Perfect Colour Made Simple. Prepared & Presented by David Harradine Photography & Training

Transcription

1 Perfect Colour Made Simple Prepared & Presented by David Harradine Photography & Training

2 2

3 3 Contents: Session One: Understanding Colour Management Why we need Colour Management Input, Display & Output Devices Monitor Choice, Calibration & Profiling Working Space & Colour Settings Session Two: Image Processing Photoshop Vs Lightroom Lightroom Catalog Management Raw & Batch Processing Digital Camera Colour Management Session Three: Fine Tuning in Photoshop Contrast Vs Colour Blending Modes Layer Masks Complex Selections Session Four: Preparing for Output Professional Sharpening Workflow Soft Proofing for RGB & CMYK

4 4 Session One: Understanding Colour Management Why we need Colour Management Colour management is the process of ensuring that as images move from one device to another, the colours stay the same, or as close to the original as possible. Colour management can often seem overly technical and confusing, but I assure you it is incredibly logical and accessible once the fundamentals are understood. When evaluating our colour workflow it s best to see it in 3 distinct stages 1. Input scanners and digital cameras 2. Display monitors, projectors 3. Output CMYK print, RGB print, specialist wide gamut print and the Web This way we can be sure we have done all we can at each stage, before moving on to the next. Or, more importantly, we resolve problems as they occur, rather than down the line where we may get a lesser result. Digital cameras, scanners, monitors and printers all have their own unique qualities and perceptions, meaning the data they produce needs to be interpreted. This is the role of colour management; to understand and compensate for the unique behaviour of our devices and maintain consistency in the appearance of image data as it travels from one device to the next. Input Display Output Device Variation In the early days of digital imaging, manufacturers tried to produce high quality monitors that all produced the same colour response from the same data. Of course this was impossible, as there were so many variables in the manufacturing process. As a result of variables such as materials used, temperature control and tolerances set, each monitor would have it s own unique response or fingerprint and however subtle the variance was, the fact was no two monitors were identical.

5 5 Once it was accepted that no two monitors could be produced identically the solution was obvious. Measure the monitor against a standard, work out how far out it is, then include an adjustment that compensates for it when displaying data. Colour Gamuts A colour gamut refers to the volume of colour a given device is capable of containing or re-producing. Put simply, the gamut of your monitor is the amount of colour it can display. Colour gamuts are, in fact, three dimensional volumes or containers, with a top, bottom and sides. The 3 coordinates of the colour gamut, map out the achievable Hue, Saturation and Brightness values for the device. Hue is the position of a colour on the colour wheel, like red, yellow, orange, purple or whatever. The hue is the broadest value of a specific colour that needs to be further refined in order to describe the colour well. Saturation is the intensity of the Hue; a high saturation colour will be very rich and vibrant whilst a low saturation colour will be pale like a pastel. A pink colour would be in the hue of red but low saturation whereas a fire engine red is also in the hue of red but very high saturation. Highly saturated colours are must commonly the ones that fall out of gamut from one device to another. Brightness is different to saturation; as a colour becomes lighter or darker, its hue and saturation do not change, just it s brightness value. If saturation is like adding or subtracting grey, luminance is like adding black or white. On Mac OSX you can view device gamuts via the ColorSync Utility, found in the your Utilities folder. On Windows you can download a gamut viewing application see link at A more professional and feature rich gamut-viewing application is ColorThink, available in both demo and full versions from Chromix.com.

6 6 Colour Models Colour models are ways of describing colour; they are languages or formulas based on the components or make up of the colours. RGB, CMYK and LAB are all colour models. Some colour models are device specific and others are device independent. The difference between the two is a very important concept to understand, and is ultimately the foundation of colour management. Device Dependent Colour A device dependent colour model describes how a given device produces a given colour, not necessarily what the colour looks like. RGB and CMYK are device specific colour models. The colours produced by their numbers will depend on the device behaviour. When a scanner scans a colour patch and decides the colour is Red 180, Green 130 and Blue 85, that s just the scanner s opinion, these RGB values are how the scanner interprets this colour. However, when these values are sent to a monitor you can be sure they will have a different meaning and produce a different colour. This is because RGB is a device specific colour model. So too the same CMYK values on three different printing presses will most likely produce three different colours if CMYK values alone are the sole communication. Imagine if we did not have rulers or tape measures. Imagine we simply went around estimating the dimensions of objects and passing on the estimated information in inches or centimetres. The ultimate value of an inch or a centimetre would not change, however, the perceived value as estimated by any number of people would vary greatly in the absence of some type of reference like a ruler or measuring tape. And this is exactly what is happening in a non-colour managed environment. RGB and CMYK values become less certain as they have been generated by any number of unqualified devices, without the use of any standard or reference. In a non-colour management environment the following happens Scanner produces some RGB values Monitor views uncompensated scanner values The colour changes The uncompensated values are then passed on to a printer The colour changes again In a non-colour managed environment the colour numbers do not change but the colour appearance does as various devices attempt to reproduce the colours.

7 7 Device Independent Colour Device independent colour models describe what the colour actually looks like, regardless of the device reproducing it. Much like a colour tape measure, which is exactly what the scientific and medical community identified the need for in 1931 when they created CIE ZYX 1931 a device independent, reference colour space. This allowed everyone s colour experience to be accurately referenced to a point that could then be communicated to others with access to this reference table. A little like two people with tape measures talking inches or centimetres as opposed to two people without. CIE ZYZ 1931 took out the guesswork for people communicating colour, which is exactly what colour management does for our devices. With the distinction between device dependent and device independent colour models in place we have the beginnings of a colour management system. The next most important components are the Device Profiles, which are the descriptions of the devices in our workflow that enable the colour management system to allow for their unique colour perception. Enter The Profile Using device independent colour models device behaviour can be measured and qualified. These measurements can then be used to adjust the data going to or coming from the device. This is exactly what a profile is and does, providing a map for the colour management system of exactly how to present the colour information to each device, to ensure consistent colour is maintained. In a colour management environment the following happens Scanner produces some RGB values Scanner profile corrects the data by comparing to reference Data is passed to the monitor Monitor profile is assessed and RGB values are changed accordingly Colours display correctly Data is prepared for print by converting to the printer profile Numbers are changed accordingly Colour prints correctly In a colour-managed environment the colour numbers do change to accommodate the various devices so the colour appearance does not change.

