This is a PAGE 3 of a copy of a thread on freegamedev.net. Once around 2010 or so, I lost a similar thread over there for my other little game, Word War vi when the forum database was lost, so I created this as a backup in case something similar happens again.
smcameron 26 Feb 2018, 13:49
Some video of the new space monsters...
smcameron 02 Mar 2018, 21:59
Anybody have any DMX lighting hardware?
I'm trying to control such lights from linux, but I do not have any DMX hardware.
If someone would like to test it, there's a program called test_snis_dmx:
scameron@sirius ~/github/space-nerds-in-space $ make test_snis_dmx CHECKING for struct termios2... not OK. We will define it ourself. COMPILE snis_dmx.c scameron@sirius ~/github/space-nerds-in-space $ ./test_snis_dmx /dev/null snis_dmx.c:236: NOTICE!!! This snis_dmx library is COMPLETELY UNTESTED as snis_dmx.c:238: I currently do not have any hardware I can test with. snis_dmx.c:240: Do not be surprised if it utterly fails to do what you want. snis_dmx: ioctl TCGETS2 failed: Inappropriate ioctl for device Failed to start DMX thread on device '/dev/null', Inappropriate ioctl for device scameron@sirius ~/github/space-nerds-in-space $
In the example, I told it to test /dev/null, which of course didn't work. If you had some real DMX hardware on a serial device, it would try to blink all the lights at 0.5Hz for 40 seconds. Seeing as how I cannot test it, it's somewhat unlikely that it will actually work, but it is within the realm of possibility.
The eventual goal is to be able to connect events within the game to lighting cues, so that e.g. being hit by enemy fire can trigger strobe flashes, red alert can turn the room lights red, etc.
smcameron 05 Mar 2018, 18:42
From the wreckage of a derelict ship, some disturbing footage was recovered...
https://www.youtube.com/watch?v=woMkFN47U-Q
smcameron 27 Mar 2018, 18:27
Development Update 2018-03-27:
https://www.youtube.com/watch?v=Jzj4CSFYD6s
charlie 27 Mar 2018, 20:42
Cool update but I can barely hear the commentary after the half way point of the video as the music and sound effects drown it out.
smcameron 28 Mar 2018, 01:36
Thanks for the feedback Charlie. I remixed the audio.
smcameron 29 Jun 2018, 14:07
Quite some time ago, I added the capability for speech recognition (using pocketsphinx) and some rudimentary Zork-like natural language processing to Space Nerds In Space. However, like a dancing bear, the speech recognition is remarkable mostly in that works at all, not because it works so well. Especially, background noise is a big problem, but also accents, talking too fast, and mumbling slightly will make it not work so well.
The speech recognition is done in a separate process and the text which it produces is fed into the game via a fifo that snis_client monitors. This means the speech recognition can be replaced relatively easily (assuming the existence of an alternative). It occurred to me that if I had an android app which could do speech recognition (should be easy enough, right?) and then forward the recognized text to an arbitrary TCP socket, then I make some little program to forward this text from this TCP socket to the fifo. In fact, I don't even need to make something, I can just use netcat for that. However, I don't really know anything about writing android apps, and it's been almost 20 years since I even tried Java. If I were to do it, there's considerable static friction before I can get moving.
I mentioned this idea to Max Redmond, a member of my local hackerspace (these days, that's HackRVA) and hey presto, now we have an android app that does exactly what I imagined.
It's here:
Source code: https://github.com/Athulus/SpaceNerdsCommunicator
Releases: https://github.com/Athulus/SpaceNerdsCommunicator/releases
To use this with SNIS, find out the IP address of one of the machines running snis_client, then run this in a terminal:
netcat -lvk -p 8080 > /tmp/snis-natural-language-fifo
Then run the app on your android device, set the IP address, and then you can hit the microphone button and talk. Android will do the speech recognition (and do a much better job of it than pocketsphinx does) and the text will get forwarded to netcat and then into snis_client, and finally to snis_server, where it will be processed. It does not have the "Computer" hot word though.
It occurs to me now that unlike the fifo, the socket is a 2-way communications channel, so it might be worthwhile to set up a special socket for getting the NL text into snis_client and use that instead of or in addition to a fifo so that responses could be sent back to the android device as text (and then maybe text-to-speech). Then you could operate the ship from a phone remotely and at least get the responses from the computer instead of operating completely blind. Not that it would be super useful to try to play the game remotely via voice control without any visible output.
