DxO Optics Pro V6.2 software review
DxO Optics Pro V6.2 review
Using DxO to process and correct my pictures
...Why not sign up for our (ad free) Newsletter to keep informed about new articles and subscribe to my YouTube Channel
The article covers mainly what Keith considers to be the key elements of DxO Optics Pro in his workflow, where he takes a few selected camera images and works on them to produce large prints.
We’ve published a number of reviews of DxO Optics Pro over the years (in fact, every version), and as the software has got more complex (in its functionality), it’s probably a bit too big to cover completely in one review. I’ve used earlier versions of the software concentrating on its ability to correct image distortions (lens and camera related) and to pull out image detail and tonality that helped me create images the way I wanted them to look.
The software is available for particular camera/lens combinations only, and comes in two different versions, supporting different cameras and lenses.
I’ll show how the software works perfectly well and can be useful if just your camera body is supported, but for the optimum results you need to work with both your camera and lens supported.
Just in case you are new to the software, it’s essentially making use of measured camera/lens capabilities to get the best in converting the raw sensor data (RAW files) that come out of your camera. Each body/lens combination is different, so that means a lot of combinations.
You can apply many of the image processing options to JPEGs from your camera too, but obviously this loses many of the advantages of shooting RAW in the first place.
The (optional) start-up screen allows you to carry out a bit of housekeeping before diving in, if you are used to previous versions, or have ‘Projects’ stored (collections of images and associated data)
I’m going to be looking at this software very much from the point of view of someone wanting to get on with processing images, so dialogues you’ll see are almost always the ‘Advanced user’ versions.
If some of the screens look a little daunting, then do remember that there is a guided ‘Wizard’ approach available that presents everything to you in a much less ‘full on’ manner.
If you are curious about how any of the adjustments look with your own images then do remember that the software is available as a free time-limited demo from DxO. I’m looking at the Mac version here, but the PC version is very similar.
Most changes in V6 are developments of existing software functionality, but do have a look at the V6 Features list from DxO if you’ve looked at earlier versions.
The range of supported equipment is growing and is listed on the supported cameras and lenses page on the DxO web site – where you can check if your equipment is supported.
There are a comprehensive set of user guides and tutorials available. I would seriously suggest you have a look at them, since the ways I’m using the software here may not match your own preferences. DxO optics Pro has been designed to work with Lightroom and other software.
It’s worth noting that the main reason I’ve stuck with DxO Optics Pro is not its integration with other software or ability to organise my workflow – it’s just plain and simple image quality. It’s the RAW conversion software I use for many of my large individual works.
Delays in software availability (for the Mac) have meant that I’ve not used it as much in the last year as I might have liked, so I welcomed this opportunity to put it through its paces, and in a few areas directly compare results with what I might otherwise choose to use.
Preparing images for processing
I’ve started, by looking at a set of images I took in 2006.
One of the reasons I always shoot RAW is that you can go back and reprocess images when software advances. It just so happened that a friend of mine wanted a print for her new home.
The photos were all shot at 1250 ISO on a Canon 1Ds – this was its highest ISO setting.
In the shot below I’m looking at a folder of folders of RAW files from 2006.
I’ve selected an image (bottom corner) to work on, and the software has informed me I need to download a ‘module’ to support my lens on the 1Ds body.
After selecting a few more images, I can see that I need two more.
Modules are the specific lens correction data that the software can use to optimise processing of any particular file from a particular camera body.
You can download modules as needed, or when setting up the software.
I also changed the mode of the software to Advanced User – lots more options, but do take your time to appreciate just what the various adjustments do, if trying out the software.
The basic workflow I’m adopting here, is to select images, decide on adjustments and then process them.
When it comes to further processing/enlargement/sharpening I’ll be working in Photoshop, but right now I’m looking to provide the most useful image for that next stage.
Note that I say ‘most useful’ since as I’ll show later – the best image for the next stage may not be the best looking.
The images I’ve selected, are shown as thumbnails below the selection – this works fine for me but there are lots of other ways of arranging your workspace.
If you don’t like my way of working, then just look at the results and find your own preferences for layout etc.
Notice the little icons around the thumbnails – these can be configured (below) to show lots of useful details about the images.
Before launching into all the different adjustment options, I should mention the collection of pre-defined preset adjustments.