8 8 A handy analogy for the profile is the prescription lens. When a person has their eyes tested the results will determine the amount of correction required in a lens. Each person requires a specific amount of correction in order to achieve his or her optimal vision. A person s lens prescription is like a profile of their vision. With a scanner, digital camera, monitor or printer the same is so, however, unlike prescription glasses the correction is a colour one rather than focus or magnification. By identifying colour imbalances we can compensate accordingly. So if a monitor is revealed to be too magenta the profile will introduce extra green to balance it out. Prescription lenses to not alter your eyes, but rather they modify the information coming in to them. So too profiles do not alter your devices, they modify the information coming from and going to them. Types of Profiles (canned Vs custom) Canned Profiles A canned profile is a generic description of a particular model scanner, monitor or printer; they narrow the variables down to your device model but not down to your specific device. Most scanners and printers will ship with canned profiles, which automatically appear in your colour settings when you load the drivers. Most printers will have multiple profiles available to describe the printer s behaviour on different types of paper, such as plain, photo and glossy. Canned profiles are produced by the manufacturers and attempt to make a broad colour generalisation about the model printer they are describing, which may be one of thousands, not the actual unit you purchased which is one of one. What they actually do is test many printers and average the results. Custom Profiles A custom profile is a profile produced specifically for one single device. A little like tailor made clothes versus clothes off the rack. A custom profile is produced using measurement data taken from YOUR scanner, monitor or printer not a similar one. A custom profile is only as good as the measurement data it is produced from, but assuming all has been measured and calculated correctly a custom profile is the most accurate profile you can have. To create custom profiles specialised colour measuring hardware and software is required, such as colorimeters to produce monitor profiles and spectrophotometers to create printer profiles. We will look at these tools in a little more detail in session 2.

9 9 The Display System When it comes to getting our display system right, there are two concepts that you may be familiar with, calibration and profiling. These are two separate processes with separate goals, so it s important to understand the difference between them. The main differences between a cheep and a half decent display are it s ability to be accurately calibrated and it s ability to preserve that calibrated state. Cheaper monitors will be all over the place. Currently Eizo and NEC are amongst the best reference displays on the market. These high-end displays actually have a separate video card installed, which is dedicated to managing calibration. This enables the most accurate and detailed calibration control. Much like a lens or even audio speakers, as you move up to the higher quality end of the market the improvements become about refinement and consistency in specific details. With a display it s things like brightness uniformity from left to right and top to bottom, reduced problems with viewing angles, wider colour gamuts and the ability to receive and preserve a highly accurate calibration. Calibrating Calibrating is physically altering the behaviour of the device to get it as close to an optimal point at possible. We do this to maximise the useful life of the monitor and to minimise the compensation the profile has to do. Calibration adjusts the actual monitor like adjusting the brightness, contrast or colour on your TV. Profiling Profiling is producing a record of the calibrated state, that the colour management system can use to compare to a known standard or reference in order to compensate and correct with total precision. It is important to calibrate your display before profiling, to minimise the amount of adjustment the profile has to make. Calibration Tools For serious colour management and display calibration you really can t go past a dedicated measuring device like the; Datacolor Spyder3 Elite X-Rite Eye One Display Pro These devices are cheaper and easer to use than ever and are essential for anyone wanting to ensure accurate colour communication from one device to another. Despite the fact that LCD displays drift much more slowly than the old CRT s did, a good regime is still to be calibrating your display at least once a month.

10 10 The Working Space So far we ve looked at device profiles, profiles that describe the behaviour of a particular device such as a monitor. Once this behaviour is profiled it can be compared to a standard reference point and compensated for. However, most device spaces exhibit two problems that undermine their usefulness as editing spaces. They are not grey balanced which means that equal levels of Red, Green and Blue will not produce neutral grey when displayed, as they should. They are not perceptually uniform which means if you mix a neutral tone by eye it will contain a colour bias in the numbers, as it should not. Once we have captured a file we will generally want to edit it somewhat, whether are subtly enhancing the existing tonal and colour information or completely re building the image, ultimately we are re-defining the data that makes up the image. So what space should this re-definition take place in? The input space seems logical as that s where the data came from, the display space also seems logical, as that s what we re looking at, and the output space also seems logical, as that s where we re going. Well the answer is D none of the above. For our working space, or RGB definition, we are going to utilize an ideal, abstract space. Abstract spaces or working spaces are the spaces we use to perform our edits in. Working spaces are device independent RGB spaces that describe an ideal colour environment, free from the various device inadequacies. These RGB working spaces blur the line a little, as to date we have said RGB is a device dependent colour model. Well in the case of the abstract space we take RGB as the model and create an ideal colour definition for editing purposes. Some common abstract working spaces that you may have crossed paths with are; srgb srgb, - small Adobe RGB medium ProPhoto RGB - large srgb could be best summed up as simple RGB. srgb aims to simulate the average un-calibrated monitor and therefore is a good choice for images intended for the web. In fact most web browsers now assume srgb of all images without profiles. Always convert your web graphics to this colour space as a last step before compression. srgb is not a great choice for your working space, although it does simplify the process of moving files between colour managed and non-colour managed applications. I like to convert all images to srgb that I have to hand over to anyone who may not really know what they are doing with RGB files, or will be viewing them in some non-colour managed application. srgb is a common default for many devices and applications as it is possibly the most widely compatible flavour of RGB being the lowest common denominator RGB colour space.

11 11 Adobe RGB 1998 Adobe 98 is medium to large and more suitable for most printing conditions. Adobe 98 is the default RGB space for all of the pre press default settings and is unlikely to cause you any grief unless you are aiming for a very specific, high end result. I would comfortably recommend Adobe 98 as a good all-purpose RGB workspace. Adobe RGB could also be considered the photographic default as most high quality displays can exceed the srgb space, but none exceed Adobe 98. Pro Photo RGB Pro Photo was original designed as archiving spaces for wide gamut scans, however more recently it has become a popular working space. ProPhoto is really only going to be beneficial if you are outputting to a super wide gamut device, and even then your file would need to contain those highly saturated colours to truly benefit. ProPhoto is in fact very dangerous if miss handled and will cause disastrous results in print if it is misinterpreted. So if you were ever supplying anyone with ProPhoto RGB files you would need to absolutely sure they knew what they were doing colour management wise, otherwise srgb may be a safer option. Due to these complications I would strongly recommend testing the limitations of Adobe 98 in your workflow before deciding you need to use ProPhoto RGB The above diagram clearly highlights the limitations of the srgb colour space, unable to entirely encompass even the CMYK gamut. The diagram also highlights the enormity of ProPhoto RGB, extending beyond the visible spectrum. As attractive as it may be, in theory, to chase these highly saturated colours that live beyond the Adobe 98 gamut. In practice it is very difficult to see a clear gain. At least in any tests I have done or seen.