Perhaps a role-play mission where you're stranded on a planet, and you have to bring the unmanned ship into orbit around your planet using voice alone. i.e. "Set a course for planet X", "full speed ahead", "how far to planet X" (repeat until the reported number is small enough), "standard orbit", "beam me up scotty". Well, it's not much of a mission, I suppose.
smcameron 07 Jul 2018, 21:10
Here's a picture of Space Nerds In Space built and running on Windows 10 using the linux subsystem (not that it looks any different than on linux):
Surprisingly straightforward.
smcameron 28 Jul 2018, 02:01
Latest dev update video for Space Nerds In Space, 2018-07-27:
https://www.youtube.com/watch?v=60oUfXtFlI4
charlie 02 Aug 2018, 00:55
That network scaling issue was an interesting one, I bet. Was it some kind of loop that should not have been there?
smcameron 02 Aug 2018, 10:17
You can read about debugging that issue here, where I spend a good deal of time chasing red herrings before figuring it out. But to skip to the "good part"...
In the most general terms, the cause of the problem was "the programmer got confused because there were too many layers of abstraction for him (me) at the moment he wrote that code." More specifically, I knew that some information needed to get to all the clients, so I called a function that sprayed the information to all the clients. However, I was calling this from a place that was already in a per-client context (already being called per client), so for each client, I sprayed the information to each client (that is to say, instead of n times, n^2 times.) I wasn't doing this for *all* the data, but for a subset of data that was particular to NPC ships (of which there are a lot, so not an insignificant amount.) so it's not quite true that the traffic literally scaled as the square of the number of clients, but it is true that it scaled as the square of the number of clients asymptotically -- the more clients there were, the closer to the true square the amount of traffic became. Also, I think it was actually number of clients *per bridge* squared, times number of bridges... but in this particular case, the number of bridges was 1.
The fix was essentially to replace the call to the function which sprayed the data out to all clients with a call to a function that just transmitted to one client. There was also some mutex confusion in there that got fixed, which may or may not have contributed to the problem.
How I found the n^2 problem was, unfortunately, really just by staring at the code and noticing it, rather than by some Sherlockian sequence of deductive steps.
smcameron 16 Aug 2018, 21:12
Dev update 2018-08-16: https://www.youtube.com/watch?v=MwmLa1IOmJg
smcameron 09 Sep 2018, 22:44
Dev update for 2018-09-09 -- mostly debugging tools.
smcameron 04 Oct 2018, 00:53
Dev update for 2018-10-03
https://www.youtube.com/watch?v=CFBD6WqT14o
sudoRmRf 21 Oct 2018, 02:13
In one of your recent update videos you discribed a theoretical MMO use but then stated that this all was not likely as it is hard enough to get 5 people to play let alone many crews. Is there some sort of limitation preventing the use of a public server running SNIS, where my three people crew could meet at my house and interact with another crew somewhere else? Or is it just simply such servers do not exist?
smcameron » 23 Oct 2018, 01:35
In one of your recent update videos you discribed a theoretical MMO use but then stated that this all was not likely as it is hard enough to get 5 people to play let alone many crews. Is there some sort of limitation preventing the use of a public server running SNIS, where my three people crew could meet at my house and interact with another crew somewhere else? Or is it just simply such servers do not exist?
Technically it can run that way, on a server on the internet with remote crew members, and I have run it that way once or twice just by myself to see how it would perform with a server in New York (a digitalocean droplet) with me in Virginia. It's not a great experience though, the latency is a bit of a problem, although it worked better than I expected -- almost normally, but not quite, occasionally sort of freezing up for a few moments at a time. I didn't try it for very long, or really try to characterize the behavior in any depth -- just ran it for a few minutes out of curiosity. Total bandwidth could be a problem as well, since each client requires approximately 100kbytes / sec, though it varies somewhat depending on how many things are going on in the game, and you could in theory design some Lua script missions to use not very many objects and drastically reduce bandwidth requirements that way.
But basically, the design has always been for a LAN game with everyone in the same room, and for a LAN game with no more than about 20 clients. With those constraints, 100kbytes / sec / client is not a big problem. Over the internet, it's a bit more of a problem.
If you want to try such an experiment yourself to see how it is, here are some instructions.
Note, those instructions are sort of mixed together with instructions about running the game with no lobby, but there's no reason you couldn't also run the lobby on the same host, and run the clients more "normally", which, if you wanted to have multiple solar systems and traverse between them, you must run the lobby server (ssgl_server), since it's part of how that works.
It is also true that "such servers do not exist", at least not to my knowledge (I don't run any.) Nothing's stopping anyone from running such a thing, but I would not advise doing so on any machine that hosts anything valuable, I'm fairly sure any determined attacker could manage to gain shell access via snis_server if they tried (not to say I know of any particular exploit, but I'd just be surprised if there weren't some bug somewhere in there that gave enough of a toehold.)
scameron 03 Nov 2018, 17:29
Six Years of Space Nerds In Space:
GunChleoc 06 Nov 2018, 21:11
Thanks, I enjoyed watching that
MCMic 15 Jan 2019, 10:40
Hello,
First, congratulations on this project, it’s quite impressive.