Try a few – see what you like. The settings values are all displayed in the side panels.
Pick one you like (or don’t like) and experiment with modifying settings – I found this a great way to get a feel for what I could change and how some adjustments interact. Making good use of presets is important if you start using this software as part of a higher volume workflow
With many adjustments it’s better to sort one out before an other. My own choice would be to deal with lighting / white balance / exposure before looking at detail / noise / lens corrections / sharpening.
Working on images
Some corrections show in the preview when files are viewed at a reduced size. Others only at 75% and above.
It pays to look at interface elements carefully to see where there are additional non-automatic or fine adjustments available.
One thing to note with DxO Optics Pro is that a lot of the lighting adjustments work exceedingly well at one of the ‘auto’ adjustment settings. Even if you’re the sort of photographer who regards any use of ‘Auto’ in settings as a dangerous sign of moral weakness, I’d say try it out and be honest with yourself when looking at the results.
As with other aspects of the software I’m going to endeavour to show some highlights of what I’m doing, rather than an encyclopedic listing of functionality.
The image below is one that I took in 2004, and I’ve used when testing RAW converters ever since. I’ve produced several 60″x40″ prints from this 11MP Canon 1Ds file,
An image from 2004? – well it just so happens that I recently had an order for a print…
I’ll come back to this image in the conclusions, but the screen shot below shows my preferred layout when working on an image. The controls to the right can be collapsed, expanded and scrolled to the aspect that interests you.
White Balance and Colour
White balance adjustment defaults to allowing selection by named lighting condition (‘Cloudy’ for example) or by using an eye dropper tool to select a neutral colour.
A careful look at the interface panel showed an advanced option (+) allowing colour temperature/tint to be selected.
There are quite a few different ways of setting the overall ‘look’ of the image from a colour point of view.
I can add vibrancy – which can be thought of as a more advanced version of turning up the saturation.
You get a more natural look to images, although it’s worth noting that in comparisons, most viewers tend to prefer more highly saturated natural colours in landscapes, and often think that ‘accurate’ versions look flat and lifeless.
The way that colour sensor data is translated into RGB values is also under more direct control here, than with many RAW converters.
The colour rendering options (below) include various generic film types as well as the useful ability to use a custom Camera ICC profile.
I don’t use this very often, but it gives a potential accuracy that I do need every so often in my work.
Not that often though, and for most users I’m firmly of the belief that ICC camera profiling is all too often more trouble than it’s worth.
‘Color Modes’ allow a degree of movement away from default contrast and saturation settings – including the optional DxO FilmPack module which allows particularly accurate renditions, as if you’d actually had a roll of Tri-X or Velvia in your camera…
The multi point tool allows for localised adjustment of hue/saturation – it works, but I might be more inclined to do such adjustments as part of my general editing later on, after raw conversion.
With higher ISO settings becoming common in cameras, digital noise processing can make a big difference.
One of the highlighted advances in DxO 6.2 is improved noise handling and processing.
As someone who avoids use of flash whenever possible, this feature is one that caught my attention.
The image below was taken at 6400 ISO on a Canon 7D (it’s part of my look at the camera at the end of last year)
…or below to see more of the image
After trying a few images from my 1Ds3 as well, I can definitely say that the noise handling is improved by a noticeable amount from my normal RAW converter (Currently ACR6.1 on Photoshop CS5)
Remember that the examples above are 100% crops, so real world performance in prints will differ – I’ve found that the DxO images sharpen up well for printing and reduction for web use.
As well as correcting lens distortions, processing with the correct lens module also allows for differences in lens performance across the field to be corrected.
This is in addition to normal generic sharpening applied to most RAW files when converted.
The Unsharp Mask option can be adjusted to match your needs – this is also of use when your lens is not supported with a lens module – it lets me get many of the benefits of DxO RAW processing for any lens I might choose to try out on my 1Ds3 (see the Moon shot below).
If I’m enlarging an image a lot, I may well turn down the sharpening a bit, having found that judicious light sharpening applied to parts of the image just before enlargement can produce less (print visible) artefacts when making large prints. This is very much something I’d suggest you experiment with if you regularly produce large prints (over 30″ across for my 1Ds and over 40″ for my 1Ds3)
When most people think about correcting lens distortions, they think of the likes of pincushion or barrel distortions.