12 12 Edit > Colour Settings The default working space we use can be set in Photoshop in our colour settings. We must have a default RGB and CMYK space. These default spaces will be utilised in the absence of any further information. A file with no profile is considered un-tagged by Photoshop and will be assumed to exist in the respective default space. Photoshop defaults to srgb for RGB and US Web Coated SWOP v2 for CMYK and has done so since Photoshop 6. Our default RGB space will be the colour space that all RGB data will reside in unless it has a profile telling the colour engine otherwise. Our default CMYK space will be the CMYK space all data will be converted to, when we go Image > Mode > CMYK. Technically CMYK is not a true working space as it s a device space. But you still must choose a default. Above we see the journey from input to display to output, as facilitated by the device independent profile connection space. In the past data would have to be scanned in a particular way, depending on how it was to be output and multiple outputs would require multiple scans. For example, one scan produced for newsprint and another produced for glossy print. Now data can simply be captured to the best of the capture device s ability and the decisions about output conversion can be made from the one, device independent, file later down the track. This is the advantage of device independence, where the user controls the device rather that the other way around.

13 13 Assigning Vs Converting When it comes to actually working with this colour management system we are often presented with the options to either assign a profile or convert from one profile to another. The difference between the 2 is significant so it is worth discussing. Assigning A Profile When we assign a profile to an image we do not change the actual RGB numbers that describe it, we only change their appearance. Therefore you could repeatedly assign a variety of profiles to an image, each profile would change the appearance of the image but the RGB values would remain the same. Assigning a profile is most common for input profiles where we want to interpret the numbers that the input device captured. Think of assigning a profile as placing a filter over a lens, it will modify the appearance of an object but will not modify the object itself. If the file is saved and closed the profile can easily be removed at a later date with no effect on the raw data. Converting from one profile to another Converting is a very different story. When we convert a file from one profile to another we are asking the complete opposite of the colour management system. When we convert we are asking the colour management system to preserve the appearance of an image in a new colour space or profile. The only way this can be achieved is to change the numbers describing the file. Converting is most common for output when we want to preserve the images appearance but describe it with numbers most suitable to the chosen output.

14 14 The Profile Mismatch Below the Profile Mismatch dialogue is telling us the file we are about to open has an embedded profile that is different to our working space. It is then giving us 3 choices for how to proceed, with Use Embedded Profile is chosen by default, as this is the policy chosen in colour settings. Using the embedded profile is the only option that will respect the files current condition. Converting will change the numbers whilst preserving the appearance so is possibly the second least destructive. Whereas Discarding will ignore the embedded profile and Assign the working space, which will change the files appearance The Missing Profile The Missing Profile dialogue is telling us the file we are about to open has no embedded profile, so is just basically a bunch of numbers in need of interpretation. It is then giving us 3 choices for how to proceed. Assign Working RGB is chosen by default, as this is the policy chosen in colour settings. A file with no profile is referred to as Un-Tagged and must have some assumption made about its colour space when viewed in a colour-managed application. The first two options above will basically do the same thing as if a file is opened un-tagged its appearance will be rendered by Assigning the current working space. The second option will produce the same appearance but also embed the profile when it s closed. The third option allows us to Assign a different profile to view the file through which is useful if we happen to know what it should be; otherwise we are forced to guess. This is why un-tagged filed are not a good idea, especially RGB ones, and are often referred to as mystery meat?

15 15 Session Two: Digital Image Processing. From shooting film to shooting jpegs to shooting raw the evolution of digital image processing has taken many twists and turns. The early days were all about getting a result comparable to the non digital process however more recently it has become all about workflow and efficiency. Early 1990 s Everyone shooting film, scanning and tweaking high res files in Photoshop. Photoshop edits are instantly (destructively) applied unless you save a copy. Late 1990 s Adjustment layers introduced in Photoshop 4 for non-destructive editing. Most users don t know they are they or why they are needed? Colour management poorly adopted in Photoshop 5, 5.5 update tries to fix. Colour management improved in Photoshop 6 and little has changed since. Early 2000 s Everyone moving to digital, Photoshop 7 introduces a file browser. Mid Photoshop 7 cycle Camera Raw plug in released. Mid 2000 s File Browser evolves into Bridge in Photoshop CS2. Discussions about Metadata and workflow begin in earnest. The power of metadata based raw processing begins to be understood. Late 2000 s Camera Raw & Bridge evolve into Lightroom to further streamline workflow. Photographers of all levels flock to Lightroom for its database management. Images not seen for years start being seen on a regular basis. Lightroom 2 introduces the Adjustment brush for selective edits. Early 2010 s Photoshop CS5 and Lightroom 3 really sing. Adobe wonder how Lightroom 4 can possibly improve on version 3?

16 16 The Lightroom interface Lightroom aims to be as intuitive and straight forward as possible to use and the Lightroom logo on the top left hand corner can even be replaced with your own logo from the Identity Plate preference setting, under the Lightroom menu on a Mac and the Edit menu on a PC. Lightroom is a module-based application, which means it has a number of distinctly different areas for specific functions. Note the five-module menu across the top right hand corner, Library, Develop, Slideshow, Print and Web. The Library and Develop modules are where the vast majority of the work gets done, with Slideshow, Print and Web there for presentation or output. The Lightroom interface will scale very nicely to larger monitors, with the side panels and menu items remaining the same size and just the centre image area getting larger. Hitting the TAB key will collapse the left and right panels and adding the Cmnd/Ctrl key will include the top and bottom panels. Same again to bring them back. The Library Module The Library module is where you will spend most of your time in Lightroom; this is where you can browse your images, rate and rank, add metadata and keywords, even make some quick, basic adjustments from the Quick Develop menu on the right. Across the bottom is your filmstrip view and in the centre is your viewing area. This viewing area can be set to Grid view (G) or Loupe view (E) simply by hitting the above-mentioned keys. The image on the previous page is in Grid view, where you can see a scalable row of thumbnails.