I will try to find enough people and hardware to test it, and I have a few questions/remarks after reading the website.
I have a github account so I can open issues if needed.
smcameron 16 Jan 2019, 02:08
github issue here: https://github.com/smcameron/space-nerds-in-space/issues/179.
The idea is you have a bunch of people in the same room playing the game.
Most likely, only one of the computers is hooked up to the giant stereo system.
If that computer is denoted as the "sound server", then that's where most of
the sound effects, etc. will go, so that when you get hit by a torpedo, it
makes a room-shaking blast from the big stereo, and not just a tiny pop from a
laptop speaker. Same idea with the "text to speech" server.
github issue here: https://github.com/smcameron/space-nerds-in-space/issues/177.
github issue here: https://github.com/smcameron/space-nerds-in-space/issues/178.
Joystick actually doesn't work that well for weapons, kind of hard to aim.
Mouse works better. Joystick or HOTAS works pretty well for NAV, rudder pedals
too. Config is in share/snis/joystick_config.txt
Mostly, keyboard and mouse should be sufficient.
github issue here: https://github.com/smcameron/space-nerds-in-space/issues/176
(see 2nd comment... that you wrote, of course.)
Huh. Nobody's ever tried to use that as far as I know. I'll have to look into it.
I noticed. :)
smcameron 17 Jan 2019, 00:40
Ok... I think it should be fixed now. Patreon also works.
MCMic 17 Jan 2019, 10:37
Yeah it worked, but the way liberapay works is a bit weird now, this is the fist time I use it since their problem with mangopay…
Surprising. I’ll have to test.
Yeah I saw that I’ll hack into it as soon as I get my hand on a joystick. (I
only have a fighting game joystick right now, does not seem appropriate . . .
https://www.amazon.com/HORI-Real-Arcade-Pro-PlayStation-4/dp/B00SULMRI4.
Ok, but I felt like putting a bit different controls on each post to make them stand out from each other.
From what I’ve computed Weapons would need only joystick (or mouse),
Engineering only mouse (or tactile), and Comms could have only keyboard (if I
find a way to click those buttons from keyboard. And a real Red alert button
would be extra cool).
I started to hack together a project to build an ISO for playing Space Nerds In Space from a USB stick (or live CD):
https://gitlab.com/MCMic/snislive
It works when built on my computer but I’m trying to get gitlab CI to build it for me and put the ISO for download.
The ISO is 800Mo and uses LXDE.
There is a slight bug with text rendering in terminals, not sure where that comes from.
The game launches and works fine. Only real problem I had is no wifi on my
computer which requires a broadcom driver only available in the AUR (not easy
to install AUR packages in the ISO, plus this one is non-free so I think it
would not be legal).
I also had some differences on where some widgets show
up in the game, making things overlap on the engineering screen. Not sure if
that’s from the resolution difference or GTK configuration or whatever.
smcameron » 17 Jan 2019, 15:55
Yeah it worked, but the way liberapay works is a bit weird now, this is the fist time I use it since their problem with mangopay…
I noticed some money showed up in my paypal, presumably you did that. Thank you very much!
I have used "Thrustmaster T16000M FCS Flight Pack" that I got cheap as an
opened box from Fry's successfully as well as the "HOTAS Warthog" (not mine)
though that one is quite expensive. One thing I should work on is the game
assumes all joystick buttons are momentary switches, but sometimes there can be
toggle switches like on HOTAS Warthog, and the game still treats them as
momentary.
Cool!
There is a slight bug with text rendering in terminals, not sure where that comes from.
The game launches and works fine. Only real problem I had is no wifi on my
computer which requires a broadcom driver only available in the AUR (not easy
to install AUR packages in the ISO, plus this one is non-free so I think it
would not be legal).
I also had some differences on where some widgets show up in the game,
making things overlap on the engineering screen. Not sure if that’s from the
resolution difference or GTK configuration or whatever.
In general, I have found that it's ok for one or two stations to use wifi,
but generally things work better with a wired setup. As for things rendering
weirdly, at different resolutions, that is quite possible. It should detect the
aspect ratio, and make some crude adjustments, but it might not be ideal. If
you look in snis_launcher, line 25, you can see some different aspect ratios.
Really though, the aspect ratio handling is kind of hacky, and the positioning
of widgets on the screen is also pretty hacky. It's kind of a hard problem.
MCMic 17 Jan 2019, 17:00
You’re welcome! It’s supposed to be a yearly donation cause of how liberapay works. We’ll see in one year if I’m still playing snis and if liberapay is still existing ^^
Nice one!
I’m not good with electronics and low-level stuff like arduino.
If I manage to play I will probably give feedback here and show what we used for input.
Ok, I’ll have to hunt for ethernet cables as well then.
I did notice some weirdness and problems in the widgets placements, I think at some point I’ll try to see if I can help on this front.
