Skylum Luminar Neo 2023-24

      Photo Review 8.9

      In summary

      Luminar Neo has gained plenty of features and functionality since our initial review, making it an even better choice for photographers who want quick and easy ways to improve their images.

      It’s especially well suited to content creators with minimal expertise in image editing but could also suit beginning photographers looking for an easy way to get started with image editing.

      Luminar Neo – Buy options
      (Photo Review Affiliate link)

      Full review

      Generative AI has been big news during 2023, with the release of Adobe’s Firefly image generator enabling a new Generative Fill in Photoshop in the second half of the year. Skylum has been quick to follow. Released in mid-November, Version 1.17.0 of its Luminar Neo software, adds a new GenExpand tool, hot on the heels of the October update, which introduced the GenErase and GenSwap functions. The new ‘generative AI’ functions have been developed using the Stable Diffusion AI platform. They are significant updates since our original review of Luminar Neo Version 1.0.2 in March 2022, so this review will concentrate on the new generative AI functions.


      The Luminar Neo workspace showing the new generative AI tools (boxed in red) and the suite of add-on Extensions.

      Also new in the latest version of Luminar Neo is a suite of seven ‘Extensions’, which were formerly sold for AU$119 for the full pack or $24.50 per extension.  From the beginning of January 2024, these extensions have been bundled into the subscription offers, along with the new generative AI tools.

      All the extensions are AI-based and the purpose of each tool is explained by its name:  Panorama Stitching AI, Supersharp AI, Upscale AI, Noiseless AI, HDR Merge, Magic Light AI, Focus Stacking AI and Background Removal AI. We’ll leave them for a possible future review.

      Luminar Neo can be used either as standalone software or as a plug-in for Adobe Photoshop or Lightroom or an extension for the Photos app on Mac OS computers. It is available as a 6.4MB download that can be purchased through monthly, annual and two-year subscription plans that include:
      – The ability to use Luminar Neo on two devices instead of one.
      – A 10% discount on Marketplace items (Skies, Presets, LUTs, Overlays. etc.).

      Trial 7-day downloads of Luminar Neo are available (and you aren’t required to provide credit card details). Users of previous versions can also update their software and discounts are available for students and teachers, new parents, charity and health care workers, low-income earners and people with a disability, first responders and members of the military – click here for details.  You can also keep up-to-date with the latest features and improvements to Luminar Neo photo editing tools.

      Who’s it For?
      The new tools make the latest version of Luminar Neo even more highly automated than its predecessor, although some features – like the Catalog function and Edit user interface –  remain largely unchanged so Luminar Neo still falls short of the functionality provided in Adobe’s Lightroom or Photoshop.

      To provide a few instances: you can’t edit metadata and the Undo function is very restricted. You can’t add text overlays to images or batch rename groups of files. Rating tools are limited to the ability to add a heart icon on favourite images or an X on those you reject.

      Fortunately, raw file processing is straightforward and older cameras are supported although functionality is also limited. Some of the same slider controls as Adobe’s Camera Raw are provided but we found they were less sensitive and the software could beslow to respond at times.

      However, the high degree of automation in Luminar Neo will be welcomed by photographers looking for quick ways to ‘improve’ their images who lack the time to learn how to use more complex applications. The new AI functions, in particular, can achieve acceptable end results with minimal effort – as long as your prompts work as desired. Like other AI-based software, success is closely linked with prompt quality and precision.

      On the positive side, Luminar Neo will be useful for content creators with little or no experience in editing, although it’s of minimal interest for experienced users of Photoshop or Lightroom (or similar applications). It’s too automated and adjustments are too limited and the new generative AI tools require constant internet access. In addition, they may not provide full functionality on Apple computers using the older macOS Catalina (10.15) and macOS Mojave (10.14) operating systems.

       What’s new?
      The latest version has been created around a modular engine in which the processing load is distributed to enable faster image processing. The developers claim modular processing allows users to apply adjustments to an image with many different tools without significant performance losses. The software will also save all edits automatically. In practice, we found this to be partially true; some adjustments happen faster than others.

      The user interface in the latest version of Luminar Neo has been updated to add the new generative AI tools and extensions to the right side panel on the existing workspace. Usefully, the left side panel also provides more information, including direct links to the Catalog and previous edits you’ve carried out with the software.

      The updated user interface in the latest version of Luminar Neo.