17 17 Sorting in the Library There are a number of methods for managing your images in the Library that are not entirely dissimilar to methods you would use in Bridge. In fact rating systems such as Stars or Colour Labels will be seen in Bridge if added in Lightroom and vice versa. Flagging Perhaps the simplest of these systems is the Flagging. Flagging basically has 3 states, Un-Flagged (default), Flagged or Rejected. To flag a photo or group of photos, select them and go Photo > Set Flag > Flag or simply hit the P key (that s P for Pick). Or to reject a photo or group of photos select them and Photo > Set Flag > Reject or simply hit the X key. Once Flagged or Rejected photos can be reset to Unflagged by selecting them and going Photo > Set Flag > Unflagged or hitting the U key. Adding the Shift key will apply the requested operation and move to the next pic in the row. When photos are rejected they become pale or ghosted when they are not selected. Then they can all be deleted either from Lightroom or from your computer by hitting Cmnd/Ctrl + Delete. Note they do not have to be selected to be deleted in this way, as long as they are rejected they will be included. Stars You can also apply a star system from one to five to further rank your photos by going Photo > Set Rating and choosing a number or simply hitting the required star number 1, 2, 3, 4 or 5. Zero will remove all stars and the left and right square bracket keys [ and ] will increase or decrease the current rating by one star. Labels In addition to Flags and Stars you can also use coloured labels to further sort your images. This may seem excessive until you have a bunch of images that you have already rated by a star system and a client then says we want 5 prints of theses pics, 3 prints of these ones and 1 each of these. Using a colour-coded system makes this super efficient and fool proof as the colours labels are independent of the stars. You can utilise coloured labels exactly the same as you would stars but from the Photo > Set Colour Label menu or with the key strokes 6, 7, 8, 9. Stacks Yet another useful feature for organising your images is Stacking. By selecting a group of images and going Photo > Stacking > Group into Stack or Cmnd/Ctrl + G you will take all of the selected pics and stack them on top of each other as you would a mounted slide on a light box. The S key will then expand or collapse the stack at will and Shift + [ or ] will move an image up or down in the stack. Cmnd/Ctrl + Shift + G will unstack the images.

18 18 Metadata Metadata, which literally means data about data, is yet another sophisticated tool to help you manage your images. The best example of metadata in action is the camera data or EXIF data as it s called. This is the information about your camera and camera settings that is saved with your images at the time of exposure. Lightroom and Bridge (and many other applications for that matter) can access this information and not only display it but sort and search on it. Lightroom has a Metadata menu in the right panel in the Library. This menu can be customised to display a range of Metadata from EXIF data, IPTC data, which is the International Press Telecommunications data. IPTC data is where you can enter you contact and copyright information and a range of other configurations of more minimal or expanded information. The easiest way to add IPTC data is to create a template with your details, which you can then simply choose and apply to a selection of images from the menu circled on the right. You can then even apply these templates on import. Keywords Keywords are another form of metadata that are especially useful for searching through old photos and very quickly finding what you are looking for. Keywords as the name suggests are single words that can be tagged onto photos like labels. There is no limit to the number of keywords you can use for any one image and like the IPTC data these keywords can be added during import. Lightroom has two keyword menus in the Library, Keywording and Keyword List. Keywording is where you can add and edit the keywords associated with a single or image or group of images and the Keyword List keeps track of every keyword you have ever used.

19 19 Filters All the afore mentioned flagging and tagging devices can also be used as filtration devices, meaning you tell Lightroom to just show you the 3 star images or just show you the images with the Keywords cat or dog. Lightroom Version 1 had a pretty basic filter and it s still there to the far right of the tool bar. However, Lightroom 2 has introduced a way more powerful filter called the Library Filter that allows you to drill down in to your criteria. For example you can say show me the 2 star images only, with China as a Keyword, shot at 100 ISO on the 35-70MM lens at f2.8 as per below. Note the > symbol to the left of the 2 stars indicating that the selection be greater than or equal to 2 stars. This can be set to equal to only or less than or equal to. You can also save custom searches from the menu at the top right. Collections Collections are another useful tool that are as quick and simple as they are due to the database structure. Collections are just like itunes play lists, they don t copy the images they simply reference them, or create little short cuts to them. To create a new collection simply hit the + icon on the collections menu, which is fifth from the top in the left hand panel of the Library. Give your collection a name and away you go. As you browse through your library simply drag and drop images on to that collection to add them. It s that easy! Images can be in multiple collections without a problem and collections can be exported as Slideshows, Contact Sheets or Web Galleries as you will see shortly. Quick Collection Then there is the Quick Collection, simply select a photo in the Library and hit B to add to quick collection, Cmnd/Ctrl + B to view your current quick collection and Cmnd/Ctrl + Shift + B to clear it. You can only have one quick collection at a time, but once your current quick collection becomes something you want to keep go File > Save Quick Collection. This will save your current quick collection as a regular collection and clear out your quick collection ready to start again. And remember, no files are being duplicated, the collections are just creating links to the one main file, so you can have as many collections as you like.

20 20 Managing your Catalogs When you first launch Lightroom it automatically creates your first Catalog in its preferred default location. You are given the option to save it somewhere else otherwise you will end up with the following file paths and names. PC - User/My Documents/My Pictures/Lightroom/ Lightroom Catalog.lrcat Mac - User/Pictures/Lightroom/Lightroom Catalog.lrcat A complete Catalog folder will include the following; A.lrcat file which is the actual Catalog or database A Preview.lrdata file which contains all your image previews after import A Backups folder where backups of your Catalog will be saved. This Catalog will serve you well indefinitely and all the photos you import will end up in there in the folders structure you imported them with. To browse and sort these photos you can navigate via subfolders, keywords and various other metadata criteria, just like you would in itunes. You can also create multiple Catalogs by simply going File > New Catalog, Lightroom will close the current Catalog then prompt you to name and save the new one. The application will then restart and open the new Catalog. If you have multiple Catalogs you will be prompted at start up as to which one you want to open, or you can choose one to always open on start up. See below the Lightroom 3 select Catalog dialogue looks vastly different from the Lightroom 2 one but still offers the same options. It is entirely up to you to decide whether you would prefer to work with one big Catalog or create multiple smaller ones. I personally like to have multiple Catalogs, but managing a single Catalog is obviously simpler at first.

