PhotoLab users often complain there’s not enough new in the annual update (not my opinion, the improvements to local corrections were substantial – if a user doesn’t feel that the single version update doesn’t offer enough improvements, s/he should just upgrade every other year).
Including blow-up capabilities in PhotoLab 7 might be enough to quiet the complainers and get some positive press. A flagship new feature.
Are you under NDA for the new features for PL7E complete (including otherwise free-standing applications that actually are in PL once the additional applications are licensed for fee – a different “plugin” for fee approach)? Unless this would work with raw files from the body or DNG raw files converted by Adobe applications, I will continue with Topaz. My only reasons for licensing the latest release are: (1) application software defects (bugs) that are fixed (once a new release comes out, DxO more or less orphans the previous release), (2) new body, lens, or raw file format support that is not available in my licensed release and that I now need (this support also could be promised in the new release, as was the case with the Nikon Z9), (3) significant new features such as yet another DeepPRIME, (4) a real catalog replacement that would work on archived media (primarily SSD/s for me these days) when attached to the computer running PL (rather like Camera Bits PhotoMechanic but rather easier to setup and use or even as good as Adobe LR for intake to Adobe PS – PL is my replacement for PS) with a firm promise that in later releases, the catalog would be convertible to whatever new format DxO might decide upon for the catalog, (5) less wallclock time for DeepPRIME, etc, processing without constantly buying the latest and greatest computer, preferably on a laptop/portable workstation, as I must do much of my work in the field.
I would argue that so called super resolution that Adobe offers and is supported by Nvidia is far from cutting edge and its quite underwhelming at this stage of the proverbial arms race in up-scaling technology. Lot of people promoting it in YouTube videos tend to be influencers paid to do so. Not to say DXO does not do the same with their products , but claims of super resolution being some kind of magic seems exaggerated compared to far more powerful technology we find in latest TopazLab GigaPixel AI for example. while I would not mind seeing up scaling AI in DXO products, I would not want to see anything less than what GigaPixel AI can do with their latest algorithms. I think since they specialize in that they have a head start, and DXO would probably be diverting valuable resources to developing something that already exits and there is a lot of competition. I would prefer to see DXO instead spend their limited resources in refining what they already have, UI teaks, speed and stability etc. And work on what they already do better than others. Because anything lower quality than Topaz at this stage would be underwhelming and since Topaz has such a big head start in this area , seems like limited resources would be better spend elsewhere in the case of DXO team.
Upscaling feature would be welcome in Photolab.
Brilliant! I could upscale my Canon R5 400 megapixel photos!
I haven’t had a close look at the quality of the result, but … PhotoLab can generate an “upscaled” image now, via the Export to Disk process;
I think the point is to do better than traditional upscaling and use all of that neural network training to do a better job upscaling.
Yes, understood, Mike … All the same, when I encountered this “feature” by accident, I was surprised to find that PL happily stretched my image … and the result didn’t seem half bad !
I have just prepared twenty images for printing to 1m40 x 1m.
I used Topaz Photo AI for resizing, this is their latest product, which combines DeNoise AI, Gigapixel AI and Sharpen AI in one app.
It is not just about resizing, it’s also about appropriate sharpening for the new size and, something I stumbled across in fixing one particularly small file - using AI to correct JPEG compression artefacts, which only became apparent on enlarging.
I don’t want to be derogatory but, DxO would really need to go some to implement something as powerful and effective as Topaz for “super sizing”. Not saying they couldn’t but, it would mean diverting resources away from what DxO have become good at and all sorts of things that are already in the pipeline for PL7,8,etc.
Funny. I thought DxO users are used to use workarounds? 
But more seriously, I think printers, especially large size printers are such a highly special kind of devices, depending on drivers, firmware updates, OS updates, paper profiles - I don’t blame any RAW converter not to diving into such deep waters. And against JPG artefacts using TIF is a, uuuhmm workaround, right? 
This kind of deep work with image processing is part of what DxO has been doing with Prime, DeepPrime and DeepPrime XD, not to mention the Lens Sharpness palette and even the Fine Detail slider. It’s an area where DxO could compete effectively. DeepPrime XD in particular works with detail recreation, which is a huge part of Topaz image enlargement.
Topaz Gigapixel is $99 (although $200 in the suite with Sharpen AI, which is an astonishing tool, and DeNoise, which is pretty much rubbish next to DeepPrime). There’s a market for this feature at a price at which DxO could compete.
Since Adobe is trumpeting its relatively lousy automated image enlargement, DxO almost has to join this game to stay competitive. Image quality is in DxO’s wheelhouse, unlike the damn DAM distraction. Oh well, thanks to the DAM, at least filtering in the image browser works in PhotoLab 6.
I completely agree with you that DxO could develop something similar if they chose to do so. However I think they would probably be better served to use their limited resources to develop other new functionality to enhance PhotoLab’ s core feature set. Or, perhaps they can take the time to redevelop PhotoLab’s extremely primitive printing functionality.
Mark
Ah yes. Truly awful. At least three different places where you can choose a profile and, no matter whether you specify borderless or not, the minute thumbnail preview always shows a white margin.
I’m assuming the Mac version of the printing functionality is probably somewhat different than the Windows version so I can’t speak to what you’re seeing. However, the Windows version is also pretty awful with very limited and poorly implemented functionality.
Mark
The first thing that strikes you is the unusual layout ,where you are expected to state how many “cells” you want to create - then you have to calculate all sorts of stuff like margins for cells and paper.
Canon provide a comprehensive layout and printing utility. It’s not simple but it has a more “trustworthy” feel, especially since the preview can be almost full screen instead of the postage stamp the PL gives.
Yep. Similar in the Windows version. Truly dreadful.
Mark
You guys should make a separate feature request about improved printing. Again this discussion is about upscaling.
(I don’t print from PL and could care less! )
I no longer have a darkroom and no longer use Cibachrome from Velvia. For routine printing, I use a Canon Pixma Pro 10 with various papers. (Not the latest and greatest, but as good as conventional papers from film, and, with the correct paper, “almost” Cibachrome.) However, neither the Canon interface nor any other interface, except an obsolete Adobe plus plugins (I do not have Adobe as a “silent partner”) allows me to carefully control the output to be what the client wants without a test run for individual images. I do have a calibrator for my screen (Spyder X pro), but this does not always translate correctly to what is printed. I fully understand that DxO does not have the resources to make such a “print driver”, but it would be “nice”. I am still looking for a mechanism to export what I see in PL6E complete on my screen to get the print correct. Do note that most of the commercial print shops I have tried (eg, metal prints that emulate Cibachrome) often do not get it right on the first attempt (or second or third if the “reliable” technician is no longer there).
You need to calibrate the printer for each ink/paper combination, creating an ICC profile.
Also ensure that the screen brightness is brighter than 80cd/m².
I have ICC profiles for my screen after calibration. Does Canon provide a mechanism for the Pixma 10? A URL will suffice.