New question:
I got an old computer (Thinkpad X200) with Xubuntu 18.04 on it. The game builds fine but then gives a black screen instead of the window content.
I looked at the output and it says GLSL 1.30 is not supported by my computer.
I checked and it’s true, this hardware does not support OpenGL 3 and newer.
Is there any hope around this problem or is it impossible to render the game on such a computer?
(I used make update-assets, would it help if I use make models instead? To build the models on the same hardware?)
smcameron » 17 Jan 2019, 21:34
New question:
I got an old computer (Thinkpad X200) with Xubuntu 18.04 on it. The game builds fine but then gives a black screen instead of the window content.
I looked at the output and it says GLSL 1.30 is not supported by my computer.
I checked and it’s true, this hardware does not support OpenGL 3 and newer.
Is there any hope around this problem or is it impossible to render the game on such a computer?
(I used make update-assets, would it help if I use make models instead? To build the models on the same hardware?)
If you run snis_limited_client (menu item 5 in snis_launcher instead of 4),
then you don't need opengl. However, you should probably only use COMMS,
SCIENCE, ENGINEERING, and DAMAGE CONTROL, the other screens (NAV, WEAPONS, MAIN
VIEW, DEMON) will probably not work very well (it won't hurt anything to try
them, but you'll see that they don't work very well.)
If you notice a red FPS indicator appear in the upper left hand corner of
the screen, it means things are running a little too slow (it shows up if
things drop below 29 frames per second). I have seen it show up when using the
limited client on the login screen, but once past that, ENG, SCI, DAMAGE
CONTROL, and COMMS seemed to be ok.
MCMic 19 Jan 2019, 11:29
Regarding Engineering screen:
What does the "Deploy chaff" button does? It’s not explained on the website.
What does the COMMS PWR slider does? It’s at 0 in preset 1 and yet the COMMS screen seems to work fine.
MCMic » 19 Jan 2019, 11:53
Regarding additionnal assets downloaded by
If so I think the best solution for me is to package them as an snis-extra-assets-git package in AUR.
MCMic 19 Jan 2019, 13:58
Finally managed to build the ISO using gitlab CI!
So you can now link to
https://gitlab.com/MCMic/snislive/-/jobs/artifacts/master/browse?job=build
which
redirects to the artifacts of the latest sucessful build.
I did not try the ISO made by gitlab yet but I see no reason why it would differ from what I build locally.
smcameron » 19 Jan 2019, 14:31
Regarding Engineering screen:
What does the "Deploy chaff" button does? It’s not explained on the website.
What does the COMMS PWR slider does? It’s at 0 in preset 1 and yet the COMMS screen seems to work fine.
So on weapons, if you point the guns at an enemy that is within range
(spinning arrow thingy is red, not blue) and press the 'N' key, and have
missiles remaining, it fires a homing missile (N is for Nuke) at them. Enemies
can also fire homing missiles at you. When that happens, it is pretty tough to
survive, and to try to make it somewhat more survivable, Comms will get a
flashing red warning in the middle of the screen about "MISSILE LOCK ON
DETECTED!" If COMMS shouts out and engineering reacts quickly enough,
engineering can DEPLOY CHAFF, which will have a good chance of throwing the
incoming missile off target and save the day. So that's what CHAFF does. If the
enemy missiles turn out to be too potent and demoralizing you can go to the
DEMON screen and type in "SET MISSILE_FIRE_CHANCE = 0" and the enemies will
never fire any missiles.
COMMS PWR does not do anything, actually.
That being said, there is an aim-able comms antenna that is normally
disabled (that is, no aiming is required by default). If you go to the DEMON
screen, there's a lua script, ENABLE_ANTENNA.LUA and DISABLE_ANTENNA.LUA
When antenna is enabled, there is a new arrow on the NAV screen showing
which way the antenna is pointed. It can only point in a half sphere at the
front half of the ship. If you try to communicate with something that is not
close by, say a distant starbase, you have to point the antenna towards it.
This means your ship has to be pointing at least kind of towards it and not
away from it, as the antenna cannot point towards the rear of the ship. Comms
is "attenuated" based on (iirc) the dot product of the vector towards the
object you're communicating with and the vector of the antenna aim. To aim the
antenna, say at a starbase, first ask SCIENCE for the BEARING and MARK to the
target. Say it's bearing 250 mark -79. Then, use the computer with a command
like "/computer antenna 250 -79". The computer will point to this bearing, and
maintain aim at this point so long as the ship is roughly oriented such that
that bearing/mark is within the front hemisphere of the ship. If the ship
points mostly away from the target, then the antenna will not be able to
maintain aim, and signal quality will suffer or go to zero. If the aim is bad
enough, your message will get dropped entirely, or, may be "distorted", meaning
some fraction of characters will be replaced with an asterisk. But, overall, I
found that it's sort of hard enough to tell what the hell is happening when
comms is working perfectly, and making comms degrade does not make the game
more fun, and it can be difficult to tell the difference between COMMS not
working well because the antenna is not well aimed vs. COMMS not working
because of a bug in the program. It seemed more frustrating than fun. So this
feature is turned off by default.