21 21 Databases & Catalogs Because Lightroom is a database we must import our images in to the database fist. However, they don t call it a database, they call it a Catalog. When you first launch Lightroom it creates a new Catalog for you. You can create multiple Catalogs, but for now lets work with one. The first thing to understand is that when you import your images in to Lightroom you are not duplicating the original images; you are just showing Lightroom where they are on your computer. Lightroom then creates a low-resolution preview in the Catalog, which will preview the adjustments you make. Only when Lightroom wants to export or output the images do the original pixels need to be referenced, but even then the image adjustments are applied to duplicate images, not the originals. In fact in many situations Lightroom can utilise the preview files, like in slide shows, web galleries and low res printing. Which is why these things can be accomplished so quickly as high res raw files do not need to be processed to produce the end result. Importing All importing is done in the Library and can be activated from a number of places, but none more easily accessible then the Import button at the bottom of the left-hand panel. File > Import Photos from disk will also do it. Seen above the Lightroom 3 import dialogue works from left (source) to right (destination). With the above files being imported from camera, converted to DNG, copied to the pictures folder, backed up to peaches and added to the Lightroom catalog. Complete with keywords and metadata template.

22 22 File Handling The first import decision is File Handling, which determines how the master high res files are managed. Seen below this drop down menu, which presents you with 4 file handling options. The first option Copy as DNG will copy your files to a new location, import them into Lightroom and convert them to DNG along the way. This is a great option for importing from camera or card reader if you are utilising a DNG workflow as it avoids the issue of having duplicate DNG and camera original files to deal with. The second option Copy is the same as Copy as DNG with the one point of difference being is does not convert the copied files to DNG. The third option Move is the same as the second but the photos will be moved rather than copied, which means a duplicate will not exist. Finally Add will import your photos into Lightroom and leave the original high res files where there are, this could be on your computer, on an external drive or disk. Render Previews Your next option is the size of your initial previews, there are 4 options here, Minimal, Embedded & Sidecar, Standard and 1:1. The main issue that informs your choice here is how much of a hurry you are in to get your images into Lightroom The previews that import the most quickly are the smallest size and the highest quality takes the longest. If you are in a hurry minimal thumbnails are the way to go and the standard previews can be created later. However, if you have the time you might as well create standard size previews from the get go. Don t Import Suspected Duplicates As the name suggests this option will check the capture time in the import files and make sure it is not already in the Catalog. This can be useful if you are importing from card/camera multiple time as you shoot more and more. Make a second copy If copying files from one place to another you have the additional option of creating a second copy for backup. This is a great option if importing on location from camera/card and you have an external drive to copy a back up to. It would be pretty pointless to make a second copy to the same drive!

23 23 File Renaming If copying files from one place to another during import you also have the option to rename them. Lightroom provides a number of naming templates including the ability to change the prefix and preserve the sequence number. Apply During Import Available to all style of imports is the ability to apply Develop Settings, Metadata Templates and Keywords. Develop Settings can be anything from brightness and contrast adjustments to black and white conversions or sharpening. Each of these options will be applied in the exact same way to every file being imported. Destination Finally if copying or moving files you will need to choose a destination to copy or move the high re negative to. Lightroom 3 now gives you a cascading file tree to navigate your hard drive and choose a destination. You can also move the files into a new subfolder and/or organise by Date, by Original folders or into one folder. The Develop Module Pictured below the Lightroom 3 Develop module gives you Navigator, Presets, Snapshots and History in the left panel and Camera Raw in the right. Also new to Lightroom 3 is a new and improved processing algorithm that is automatically applied to any files with no current settings. Lightroom 3 has added some new presets for you to explore including some cross processing options that are worth a look. You can also, easily create your own by clicking on the + symbol to the right of the Preset menu, naming the preset and choosing which adjustments to include.

24 24 History and snapshots work pretty much the same as they do in Photoshop, allowing you to return to any point in time or to record a given moment for even easier access. Mousing over individual Presets, Snapshots or History states will preview them in the Navigator windows. But unlike Photoshop these history states remain forever, they are not trashed after a restart. The right hand panel IS Camera Raw, with the additional Previous and Reset buttons at the bottom, which allow you to apply the previous settings to the current image or to remove all settings. The Basic Tab Exposure Exposure is a lot like the white slider in Photoshop s levels, only it goes both ways. It has a range of 8 stops (4 either side of your starting point) and will initially lighten or darken just your highlights. The adjustment will gradually spread to your mid tones and shadows as it becomes more extreme. Exposure is set to 0 by default. Recovery & Fill Light Recovery & Fill Light are a lot like Photoshop s Shadow/Highlight sliders, allowing you to recover detail in highlights and open up shadows, without affecting other parts of the tonal range. However, they are more powerful here because they are working on Raw data. Due to the nature of linear gamma Raw files the recovery slider is significantly more powerful then the Fill Light as there is so much more detail in the highlights to extract. Both sliders are set to 0 by default. Blacks The blacks slider works much like the black point slider in Photoshop s levels, darkening the shadow end of the tonal scale and rising to the 3/4 and mid tones with large adjustments. If your Raw files are too dark from capture you miss out on any of the benefits of the shadow slider. However, if you leave a nice little gap on the left hand side of your histogram you can neatly dial the black point to exactly where you want it to start. The Shadow slider is set to 5 by default. Brightness The brightness slider is similar to Exposure but focuses it s results into the midtones. When recovering highlight detail with the exposure or recovery sliders the perfect counter balance can be to restore the mid and 1/4 tones by raising the brightness. The brightness slider is set to 50 by default. Contrast The contrast slider works like a mild S curve, locking down the black and white points, then darkening the 3/4 tones and lightening the 1/4 tones. Due to the nature of digital capture there are few Raw files that will not benefit from a contrast boost so the contrast slider is set to +25 by default.

25 25 Clarity Introduced CS3 the Clarity slider is kind of like a sharpening function that boosts local contrast without overly sharpening edges. What it really does this will make more sense later on when we look more broadly at how sharpening works. Saturation & Vibrance The saturation slider is much like the Photoshop s saturation slider. It can be used to push the saturation of colours higher and its effect on the image data can be clearly seen in the histogram. Vibrance is a much smarter version of Saturation. Vibrance will detect if certain colours are becoming too saturated and back off on them to avoid them clipping, it will also detect and protect skin tones ensuring they do not end up looking over cooked. Vibrance is set to 0 by default and is best visited before you resort to Saturation.

