Page 3 of 5

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 8:52 am
by Detritus
JoshParnell wrote:55 89 e5
Hrrm... :think:
JoshParnell wrote:b8 39 05 00 00 5d c3
Hrrm... :think:

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 10:23 am
by Achati
I am voting for option Z too, hoping that once it just "works" you will improve LTSL's grammar and syntax to make it the best for you to work with
(i wold love to use python like light syntax)

so in memory compilation means probably that its only compiling all LTSL when loading the game and then storing that in memory until its needed
it sounds like editing LTSL while the game is running and then letting it recompile would be possible then
also caching the compiled blobs if nothing changes in LTSL

although LJ is probably slightly easier for modders because there is libraries and stuff for lua that wold have to be either ported to LTSL or i dunno
can you interface with other languages from LTSL ?

and thanks for staying (relatively) sane and keeping us updated

PS: i would totally run LTOS as a custom WM/extension of archlinux or something similar

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 10:44 am
by Talvieno
Personally, I think The Nuclear Option sounds sexy, but would much prefer if we didn't have to use it. :) If it's what it comes down to, and it's a guaranteed failsafe solution, then I'd say... let's do it. :P Better than continuing to search for another option in vain.

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 10:50 am
by JoshParnell
Detritus wrote:
JoshParnell wrote:55 89 e5
Hrrm... :think:
JoshParnell wrote:b8 39 05 00 00 5d c3
Hrrm... :think:
See below..
ruok wrote:1337 post indeed Josh :geek:
:clap: :clap: :clap:

Hahaha somebody got it woohoo!!! :D Great first post, welcome to the forums ruok :thumbup:

(More responses on the way, but for now, this ^ . I just woke up. Yayy Monday :squirrel: )

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 11:32 am
by kostuek
Talvieno wrote:...If it's what it comes down to, and it's a guaranteed failsafe solution...
Is it? I'm kind of sceptical about this. A "mature" ( with many many years of development in it) c++-compiler does a huge amount of optimization while compiling. So the task is not just to convert some script language into machine code, wich is hard enough, but to do it efficient. Or you may end with a very fancy but also very slow machine code.
So, personally, I hope it never comes to "option Z".

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 11:50 am
by Talvieno
Well, I don't know whether it is or not, Kostuek, hence why I included the "If". :)

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 12:45 pm
by alpan
I appreciate that you're still shooting for a 100-ship battle as a benchmark (I think I kind of remember such a target from the initial KS, or the early dev logs). Not because the number is particularly meaningful to me, and certainly not because I expect to be in such battles very often; on the contrary, that you're targeting a scenario that most players will be engaged in at most (lemme guess...) 10% of the time pretty much means the remaining 90% will end up performing very well. Not many software developers approach their work in the same way a civil engineer approaches their bridges. The continued emphasis on robustness is refreshing.

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 12:50 pm
by Cornflakes_91
You missed a 0 there :ghost:

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 1:14 pm
by Detritus
0

There ya go, I feel so helpful. :ghost:

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 1:29 pm
by Achati
there is also the option to use the AST and turn it into a format useable by llvm, that is slower but it will give you the optimization

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 2:03 pm
by Lum
Yeah, please call Hello Games and ask them why their game runs so smoothly, that you must, no, need, no, crave for that knowledge... :twisted:

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 2:04 pm
by Silverware
Lum wrote:Yeah, please call Hello Games and ask them why their game runs so smoothly, that you must, no, need, no, crave for that knowledge... :twisted:
Could it maybe be because they removed all of the things that use CPU?

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 2:06 pm
by Lum
It was sarcastic.... :ghost:

Or... NMS performs now in another level which I don't know... :problem:

Re: Sunday, February 12, 2017

Posted: Mon Feb 13, 2017 6:07 pm
by Miklos
That nuclear option sounds awesome tbh

Re: Sunday, February 12, 2017

Posted: Tue Feb 14, 2017 3:24 am
by BFett
Cornflakes_91 wrote:
BFett wrote: I'll see you in August. :P
How many times similar statements of you have been wrong? :ghost:
Theoretically? 1, though technically zero since I've always said that if Josh starts working on features then it will take 6 months to finish LT. Since Josh hasn't worked on features and instead has been working on the "Fundamental Problem", I have been wrong exactly zero times.

If Josh actually starts working on finishing up LT, we can start the timer and see how far off my guess has been.