plugin), spot healed the catchlights to look a little more natural, cropped - all told about one and a half minutes. There were some skin blemishes that I worked on with the spot healing tool for another 5 or so minutes (very carefully as it was a portfolio shoot).