26 26 Sync Settings Make adjustments to one single image then choose other similar images that will benefit from same, then simply click Sync Settings at the bottom right of the library module and all the settings from the first image are applied to the others. Notice when a group of images are chosen in the library the unselected images are dark gray, the selected images are light gray and one of the selected images is even lighter gray. This lighter gray image is known as the Most Selected image and is the one that will have its settings applied to others within the selection. Clicking on another image within the selection will make it Most Selected. You can also copy and paste settings from one image to another from the Photo > Develop Settings menu or with a right click on the source image thumbnail. The Calibrate Tab Lightroom and Camera Raw have generic profiles for every supported camera model. These profiles are automatically assigned to your raw files and the calibrate tab lets you further edit and save these profiles as your new camera default. Over time these generic profiles have been modified by Adobe and Lightroom 2 introduced extra profiles that attempted to mimic the appearance of the various camera manufacturers styles or looks. You will see different option in the profile menu depending on what brand camera you have. Here we see the Nikon options. Basically the idea is to choose the profile you feel produces the best result before making any adjustments to your image. If you find a profile that you want all your images to default to you can set this as your default by choosing Develop > Set Default Settings in the Develop module. Truth be known I rarely visit the Calibrate tab and tend to rely on the defaults as my starting point to adjust from. However, when some images contain colours that are a little too saturated to preserve detail I often check to see if Camera Neutral will give me a better starting point.

27 27 Syncing Adjustments with Camera Raw Adjustments made in Lightroom or ACR are cross compatible. Which means you can do half your work in Lightroom and continue working in ACR or vice versa. However, to ensure this works properly you must tick Automatically write changes to XMP, under the Metadata tab in File > Catalog Settings. This is turned off by default so be sure to turn it on for each Catalog that needs to move between ACR and Lightroom. Adobe turns it off by default because it makes Lightroom run slightly faster, I have it on all the time and have see little difference performance wise with or without. The extra benefit of turning it on is it creates a back up of your image adjustments with the images themselves rather than relying just on the database.

28 28 Session 3: Fine Tuning in Photoshop Tonal Adjustment Even though Photoshop can see and manipulate 256 levels of brightness in an 8-bit image it is useful for you to be able to see your image as at least 5 levels. 1. Shadows levels /4 tones levels Mid tones levels /4 tones levels Highlights levels These figures are not set in stone they are more of a guide and are useful to get you seeing your images more as Photoshop sees them. This will make you more skilled at selecting your image adjustments and understanding exactly what they are doing. Curves Curves are the most sophisticated tool for fine tonal and colour adjustment, so lets jump straight in to them. The default curve is a 45-degree line with black in the bottom left and white in the top right in RGB or light. All curves initially look like this, regardless of the image. The curve then bends to reflect the nature of the adjustment you have made. The adjustment above left, flattening the curve, would result in reduced contrast in the image as both the black and white points are decreased. The adjustment above right, steepening the curve, would result in increased contrast as the black and white points are increased.

29 29 The real power of curves begins when adjusting points along the curve as demonstrated below. Here we see the classic S curve which will boost contrast in the mid tones, whilst locking down our Black, Mid Grey and White points. Much like having many sliders along the base in levels, here in curves we can make adjustments at any point along the tonal range, giving us total control of our image data. Curves, like the histogram, are not unique to Photoshop, and can be found in many scanner drivers and more recently raw processors for digital photography. With Curves open you can run your cursor over the image and click on a point to reveal where it s located along the curve. Cmnd/Ctrl clicking on that point will place an anchor point on your curve. Holding down the control key and tapping the tab key will cycle between points on the curve. However CS4 has changed this to the and + keys. Then for super fine control, the arrow keys, left, right, up, and down, will adjust that point by one level at a time and adding the shift key will change it to 10 levels at a time. Just as the S curve will increase midtone contrast by making the midtones steeper, making any other part of the curve steeper will increase contrast in just that area. This is the real power of curves, the ability to apply an adjustment to one area and then remove it from another. Target Adjustment Mode CS3 added many new features to the curves dialogue box, with a histogram, clipping preview, individual channel overlays and a bunch of presets. However, the core behaviour of the curves dialogue, as described above, did not really change. Below the new look CS3 Curves dialogue box and the small but powerful new feature in CS4 s curves dialogue, the targeted adjustment or on image adjustment tool. Click it to select it and then simply click and drag up or down on your image. Clicking on the image will activate the corresponding points on the curve and dragging up or down on the image will drag the curve up or down.

30 30 CS3 s New Look Curves CS4 s Small but powerful new feature Checking show clipping will provide live clipping previews as you adjust your shadows and highlights, or holding down the option/alt key and sliding the levels like black and white sliders at the bottom of the curve will give you a levels style clipping preview. The presets menu above has a bunch of preset curves for you to experiment with. And you can also save your own presets, something you ve always been able to do, but now they re nice and accessible right there in the Presets drop down menu. Improved Brightness/Contrast Previous to CS3 Brightness and Contrast would brighten or darken every single pixel by the same amount causing a weakening of shadow or highlight detail depending on which way you moved. This basically made it unusable! However, since CS3 the tool has been vastly improved, making an increase in brightness more akin to raising your1/4 tones in curves a decrease more akin to lowering them and contrast now works like an S curve. If you want to see how bad it really was check the Use Legacy check box below preview and it s back to the dark ages.

31 31 Colour Correction by the Numbers A colour cast is introduced as a result of an imbalance in one or more of the colour channels, correcting colour is about bringing the values back into balance. So unlike tonal adjustment, where we adjusted all channels as one (the composite), colour adjustment involves adjusting individual channels by different amounts. Traditional colour correction was all done by the numbers, the objective was to set correct highlight and shadow values and the rest of the image would fall in to place. The highlight and shadow points are the brightest and darkest points in an image that you wish to preserve detail, not the extreme black and white points. Common aim points for this style of colour correction in CMYK were 3C 2M 2Y 0K for highlights and 80C 70M 70Y 70K for shadows, with neutral targets being equal magenta and yellow with a slightly higher cyan, due to cyan s inherent weakness. This style of colour correction is what inspired the black and white eyedroppers available in the Levels and Curves dialogue boxes. The eyedroppers are used to set shadow and highlight points, and can have their own versions of black and white set by double clicking on them, so we can easily translate the CMYK method in to RGB. By default the eyedroppers are set to R0, G0, B0 for black and R255, G255, B255 for white. Because the droppers will force a neutral highlight or shadow into the area they are clicked on they will not only improve contrast, but also help to remove colour casts, as setting the shadow and highlight to neutral should reduce or remove any colour imbalance throughout the tonal range. However, setting shadows to 0, 0, 0 and highlights to 255, 255, 255 is a little extreme and risks clipping detail, so we peg the numbers back a little to ensure we retain shadow and highlight detail. Double click on the droppers in levels or curves to reveal the colour picker, here you can set your preferred values for these tonal points.

