haraldthi

I need to grade (develop) several photos to the same standard, but run into a few problems. The right way to do this for me, would mean being able to overwrite the profile I have laid on all the pictures, and have all the pictures loaded with said profile update automatically. Instead, Affinity Photo rejects saving a profile with a previously used name, so when updating the profile I need to give it a new name, and then change all the photos to use that new profile. Which is both tedious and messy. I understand the need to protect users agains sudden and unwanted changes, but it goes too far. It slows down a whole class of use cases. I also see it could be kind of slow to update a large number of pictures at once, but the advanced thing to do here would be a "lazy" approach, where the changes wouldn't be applied to the pictures before needed. As in viewed, saved, exported and so on. (But for me, either way is an improvement.) I also tried the snapshots in the develop persona, and while I see that they have a limited use they are impossible to rename. Which makes it difficult to remember where I was when I took that snapshot. Fix, please?

Yep. It may be rather nerdy, but then again, these people are okay with that. :-)

Ah. What do they stand for, politically? It's a rather radical movement in free software. It started with competing versions of the UNIX operating system releasing different versions of their tools, and everyone was pretty confused because what the software houses thought of as their unique strengths in the market developed into every system it's own ways of doing things. So people with time on their hands (hobbyists, students and academics) made their own tools and released it so everyone could use it freely, with the one exception that if you changed the tools or built further on them, you had to release that too freely and under the same terms. Linux is an example of that. Linus Torvalds, at the time a finnish student in informatics, wasn't pleased with the current crop of operating system kernels and their limitations on how he could change them at will, so he made his own one under a GNU license. It was pretty small and limited at the start, but it caught on and spread via bulletin board systems (the internet of the time). It was a missing piece for a fairly complete set of tools under the GNU license so it fit in and made the package whole. They finally had a whole, freely available system they could tinker with and change as they wanted. Linus is more pragmatic than many in the GNU movement, which is considered more religious, but they have a working relationship. Even if the GNU movement has released its own kernel, HURD, over time. So it really started out as a system for tinkerers, but grew and now works fairly well. For consumer oriented programs, the freely available tools work as a lowest common denominator, but when it comes to raw computing they're pretty much at the forefront. I very much like that there's freely available tools I can go to when I need to do something I don't do much, but when it comes to the tools you work with and live by, paying a little more may very well be worth it. Your paying for software helps people develop their software further. So as far as it doesn't go into exploitation and monopoly practices, you get what you pay for.

It's a clear enough answer to me. They need to know that enough people will buy it. Based on statistics. Serif will do what they consider best for earning more money. And be able to hire more staff, so that they can grow even faster. If that means releasing an ipad app in a market they deem ripe for it, they're betting that it will be a strength. Just the same way as if they were developing a desktop app. If they considered the ipad app a loss, they wouldn't do it. With many of the libraries being the same it's long term gain, even if it may have hurt the performance short term. What they're effectively saying is, you can't harass them into developing for a platform like Linux. They're thinking about it rationally, so you need rational arguments. And the most pressing argument is statistics, because that will tell them if they'll have the needed economic gain or not.

