Ben

Thanks for this. We did a lot of the compression work early on, but it has been written to allow it to be extended to new compression methods. Looking at the figures, clearly it'll be worth us looking into this again. Either ZStandard or LZFSE might be our next move. The timings for LZ4 look promising if the sacrifice of compression ratio is not too much. Of course, timings and ratios are subjective. I did all my testing using our own data format based on real life raster sources (or photos as normal people call them). The numbers might look different with the sort of more random data patterns you get from images. We already do a certain amount of work separating channels, tiling, and scanline prediction before compression, so it'll be interesting to see what further gains we could make using an alternative compression library.

We currently don't offer any way to change shape types, other than going from a Circle to a pie/donut (as those shapes are easily relatable). Other shapes, not so. Some shapes are projected into the original bounds, and others are based on a circle projected into the original bounds. Changing between the two shape types will not necessarily result in a shape with the same geometry extents. So, it's more predictable to drag out a new shape.

The Windows guys are looking at where the higher res icon is/has gone. Looks nice on Retina Mac though.

The tooltip was missing for some reason. Any warning on the export options tells you that the selected items have different values for this setting. i.e., one is PNG, another is JPEG, for example. Changing the value will apply the change to all items in the selection, so it is warning that their are currently differences.

@Oval Some people were making the comparison, but your point was that the 'original' Photoshop was written by two people. We developed the first version of Designer and Photo with 5/6 people. Cast your mind back - was that original Photoshop 2/5 as good as the first release of Affinity Photo? As I recall Photoshop didn't get particularly good for 3 or 4 major versions. Another issue is anyone who thinks that just throwing more people at the problem will turn faster improvements. That is a bad assumption. The current team has an in-depth knowledge of our code base, and our development strategy. Adding new people doesn't instantly equate to faster development. In fact, in some cases, adding new people can have a detrimental effect as they take time to get up to speed, and have a greater chance of introducing bugs.

Do you want to elaborate on this? Is it the font size?

The drag-clone is going to be overhauled in 1.7. The new approach will allow you to start a drag-clone with Alt, then release Alt to allow snapping - the clone will remain. You will also be able to use Cmd - the new method will allow you to toggle the Cmd key at any time to add or remove the clone while dragging. So you could start a drag, then make it a clone later by pressing Cmd, or remove the clone by releasing Cmd. This will work independently of Alt to affect snapping. So, the better approach will be to use Cmd instead of Alt to control cloning while dragging. Really the Alt method was only there for people who couldn't reach that little further to use the Cmd key (or those who could only cope with the "Mac way" of using Alt to make a copy). Something I've never been too happy with, but we've accommodated users that insisted it had to be Alt to the detriment of our standard convention that Alt only ever affects snapping.

Hi. Currently, when cloning with either Alt or Cmd you will get a cloned object done through a Duplicate command at the start of the drag. As you drag, Alt will then be cancelling snapping, so you can release Alt to allow snapping again. The clone will still be there, as that command has already been done. If you start the clone drag with Cmd, it won't affect snapping. As far as snapping goes, whether you clone or not, snapping to pixels should work exactly the same as if you just dragged an object. If there are other candidate objects on the page, it's possible that you have snapped to then (as a result of bounding box mid points, for example) that could result in half pixel positions. Turn off all other snapping, and just leave snap to pixel on, to see if this is the case. The only way it should be possible to end up with half pixel values is if there is other snapping occurring. If you have an example document that shows snapping going wrong, I can look into it further.

Sounds like you really need to use Compounds to create a boolean of your vector shapes. Rasterising is not really the "Designer" way, because you'll lose edibility and precision.

This file can be rescued if you open it using the latest Beta version. It should be able to recover an older revision of the file. You should then do a Save-As and save it to a new filename. You will have lost some of your more recent changes, but some of the work will have been recovered. Next, we need to try and work out what caused your file to go wrong. The file appears to be the correct length, but the end of the file was filled with 0's, which I've only seen happen with NAS storage. Failed USB storage usually results in a short file (which isn't the case here). We've yet to identify issues with locally attached (SATA) storage, and these reports are quite rare. Judging by what you've said, Affinity didn't report a failure to write the file, so no error code has been generated that we would have been able to pick up on. That said, if we can ever reproduce this we'll have some clues about how we can mitigate for this failure. Can you give me a description of your setup, and a little information about what you did when you saved the file and closed Affinity. Thanks.

CropTool has been rewritten, and will include inbuilt and user presets. Will beta in 1.7.

Yeah - it was spam. I redacted the link when I made my post, but Miguel was right to remove the post.

Not really, that link is talking about (natural) noise in sound. The white/pink/brown is descriptive rather a reference to actual colour/chroma.

This has been done, but didn't make it into the current round of Betas. It will Beta after the next major release has been made.

Not entirely - it's because the App Store gives us the sort of global exposure that's hard to achieve any other way. Plus, it's a very well run store.