rvst

After reading your most recent reply, I went back and read the first post. I think the two images you posted incorrectly give the impression of an installation error. However, you don't appear to have an installation error with Affinity software, reading the other posts in this thread. They installed correctly. The installation error in the second image above comes from an unrelated component and is a bit of a red herring. Looking at the first image, it appears you're trying to execute the Affinity application by directly invoking the executable in "C:\Program Files\WindowsApps\SerifEuropeLtd.AffinityPhoto2_2.0.0.1640_x64__3cqzy0nppv2rt\App\Photo.exe" You can't do this with Windows apps. They're sandboxed and the ACL on the files will prevent direct execution. You need to run Affinity apps either by clicking on the icon in the Start Menu, or by using the aflaunch.exe app posted by Mark Ingram.

I have a Galaxy Note 9 and I regularly use DNGs from the phone. They do not suffer from this problem.

I don't have a solution, but I can confirm the same behaviour here on both Affinity v1 and v2. Other applications display this DNG with what appear to be around a +3.5EV exposure difference. Not seen this behaviour before.

@Mark Ingram I found the command line switches on aflaunch.exe made it clumsy to use in some cases and impossible to use in others (where the 3rd party app will not allow the user to specify arguments to the external editor to choose which app is invoked and where you're wanting to use, say, Designer instead of Photo; an example would be XNViewMP). I hacked together my own launcher and I figured I'd share the code with you since it's easier to use and works with all Affinity apps and every 3rd party app I've tried. It's polymorphic in the sense that the launcher application will choose the Affinity app to launch as a function of its own basename. I imagine you're using similar code to launch the app in your aflaunch, so this code is simply to highlight a little functionality which does away with needing to pass command line switches to decide which app to invoke. Rename the executable AffinityPhoto2.exe and it will launch Photo.exe. Rename it AffinityDesigner2.exe and it will launch Designer.exe, etc. I think this would be a useful enhancement to your aflaunch.exe - one executable can launch all 3 Affinity apps without requiring command line switches. The code below is a hack, but feel free to use any of it if you wish - I assert no copyright. Caveat emptor - there may be bugs. #include <ShObjIdl.h> #include <atlbase.h> #include <appmodel.h> #include <strsafe.h> #include <processenv.h> #include <tchar.h> #include <wchar.h> PCWSTR CommandLineArguments() { // GetCommandLineA returns quoted arguments if there are spaces // Check if the first character is a quote // Then find the first occurrence of '\" ' // or if unquoted, ' ', to locate the beginning of the arguments LPWSTR cmdline = GetCommandLineW(); PCWSTR s = nullptr; if (*cmdline == '\"') { (s = StrStrW(cmdline, L"\" ")) ? s += 2 : nullptr; } else { (s = StrStrW(cmdline, L" ")) ? s++ : nullptr; } return s; } HRESULT LaunchApp(LPCWSTR AUMID) { HRESULT hr = CoInitializeEx(nullptr, COINIT_APARTMENTTHREADED); if (SUCCEEDED(hr)) { CComPtr<IApplicationActivationManager> AppActivationMgr = nullptr; if (SUCCEEDED(hr)) { hr = CoCreateInstance( CLSID_ApplicationActivationManager, nullptr, CLSCTX_LOCAL_SERVER, IID_PPV_ARGS(&AppActivationMgr) ); } if (SUCCEEDED(hr)) { DWORD pid = 0; PCWSTR args = CommandLineArguments(); hr = AppActivationMgr->ActivateApplication(AUMID, args, AO_NONE, &pid); } } CoUninitialize(); return hr; } WCHAR *GetErrorMessage(HRESULT lResult) { static WCHAR buf[512]; DWORD cchMsg = FormatMessage( FORMAT_MESSAGE_FROM_SYSTEM | FORMAT_MESSAGE_IGNORE_INSERTS, NULL, lResult, MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), buf, 512, NULL ); return cchMsg ? buf : nullptr; } LPCWSTR basename(LPCWSTR name) { WCHAR* p = StrRChrW(name, NULL, L'\\'); return p ? p + 1 : name; } LPCWSTR GetAppAMUID(const WCHAR* argv0) { LPCWSTR exename = basename(argv0); if (lstrcmpiW(exename, L"AffinityDesigner2.exe") == 0) { return L"SerifEuropeLtd.AffinityDesigner2_3cqzy0nppv2rt!SerifEuropeLtd.AffinityDesigner2"; } else if (lstrcmpiW(exename, L"AffinityPublisher2.exe") == 0) { return L"SerifEuropeLtd.AffinityPublisher2_3cqzy0nppv2rt!SerifEuropeLtd.AffinityPublisher2"; } else { // default in case the user changes the executable name to something unrecognizable return L"SerifEuropeLtd.AffinityPhoto2_3cqzy0nppv2rt!SerifEuropeLtd.AffinityPhoto2"; } return nullptr; } int wmain(int argc, WCHAR *argv[]) { LPCWSTR appname = basename(argv[0]); HRESULT hr = LaunchApp(GetAppAMUID(appname)); if (FAILED(hr)) { wprintf(L"%s: %s\n", argv[0], GetErrorMessage(hr)); } return FAILED(hr) ? 1 : 0; }

This will be fixed in an upcoming release. Serif will be making a "normal" installer, so you only have to use the helper application until then.

