|
Socoder -> Language Tests -> JSE - Optimisationalism 5 |
|
| Tue, 22 Apr 2025, 02:04 | |
|
Dan |
Same PC as my previous post.
|
| Tue, 22 Apr 2025, 03:03 | |
Jayenkai
|
Oooh, that's some good squeezin'!!! Hopefully I've not completely knackered anything along the way, doing all this. -=-=- ''Load, Next List!''
|
| Fri, 25 Apr 2025, 07:59 | |
|
Jayenkai
|
OK, v.3gio is now uploaded. You might find it to either be faster or slower, depending on your browser. In Safari and Firefox it's a fair amount quicker, but in Edge it's slower than the original, which is NOT a good direction to be going in. Hmmm.. Might have to spend a bit of time in that-there Windows for a bit, and see what nips and tucks I can get out of it. -=-=- ''Load, Next List!''
|
| Fri, 25 Apr 2025, 19:54 | |
|
Dan |
|
| Fri, 25 Apr 2025, 22:09 | |
|
Jayenkai
|
Yeah, that's more or less what I'm seeing. Though the work has been great on Mac, it's been pants elsewhere. A fascinating difference in JavaScript inner workings. Ok, more to do! Keep optimising, Jay!! -=-=- ''Load, Next List!''
|
| Sun, 27 Apr 2025, 08:20 | |
|
Jayenkai
|
v.3gip of JSE has now been uploaded. After a few days more work, and a fairly significant amount of hefty rewriting of the fundamentals of the language, things are.. .. Kinda the same'ish. It's not really been the outcome I was hoping for, if I'm honest. Let me know how you fare. : Download | Suno Link One thing I've been frustrated over, in recent tests, is that if you open a different project, the engine seems to slow to a crawl. Hitting reload in the browser will fix this, but.. .. Damned if I can figure out the reason for it. .. *angry man shouts at clouds* -=-=- ''Load, Next List!''
|
| Sun, 27 Apr 2025, 12:08 | |
spinal
|
-=-=- Check out my excellent homepage!
|
| Sun, 27 Apr 2025, 13:01 | |
|
Jayenkai
|
Yep.. Waste of .. approximately 2 days super-tough work. Hurray... -=-=- ''Load, Next List!''
|
| Mon, 28 Apr 2025, 15:31 | |
|
Jayenkai
|
v.3giq, and another hefty jump in speed, with hopefully no odd drawbacks. I've been using a few programs to keep an eye on whether things are still functioning as they oughta be, but if you've got any coding you think shows any flaws in my recent tweaks, do please let me know. Over on the Quest 3 headset, the current version is running about 1/4 or so faster than it did a couple of weeks ago, dropping from 364 down to 275, which is a fairly hefty drop, all things considered. Like I said, I've tried keeping my eye on things, ensuring nothing I do breaks anything, but if you think I have, please do let me know ASAP. -=-=- ''Load, Next List!''
|
| Mon, 28 Apr 2025, 16:17 | |
|
Dan |
The latest test est st t
|
| Tue, 29 Apr 2025, 01:27 | |
|
spinal
|
Nice!
-=-=- Check out my excellent homepage!
|
| Tue, 29 Apr 2025, 05:34 | |
|
Dan |
Something is broken, tho ...
|
| Tue, 29 Apr 2025, 05:35 | |
|
Jayenkai
|
Yeah, I was worried that might be the case. What's broken?
-=-=- ''Load, Next List!''
|
| Tue, 29 Apr 2025, 05:42 | |
|
Dan |
I do not know what exactly. I have had a chip8 emulator running. But now it's not. Well ... it does run, it just does not display anything. Well It does, but ... not like before.
|
| Tue, 29 Apr 2025, 05:46 | |
|
Jayenkai
|
Fancy sending the code across so I can track it down?
-=-=- ''Load, Next List!''
|
| Tue, 29 Apr 2025, 05:47 | |
|
Dan |
JSEChip8_txt
|
| Tue, 29 Apr 2025, 06:00 | |
|
Jayenkai
|
Gosh, that's a big difference. OK, I'll dig into it. Thanks for the code!
-=-=- ''Load, Next List!''
|
| Tue, 29 Apr 2025, 06:23 | |
|
Jayenkai
|
 This may take a while..!! -=-=- ''Load, Next List!''
|
| Tue, 29 Apr 2025, 08:15 | |
|
Jayenkai
|
You'll be happy to know that I've now fixed the issue, which occurred between 3gim and 3gin, and was to do with me incorrectly accessing the wrong value during the IF command. Good grief, why that didn't show up earlier, I've no idea. .. ... Now... why it doesn't work in 3giq, I've no idea.. Step by step.. Fix by fix!! -=-=- ''Load, Next List!''
|
| Tue, 29 Apr 2025, 10:16 | |
|
Jayenkai
|
Getting there, but slow progress. .. or rather, the engine slowing back down again!!! Further tweaking required. -=-=- ''Load, Next List!''
|
| Tue, 29 Apr 2025, 13:43 | |
|
Jayenkai
|
v.3gis is now uploaded OK, did a whole lot of shuffling, and tweaked a few things. The arrays are again optimised a teensy bit, and the IF bug I found earlier is done. The bigger issue between 3gip and 3giq was that I stopped clearing out the argument array. This caused all manner of issues that none of my own games were showing up (that I bothered to test!). Oddly, I found that specifically setting JSE_argn[0]=0 was about the same speed as clearing out the entire array with JSE_argn=[], so in the end I threw those back in again. Anyway, give it a go and let me know if your little program is running like it should, now. I do hope it is. Meanwhile, Optimisationalism 5 is now running about 2% slower because of this tweak. I'll see if I can find a quicker way of doing this. I'm currently thinking that one single JSE_arg array will solve a lot of problems, instead of having JSE_arg, JSE_argn, JSE_argst, as well as a RefStack to deal with, too. Yeah, if one single =[] is quicker, then... See, this is why Optimising is important! -=-=- ''Load, Next List!''
|
|
|
| |||
|
| ||||