Replying to:Ralph (Developer)theres always a new update my todo list is larger than wikip
Ngl for few months I have no idea what’s planned next and wanted to ask if you’re gonna do some teaser since it’s been quite a while… Then turned out you just fixed 3.1.x like week ago or two
Replying to:arcionekNgl for few months I have no idea what's planned next and w
3.1.9 was actually gonna be 1 last quick fix (mainly to correct all the text alignment, as LibGDX corrected their Label issue)
but while I was waiting for that ^ I accidentally dabbled with improving lag during tweening calculations (and successfully did so) which got me thinking about how to make even *less* lag with movieclips (Graphics aka Single-Frame MCs only) by allowing them to be cached (similar to Flash’s cache-as-bitmap checkbox)
so now i’ve been doing that and it’s a massive headache
but if used right MCs that are cached (good for backgrounds or a car or a gun with 1000 nodes for example) will be VERY quick to render, as they’re images and not vector while cached
Replying to:Ralph (Developer)3.1.9 was actually gonna be 1 last quick fix (mainly to corr
Is there any way to compact the node calculations so that you can increase the node limit or simplify so that it mandates less CPU and RAM power? And also how much ram and cpu does SN use up when you have a full frame (of 6400 nodes) or one fig of 384 nodes?
Replying to:Ralph (Developer)i'm always looking for ways to optimize, so mayhaps one day
well GPU’s are pretty good at calculating many vectors since they have obscene amounts of cores each preforming very minor calculation. problem is most phones and tablets dont have a proper GPU more than integrated graphics so we’re left with mostly CPU’s doing all of the work and as we know its a big pain in the ass
2020-08-16 15:08:16 UTC 
![[Abandoned]](https://sticknodes.com/wp-content/themes/dw-minion/assets/img/wpsn/default-avatars/03.png "[Abandoned]"){kind=avatar}
Wait there will be a new update?
theres always a new update my todo list is larger than wikipedia
Ngl for few months I have no idea what’s planned next and wanted to ask if you’re gonna do some teaser since it’s been quite a while… Then turned out you just fixed 3.1.x like week ago or two
3.1.9 was actually gonna be 1 last quick fix (mainly to correct all the text alignment, as LibGDX corrected their Label issue)
but while I was waiting for that ^ I accidentally dabbled with improving lag during tweening calculations (and successfully did so) which got me thinking about how to make even *less* lag with movieclips (Graphics aka Single-Frame MCs only) by allowing them to be cached (similar to Flash’s cache-as-bitmap checkbox)
so now i’ve been doing that and it’s a massive headache
but if used right MCs that are cached (good for backgrounds or a car or a gun with 1000 nodes for example) will be VERY quick to render, as they’re images and not vector while cached
enjoy that wall
should be done this week
Now you got me jumping in my chair..
stop youre going to break the chair
Is there any way to compact the node calculations so that you can increase the node limit or simplify so that it mandates less CPU and RAM power? And also how much ram and cpu does SN use up when you have a full frame (of 6400 nodes) or one fig of 384 nodes?
i’m always looking for ways to optimize, so mayhaps one day i’ll go big brain mode but…i mean, even Flash lags and it’s, well, Flash
vector-based art is CPU intense
well GPU’s are pretty good at calculating many vectors since they have obscene amounts of cores each preforming very minor calculation. problem is most phones and tablets dont have a proper GPU more than integrated graphics so we’re left with mostly CPU’s doing all of the work and as we know its a big pain in the ass
okay so wouldn’t that basically be a gateway towards having images as sticks too tho.
yes, actually
…
but stop