That has nothing to do with comms power though... which comms power doesn't
do anything because I couldn't really figure out something good for it to do,
and realistically, a radio just doesn't need that much power to begin with. If
you have a good idea about what comms power ought to do, let me know.
smcameron 19 Jan 2019, 14:34
If so I think the best solution for me is to package them as an snis-extra-assets-git package in AUR.
No, the web server has more assets than are in that github repo. I made the
web server because git is not especially good for storing assets and I realized
if I continued to use git for that, I would soon use up my allocation of free
space on github.
smcameron 19 Jan 2019, 14:34
Finally managed to build the ISO using gitlab CI!
So you can now link to https://gitlab.com/MCMic/snislive/-/job ... ?job=build which redirects to the artifacts of the latest sucessful build.
I did not try the ISO made by gitlab yet but I see no reason why it would differ from what I build locally.
Cool!
Does this auto-sync with my repo on github? I guess it probably does.
I have prepared a patch to link to this, but I'm not going to do it until
you at least try out the ISO image. I mean, *someone* should try it before we
inflict it upon the world.
MCMic 19 Jan 2019, 15:34
It is manual built only, to avoid taking too much power and spacedisk of gitlab for no reason.
But when built, it builds the AUR package which pulls latest the commit from github master. The name of the file contains the date it was built on.
Fair enough.
I’ll keep you updated when I tested it.
You may want to link to the project webpage as there are some
information/instruction as well as the download link, not sure.
Also if you are editing the website source, it would be good to fix the
width of the webpage which is way larger than my browser, and this is
extra-annoying from mobile as well. From what I’ve seen the best fix is to put
"overflow:auto" as css style for the pre tags. It’s those pre blocks which are
larger than anything else.
No, the web server has more assets than are in that github repo. I made the
web server because git is not especially good for storing assets and I realized
if I continued to use git for that, I would soon use up my allocation of free
space on github.
Hum. That is not practical for packaging, the PKGBUILD needs to be exaustive
regarding source of data and their checksum (well git repos are an exception
regarding checksum). Could you provide those in a single archive so that it
can be added as a source of the package?
And if the github repo with assets is obsolete you could mention that in its readme.
Intuitively I would have expected a range setting like the one for the
radar, which requires more power to talk to far starbases. But maybe it makes
no sense and radioing far does not take more power, just more time?
In the end the solution may just be to remove the comms pwr bar from engineering.
MCMic » 19 Jan 2019, 15:38
Note: I can add you to the project in gitlab and give you enough rights to trigger builds if you want.
You can log into gitlab.com using your github account and it will create you an account automatically.
smcameron 19 Jan 2019, 20:13
...
Also if you are editing the website source, it would be good to fix the
width of the webpage which is way larger than my browser, and this is
extra-annoying from mobile as well. From what I’ve seen the best fix is to put
"overflow:auto" as css style for the pre tags. It’s those pre blocks which are
larger than anything else.
My web-site editing skills are pretty rudimentary (which i how I like them). I'll poke at it some, but don't expect much.
No, the web server has more assets than are in that github repo. I made the
web server because git is not especially good for storing assets and I realized
if I continued to use git for that, I would soon use up my allocation of free
space on github.
Hum. That is not practical for packaging, the PKGBUILD needs to be
exaustive regarding source of data and their checksum (well git repos are an
exception regarding checksum). Could you provide those in a single archive so
that it can be added as a source of the package?
And if the github repo with assets is obsolete you could mention that in its readme.
Hmm, a single archive defeats the purpose, which is to not download things
more than necessary. Checksum (md5sum) and locations of assets are here:
https://spacenerdsinspace.com/snis-assets/manifest.txt
Prepend https://spacenerdsinspace.com/snis-assets/ before all the paths listed in that file.
Or, don't include the assets in the package, and include
util/snis_update_assets.sh in the package, then run that script to fetch the
assets separately. The whole point is the assets change from time to time, and
you shouldn't have to re-install just to get new assets, and you shouldn't have
to download the assets you already have.
Or, you could make your own archive, by running snis_update_assets.sh and packing up whatever it grabs.
MCMic 19 Jan 2019, 20:37
The build step of the package is supposed to run offline, as far as I know. I’ll see what exactly is missing and if I can think of an easy solution.
I tested the ISO from gitlab.com and it works as fine as the one built locally.
I failed to boot it from an UEFI computer but that may be my inexperience with UEFI, I’m gonna try some more.