      Unfortunately, there’s still no Help button for quickly finding what a selected tool can do and how to use it. As we noted in our last review, almost all applications provide built-in Help assistance and it’s particularly important in software targeted at less sophisticated users.

      In common with other AI-based image editors (including the latest versions of Photoshop), how well each of these functions will work depends largely upon the nature of the original image you’re working with as well as the clarity of the prompt you use to specify the adjustment(s) you want. The basic capabilities of the three new generative AI tools are outlined below.

      GenErase, an advanced Erase tool for eliminating objects like vehicles, street signs, utility poles, unwanted people or animals or any other item that can be easily selected from your image. It’s simple to use; you simply paint over the area you want to remove and the software does the rest, based on the rest of the image.


      In GenErase mode you simply paint over the section of the image you want to remove and click on the Erase button. Brush width is adjusted with a slider in the control panel, which is outlined in red.


      Clicking on the Erase button darkens the screen for a few minutes while the software does its work. Tips for using the software are displayed below the image being worked on.


      The end result.

      If you’re not happy with the first result the software delivers, you can click on the Erase button again for another attempt. You could also refine the edges of the area selected for replacement. Both processes can be repeated until you’re happy with the result.


      A second attempt after small adjustments that made the replacement area slightly larger.

      For comparison, we gave the same task to Photoshop CC, in this case using the Generative Fill tool. No prompt was provided so the software only had to create new content based on the rest of the image. Photoshop provides three options to choose from, as shown in the screen grab below. We selected the first option.


      This screen grab shows the user interface provided in the latest version of Photoshop CC, when asked to replace the same area in the image as we used with Luminar Neo.

      GenSwap, which lets you replace or swap objects within the image frame or seamlessly add in new ‘AI-generated’ elements like trees, animals, birds or structures. This tool is similar to GenErase in that the part of the scene you want to swap out must also be ‘painted over’. As with Gen Erase, the toolbar is below the image you’re working on.


      The workspace for the GenSwap tool.

      GenSwap asks you to input a prompt (circled in red in the screen grab below) describing what you’d like to ‘swap in’ to replace the area selected. Tips from the software include the need to keep it simple, express the concept clearly and avoid the use of verbs such as ‘change’, ‘insert’ or ‘remove’. Clicking on the Swap button sets the software to work, darkening the screen for a minute or so while the new content is generated.


      We’ve asked the software to replace the vehicle remains with a ‘camp fire’.

       When using this tool it can be worth making the selected area somewhat larger that the item you want to swap out.  As before, success depends on how well you’ve crafted your prompt. Usually the simpler and more specific it is, the better the software will be able to create a suitable match. Even so, you may also need several attempts to obtain a satisfactory result, as the screen grabs below demonstrate.


      First attempt.


      Second attempt.


      Third attempt.

      GenSwap can also be used for sky replacement – although you need to be careful when selecting an image to work on as ambiguities on the horizon will often be ‘developed’ into strange things.


      Sky selection in GenSwap for sky replacement. Note the simple, unambiguous prompt (circled in red).

      The first attempt, using ‘night sky’ as a prompt, added clouds and trees and converted the circle of sky-viewing seats into a rather unrealistic building.

      So we tried a more specific prompt, asking for a ‘dark night sky with stars’. No change was made to the selection area, which meant the software once again turned the circle of sky-viewing seats into a building.

      Because each iteration of the GenSwap tool is based upon the last image it created, the only way to stop the software turning the circle of sky-viewing seats into a building was to go back to the previous image with the campfire, before any adjustments had been made to the sky.


      This time the selection area was reduced in size to exclude the horizon and sky-viewing seating.

      Two results from repeated iterations of GenSwap are shown below, each with the new sky blending gradually into the rest of the scene. Aside from the slightly unnatural appearance of the stars, overall the software did a pretty good job.

      Once again, we gave the same tasks to Photoshop and using the Generative Fill tool, this time with the same prompts as we gave Luminar Neo. The results were interesting.


      Because the area we selected was slightly larger in this image, Photoshop’s Generative Fill tool created larger camp fire inserts.


      A second try allowed us to choose from six options.


      For the sky replacement we repeated what we’d done for Luminar Neo and selected the entire sky.


      Like Luminar Neo, Photoshop converted the circle of sky-viewing seats into a building but it produced a slightly more realistic-looking sky.