V2 software is installed as a UWP app and so it lives in a sandbox. You can't access it or execute it directly. You can only click the icon in the start menu to launch it. This is intentional on the part of Windows. Because of this, one needs to use a "helper" app to launch it when integrating with a 3rd part program. You can find one here:

Very positive review for Photo V2 released yesterday on Digital Camera World https://www.digitalcameraworld.com/reviews/affinity-photo-2-review

I think you'll find the vast majority of Affinity users love the product, likely even the most vocal of complainers on the forum. I've bought all three apps, even though my usage of Publisher is very low. They're just very accomplished pieces of software and their affordability and features make me want to support Serif by buying their software. Serif have built a feature-rich product suite and sell it at a sensible price, instead of the crazy subscription prices that some other vendors charge that only those who actually use the software for professional purposes can justify. In their blind rush to transform themselves into a SaaS company, so many vendors are trying to switch to a subscription. But what these vendors don't realize is that users cannot and don't want to absorb so many subscription fees. Affinity have realized this and thus have a very loyal following. Personally, I did not choose Affinity instead of Adobe. I simply cannot justify the cost of Adobe since I'm a hobbyist and not a professional. My choice was between Affinity and non-consumption (probably similar to some other users). That said, the V2 release has been less than smooth with the installation issues, other bugs and some questionable UI design choices. Some of the UI design is worse than V1 (many posts have covered this). The stability of the V2 software leaves something to be desired: my Windows event log is riddled with errors from crashes and there are tons of regressions in the software. It feels like a beta release in this iteration, which a release-quality version shouldn't feel like. But I love the software and I want to keep using the software. So I report the bugs in the forum. And I may even occasionally rant about a particular design choice. It doesn't mean I don't like the software. These bug reports and feedback, even those posts where people are ranting, provide the information necessary for Affinity to keep improving the product.

View-->Studio is for version 1 of Affinity Photo You find the option to show the Layers window in Window-->Layers on V2 of Affinity Photo.

Here's a list of tools for iPad: https://www.handyrecovery.com/best-ipad-data-recovery-software/ They're all commercial, but several have free trials that can recover the data. The first one in the list can recover up to 500MB in the trial version

I reported this in the wrong forum. Linked here as well and confirmed by @lphilpot

No, I hadn't - thanks for pointing that out. I'll link this thread Here you go clip0052.mp4 Platform: Windows 10 Professional 64 bit

You should post the file you're trying to export as well as the exported file - we can't tell anything from the exported file alone.

There are a number of commercial and free tools that allow you to recover files. Your mileage may vary depending on whether or not you have written data to your disk after the files have been deleted. You didn't specify if you're on Mac or Windows. Here's a link to a free one for Windows: https://www.iobit.com/en/iobitundelete.php You should stop using your computer until you have recovered the files. When you delete a file, the actual data isn't deleted, just the "map" that strings together the various blocks that makeup the file. Those blocks are marked as free space, but the data is not deleted, so it can be recovered by specialist software. However, the more you write to the drive, the likelier the operating system is to choose one of those disk blocks that contain your data and has been marked as free space, permanently destroying that file. So stop using your computer, download some recovery software and your file might be recoverable

I should probably give DxO a look again. I last bought a copy around a decade ago and it was a little bit of a one-trick pony at the time. Based on your comment and the length of time since I last had a copy, it might be a worth a spin

I have On1 Photo RAW and I'd struggle to recommend it. The UI is absolutely awful. Every time I use it I'm confused about where things are located. It seems to have worsened across versions. I still use Lightroom 6 in preference to On1 - I really only use its Portrait AI now which does a pretty good job as long as you don't crank the settings up too high. Their upgrade pricing sucks. There's barely a discount. I didn't even get a single update before they moved from the 2022 version to the 2023 version and charged the full upgrade price. So if you use On1, bank on paying $100 per year for an upgrade and getting few to no free updates before the next version they term "major version" is released, which often has functionality one would expect only of a minor release. Overall, On1 feels like a subscription product with once yearly updates priced too high for what it does. I chose not to upgrade from the 2022 version to the 2023 version.