I also noticed I have no sound once booted from the USB stick. I may need to include pulse audio, I’m gonna look into that.
[EDIT]Hum, in the end the solution on the UEFI computer was just to wait, it
seems the black screen I get is the boot menu, and waiting for long enough it
just starts the first menu option and boots the system. And it had sound, so
there is sound in the system after all. I think I will still add pulseaudio and
pavucontrol to the iso to be able to set up sound on any setup.
smcameron 21 Jan 2019, 21:48
MCMic, I sent you an email with a link to an archive of all the SNIS assets.
In case you didn't see it it's here:
https://spacenerdsinspace.com/snis-asset-archives/
MCMic 24 Jan 2019, 13:56
What is the difference between "disembark" and "eject" passengers? I think I did the wrong one and did not get paid.
smcameron 24 Jan 2019, 16:51
Heh. I had to look at the code to find out. This is not a very frequently exercised feature, I would say.
I think the difference is that "ejecting" a passenger means delivering them to someplace they weren't intending to go, while "disembarking" them means delivering them where they intended to go. If you disembark passengers, only those intending to go to the starbase you're currently docked at will leave the ship.
If you eject passengers, the ones being ejected at the wrong starbase will deduct money from your wallet in the amount of the fare. You will still get paid for passegers "ejected" at their correct destination. So for passengers ejected at the wrong desitnation, not only you didn't get paid, you got fined. This should probably be made more obvious somehow.
Here's the relevant "disembark" code. sb->id is the starbase you're currently docked at.
MCMic 24 Jan 2019, 23:53
Thanks for the explanation.
I feel like fares earned and fines paid should produce some output indeed
(since the station did tell me it charged me for refuel for instance, when I
got no output after ejecting passengers I thought I did not get paid. Of course
this is also because I was testing the game alone and have no crew to monitor
the money account and such).
I hit a problem with my gitlab built ISO, with all the assets the ISO is now
more than 1Gio and gitlab.com refuses to host it so it gets trashed…
MCMic 28 Jan 2019, 12:38
I finally was able to play the game with a whole crew this week-end.
We tried and failed space pox, then went back to saving planet erph which we
won, then tried space pox a few other times and lost.
(spoilers about SNIS missions)
First time we missed the box, mined the derelict of the lunear, found
nothing, went through the wormwhole, did not understand what to do after. Then
a few times we sent the mining bot on the box and died quickly after.
Then we used the tractor beam to take the box through the wormwhole with us and
it worked. We had the vaccine. We were not sure where to go, so we tried to
hail the starbase to get repairs and fuel, and information.
There was a weird bug here, the starbase appeared as SB-00 but if we hailed
SB-00 it would answer that we are way too far to dock, while we were just next
to it. Clearly I think the SB-00 we were hailing and the one next to us was not
the same one. So after trying a few things we finally decided to go back
through the wormwhole but we got short on fuel at that time. In a desperate
move we hailed a starbase (sb-00 I guess) and asked to be tracked. It worked,
and we managed to dock, and to see that this starbase was actually next to the
Bonx planet we’ve been searching, but at that point the last member of the
icarus died. We tried to search a bit for icarus around Bonx and did not see
it. (But do not tell me where it is we will try again sometime)
Despite the station hailing bug, the main problem we had was how to know
where an object which is not in range of scanners is. For instance the Bonx
planet, or the Icarus station, how do we know it’s our side of the wormwhole,
and how do we know where it is. We tried hailing Bonx or Icarus with no
success. Actually it was the same for planet erph, we found it because we
could actually see it from the window, but it was kind of random «Hey here’s a
planet let’s try to see if it’s this one».
So either we missed a way of doing that, or the game would need something
like this, maybe being able to ask starbases for close planets and other
starbases names/positions?
We had a problem with TTS, I activated it only on the server/main screen
which was plugged into the sound system, but some things were silent, like
mining bot deployment, or computer answers. This means we were basically unable
to use the computer since it did not answer. (but things like "docking system
engaged" was correctly spoken) So what did I do wrong, should TTS be checked on
all stations?
Speaking of the computer, it feels a bit like cheating since we can ask it
do drive for us, mostly. So I was thinking maybe this is what should be
controlled by the power slider for COMMS screen. A low power computer would
only be able to give manifest and hail stuff, while a fully powered can set
courses.
We also missed a way to add currently selected object as a waypoint in the
Science screen. And in some cases the SCI arrow was not there in the Navigation
screen, not sure why (even with a selected object, which is why we tried to add
a waypoint for it I think. But maybe this was the buggy sb-00 again and it’s
the same bug).
All in all the game is quite fun, I think the least interesting job is
engineering screen, since he goes from doing nothing because there’s nothing to
do to doing nothing because there’s not much he can do giving the "all failure"
situation. (at least at our level of playing). A really good improvement would
be to allow the engineer to add presets of his own. This would allow him to
prepare presets for fighting, warping, that kind of stuff and ajust them as we
go. We did not use the chaff and maybe we should have, not sure.