      However, when we reduced the selection area to exclude the horizon and sky-viewing seating and asked for a ‘dark night sky with stars’, the result was disappointing. The sky area wasn’t darkened adequately and when we tried reducing the brightness of the selection layer the transition between it and the background layer became obvious. Luminar Neo delivered a better result in this case.

      GenExpand, the most recently-added tool, is used to extend the boundaries of existing image by adding in new AI-generated content, based on analysis of the source image. It can be used to change the aspect ratio of an image as well as create wider landscapes out of vertical shots, as we’ve done in the example shown here. Photoshop has a similar tool, named Generative Expand and we’ll compare the results we obtain with both applications, based on the same source image.

      To give you some idea about how much images can be expanded, from the original 20MB source image, a JPEG with 3888 x 5184 pixels, Luminar Neo’s GenExpand created an image with 9706 x 5184 pixels that was 43.1MB in size. Photoshop’s Generative Expand tool produced an image with 8520 x 5184 pixels that was 28.9MB in size – although we could have made it wider with additional expansions.

      There are a few pre-requisites for obtaining usable results, regardless of which software application you use. It’s best to work on one side of the image at a time and expand in small increments because the software has to fill the new space with AI-generated content that fits in with the original content in your source image, which provides much of the source data (with the rest coming from the web).

      The software works best with source images that have well-defined structures and textures, especially around the edges. Forest scenes like the one we’ve used have plenty of potential but the software will also find it easy to expand areas of sky, sand or sea.

      When images are expanded, new content is generated and added. Consequently, it will take several minutes to create the fill-in data and you’ll end up with files that are larger than your original source image.


      Start by expanding one edge of the image and inserting the prompt to tell the software what to insert. In this case, the prompt was ‘more forest’.


      Once the software has filled the new area with content you’re happy with, click on the Expand button and move to the other side of the image, repeating the process. There’s no need to change the prompt in this case.


      From this point it’s simply a matter of working on alternative sides of the image, keeping the areas you’re expecting the software to fill relatively small, as shown in the screen grabs below.

       

       

      When you’re happy with the results you’ve obtained, clicking on the Save button will lock in the changes and generate a new image.

      With Photoshop, you open the image and use the Crop tool to extend one side of the image then right-click on the blank area and select Generative Expand from the dropdown menu. We prompted the software with the same ‘more forest’ request as we used with Luminar Neo.


      We prompted the software with the same ‘more forest’ request as we used with Luminar Neo.


      The panel on the lower right hand side shows the number of iterations of GenerativeExpand that were made to produce the end result.

      The final step in the Photoshop process is to flatten the image and save it.

      We tried asking both applications to generate new fill for larger extensions.  Luminar New limited how much you could drag out each side but Photoshop seemed capable of handling much wider extensions and ended up delivering better results than we obtained with small incremental expansions.

      General Advice
      Regardless of which application you use, the clarity of your prompts will generally be the key to successful outcomes. So if you’re obtaining poor results, the first thing to do is try a different prompt.  The main things to focus upon when writing prompts are as follows:

      1. Keep them simple and very specific.
      2. Avoid instructional prompts, such as create, alter, insert, remove or produce. Descriptive adjectives and simple nouns will always give you better results.
      3. Use photographic terminology when appropriate.
      4. Use defining words to provide specific information about what you want. Specific prompts such as a black cat, a cloudy sky, geometric buildings, happy children, birds in flight will enable the software to close in on what you want.
      5. It’s OK to leave the prompt field blank to see how the software will handle the in-fill or replacement. In most cases it will fill the selection with a look similar to the surrounding image pixels.
      6. If the software doesn’t create a result that satisfies you, try again or start afresh with different prompts.

      Saving Results
      Images edited with Luminar Neo’s generative AI tools will always be saved with the relevant extension indicating which tool was used. Thus, images created with GenErase will be renamed with the genErase extension, those with GenSwap will have the genSwap extension and GenExpand-generated images will gain the genExpand tag.

      You can choose which file format they’re saved from popular options including, JPEG, PNG, TIFF, JPEG-2000, Photoshop (PSD) and PDF. You can also select the size and resolution for each image you save and choose between the sRGB, Adobe RGB and ProPhoto RGB colour spaces.

      Compared to Adobe Photoshop
      While Luminar Neo offers a very comprehensive suite of image editing functions, its high levels of automation target it at a completely different type of photographer. Photoshop is a fully professional application that provides much more adjustability plus integration with other applications in the Adobe CreativeCloud suite.

