Oolite Bulletins

@phkb Thank you so much!

The newest version is now available via the extension manager:

07/10/2018 - Version 2.9:
* better handling of the alpha channel

Prester John wrote: ↑Sun Oct 07, 2018 6:00 pm

You just need a lot more memory on your PC, for all these graphics... Right?

I'm not really sure. I know that (some) of these effects are done using "shaders", which somehow offloads the processing required to a graphics sub-system away from the general processing unit. And I recall (from when I last paid attention to such matters, in the 1990s) that graphics cards sometimes (generally? I dunno) have separate memory to the main processor. But how all that adds up, or what bits go where is just a black hole to me.
If I've got work to do on the laptop, it involves text, and occasionally changing the font of the text (but generally, it's the words that matter, not the appearance), so as long as my computer does that, the rest is someone else's problem. Where's my Perilomatic (R) Sunglasses?

@RockDoctor: °sigh°
@Prester John: Not so much memory. It's more a graphics card which can handle this.

@gsagostinho:The rings are looking great now, gsagostinho!

And after taking a look at the shader - some ideas:

1. It could be worth using Oolites fogUniform for the atmosphere instead. Then you can get rid of quite a bunch of instructions easily. Shift the work to Oolite .-)

2. Let's help the compiler to do the right thing. could be simplified to
3. Another one to help the compiler. doesn't need the max(). The value will never get negative. And most compilers will reuse the min(1.0, distance/5000000.0), so I left it in. But there's no guarantee that the compiler will reuse it... <mumbles something about stupid things>


4. You could also get rid of some branching could become A similiar approach could be used for the threshold calculations.

It's just a start, I know, but better than nothing.
Optimizations will take eat time.

Edit: Optimized further, got rid of logical_if.

Thank you SO much, Svengali! Really appreciate this, I have just incorporated your suggestions and they work exactly as they should, so I just made a new release.

I just made some more updates:

- more optimizations
- rings get more transparent when entering an atmosphere

13/10/2018 - Version 2.10.2:
* some more minor optimizations

There seems to be a fault in the latest version. Here's a shot using a plain 1.86 and just the System Rings OXP (so no chance of any conflicts), at Riedquat (G1)

(unless, of course, this is my graphics card going crazy on me...)

Thanks, I think I know what happened and it should be fixed now. Could you please update your OXP and try again?

Ah, much better! Thanks!

After having my ship destroyed, I got the following exception:

Warning: removeFrameCallback(): invalid tracking ID.
Active script: System Features Ring effect script 2.10.3
systemfeatures-rings-effectScript.js, line 30:
removeFrameCallback(this.$fcb);

Milo wrote: ↑Fri Jun 19, 2020 4:07 pm

After having my ship destroyed, I got the following exception:

Warning: removeFrameCallback(): invalid tracking ID.
Active script: System Features Ring effect script 2.10.3
systemfeatures-rings-effectScript.js, line 30:
removeFrameCallback(this.$fcb);

What did you do? (So I can not-do that.)
I've never had a problem with pelting through a ring at torus drive speeds or under full military injectors - but I always flinch as I get the ground-rush which BASE jumpers describe.

I haven't looked into it yet, but it seems to be happening around the time I arrive at the station after my ship is destroyed (I'm testing shields so this is happening a lot...). I can more consistently reproduce it now by loading a saved game after being returned to the station after getting my ship destroyed. I'm using Ship Repurchase, unsure if that makes a difference.

Milo wrote: ↑Fri Jun 19, 2020 9:13 pm

I'm using Ship Repurchase, unsure if that makes a difference.

I've never used that one. Which makes it a potential target, straight off.