line-weight

Over what kind of timescale would that degradation apply? Or in other words how likely is that it would be having a significant effect on my Spyder5? Any "colour calibration for dummies" tips - ie devices / software that will give a 'good enough' result for those of us who don't have highly technical requirements? My aim is really just to know that the colours I see on my screen have a reasonable chance of being as similar as possible to what others will see on theirs.

I'm not sure how true it is, but I seem to remember reading that older calibration devices degrade over time and therefore become unreliable.

When this happens to me and I break into a cold sweat... it usually turns out that I have done something like inadvertently saving the drawing into the wrong project folder. So, when I go and look for the last-saved version, and/or the backups, in the location they should be, they appear to be several days out of date. Do some kind of search for "recent files" (or "open recent" in VW) and you might find that your drawing and all its backups are fine, they are just located in a different place from where you thought they were. Another thing that can happen... I have a crash and then open a backup file, get distracted and start working on it without re-saving it as the main file. Then think it's disappeared but in fact it (and its backups) is nested within the backups folder for the original file.

haha, I'm glad it's not just me, then!

It's ridiculous that to edit a roof object, you have to choose a command "edit group". How is a user supposed to guess that? Not only is a roof object not called a group, but it doesn't behave anything like a group.

One small issue I have run into with my multiple display setup on an M1 mini: the display that I have connected via a Displaylink adapter works fine but it looks like I can't do colour calibration on it. I have a datacolor spyder5 thing that I use and it can do the other two monitors but not the displaylink-connected one. Not really a big deal for me... I never feel very confident I know what I'm doing with colour calibration anyway.

Should work fine, and this video might help you

Same here. You can have it run along a multi-segment polyline on the flat, or you can have it run along a single segment line with a change in elevation. Why not both!

Here's the announcement from VW in March. I have highlighted certain parts in bold. This was released with stuff that doesn't work, and 4 months later, it still doesn't work.

I think that what you mean is when the inclines are non-parallel in plan - an incline can always be expressed as a fall in one direction - but sometimes non-parallel inclines will connect fine and sometimes they won't, in my experience.

It's a right pain, because drawing a multilayer roof buildup in the kind of configuration the OP wants is really tedious to do manually, and the roof face tool, if it actually worked, would make it loads quicker, and loads easier to edit subsequently. But in industry-leading, BIM-ready Vectorworks, a functional roof face tool is just a bit too much to ask.

I have not tried for very long, but I could not get them to join properly. My experience with roof faces is that they have quite a lot of bugs and sometimes connections work properly, sometimes they don't. I mostly only use them for quite simple roofs. For something like the dormer you are building, it's likely that I'd end up building it manually from solids.

Roof face needs to look something like this in plan

The problem I describe is covered in some detail here (need to read beyond the first few posts):

Unfortunately, no such magic happens when viewed in OpenGL or rendered in renderworks, which can cause all sorts of problems with gradual curves.

....maybe you have to set your units to microns and scale all your drawings up x1000?

I've seen this given as a reason before - and can understand that it might be why we have to wait for things like reveals in multi layer walls to behave properly - but it's no excuse for failing, for years on end, to fix basic dysfunctions like sills being drawn the wrong way round. It's also not really a plausible excuse for essentially making no improvements *at all* - again for years on end.

The window & door (and most other tools) certainly don't seem very localised for the UK market to me - most of the time I have the impression I am using a US-focused product. Especially with windows and doors I'm often guessing what settings actually do because they use building terminology much of which is meaningless to anyone used to UK terminology. This in itself is annoying and confusing.

Yes - Windoor is for Australia or NZ I think? But I think there is also something else again for the German or Austrian market? There may be others.

Maybe. But then you might expect that it would be in the purview of Vectorworks to provide something at least as good as what can demonstrably be created by a maker with much more limited resources than they themselves have.

It's clear that there are plenty of people out there (including existing VW employees) who could build massively improved tools and UI for Vectorworks. The issue is that whoever the people are, who ultimately are in charge of strategic decisions at VW, either just don't get it, or have made a commercial decision that they simply don't need to offer us these improvements in any great hurry whatsoever. It's bizarre because (for example) there are already improved window and door tools out there, that are available only to certain regional distributors. We are never given an explanation as to why they can't be offered to all Vectorworks users - not even as a paid extra. I think there are many of us who would probably pay extra for them, even though they already ought to be part of the basic functionality of the programme, just for the time and pain saved. I honestly don't understand it. In the past I've wondered if we could somehow crowdfund some improvements rather than waiting for ever for VW to sort things out themselves.

My experience is that there are a bunch of problems with the datasmith export to TM and it's unusable for me at the moment. More on this thread

I had a problem with a Logitech mouse which wasn't recognised by "Logi Options" which you have to use instead of Logitech Control centre if you're on an M1 mac (or any Big sur mac?). The logitech website said it was compatible but it isn't. After about 2 painful months of back and forth with their customer services I got a mouse that does work. That was the 3rd mouse they sent me.

Thanks for the update. I'll see what happens when SP4 is out but it sounds like unfortunately it'll remain unusable for me for now.

Unfortunately, a month down the line I've heard no news of any progress on the problems I describe upthread.