      Luminar Neo is very much an amateur/hobbyist application. It lacks the organisational flexibility of Adobe’s Lightroom and the Undo function is very basic and quite limited in scope. You can’t add text to images; nor can you view or edit image metadata. There’s no way to batch rename images and exporting multiple images is very slow. Adobe’s software is far superior in all these respects.

      Luminar Neo will win out for photographers who want quick and easy ways to achieve what would otherwise require complex editing processes – and a considerable amount of prior training. As such it’s a capable application that is competitively priced for what it does.

      It’s also easy to open a saved image with prior generative AI editing and apply one of the many creative effects Luminar Neo includes by default. The image below was created by using the Sky Replacement feature after we had used GenSwap to swap in a camp fire in an original image.


      This image was created in Luminar Neo from our GenSwap image by adding in one of the sky replacement presets the software provides. 

      On 1 January 2024, Skylum transitioned Luminar Neo to a subscription-only base, although ‘Lifetime’ licences are still available and the company will continue offering support and updates to customers with existing Lifetime plans until the end of the year. You can purchase a Lifetime licence with Extensions and one-year access to Generative technologies for AU$299. Existing purchasers will also retain unlimited access to Extensions they’ve already purchased.

      Customers with 2023/24 Creative Journey Passes will be able to access the generative AI tools until 31 December 2024. Click here for details.

      The table below compares the subscription plans of Adobe and Skylum, as of 1 January 2024.

      Subscription period Cloud storage included Additional comments
      Monthly Annual
      Luminar Neo AU$14.95 AU$119 No 2-year subscription available for AU$179
      Adobe Photoshop Photography Plan AU$14.29 AU$171.47 20GB Includes Photoshop, Lightroom, Lightroom Classic, Photoshop Express and Adobe Firefly
      A$343.07 1TB
      Photoshop Plan AU$29.99 A$343.07 100GB Includes Adobe Express and Adobe Fresco
      Creative Cloud All Apps Plan AU$79.99 AU$871.07 100GB covers 20+ creative apps and services

      For those who don’t need the additional Cloud storage bundled with Photoshop, Skylum’s pricing for Luminar Neo looks like good value. However, costs will increase if you add any ‘creative assets’ that are not included in the standard Luminar Neo bundle. Options include additional skies, overlays, textures and backgrounds as well as presets and LUTs (look-up tables) for achieving specific effects. You can browse through the free and paid options here.

      Unlike Adobe, Skylum doesn’t provide a smartphone or tablet version of Luminar Neo. File sharing via a smart device requires the Luminar Share app, which provides a bridge for importing or exporting photos directly to external devices.  It’s available through the Apple App Store or Google Play.

      Conclusion

      Please Login or Register to access the Conclusion.

      Luminar Neo – Buy options
      (Photo Review Affiliate link)

       

      SPECS

      Build: 1.17.0
      Hardware
      : CPU Intel Core i5 or AMD Ryzen 5 or better, OpenGL 3.3 or later compatible Graphics Cards
      Systems compatibility: Windows 10 version 1909 or higher (only 64-bit OS) or macOS 11 or higher operating systems
      Display support: Minimum 1280 x 768 size or better
      Disk space requirement: 10 GB free disk space; SSD for best performance
      Minimum RAM:  8GB (16GB+ recommended)
      Supported export formats:  JPEG, PNG, TIFF, JPEG-2000, PSD, PDF
      New Raw formats supported
      : Nikon Z f (standard or lossless compression only), OM System TG-7, Leica M10 Monochrom, Leica M11 Monochrom and Leica Q2 Monochrom
      Computer interface: Internet access to download software and access generative AI features
      Tethered shooting: Not supported
      Editing functions: Brush Tools, Layers, Camera & Lens Corrections, Noise Reduction, Keyword Tagging
      Export to social media: Supported via Share Image button
      Mobile device support: Android and iOS apps available; Can accept images from smartphone via QR code link; Mac users can use it on the iPad Pro 2021 or later
      Trial downloads available: Yes
      Distributor: Skylum
      Luminar Neo – Buy options
      (Photo Review Affiliate link)

       

      Rating

      RRP: Subscription plans: AU$14.95 monthly, AU$119 annual, AU$179 2-year subscription; Lifetime licence + Extensions + 1-year access to Generative technologies is AU$299

      • Features: 8.9
      • Ease of use: 9.0

      BUY

      Luminar Neo – Buy options
      (Photo Review Affiliate link)