Question How to improve appearance of a moving system? (Spawning behavior)

I have a beam effect that cuts across the environment:



Each frame, it potentially moves, striking a different part of the environment. This moves the Visual Effect which generates the fire/smoke effect you see here.

However, any time the Visual Effect moves even reasonably fast, it results in the fairly ugly effect you see here, where the system spawns new particles no faster than whatever the framerate is. This results in gaps between each particle, which looks bad.

I can somewhat improve this by increasing the spawn rate, but that still produces gaps. I can also increase the particle size to cause more overlap, but eventually the particles are too large to be reasonable.

I was just wondering if there were any other common approaches to produces "smooth" emission over a distance for things like this? I have the same issue on a missile, where it leaves a smoke trail behind. It ends up leaving discrete "puffs" of smoke behind it, rather than a reasonable looking trail of smoke:

 

I am also interested in this, but I want to add a bit more to this question as it is also visible in trails/ribbons.
For example there is very common use case where character has sword, it generates swing trails during movement and the same problem can be observed - there are no nice, round ribbons, but instead square and broken lines.
I can't even count how many games I have seen like this, is there solution to this? (except manually sampling and interpolation everything)

 

Is this different and in addition to sub-frame interpolation? Or just an application of it?

I'd also like to reiterate that I think the voting idea is flawed. It's a measure of (a) how many people find the roadmap page (b) realise that you can vote (c) bother to vote.

None of this correlates with the true utility of a feature.

I'm sick of voting and have given up because it makes me enter my details afresh every time and there's no indication of what I've already voted on.

I also only ever used to vote when I was in a bad mood so most of my comments were along the lines of "do you really need to ask why this feature is important!?"

 

This.

 

That looks pretty interesting. Thanks, Vlad. I'll try it out.

 

I think the flaws with the feedback page are minor and fixable:

1. Make it clear what I've already voted on
2. Don't send the damn verification email every time
3. Don't mush together voting and more detailed feedback. A vote is a simple vote (Uservoice used to do this fairly well. Additional feedback should be optional and in addition to votes.
4. Get the messaging clearer around the kind of feedback you want. At the moment it sounds a bit like "If you want this feature, here's a text box where you have to convince us it's really worthwhile or else we won't take your submission seriously" while from what you wrote above it's more like "if people explain their use case to us it helps us figure out a way we can cover the largest number of different use-cases as possible".

Just review the copywriting on this page and give it careful attention from people experienced in writing this kind of copy.

 

Sorry for bringing roadmap feedback here, but as the problem was already mentioned I will add my two cents:

I strongly agree about first 3 points above and voting must be more clear, but I personally think the worst part is that I still do not know how should I treat the roadmap (the key factor to bother). Is this just thing for developers and it's used as "window" for unity users?
Do we actively participate in feature selection? Does this board even matter? Simply speaking, is this important board and feedback is considered as very critical or just bonus page to let people whine about missing features?

If roadmap page is important, then it's a bit useless, becuase most likely 80% of people has no idea it exists or can vote (I think before it was possible to see number of votes and it was somewhere between 100-200 for most voted ones) and even if they know, they probably think this vote is not important or does not matter, becuase as @andybak said, page is not user friendly and makes an wrong impression of "you need this feature? pff, you can try to convince us, well whatever".
Providing explanation and email every time is also barrier stopping me from voting, I lost interest at least 3 times because of this, for example when I need to explain why I need point light shadows in URP I simply lose my will and I going to do something else

Otherwise, if roadmap page is not important, then why should we even bother?

What I am trying to say is you need to present the page in the way it should be - if this is the best place to put feedback, then advertise it everywhere this way, if votes are important then try to get a lot of them and throw roadmap into face of users and say "tell us here if you are missing features!".
Additional note/example - there is card "Submit a new idea" and I think I did it once, but there was no feedback (I don't know if it was already known or card was actually made or no one just bothers to read it or maybe it didn't work) in result I lose even more interest and trust to this page.

 

Hah, yeah I was really annoyed that time, because of missing feature. Anyway thank you for going in depth about roadmap, I think I understand now the role and reasoning behind all of this and definitely I will try to send more votes there. Thank you for your time!

 

On sub-frame interpolation and the overall choppiness of particle/particle strip movement:

An essential part of any future solution - whether it is sub frame interpolation or otherwise - would be to provide a UV mode that is set to tile in world space. I have seen several attempts to create this using the Custom UV option and creating a custom attribute, TexCoord, but none have achieved consistent or reliable behavior.

Another UV mode that would be appreciated would be Tile Per N Segments. Currently the Tile Per Segment option has very few use cases.

I see this as related because I often want my UVs for particle strip quads to correspond with smooth-spawning. UVs get distorted very easily when using particle strip quads with the discrete spawning