Many lenses exhibit more complex distortions – particularly cheaper ‘kit zooms’. With the correct lens module, a wide range of distortions can be ‘fixed’.
It’s worth noting that if you are wanting to set up shots for creating your own lens correction data (for ACR6.1 for example) then it can be difficult to light targets and get the correct camera/target positioning. I’m not saying that it can’t be done, just that it requires a level of precision and care that you might not want to bother with (I’m including myself here) – Then again, and I’ll come back to this, DxO has a wide range of lenses supported – but you may find significant gaps…
Interiors show the equivalent of a very wide angle rectilinear lens, although perspective like this is something to be used with care. (mouse over to see)
For some images, the option to partially adjust the geometry is useful.
It’s important to remember that as you use wider and wider lenses, there are always going to be different effects visible when the lens projects the 3D world onto a flat sensor.
As to whether the effect is noticeable/desirable, it’s up to you. The range of adjustments do make it easier to avoid the stretched heads effect if there are people in the shot.
There are other image adjustment tools, such as the ability to force lines to be parallel.
The shot below was hand held with a 17mm shift lens – normally for my architectural work I’ll always use a tripod, but for this shot of the 1923 Lutyens designed war memorial in Leicester I didn’t have one with me (I was out visiting someone, and the TS-E17 just happened to be in my bag ;-) )
Note that the TS-E17 lens is not a supported one.
It’s not far out of true, and with a structure like this, it’s not hard to find lines that should be horizontal.
You can zoom in for more precise placement.
More extreme corrections are possible, but this is perhaps an operation I’d prefer to carry out after I’ve processed the image.
There are many more ways of adjusting images which may or may not be of interest. This is the sort of thing you just need to experiment with, and the demo of the software does give you a month’s use.
It’s important to note that if your camera body is supported, then any picture taken with any lens can be processed.
Here’s a 1Ds3 shot of the Moon, not long after sunset, with a 2000mm f/10 lens (aka a Celestron 8″ telescope) Just a very quick test with an adapter – 1/800 at 1600 ISO since the telescope wasn’t guided and the atmosphere was not at its most stable.
You can see I’ve slightly modified the tone curve to get a bit more contrast. Noise reduction settings were set automatically by the software for the 1Ds3.
The Moon – some extra sharpening applied for web version.
Black and White
Over the last few years I’ve looked at a lot of ways of converting colour to black and white. One thing I’ve learnt is that optimising your colour original image to ‘look good’ can often not produce the best black and white version.
If I apply a strong lighting correction, it’s all too easy to get an image that (for colour) is somewhat ‘overcooked’ and pushing towards the (currently all too popular IMHO) HDR look.
However, when it comes to black and white, I’m often looking to emphasise form and structure, so having the strong colours and very effective shadow lighting can be of use.
Perhaps not too strong for a colour version of this image, but I’d probably use ‘medium’.
Depending on how I wish to process the image further, I’ll add in an amount of sharpening and correct for lens distortions.
I’ll not do the conversion to black and white here in DxO, so I may create a few different versions of the image, in colour, to see how it looks after conversion.
If this seems a slow way of doing things then yes, it is. This is the approach I’ll take when working on an image for a large commissioned print.
The picture of Fish Creek falls (near Steamboat Springs, Colorado) is one that I used to produce a 60″ high print, a few years ago.
I returned to the original 1Ds RAW file when looking at DxO 6.2 to see how raw processing had changed.
One one particular improvement I noticed was that lens corrections for lens softness produced much smaller sharpening halos.
This has the benefit that I can use the lens correction at ‘full strength’ for more of the image, when I know that I’m going to be resampling to a larger file size (the 1Ds is ‘only’ 11MP)
The version to the right has been processed in DxO and then converted to B/W with the Nik Silver Efex Pro plugin.
It’s had no other work on it, other than shrinking to fit here, and a bit of sharpening for web display.
For a large print, there are areas of the image I’d work on further, such as bringing out a bit more structure and detail in the falling water.
I’d also look at making sure the patterns and structure of the rocks would show up in the shadows – it’s a lot easier to do this if you’ve got the RAW conversion the way you wanted.
When looking at the examples I’m using in this article, please do remember that you are seeing compressed JPEG versions of screen shots, often resampled to fit.