Hi. I'm new to Affinity Photo and has been testing it out for a few days now. One of the areas that interest me is the HDR merge, as it allows me to do picture with a colour quality and contrast range I haven't been able to before. I've been testing it as I usually do, that is to give it a nearly impossible task and see where it breaks. And sure enough, it has its limitations. Yesterday I went with my wife to the beach, and with semi-cloudy weather and a low evening sun reflecting in the water I saw it as a perfect opportunity to test HDR. With my wife whittling about with her own creative ideas, I tried to capture her and the surrounding sea as best as I can. Or the low flying swallows against he sun. The problem being of course that a Sony a6000 rattling away the best it could, couldn't possibly capture everything that moved about without causing Affinity Photo to get royally confused. In this environment, ghosting becomes a serious issue. Even when you let Affinity do aligning and ghost removal, and even when you turn on the "remove shadows" function. You need to turn on the clone brush tool, as advised in your fine tutorials, and go at it. The problem is you need to choose the right picture as your base and when all the movement in the picture makes Affinity Photo (hereby called AP) confused about what moves and what is still, the pictures aren't in sync. That's what makes the ghosting issue first hand, and that's what making it difficult to paint properly. Half the time you can't paint with one picture close to another, because the sync problem goes the wrong way and every time you get close to the edge, you get the edge before you really need it. (And the other side of the edge is too dark (or bright) to be painted with that picture.) I guess the proper solution would be more image recognition and morphing the images making up a merge to fit. But that would be difficult to do, computing wise. And I guess you would need the more manual approach anyway: To be able to sync up the pictures so they match. And being able to use masks to define what sync to make for that part of the picture. Not only would that make it possible to make well synced pictures in difficult conditions, even if it would mean a lot of manual work. It could serve as a base for a more automatic approach for morphing and syncing pictures. The ideal would of course be to have automatic routines define a sync you could later manually adjust, and this would be a roadmap towards that. I would also like to see menus such as "Open new HDR merge" use the settings last used instead of going back to defaults all the time. I want to at least try to remove ghosts automatically and see how it goes, but going directly into tone mapping is a complete goner for pictures like these. It needs to be clone painted first. And that's two clicks per picture that could be saved. More than that, I often experience crashes. Not being able to save in tone mapping mode is a risk I don't want to live with, but at least being able to save before going into tone mapping has saved me more than once. I just wish AP could remember the photos going into the merge when saving, as you sometimes want to do further editing with the clone brush. In other cases, pictures can cause AP to crash always, so you don't get much of a chance. Another thing that could be fixed is, I believe, with Windows only. AP does a lot of file reading, and with a proper UNIX system at the bottom, the system caches files in memory so that they don't have to be read from disk so often. Windows doesn't do this, or at least it does so very poorly. The result is, the processors constantly has to wait for disk reads. And with slow, old-fashioned disks such as mine, this cuts performance to half. If you, for the Windows version, could cache the files being read to memory, or at least not read them from disk that often, it would improve performance. For long running jobs, it would also be a help to do things in the background. That would mean you could work on another picture while the first were chugging along. Not something of major importance, but there it's said. As others have noted, it would also be nice to combine bracketing features such as HDR panoramas (often used as backgrounds for 3D rendered images), focus bracketed HDR pictures of landscapes, and of course focus bracketed HDR panoramas. (Anything more?)

I can add to that: Basic drag and drop functionality (in windows, at least). I shoot in ARW, a raw format the Windows file browser can't handle (along with many others) so choosing what picture to merge is a walk in the dark. I have to look at the file numbers, and getting it right requires double checking. I use the IrfanView thumbnail viewer to browse pictures, and being able to drag pictures from there onto the merge window (so I can see what I do) would be a great plus.

Only statistics will say if there's a chance or not. Serif does well in considering what will give them the most out of their efforts, but Linux is both maturing and growing as a market, and that makes it more a question of time than an "if". *When* is the market large enough and mature enough to be worthwile? If there's a platform that's going to lose out today, I think it'll actually be Microsoft, because they have almost entirely lost the enthusiast market (except for gamers) and etnhusiasm is what drives a market forrward, long term. Not a "I don't care, I just need something that works". Ask IBM what happened to their dominant market position in the 80s, and you'll find out how that works. Microsoft and the PC market almost completely undermined them, because they were seen as a behemoth unable to move or do anything of interest. Today, Microsoft is that behemoth, and their abandonment of proper ethics gives them a rumor they can't survive very long. So, I'll wait and see how the market grows. And the first programs that works for me on Linux, I'll buy and use.

Seriously, I'm just adding a vote. If Affinity Photo was released on Linux, of the Debian/Ubuntu variety, I would not be slow to buy it. Why? Macs are good for many things, but their effort towards simplifying things make them go seriously wrong sometimes. They're also expensive, in a bang per buck perspective. Which means I'm sometimes stuck with windows, which fails so bad in my perspective that I won't even mention it. I mainly work in sound and video, and that sometimes also means photo editing, and the free alternatives on Linux are not enough for professional use as far as I see it. Linux is developing towards a meaningful media platform, but lacks in areas like photo editing. So I hope to one day not have to boot back and forth. Enough said?