Ben

Yes - I think that expectation is the problem for many users venting frustration.... and the reality is that the chances of us creating your perfect app is pretty remote (if that is your expectation). Be realistic - AI is a costly app, has been around for decades, had quite a large development team and still a large portion of the more useful functionality that people expect is provided by third party additions. We are already putting a lot of such functionality directly into Affinity, and trying to do new things. At some point we have to make a choice - time is a limited commodity. We are also aware that just throwing more people at the problem isn't the solution. The reason we are a slim team is that we are a capable team, and we can trust each other. If we doubled the team size over night just to try speed through loads of features - the result might not be all you hope for.

As I stated in another thread- I already added true circles when I did the construction geometry feature for the Node tool, but this prototype feature is currently revoked. This involved all the maths for circles and ellipses - plus the ability to define a circle by two or three points, and the rather complex maths for performing snapping through arcs. Our current circles are, as stated, only a Bezier quadrant approximation. As such they are considerably inaccurate as far as technical style drawing and snapping goes. True circles will be represented correctly in terms of the maths. The rendering of them will be handled appropriately such that errors will be imperceivable. Export, of course, will require certain choices, depending upon the output format - whether to export as a circle, ellipse or Bezier approximation. There will be improvements to circles soon. I hope it will be some time in the 1.7 cycle, but I can make no promises. It will require new ways of handling them that is a departure from our current box-based approach. This means a lot of tooling changes to accommodate this.

How about we keep this thread civil, please..?

It'l be fixed in the next Beta anyway

That is precisely NOT what I was saying. That's what "balance" means. We realise that both old and new customers want the same things. So, features that existing customers want will also appeal to new customers. You need to understand that it is those new customers that are essentially funding the continued development that you expect. While filling out the road map, we also need to keep a firm eye on the changing world, and look to add features that will set us apart. If we stop innovating, we will stop attracting new customers, and the development of the features you want will also stop.

Obviously, we have to consider new customers. We don't have a subscription model, so if we don't get new sales - existing customers don't get updates either. There is far too much overlooking of "basic" features that we have added. It's so easy to spot the few things we haven't yet done, and ignore the many things we have. We have refined many tools in 1.7. There are a lot fo internal changes - many that average users will miss, but professionals and experienced Affinity users will spot. There is a balance that we need to strike. Refining existing tools, adding people's wish list for replicating tools from other apps, and most importantly innovating and moving software forward. As @JET_Affinity says - what's the point in us just cloning your other favourite software?? You probably already own/use that anyway. And, as I've said many times in the past, if all you want form Affinity is a cheap clone version of Photoshop or Illustrator, you are going to be disappointed.

Surely not if he is reference Gimp as the example. I'd imagine he's getting a raster selection from a shape mask and shrinking/growing that. Of course - @PAULRFONTENOT needs to read up on Affinity and learn the difference between vector and raster.

ok. I think this thread has gone far enough. @Jowday - as has been pointed out, the OP is not particularly constructive and suggests that no work has been done, at all, in the past years. That is absolutely not the case, and the evidence is clearly documented on this forum. Numerous items from our road map have been completed since our initial release - these have also been balanced with ever changing industry expectations. The OP alludes to basic features being missing, but does not state what these basic features are. Without such information, it is nothing more than a rant, and gives us no insight into what this particular user expects. That is helpful to no one other than the person letting off steam. I'd also add that what many people view as "basic features" are often those that didn't appear in other software for many, many years, or are actually quite niche, or very often are not actually that "basic" - certainly not in terms of their implementation. It takes considerable time to implement any tool from point of view of performance, precision and usability. Our decision not to put out half-baked solutions is one we stand by. There is plenty of evidence in other software where features have been rushed out and bad choices have become entrenched. It's easy to make blanket statements about how we are not meeting your expectations when you don't have the complete overview of what we currently have and where it has to go in the future. We also have many customers who are very happy with the direction we have chosen. As has been stated - you are ready to compare our performance to other development teams that are ten or even fifty times the size, and with software that costs more than ten times as much. While the comparison is flattering, I'd ask you to consider how your expectations of parity with those you are comparing us to can possibly be realistic.

As has been said - there are some improvements to snapping and the new Point Transform Tool. But - I am very aware that we need to do something more with circles. We did have a prototype feature that used perfect circles (not Bezier approximations). This was put on hold util we can refine the UI/UX. Unfortunately it probably won't be in 1.7.0 due to time constraints. It might make it in soon. In a nutshell, we need the concept of true geometry - circles and ellipses. Else, we are still dealing just in Beziers with all their considerable inaccuracies. This is actually more of a technical drawing feature, and not so common in illustration. Te key thing is precision - of geometry and affecting snapping - and that will include the potential for tangents and normals to circles and ellipses (a feature that was in my prototype). It has not been forgotten. As for aligning objects to a curve - I can't make any comment.

While smart icon generator seems "smart" all it appears to do is extrude a 2D image through an isometric plane, and replicate the edge pixels on the other faces. It's an illusion, more than real isometric.

I think @PAULRFONTENOT is still confusing raster and vector functionality. Affinity is not Gimp. I think you need to read through the help and tutorials to familiarise yourself with the app a bit better. Otherwise, we are just going to be telling you information that is already available.

I've got to admit - I'm not exactly sure what you are asking for? Do you mean just he ability to resize objects???

Seriously??? My thumbs are so tired from just being "twiddled".

I can confirm that the file is unrecoverable. There is too much data missing. Could you give me a bit of information about your setup? OS, where you are saving your files, etc. Thanks.

Also - while 86.603 sounds like a good approximation - that's actually quite inaccurate if you start to rely on our snapping. So - if I were to try to align a set of objects relying on that scaling, but applying various rotations to fit to the three isometric planes - you'd find that there would be a significant error, and that error would grow as you snap one object to another. Our snapping system has a certain tolerance when determining equality in snapping lines. The best solution is to use our cube system to maintain the highest precision. If you were to do the SSR transformation yourself, you are better off using "*sin(60)" in our transform panel to scale the object. That will produce a scaling that is as accurate as possible (for computer maths).