The COMMS screen was busy enough so I do not see the need for this RTS
feature (we did not try it). From what I’ve seen in your video about it this
feels a bit like if the comm guy was playing something else on his phone while
on the bridge. The idea of guiding a fleet of NPC ships around us would be
something else though ^^. Are ally ships something we can do in scenarios?
An other question, is there any way to go back to initial solarsystem state
once a mission is started? After winning saving planet erph, we docked at the
starbase, which told us to go away because it was zarkon, so we went away but
there was nothing in the solar system anymore because of the mission beginning
clear. As I did not know how to do that we just played missions from this point
on. The mainmenu is quite handy and there should be a way of launching it
without going demonscreen. (maybe a key binding on mainscreen?)
I will send a patch with a joystick configuration for a second joystick we used for weapons. We never had the case where we used the wavelength feature, is it useful sometimes? It always felt like fights were too short to be able to look into that. Except the zarkon but we had most of them with missiles.
Is there any way to buy torpedos and missiles?
Torpedos are really hard to use.
Is it possible to take down a starbase?
When COMMS screen puts the science screen on the main screen we do not see the details of the selected object nor the 3D view of it. Is that known/expected?
Also not sure if this is possible but it would come in handy if the cursor
of the displayed screen was shown, allowing the scienctific to point at things
while explaining stuff for instance (But I get that this might be impossible by
design).
I get a lot of errors in the output about files that could not be found, for
instance "/usr/share/snis/textures/green-spark.png", am I missing files? I
can’t find those in the git repo either. I’m using the AUR package on
ArchLinux.
What is the sound server role checkbox for?
I failed to launch any lua script, it says it cannot find the
script, but the path it’s showing starts with share/ and not /usr/share/ so I’m
thinking maybe the prefix is missing in the script path in the code?
If I look into /usr/share/snis/luascripts I only have 7 scripts and I do not
have the SAVING-PLANET-ERPH.LUA for instance, are some missions missing from
make install? (It would be good if missions could be separated from other
luascripts in some way, so that we can know which scenarios are available to
play and which scripts are just testing or tools)
Is there any recommendation on controls? I expect a joystick for weapon control
would come in handy, are there other posts which profit from specific control
devices
The first time I tried to build the AUR package it failed on a cp of the TTS
script, saying bin dir did not exists. I’m thinking this is some random problem
depending on the order the make threads are run.
The liberapay account is broken since they changed money operator and you did not configure the new one (it’s more like their money operator ditch them...)
I have a github account so I can open issues if needed.
The liberapay account is broken since they changed money operator and you
did not configure the new one (it’s more like their money operator ditch them…)
Ok... I think it should be fixed now. Patreon also works.
Joystick actually doesn't work that well for weapons, kind of hard to aim.
Mouse works better. Joystick or HOTAS works pretty well for NAV, rudder pedals too. Config is in share/snis/joystick_config.txt
Mostly, keyboard and mouse should be sufficient.
Ok... I think it should be fixed now. Patreon also works.
Joystick actually doesn't work that well for weapons, kind of hard to aim.
Surprising. I’ll have to test.
Mouse works better. Joystick or HOTAS works pretty well for NAV, rudder pedals too. Config is in share/snis/joystick_config.txt
Yeah I saw that I’ll hack into it as soon as I get my hand on a joystick. (I only have a fighting game joystick right now, does not seem appropriate . . .
https://www.amazon.com/HORI-Real-Arcade-Pro-PlayStation-4/dp/B00S
There is some code to enable custom built control panels, however it is not
well exercised, so probably has some missing features. There is a little
library in snis-device-io.h
and snis-device-io.c
and an example of using them
in device-io-sample-1.c.
The idea is you build a little control panel using
(say) an arduino and write a little program to communicate with your arduino
via the serial port to poll the state of control panel buttons, toggle
switches, potentiometers or whatever you dream up in the way of controls, and
then translate this state into actions transmitted to the game as defined by
snis-device-io.* The little library takes care of setting up a socket to
communicate with the game via snis_device_io_setup(), and you transmit actions
to the game via snis_device_io_send(). At this point, the traffic is one-way --
that is there is no information from the game sent back to the custom control
panels -- which would mean creating an engineering screen with a bunch of LEDs
to show power and coolant status would not be possible with the current code,
though it could be expanded of course -- there just isn't demand for it (except
my own demand, which has so far not been strong enough.) The farthest I got in
the way of building a control panel is one time I made torpedos be fire-able
via a toggle switch and an arduino: https://www.youtube.com/watch?v=w6vwbhHMmI4
Mostly, keyboard and mouse should be sufficient.
