123
-=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- (c) WidthPadding Industries 1987 0|518|0 -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=-
Socoder -> JSE -> JSE - Jay's Scripting Engine

Page : 1 2 3 4 5 6 7 8 9 Prev 10 11 Next 12
Posted : Thursday, 22 July 2021, 07:55
Pakz
Ok, that same line seems to work. But I wanted to do it like below.



edit: You do not appear to have a Continue command either? I use these quite a lot for jumping to the next cycle in a loop.



Above, if I got it right. You do not check the same entities in the loop.
Posted : Thursday, 22 July 2021, 08:26
Jayenkai
Your first issue might be end if vs endif.. I'll add that into the parser checks. I've written it as specifically endif because I'm aiming for SingleWordCommands.. But, yeah, people do write "end if", so I'll have to add a check for those.

Second issue..
Continue/Break are commands I tend not to use nowadays. One of those things where it didn't do what I expected it to do in a different language, because different languages do things in different ways.

IIRC, the language I fell out with it in, when I used it, it stopped the loop dead in it's tracks and exited it. Which is what I would expect Break to do.
I found that very frustrating, so opted to do other methods instead.

But, yeah, it should be in.
Continue = Goto Next, continue the loop
Break = Goto the line after Next, abandoning the current Loop

Is that right?! Does that "feel" like the right method to everyone?

-=-=-
''Load, Next List!''
Posted : Thursday, 22 July 2021, 08:32
Pakz
Break seems like something I would use also
Posted : Thursday, 22 July 2021, 12:08
Dan
Hmm other languages use exit or exit loop instead of break.

AutoIt, for e.g. has an optional level parameter, and with it you can exit multiple for-nexts at once.



https://www.autoitscript.com/autoit3/docs/keywords/ExitLoop.htm
Posted : Thursday, 22 July 2021, 13:07
Jayenkai
Eek, a parameter could be tricky..
.. maybe ...
I'll test it and see.

I've got the basic version "mostly" working, but it can be glitchy in the wrong places. It's fairly easy to trip it up, right now.
.. but, hey, what else is new!!
I'll post the results later/in the morning, depending on how twisted in knots it all gets!

"Exit" will be added as an alternative.

-=-=-
''Load, Next List!''
Posted : Thursday, 22 July 2021, 15:41
Jayenkai
Ok, I think I've fixed Pakz' "if then ;" issue. It should know what's going on, now. I've also (I think) got "end if" compacting down to endif.
I don't *think* either of those should cause any issues, but as always, let me know if I've knackered something.

Meanwhile..

  --v

Exit/Break, Exit(n)/Break(n) and Continue should all work fine.
Currently undocumented.
Lazy Jay..

Be sure to refresh to get at least v3gec

-=-=-
''Load, Next List!''
Posted : Sunday, 25 July 2021, 08:52
Jayenkai

v3geg


(As always, might need a refresh to catch up)

The obligatory set of speedups, which have been gradually implemented over the past couple of days.

This version broke quite a lot of functionality, earlier, as minus numbers because infinitely more chaotic than they already were!
But a little to-and-fro'ing and everything got back to where it oughta be.
.. I hope!

As always, be sure to stress test it with anything you can think of.

I don't think anything has changed other than speedups.

If you'd like to check the Optimisationalism 2 test, you should (hopefully) find your numbers have improved a fair bit.
Let me know!

-=-=-
''Load, Next List!''
Posted : Sunday, 25 July 2021, 12:04
Dan
Yes, the numbers on the PC are now between 102 and 130 (before it was around 150)

And 180 for android GN20
Posted : Thursday, 29 July 2021, 07:36
Jayenkai
  --v


-=-=-
''Load, Next List!''
Posted : Thursday, 29 July 2021, 09:10
Pakz
That is impressive!

How do you rotate that text into 3d? Is it a kind of fake 3d?
Posted : Thursday, 29 July 2021, 09:21
Jayenkai
No, just the simple SetSize command with a Cos attached, so their width is reduced when near the edge, and backwards, too.
If I used FontSize, that wouldn't do the trick. Size works because it can scale x and y independently and also invert the x.