Comparing your video to what normally happens - first you see "Loading fonts", which you're seeing, then rapidly after that, "Loading data", which you're not seeing during this long wait. So something is blocking during the launch process. You resource usage looks very low. It's not nailing the CPU, GPU or SSD, so I doubt the problem is your hardware This looks suspiciously like a blocking system call or the program waiting on something and timing out. A blocking system call is a low level operating system function, such as one to read from a network socket, that blocks and waits if there's no data available. Generally programmers will try to avoid blocking calls when a user interface is displayed, since it can lead to an unresponsive program. However, Serif might be using one during the application launch process, as the UI is not displayed yet (the code for a blocking system call is a easier to implement than an asynchronous variant - forgive me if you know this already and I'm teaching granny how to suck eggs) It could also be something like a DNS timeout or it's getting a file read error on an operating system resource it needs to load during startup, so retrying multiple times and eventually giving up (it wouldn't hurt to do a "DISM.exe /Online /Cleanup-image /Restorehealth" followed by a "sfc /scannow") Hopefully Affinity has the ability to create a debug log. This will probably point to where the problem is. @Patrick Connordoes Affinity have the ability to create a debug log to assist in troubleshooting this problem?

I tried this out myself, since this is a technique I use often to refine the edges of a luminosity mask when doing a sky replacement. It works the same for me as it does in V1. That is to say, I can paint in white or black on a mask in overlay mode. Using white will lighten any grey but not touch pure blacks and using black will darken any grey but not touch pure white Video to show what I'm seeing clip0052.mp4

Yeah, it looks like it. There's clearly no "ultérieure" version as version 2.0 is the latest

This is a right pain in the ass that bites me regularly. I activate all privacy options possible in browsers and only use Firefox or Brave. I then have a raft of addons to block trackers, randomize the user agent, the canvas fingerprint, the audio fingerpint and pretty much everything that modern tracking software uses to identify a computer, as well as automatically deleting every cookie and local storage set by a site as soon as I close the tab. So to these websites, every time I hit their website, it looks like a new user to them. I had one of the data companies who I subscribe to for work accuse me of sharing my password with 76 different people because I do this and their dumb software doesn't even bother to check the IP address to see that all these 76 "different logins" come from the same IP !! I think they felt pretty stupid when I pointed this out to them.

I've noticed an odd bug in the user interface. Open any file in Photo in one window as long as it's not a RAW file. Leave the persona in Pixel Persona Open a RAW file in another window and switch to Develop Persona leave the Develop/Cancel buttons up, that is to say, don't develop the photo. Now, switch back and forth between the two windows by clicking on the window title bar. Every time you complete the switch from one image to another, the image is offset more to the right or to the left. Eventually, both images will virtually disappear off the screen, one to the left and the other to the right

.

Oh, interesting result too from my perspective. I'm using an older Nvidia gaming driver and not the studio driver (and a newer generation of the hardware). I'll have to update the driver and see if that makes any difference. I did a clean boot and ran the benchmarks again to see if the behaviour is consistent, which it is. I can reproduce this result every time. EDIT: after updating to the 526.98 Studio driver instead of the older game driver, the behaviour is identical. I also updated my AMD chipset drivers to the ones released on the 21st of this month. So it's not the graphics driver or the chipset drivers causing the performance hit.

I've just been playing around with Darktable. It's a very impressive piece of software for an open source package, with an enormous amount of settings one can tweak. I particularly like some of the adjustments (for example the contrast equalizer, tone equalizer and the colour balance RGB which allows one to adjust vibrance/chroma/saturation according to tonal range). I was able to get a great develop on a very difficult RAW quicker than with some other software due to the fine grained control it gives of many settings. It's not the most intuitive UI and it does a couple of odd things here and there (for example, I had red artefacts on some clouds on a brilliant blue summers day which I had to eliminate by going into the input profile setting and telling it to clip the output gamut. Wasn't obvious to work that one out (what was causing it that is). It doesn't do a terrible job with the initial settings it applies to a raw file either, although some of them tend towards too much of a "processed" look

Yes, Samsung Magician STILL doesn't recognize it. These higher performance nvme SSDs run very hot. I have an aftermarket Sabrent heatsink on mine as well as a Noctua 40mm PWM fan mounted on a very long standoff screwed into the M2 mounting hole for the SSD so it blows directly down onto the heatsink. Mine runs at 34 celsius during a normal load. Before the heatsink / fan combination it idled at 50 celsius and temperature spiked up under load. I probably would not buy a 980 pro again, despite having had many Samsung SSDs. The first one almost trashed all my data - it had hundreds of block read errors, so I RMA'd it. I keep wondering if the excessive heat was the cause of the drive going fubar. Inconveniently, it was my system disk, so I had to buy another SSD before I could RMA it. So now I have two 980 pros for my sins. The second one seems fine after using it for a few months, but these 980 pros use TLC flash, which has slower write speeds than SLC flash. After the write ahead buffer is filled, write speeds slow down considerably. I wasn't suggesting this particular model of SSD, just suggesting a modern PCIe 4.0 nvme SSD, as they're a lot faster than what he has now - there are quite a few with similar performance specs. I just quoted the 980 pro since that's what I ran my benchmark on to check what speed it does.