DxO PhotoLab 4 comes with AI-powered high ISO noise reduction

Oct 21, 2020

John Aldred

John Aldred is a photographer with over 20 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.

DxO PhotoLab 4 comes with AI-powered high ISO noise reduction

Oct 21, 2020

John Aldred

John Aldred is a photographer with over 20 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.

Join the Discussion

Share on:

DxO’s PhotoLab has seen a lot of improvements over the years. Their latest iteration, PhotoLab 4 comes with some pretty impressive high ISO noise reduction along with a host of workflow speed and other features. These improvements allow you to customise the user interface and provide easier access to your more commonly used tools, as well as adding a few new ones.

YouTube video

The big advantage DxO has over most of the competition is that they’ve kind of become a standard for benchmarking cameras and the images they are capable of producing. This means their engineers have used pretty much every digital camera ever made and worked with the files they produce. It also means they have access to a database of millions of images. Images that they’ve been using to train their deep learning AI.

This leads us to the DxO DeepPRIME AI engine built into PhotoLab 4, which denoises your images during the raw file demosaicing process to maximise image quality in the final output and offer increased noise reduction while retaining as much of the original colour and detail as possible at higher ISOs and in low light environments.

And it works on really old cameras with pretty terrible high ISO performance, too! Like this Nikon D300S at ISO2500 (I shot a D300S for years, no way I’d ever take it this high!). Don’t worry, though, it has camera profiles to support raw files from all the latest cameras, too, like the Canon EOS R5 & EOS R6, Nikon Z6 & Z5, Panasonic S5, etc. and over 60,000 camera & lens combos for lens correction.

It even manages to handle astrophotography quite well, being able to distinguish between the difference between high ISO noise and stars rather well!

It also manages to pull back some pretty impressive detail in the shadows of images that’s otherwise quite difficult to make out, too.

It’s making me want to dig out my old Nikon D100 and see how well it handles anything over… Well, ISO200 if I’m completely honest (yeah, it really wasn’t great). Anyway, as well as the new Deep Learning AI-powered noise reduction, DxO PhotoLab 4 also sees a bunch of new workflow improvements to help you speed up your editing process.

First up is DxO Smart Workspace. This splits up all of the modifications you can do to an image by editing type (Light, Colour, Detail, etc), as well as offering a new favourites system to only display your most… Well, your most favourite modifications. After all, if there are things you use on every image and other things you never touch, there’s no reason to see them all at once, is there?

You can also limit the corrections and adjustments to only show by “Active” status. That is to say, if you’ve changed a setting, then they show up. If they’re still at their defaults, then they don’t, which means it really easy to see exactly what you’ve already done to an image without having to go hunting for it. And to make the hunting even less of a pain, there’s a new search feature (and, yes, it understands what the tools are called in other applications!).

PhotoLab 4 finally introduces some hefty batch renaming tools, which let you edit, search and replace filenames to manage your image library more efficiently. you also get selective copy and paste of attributes applied to an image. So, if you’ve just set a custom white balance at the beginning of your session, or you’ve pulled back the highlights to claw back those blown whites on your bride’s wedding dress, you can now selectively copy and paste those individual changes onto all the rest of those images without affecting everything else.

The edit history has also been expanded, allowing you to not only undo individual steps as far back as you like, but also allows you to deep-dive into presets you might apply to see each individual thing that preset changed in the process. You can even create a new version of an image from the edits history tab, too.

Watermarking (whether you’re for or against it is another debate and a different post entirely) has also been expanded and now offers the ability to use both text as well as images simultaneously using a variety of blend modes. So, if you want to brand with your logo, but have specific text for a session, now you can. you can also save watermark presets. And they don’t have to be enabled while you’re editing, you can work on your images at will, and then just have it bring them up when it’s time to export (very handy if you’re just sending over a bunch of web-sized proofs to a client that they haven’t paid for yet!).

Overall, it looks like a pretty significant upgrade over the previous version, and might just help you to both pull out an extra stop or two of ISO from your camera at a quality you deem acceptable, as well let you spend less time sitting at the computer after a shoot!

DxO PhotoLab 4 is available to buy now for $99.99 (or $49.99 to upgrade from a previous version) for the Essential Edition and $149.99 ($69.99 to upgrade) for the Elite Edition until November 19th. After that, the regular prices for Essential and Elite go up to $129 and $199 respectively, with the upgrade cost going up to $49.99 and $69.99. If you’re not sure if you want to buy yet, you can also download a free 30-day trial.

Filed Under:

Tagged With:

Find this interesting? Share it with your friends!

John Aldred

John Aldred

John Aldred is a photographer with over 20 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.

Join the Discussion

DIYP Comment Policy
Be nice, be on-topic, no personal information or flames.

Leave a Reply

Your email address will not be published. Required fields are marked *

2 responses to “DxO PhotoLab 4 comes with AI-powered high ISO noise reduction”

  1. Amanda Rain Avatar
    Amanda Rain

    noise and grain is the one thing I detest in a photo it is my worst enemy

  2. JP dJ Avatar
    JP dJ

    DxO does not benchmark to be the nice new kid on the block of digital image processing (DIP).
    DIP starts with building an understanding of how a specific camera’s raw file needs to be interpreted. A sensor is analogue and the bits in the raw file follow from Analog-to-Digital (AD) conversation. Was this done linearly? Are the individual photosites in the sensor linear? Then you have figured out your (say) 14 bits raw data. Now you DIP needs to start inventing RGB pixels from the raw data (from monochrome photosites – a sensor has no pixels – a pixel has RGB values for a single point, raw data does not). Did the camera give additional rows and columns of photosite values? No? Bummer, now DIP needs a separate edge algorithm. Yes? Great – easy.
    The assessment of each camera to reverse-engineer its way of creating a raw file that implies how to DIP it, is what’s behind the “benchmarking”. This might even be firmware dependent. As lens flaws can be corrected up to a point in DIP, these are assessed as well.
    Now you have the knowledge for Photolab 4 or Camera Raw 10. This is why it can be critical for your DIP app to support your camera/lens combinations.
    Some smartipants at DxO one day said, let’s publish our assessments as benchmarks and generate traffic to our websites or else we never sell our apps.
    Now imagine that your raw photos are what a software developer thought they should be, based on a couple assessment shots with a version of your camera. Or, worse a greatest common divisor of most camera profiles called Adobe default. And understand that the pixels on your display from the raw file are the result of de-Bayerization and demosaicking – which could have been done worse or better than what a camera manufacturer would do with their inside in knowledge: my camera”s raw files gave color processing noise from DIP with app A, not the manufacturer’s, not with app C and not with app A when I used adapted old lenses.
    May I should try Photolab again.