Ok, but I felt like putting a bit different controls on each post to make them
stand out from each other. From what I’ve computed Weapons would need only
joystick (or mouse), Engineering only mouse (or tactile), and Comms could have
only keyboard (if I find a way to click those buttons from keyboard. And a real
Red alert button would be extra cool).
I started to hack together a project to build an ISO for playing Space Nerds In Space from a USB stick (or live CD):
https://gitlab.com/MCMic/snislive
It works when built on my computer but I’m trying to get gitlab CI to build it for me and put the ISO for download.
The ISO is 800Mo and uses LXDE.
I noticed some money showed up in my paypal, presumably you did that. Thank you very much!
There is some code to enable custom built control panels, however it is not
well exercised, so probably has some missing features. There is a little
library in snis-device-io.h and snis-device-io.c and an example of using them
in device-io-sample-1.c. The idea is you build a little control panel using
(say) an arduino and write a little program to communicate with your arduino
via the serial port to poll the state of control panel buttons, toggle
switches, potentiometers or whatever you dream up in the way of controls, and
then translate this state into actions transmitted to the game as defined by
snis-device-io.* The little library takes care of setting up a socket to
communicate with the game via snis_device_io_setup(), and you transmit actions
to the game via snis_device_io_send(). At this point, the traffic is one-way --
that is there is no information from the game sent back to the custom control
panels -- which would mean creating an engineering screen with a bunch of LEDs
to show power and coolant status would not be possible with the current code,
though it could be expanded of course -- there just isn't demand for it (except
my own demand, which has so far not been strong enough.) The farthest I got in
the way of building a control panel is one time I made torpedos be fire-able
via a toggle switch and an arduino: https://www.youtube.com/watch?v=w6vwbhHMmI4
In general, I have found that it's ok for one or two stations to use wifi, but generally things work better with a wired setup.
As for things rendering weirdly, at different resolutions, that is quite
possible. It should detect the aspect ratio, and make some crude adjustments,
but it might not be ideal. If you look in snis_launcher, line 25, you can see
some different aspect ratios. Really though, the aspect ratio handling is kind
of hacky, and the positioning of widgets on the screen is also pretty hacky.
It's kind of a hard problem.
make update-assets
, those are the ones in the https://github.com/smcameron/space-nerds-in-space-assets git repository?
Regarding additionnal assets downloaded by
make update-assets
, those are the ones in the https://github.com/smcameron/space-nerd ... ace-assets git repository?
Does this auto-sync with my repo on github? I guess it probably does.
I have prepared a patch to link to this, but I'm not going to do it until
you at least try out the ISO image. I mean, *someone* should try it before we
inflict it upon the world.
That has nothing to do with comms power though... which comms power doesn't
do anything because I couldn't really figure out something good for it to do,
and realistically, a radio just doesn't need that much power to begin with. If
you have a good idea about what comms power ought to do, let me know.
Or, don't include the assets in the package, and include
util/snis_update_assets.sh in the package, then run that script to fetch the
assets separately.
for (i = 0; i < npassengers; i++) {
if (passenger[i].location == b->shipid && passenger[i].destination == sb->id) {
send_comms_packet(sb, npcname, ch, " PASSENGER %s DISEMBARKED\n",
passenger[i].name);
ship->tsd.ship.wallet += passenger[i].fare;
/* passenger disembarks, ceases to be a passenger, replace with new one */
update_passenger(i, nstarbases);
}
}
And here's the relevant "eject" code:
for (i = 0; i < npassengers; i++) {
if (passenger[i].location == b->shipid && passenger[i].destination == sb->id) {
send_comms_packet(sb, npcname, ch, " PASSENGER %s DISEMBARKED\n",
passenger[i].name);
ship->tsd.ship.wallet += passenger[i].fare;
/* passenger disembarks, ceases to be a passenger, replace with new one */
update_passenger(i, nstarbases);
continue;
}
if (passenger[i].location == b->shipid) {
send_comms_packet(sb, npcname, ch, " PASSENGER %s EJECTED\n",
passenger[i].name);
/* Player is fined for ejecting passengers */
The "update_passenger" code looks like this:
static void update_passenger(int i, int nstarbases)
{
static struct mtwist_state *mt = NULL;
if (!mt)
mt = mtwist_init(mtwist_seed);
character_name(mt, passenger[i].name, sizeof(passenger[i].name) - 1);
passenger[i].location = nth_starbase(snis_randn(nstarbases));
do {
passenger[i].destination = nth_starbase(snis_randn(nstarbases));
} while (passenger[i].destination == passenger[i].location && nstarbases > 1);
passenger[i].fare = compute_fare(passenger[i].location, passenger[i].destination);
}
Basically, the passengers get re-used, and placed on random starbases intent on travelling to other different random starbases.
CONTINUED ON PAGE 4.