Reversing to Orion - project 1oom

[induktio]

I haven't looked at this project in detail yet, but if you're looking a way to generate static web sites, maybe the simplest option is to use Jekyll. It should be supported by Gitlab as well as Github.
https://jekyllrb.com/docs/github-pages/
http://jekyllbootstrap.com/usage/jekyll-...start.html

[1oom_aaron]

The project coder replies:

if you're looking a way to generate static web sites

I'm not. $EDITOR is an easy enough way given the simplicity. Adding Jekyll to the mix does not help any (without templates), plus installing a programming language runtime and package manager to check out the edited website is not an option in my workflow.

If the homepage is too 1993 to bear, consider whipping up a better one using whichever method you prefer. Just keep CSS, JS and 3rd parties out of it.

For the game engine part you can indicate when you start a game that a player is an AI player and the game engine can apply whatever cheating it needs, and the AI clients can have additional cheat functions on the API.

True enough, words eaten. If you wish to see it happen and can hold a C compiler from the right end, fork and hack away! My time is tied with bug fixing and UI improvements until v1.0. I've never written a networked program before so someone with experience in those matters really should take over that part/phase of the project.

EDIT:

So a client connects to the server, says "I'm AI, let me cheat". What sort of anti-Turing test does the server apply to verify this? Remember that this is all Free Software so a human player who wishes to cheat can see and use all the client and server code.

On a different topic, I have begun to collect the OSG Errata. Contributions welcome.

[Jeff Graw]

I've got a save game that will (typically) crash after turn processing. Also, at a certain point miniaturisation makes high end warp drives more space efficient than low end ones (which probably isn't a good thing in and of itself) but since retros are always selectable this can allow the creation of a design with "negative" space (which ends up looking like an overflow in the UI).

https://www.dropbox.com/s/hgxk460m9l5mqo...6.bin?dl=0

[1oom_aaron]

The project coder replies:

I've got a save game that will (typically) crash after turn processing.

Thanks. I assume this was taken with v0.6-34 or older. Versions v0.6-36 and newer are deterministic by default and "(typically)" should never be needed. The homepage FAQ has instructions on what to do when encountering a bug; they are still worth a read.

Also, at a certain point miniaturisation makes high end warp drives more space efficient than low end ones (which probably isn't a good thing in and of itself) but since retros are always selectable this can allow the creation of a design with "negative" space (which ends up looking like an overflow in the UI).

Got a save?

[Jeff Graw]

Save was taken with v0.6-36. It crashed on an earlier turn, but on reload I was able to get through it until it crashed again a couple turns later. I haven't much tested that save to see if the crash is fully deterministic on it though. It's possible I made some slight change earlier that prevented it from crashing in the original location. Or perhaps it's not fully deterministic after all :P

The save I provided exposes the retro bug. Sorry I wasn't clear that it applies to both issues.

[1oom_aaron]

The project coder replies:

Save was taken with v0.6-36. It crashed on an earlier turn, but on reload I was able to get through it until it crashed again a couple turns later. I haven't much tested that save to see if the crash is fully deterministic on it though. It's possible I made some slight change earlier that prevented it from crashing in the original location. Or perhaps it's not fully deterministic after all :P

I jumped to conclusions here, sorry about that. This turned out to be a case of stack corruption. Whether that is always deterministic... I'll choose to not think about.

The save I provided exposes the retro bug. Sorry I wasn't clear that it applies to both issues.

No worries. Both fixed now in v0.6-43. Also noticed the costs were off, fixed that too.

Three bugs with one save. Not bad!

[Jeff Graw]

How about an option that engine space/cost rounds to int after all engines are placed instead of at the per-engine level?

That low end engines become larger than high end engines at high propulsion tech level, besides being really counter-intuitive, is a moderate game design issue. It means that there is less meaningful choices in ship design, and it's almost certainly an oversight.

EDIT: Or is low end engines costing more at a certain point more of a quirk with how power works than rounding?

EDIT 2: Yup, looks like it. In fact, rounding cost afterwards looks like it could make things worse in a lot of scenarios. I guess it's just an unfortunate effect of how power works and how the various engines are balanced.

[1oom_aaron]

The project coder replies:

How about an option that engine space/cost rounds to int after all engines are placed instead of at the per-engine level?

That low end engines become larger than high end engines at high propulsion tech level, besides being really counter-intuitive, is a moderate game design issue. It means that there is less meaningful choices in ship design, and it's almost certainly an oversight.

EDIT: Or is low end engines costing more at a certain point more of a quirk with how power works than rounding?

EDIT 2: Yup, looks like it. In fact, rounding cost afterwards looks like it could make things worse in a lot of scenarios. I guess it's just an unfortunate effect of how power works and how the various engines are balanced.

I guess this solved itself. Adding such optional rounding would add many ifs, the benefit is questionable and I'm wary on tampering with economic/design/etc models.

--

On a completely unrelated note and intended pun: does the music have correct tempo?

I lack the sense of rhythm and/or software to verify this. The option -xmidtpq / xmid_ticksperq controls the tempo, the default value is 55. I'd like to get the correct value determined and the option removed.

I'd suggest recording two loops of the main menu music with MOO1 and 1oom (with xmidtpq values near 55) and trying to align the recordings with an audio editor. Report which one is closest.

If you have the software and about 15 minutes to spare, consider lending the project a hand.

[1oom_aaron]

1oom v0.7 has been released. Changes:

• game is deterministic by default
  
• more options for planetary governor: (see doc/usage_common.txt)
  • spend rest on research/ships/reserve
    
  • ecology mode
    
  • build star gates
    
• new -yearsave option for yearly saves (1oom_save_YYYY.bin)
  
• Ctrl-F5 takes a screenshot
  
• new -uiextra UI features:
  • 1.40m-like research bonus lights
    
  • extra menu (click bottom-left corner)
    
  • message filtering (via extra menu)
    
• new "planet X has finished terraforming" message (off by default)
  
• about 30 bugs fixed, notably:
  • many memory corruption crashes fixed (C is fun)
    
  • ETA is now correct
    
  • spy cost in Races screen matches actual costs
    
• doc/pbxin_fixbugs.txt expanded, including fix for Oracle Interface
  

Source code: https://gitlab.com/KilgoreTroutMaskReplicant/1oom
Binaries: https://gitlab.com/KilgoreTroutMaskRepli.../tags/v0.7
Homepage: https://kilgoretroutmaskreplicant.gitlab.io/plain-html/

Test. Report.

[Juffos]

Greetings

I earflapped and stopwatched the music tempo from the main menu and couldn't figure any meaningful difference. However, the opening sequence... I made this video to showcase my findings - the music just doesn't seem to work.

https://www.youtube.com/watch?v=cDZr73N-SYg