32 32 Better RGB numbers for shadows are 13, 13, 13 and for highlights 243, 243, 243. These numbers will translate to a 95% and 5% dot on the press respectively. These numbers are a little conservative, but safe for most printing processes. With some testing and/or discussion with your printer you may fine tune then further for a specific output. Flesh Tones. Once you have set your highlights and shadows your image will have no doubt improved, particularly in contrast, but hopefully in colour as well. However, these steps alone are often not enough to perfectly nail the ideal colour renditions of elements with in your image, so some tweaking is also in order, especially in wellknown colours. Working in RGB for maximum flexibility but using the CMYK numbers as our guide, the following rules of thumb should help you ensure colours appear in print as the average viewer would expect them to. Skin, Caucasian Yellow roughly equal to or slightly higher than Magenta, Cyan about one fifth the Magenta and no Black. The higher the Yellow is to Magenta the more warm and tanned the skin will appear but ideally no higher than about 15 points. Skin, Asian/Hispanic Yellow will generally be at least 10 points higher than Magenta, and Cyan around one quarter to one third the Magenta and again no Black. Skin, African Is the most varied, and can have a much higher Yellow to Magenta or can equal Yellow and Magenta depending on the person. However, there is always a significantly higher Cyan than any other flesh tone, up to ½ the Magenta. And Black may be up to ½ the Cyan or higher in mid tones and shadows. When analysing skin values for colour corrections it is a good idea to look for a nice smooth quartertone, just below a highlight but above a midtone. It is also useful to take your reading from the neck, especially with women, do avoid getting false readings from any makeup. Memory Colours, Grass & Leaves The natural green of grass and leaves should always have Yellow as it s highest component, Cyan as it s second highest and Magenta as it s lowest, as Magenta kills green. The higher the Yellow the more warm and lush the grass will appear, the lower the Yellow the cooler it will appear. But as soon as Cyan become higher than Yellow it s wrong, grass is never Blue. Blue Sky Equal values Cyan and Magenta in theory make Blue, but because of the inherent weakness of Cyan ink they make more of a purple in print. A blue sky will have Cyan as it s highest value with Magenta below it and Yellow often absent. The lower the Magenta the lighter the Blue the higher the Magenta the deeper the blue. The addition of Yellow will make the Blue appear duller and more overcast.

33 33 Blending Modes Photoshop s blending modes ordered in to logical blocks, each with their own general flavour of blending results. They are as follows; The Darken set, which will darken blended pixels when applied The Lighten set, which will lighten blended pixels when applied The Overlay or Contrast set which increases contrast in the blended pixels. The Comparative set which is designed for comparing and combining images. The Influence set, which takes just one aspect of your layer to influence another, such as colour or luminosity, as already addressed in session two. Blending Modes pop up in a variety of locations in Photoshop, however, I would like to focus on the ones in the Layer palette for now and what you learn there will apply to Blending Modes found elsewhere, such as the brush tools. A Blending Mode really has no effect on any one individual layer, it s more the way one layer interacts with another that is affected, hence the term Blending. I find the Blending modes fall in to two categories, useful and not so. However, simply understanding what they do may inspire you to find a use for some of the ones that fall in to the second category. Over time I am finding more uses for Blending Modes that I have, in the past, ignored. Blending Mode Tip s & Techniques To fine-tune your sharpening a duplicate sharpened layer can be blended with either Lighten or Darken depending on which side of the sharpening halo you want to preserve. For further fine tuning 2 sharpened layers can be used, one set to Darken and one to Lighten, then you can dial the opacity to taste. For super fast tonal adjustment an empty adjustment layer blended with Screen will brighten, Multiply will darken and Overlay will boost contrast. And 38% opacity is very close to one F-Stop, which you can halve or double accordingly. For a more powerful version of the Dodge and Burn tools create a new layer above your image and fill it with 50% grey and set its blending mode to Overlay. Now painting with black will burn and painting with white will dodge. The advantage lies in the fact that you are not painting on the original image, so you can undo or re-edit settings at any time in any order. Soft Light is a softer alternative to Overlay and a blending mode you will find many uses for. I have a 38% curves adjustment layer set to Soft Light as an action and use it a lot. Sometimes19% is enough to give your image a boost.

34 34 For a soft edge, high contrast portrait, duplicate the background layer and gaussian blur it by about 15. Now change the blending mode to Overlay and reduce the opacity of the layer to taste. For further effect you can lighten the blurred layer with levels or curves to make the highlights glow. For an alternative to Unsharp Mask duplicate your background layer and run the hi-pass filter over it at a radius of about 2.0 Filter > Other Hi Pass. Now change the blending mode to Overlay and you have sharpening without halos. You can also try Soft Light or Hard Light for softer or harder results. When aligning multiple exposures the Difference and/or Exclusion blending modes can be useful for ensuring your registration is as accurate as it can be. You can cycle through the layer blend modes with Shift and + or this will not work when you have a brush tool selected, instead it will cycle through the brushes blend modes options. Layer Masks Now that we understand channels lets have a look at masks, this may seem like an abrupt change in topic, but stick with me and you ll shortly see it s not. Adding a layer mask is where the control really kicks in and you start to work with incredible freedom and ease. You can add a layer mask to a layer (not the background) by going Layer > Layer Mask where you have a choice of Reveal All or Hide All. Reveal all will produce a white mask, which leaves the entire layers contents visible, and hide all will produce a black mask, which will completely hide the layers contents. Once you have a layer mask in place painting with black or white will further hide or reveal the layers contents. Painting with 50% opacity will paint on the mask with 50% grey, which will result in 50% opacity of the layers contents. Sounding familiar? The following technique will reveal the usefulness of the layer mask, whilst showing you a good technique for reducing blemishes and coarse or blotchy skin. 1. Take a portrait style image where the subject has coarse or blotchy skin. 2. Duplicate the background layer by going Cmnd/Ctrl J. 3. On the dup layer go Filter > Noise > Gaussian Blur radius 5 so you can see it 4. Select Layer > Layer Mask > Hide All to add a black layer mask to the blurred layer, this will hide its contents completely. 5. Now paint with white and a soft edged brush to reveal the softening effect where you want it to be applied, just like applying make up with a brush. 6. If you accidentally apply some blur to an area you do not want it to be no problem just paint over it with black to hide it again.