It’s sometimes difficult to show on the web, what appear as subtle differences or enhancements in a file opened in Photoshop, on a good quality calibrated monitor.
Have a go with the demo and see if the differences that this software makes matter to you?
Once you’ve selected the images you want and decided the settings for them, it’s time to process them to whatever formats you need.
Note – although I’ve concentrated on RAW processing options, the software can process and correct JPEG files from supported cameras – obviously this restricts what you can do to some extent.
There are a number of modifiable preset options available, and you can add additional ones if need be.
In this example, I’ve just ticked the TIFF option, so each conversion will produce one TIFF file.
If I’d ticked other options then I’d get more than one output file per conversion.
Note the availability of the DNG format. A DNG generated by DxO Optics Pro is a 16-bit linear DNG format file.
I don’t use this myself, since I always retain the original camera RAW files, but it may have uses for your workflow, where you want to process your files with Adobe Camera Raw for example.
Once you are happy with all your settings, you can just set the software going, to do its stuff.
In the past, I’ve found the DxO software somewhat sluggish in performance on some of my Macs, but this version makes good use of the performance available in my 8-core Mac Pro (OSX 10.6.2) and processed the 27 RAW files in the display below, in just over 7 minutes.
As images are processed the thumbnail status is changed to ‘processed’.
During processing you can return to working on selecting or setting adjustments for images.
There is another tab entitled ‘View’ where you can look at images and see how the processing has gone.
For myself, I’ll be opening images up in other applications, but it does allow for a quick check that everything has gone as planned.
As I mentioned earlier, it’s worth getting to know the range of presets supplied and also to consider making your own, once you find some settings you really like for a particular purpose. For example I know a wedding photographer who uses a particular setting to bring out detail in white wedding dresses…
The software remembers all your adjustment settings, in either its own database or as external ‘sidecar’ files
I’ve deliberately looked at just a subset of what the software can do – concentrating on what I still feel it does best, turning my RAW camera files into files I can work on to make my prints.
DxO 6 for the Mac has taken considerably longer to get to market than I suspect DxO would have liked, but now it’s here I have some software that is of real use to both my day to day and specialist print work.
I find the interface design and general layout of the software feels much better than earlier versions. With V6.2 I had that ‘what does this little icon/button mean?’ feeling a lot less than in years gone by.
I have to note that, as I mentioned in my review of DxO 5.3, support is still not available for my Canon EF14 2.8L II lens, and this represents what may still be a show stopper for some potential users. DxO are working hard to add support for different cameras and lens combinations, but as I find with my 14mm, even one of Canon’s top ‘L’ lenses (OK at £1800 perhaps not a top seller) is not supported with Canon’s flagship camera body (the 1Ds Mk3)… Of course I do use the 14mm with the 1Ds3 and DxO, but I’m not quite getting the best out of it.
It does grate a bit too, that the software insists on suggesting I download the module for a 14mm Sigma lens when EXIF data says Canon 14mm (DxO tell me that they are looking into this…)
The help menu tab in the application is of somewhat variable usefulness, returning some irrelevant Mac OSX related items when putting a query about ‘resizing’ into the search box.
On-line help is of more use, taking you to the PDF manual and tutorial options. Overall the in-application help did seem a little lightweight in comparison to many other applications I’m used to using on my Mac.
This is one area where I can’t comment on what the Windows version looks like.
Once past these niggles, let’s have a look at what I consider the software is really about – image quality.
Do remember my previous caveats about images on the web when looking at these comparisons.
As you might know, I rarely ever do comparative reviews, since they are difficult to do in a meaningful and accurate manner. I’m not wishing to get into the ACR vs DxO lens correction battles, but I’m going to show a few examples of comparisons with ACR6.1 (via Photoshop CS5) where I feel that there is a difference that is of importance in my work.
As they say YMMV… so if these example look of interest, then try it for yourself.
I’ll start with an image that makes an excellent (IMHO) large print and illustrates some of the difficulties that a RAW converter has to deal with in real world images with a wide range of lighting
The area around the sunset saturates pixels – there is just more light than some pixels can handle. This means that a particular R, G or B pixel will put out its highest value – 255 if we are working at 8 bit.
Fortunately for those of us liking to photograph sunsets, a lot of work has been put into recovering as much useful information out of these highlight areas.