-=-=-
''Load, Next List!''
Posted : Thursday, 29 July 2021, 13:25
Pakz
I was experimenting a bit with jse. I noticed there is no way to erase pixels in the editor once you placed them?(for the transparent ones) Is there a erase shortcut that I am not aware of?
Posted : Thursday, 29 July 2021, 13:32
Jayenkai
Aah, good call.
Sorry, that's another one of those "I've been doing it so long, it's second nature to me" things.
Magenta (255,0,255) is traditionally (for me, at least!) transparent.
In the colour picker, there's a lonely Magenta on its own at the bottom.
This is your transparent colour.
I need to make that more obvious.

-=-=-
''Load, Next List!''
Posted : Saturday, 31 July 2021, 05:29
Dan
I have had the code from this video:

View on YouTube

converted to BlitzBasic.

Now i thought to try it in jse, and it worked without modification.
(Well, i had to change the delay into Wait ... )

Here it is.

  --v

Comment out the first wait to see the drawing in full speed.
Comment out the second wait to see the full speed dia-show
Posted : Saturday, 31 July 2021, 05:38
Jayenkai
Neato!!!

Comment out the first wait.
Replace the second wait with a Flip
and change DR's step size to 0.05

-=-=-
''Load, Next List!''
Posted : Saturday, 31 July 2021, 06:22
Jayenkai
Actually this is a really great example of variable-slowdown..

Compare..

> Reveal 🔎

vs

> Reveal 🔎

-=-=-
''Load, Next List!''
Posted : Saturday, 31 July 2021, 09:04
Dan

Posted : Sunday, 01 August 2021, 13:09
Dan
Hmm, have you noticed that, when printing the text, at the 1st line (the top one) there are few pixels cut off ?
Like the position of the font is not right.
The plotting of pixels works ok.

  --v

Posted : Sunday, 01 August 2021, 15:28
Jayenkai
Yep, it's been noted, and in fact the whole engine had a hefty realignment thing added, to make sure everything lines up nicely.
Still happens, though. *shrugs*
Depends on your browser, and it's one of those "You git!!!" browser compatibility issues that's been winding me up.

-=-=-
''Load, Next List!''
Posted : Monday, 02 August 2021, 05:41
Jayenkai
  --v

Weeeeeeeee...

-=-=-
''Load, Next List!''
Posted : Tuesday, 03 August 2021, 01:02
therevillsgames
Re: variable-slowdown

On my work laptop the first example with the variables runs faster at 180, whereas the one with hard to read (eg less variables) only runs at 127

Edit Ah... It's not fps, but time per frame so the hard to read is faster, my bad.
Posted : Tuesday, 03 August 2021, 04:05
Jayenkai
Yeah, I tend to use Mills onscreen. If you want FPS and "CPU Load-ish" values, the FPS toggle button is always available.

-=-=-
''Load, Next List!''
Posted : Tuesday, 03 August 2021, 06:08
Jayenkai
  --v

Note : Stress tester, ahoy!!

-=-=-
''Load, Next List!''
Posted : Tuesday, 03 August 2021, 14:44
Jayenkai
  --v

This example shouldn't be too taxing on most systems, but I think it depends on how well the browser handles blend modes..
*shrugs*

-=-=-
''Load, Next List!''
Posted : Tuesday, 03 August 2021, 17:49
Jayenkai
Been playing with Dan's Amiga code again, and managed to squeeze a bit more out of it.

Removed all references to Float() since those are already assumed by JSE. (Note : You'll have to Int/Round/Ceil/Floor if you want crisp numbers, here)
Float was slowing it down, especially with so many instances.

I also reduced the number of loops it does, but replaced the pixels with lines so it's not so obvious.
The result is quite nice!


> Reveal 🔎

-=-=-
''Load, Next List!''
Posted : Wednesday, 04 August 2021, 01:59
Dan
That looks great ! (all 3 examples)
Page : 1 2 3 4 5 6 7 8 9 Prev 10 11 Next 12