35 35 Left we see the soft layer and it s layer mask icon. Note the black corners on the layer mask icon indicating the layer mask is active for painting on. If the actual layer were active you would be painting on the image pixels rather than on the mask. Thus causing destruction. You can continually hide and reveal by painting black over white over black as long as you like with no limitations or image degradation. Tips: The D key will return your foreground and background colours to black and white and the X key will swap them back and forth. Option/Alt clicking on the Mask thumbnail will show you just the mask. Shift clicking on the Mask thumbnail will temporarily disable. Alt dragging the mask to another layer will copy it to that layer Shift + Alt dragging the mask to another layer will copy and invert it So a mask is basically an Alpha Channel but accessible in the layers palette. What s white is selected, what s black is protected and what s gray has partial transparency, the darker the gray the greater the transparency. That should make for less to remember, as a Channel, a Mask and a Selection are all basically the same thing. An even quicker way to get a layer mask in place is to simply click on the layer mask icon at the bottom of the layer palette. This will give you a reveal all (white) layer mask by default, or hold down the alt/option key for a hide all (black) one. If you have an active selection present when you add your layer mask the selection will be automatically applied to mask, making the selected pixels white in the mask and the unselected ones black. And if you have applied feathering to your selecting this will produce a soft edge to mask, represented by gray. * Introduced in CS3 - Quick Selection Tool Imagine a magic wand that you could click and drag through an area of colour and it would continually update its selections boundaries based on the pixels you dragged it over. Well that s pretty much what the new Quick Selection Tool tries to pull off, making for lightening fast selections of suitable areas.

36 36 Obviously its effectiveness is image dependent, so it may not work perfectly every time. But when it does, it s pretty cool. The tool can also be set to sample all or one layer and to select, add to selection or subtract from selection. * Big Improvements in CS5 - Refine Edge Introduced in CS3, refine edge was a great addition for refining the edges of your selections after you hade made them. However, its biggest limitation was that all adjustments you made were applied globally along the entire edge of your selection, regardless of whether it was a hard or soft edge. Well enter CS5 s Edge Detection and all that has changed. Now you can apply global refinements with the sliders followed by local adjustments just to the wispy areas. You can still Adjust the Edge globally with Smooth, Feather, Contrast and Shift Edge. But check Smart Radius and the real fun begins. Use the Radius slider to define the width of the edge you want analysed for refinement. Then, use the Refine Radius Tool to brush wispy areas like hair and it will do it s best to preserve the hair and remove any offending background. There is also a Remove Refinements Tool below the Refine Radius tool that will remove any refinements that have gone too far. Finally, check Decontaminate Colours and adjust the amount and any remaining stray pixels from the old background will have their colour changed to better match.

37 37 Soft Proofing The real payoff of a tightly colour management workflow is the ability to simulate a variety of output conditions and settings in the one file without hurting it. This is known as soft proofing or proofing on screen. Those who doubt colour managements effectiveness would be horrified at the thought of going to press with out seeing a Hard Proof, but those who have embraced the solutions of the 21 st century regularly demonstrate the power of colour management by doing exactly that. In theory the combination of a well calibrated and profiled monitor and well chosen workspace should be showing you exactly what the file at hand looks like. Which is pretty much what it is doing, however, what it is not doing is showing you what it s going to look like in print. Especially if you haven t told the colour management system exactly where you are printing. Utilising Photoshop s Proof Set-up command under the View menu we can simulate and manipulate a wide range of output scenarios before signing on the dotted line and choose the one that serves us best. 1. The Custom Proof Conditions option allows you to choose a previously saved set of soft proofing conditions. 2. The first step in creating a new soft proof is to choose a profile to use to soft proof or Device to Simulate. 3. You can also opt to Preserve Numbers, which will show us how a supplied CMYK file would print on a press other than the one it was separated for, if we did not do a profile conversion. This is essentially like assigning a profile, however very convenient to have in the soft proofing dialogue box. 4. Then we can choose the Rendering Intent that produces the best result for our conversion by making a comparison. Remember the luxury of selecting your rendering intent by eye is highly dependent on all the display variables we discussed in session two. 5. Black Point Compensation is an Adobe addition to colour management, which resolves the gulf between the depths of black your monitor can display compared to ink on paper. BPC scales your source black to your destination

38 38 black when converting to ensure the shadow detail is not pushed too far one way (too dark) or the other (too light). 6. Simulating Paper Color and Ink Black will dull the monitor at either, or both ends of its dynamic range to produce the most accurate soft proof possible. It should be noted that the dulling of the monitor white can be quite disturbing and it s best to hide anything white on your desktop like menu bars etc, as they do not get dulled and can make your proof look awful by contrast. Sharpening Workflow Professional sharpening ideally involves up to three processes, capture sharpening, to compensate for any softening during the capture process, creative sharpening to enhance the appearance of the image and output sharpening, which is intended to safe guard against any softening that may occur at the output device. Capture and creative sharpening are best done at original file size to create an independent master file that can be optimised for a range of output conditions. Output sharpening, however, must be done ONLY at the final output size. Unsharp Mask (USM) There are a variety of sharpening methods available, ranging from Photoshop filters to 3 rd party plug ins, with the most widely used technique and the basis for all others being the Unsharp Mask filter. The way the Unsharp Mask filter works is to look for edges of contrast within your image and attempt to increase their contrast to create the illusion of greater sharpness. The Unsharp mask filter has three separate sliders that each contributes in their own unique way to produce the overall result. Amount controls how much the edge contrast is increased by, like a volume slider. Amount goes from 0 to 500. Radius controls how far from the edge the increase in contrast fans out, creating smaller or larger halos. Radius goes from 0.1 to 250. Threshold determines how much edge contrast must be present before any sharpening will be applied. A low threshold will sharpen all edges, and a high threshold, will sharpen only the highest contrast edges. Threshold goes from 0 to 255. Commonly you would increase the threshold on a portrait image to avoid sharpening the low contrast areas like skin texture or blemishes whilst ensuring sharpening of higher contrast areas like eyes and teeth. Where as with an image of trees or other textured surfaces, where you want to sharpen all the fine detail as much as possible, you would opt for a lower threshold, possibly as low as 0.