This relies on knowing that not all pixels may be saturated and doing some tricks to try and work out what would have been there.
If you are familiar with Adobe Camera Raw, then this is what the recovery slider is for.
Using just the default settings, the area around the sunset exhibits some weird effects – I was there and there was not a green patch of sky.
Now it happens that ACR works rather well at this and can produce a much better conversion – indeed I use this software regularly to process images as part of my work.
What about DxO 6.2?
Well, this is the result of using light (auto) highlight recovery and medium auto lighting.
It’s different, and in the context of making a very large print, a better image to work on, for the look I’m after.
This second example (both samples at 100%) shows the best version of the neon tubes that I could get in adjusting a high ISO shot from my 1Ds when processing with ACR
Note the green ‘cores’ of the tubes. It’s a night time shot from a 1Ds at 1250 ISO.
Below is a DxO version that also shows differences in noise handling.
There are a number of other examples I’ve got, that show aspects such as the reduced shadow noise and ‘stairstepping artefacts’ (see below) in DxO conversions.
The example below shows how fine detail of some railings (@300%) is rendered differently in DxO (left) – whether you’d notice this in normal image use is an entirely different matter…
Note, If you think this looks bad, it’s a tiny crop from an image. It’s also much less than the sorts of problems you’d get all over the place if your camera sensor didn’t have an anti-aliasing filter in front of it. Some complain about such filters, but for most of my work – I’d rather it was there than have to pore over images looking for artefacts. If you don’t like it, then buy a MF digital back…
These suggest to me that DxO is probably still ahead in many aspects of processing RAW camera files.
But, and it’s a big but, do these advantages reflect in real world images as you might wish to use them?
This is where many of the forum discussions I see, generate heat but little light – it all depends on if your chosen equipment is supported, what you want to do with your images and what other software you use (not to mention how much money you have to buy stuff).
In some ways I perhaps get less benefit from the software by using good quality lenses, where the distortions and softness are (I hope) noticeably less than cheaper ‘kit’ lenses.
Only you can decide if the benefits and ways of working that DxO Optics Pro offers fit in with what you do. For myself, having the option to process files in a different way, with a different ‘look’ is of real benefit. The difference in image quality is a tangible benefit to my business.
It’s been a while coming, but at it’s core, DxO Optics Pro V6.2 is currently the best solution I have for processing many of my images.
Top quality RAW image processing software that integrates lens correction and image adjustment.
Latest version includes very good noise reduction for high ISO images.
Works stand-alone or can be integrated with other software solutions such as Adobe Lightroom.
The software has a lot of functionality, and once you move to advanced modes it can be quite complex, but well worth getting the free demo to see how you like what it does.
Can be purchased in download form.
See Update Nov. 10 – V6.5 HDR processing
- Mac OS X 10.5, 10.6 (Intel processor – only)
- Memory requirement : 2 GB RAM (3 GB RAM are recommended for processing images which were taken with a sensor of 20 Megapixels or higher.)
- Disk space : The installation of this software requires a minimum disk space of 400 MB. Additional disk space may be required and will vary with the DxO Optics Modules you select.
- Pentium 4 or Pentium Dual-Core or Pentium M or Pentium 64 bit, or AMD or AMD Dual-Core processor or AMD 64 bit
- Microsoft Windows XP SP2 32 or 64-bit, Microsoft Windows VISTA 32 or 64-bit, Microsoft Windows 7 32 or 64-bit – The use of a 64 bit system is recommended for processing images which were taken with a sensor of 20 Megapixels or higher.
- Memory requirement : 2 GB RAM (4 GB RAM are recommended for processing images which were taken with a sensor of 20 Megapixels or higher.)
- Disk space :The installation of this software requires a minimum disk space of 400 MB. Additional disk space may be required and will vary with the DxO Optics Modules you select.
Never miss a new article or review - Sign up for our occasional (ad-free) Newsletter
Enjoyed this article?
All the latest articles/reviews and photo news items appear on Keith's Photo blog
We've a whole section of the site devoted to Digital Black and White photography and printing. It covers all of Keith's specialist articles and reviews.
For All about using tilt and shift - articles/reviews about tilt/shift lenses
Articles below by Keith (Google's picks for matching this page)
...Why not sign up for our (ad free) Newsletter to keep informed about new articles and subscribe to my YouTube Channel