A plea for SI units

[icomeinpeace]

I recently got Terra Invicta and have been playing for about 50 hours. Fantastic game so far!

My one big annoyance up to this point is the use of "kps" (which I, as a European used to the SI system, read as kilopond-seconds) for Δv. Especially since this game makes an effort to take hard sci-fi seriously in regards to e.g. transfers I feel this is pretty immersion breaking. Can we please get (at least an option for) the "correct" km/s instead?

Also in some places acceleration is given in gram-seconds (gs) which is equally messed up. I can see the appeal of using gees as a familiar unit, but then perhaps write it as (milli)gees instead? Ideally in combination with the SI unit if there's enough space: "0.314 m/s² (32 milligees)". Or equivalently "314 mm/s² (32 milligees)".

Here's some discussion about using "gee". (I have a hard time believing anyone would actually use "g" in a technical text today since that's an SI unit for mass.)

Well, that's my plea. Thanks for the game!

[icomeinpeace]

Oh, and another one: "sec". The SI unit symbol for the second is "s".

[icomeinpeace]

I also noticed when starting a Skirmish (also perhaps at start of spaceship battles in game, something I haven't done yet) that the initial speed is set to a number of "mps". Miles per second?

This is the kind of thing that causes missions to Mars to fail!

[StrykeSlammerII]

Not being used to the SI system myself, I find g (or gees or g's) a common usage for standard gravity.
The opening paragraphs of Wikipedia's articles on "Gravity of Earth" and "standard gravity" both refer to `g` as a constant... but the "g-force" article clearly specifies that "The unit g is not one of the SI units".

Adding a display option for more strict usage of SI units makes sense to me, I'd upvote a feature request for this in the Discord

I will question whether TI counts as "a technical text" but not enough to argue that point

[icomeinpeace]

I don't mind the use of gees either as long as it cannot be confused with the standard units. Using "g's" rather than "gs" would be an improvement.

The use of p instead of the division slash does not fly in SI-land though; it's just wrong. Neither does omitting the actual unit (as in kps - kilo-what per second?).

I didn't know there was a Discord where feature requests can be made! I'll look into that. Thanks!

(Also, sorry for the late reply; I wasn't subscribed to the topic!)

[icomeinpeace]

I poked my head into the Discord but honestly, I'd rather spend my time playing the game than repeat what I've written here on Discord.

I assume that e.g. "kps" is something that just happened to carry over from some internal variable name, and that when someone from Pavonis reads this thread they'll think "Huh, I guess we didn't put much thought into that; let's add this to the to-do list to make the game look better!". If not, well... I guess it's not the end of the world.

(BTW, as a programmer messing with various technically oriented software, I've seen multiple times how programmers use unit names as part of variable names. Don't do that! Rather than "EV_kps" and "thrust_N", call the variables "EV" and "thrust", and just use the base SI units for the values. I.e. an EV of 19.2 km/s would be stored as 19200. Then add prefixes or do whatever conversions the user wants at display time. Makes things much simpler in the long run!)

[StrykeSlammerII]

Since you called out "gees" as a spelling, I've since noticed some places where the game spells it out like that. Apparently "gs" is ok but "milligs" is not?


"kps" feels to me like a possible Americanism. I recognize "ps" as "per second", though abbreviating "km" as just "k" here does seem a little weird. Maybe "kmps" is just too long? Or maybe it came from "klicks per second"?

A quick google search for `"kps" rocket` didn't find many relevant hits, *but* the first one (https://www.spacecentre.nz) says "The speeds of rockets is normally measured in meters or kilometers per second (m/s, km/s or kps)."
There were a few other relevant hits, including a Wired article from 2012, so kps definitely did not originate with TI

I'm still in favor of a "strict SI display" option!

[icomeinpeace]

Yeah it feels like like a bit of a mess. I saw a report in the Discord that antimatter production was reported in a's, which apparently was supposed to be ag (attograms). I hope they find the time to implement some global handling of the display of units.

"kps" feels to me like a possible Americanism. I recognize "ps" as "per second", though abbreviating "km" as just "k" here does seem a little weird. Maybe "kmps" is just too long? Or maybe it came from "klicks per second"?

I'm sure it is: mph -> kph (which I see disturbingly often; it's equally jarring every time) -> kps isn't exactly far-fetched. It's frustrating to me: if you switch to metric units, why not also switch to the well-thought-through and unambiguous SI system completely? E.g. it's pretty hard to do unit analysis on "kph"... Also, no conversion factors needed on calculations if base SI units are used.

Time to reach velocity v using acceleration a? t [s] = v [m/s] / a [m/s²]. It's easy to check that the units match (i.e. that I haven't switched v and a around in the calculation etc). Need to calculate motor power? Well if you have the rotational speed in the base SI units of radians per second, and the torque in newton-meters, just multiply them together: P [W] = ω [rad/s] × T [Nm].

The more we see things like "kps", the worse the understanding of physics in general will be.

Thanks for the support!

[StrykeSlammerII]

Regarding antimatter production: yes the prefix 'a' is used (among others), but the unit of space resources is elsewhere noted to be decatonnes, seemingly based on 1 boost = "1 decatonne into LEO".
This makes antimatter prefixes confusing for everyone


I haven't been able to easily find an etymology for "kps" in American rocketry so I won't comment further there
I would love it if anyone could share some additional reading on this topic.


Maybe I read these metric units backwards, because I don't have the same problem with unit analysis. But then, I am more used to 'p' -> 'per' -> divide than 'pico-' scale prefixes.
I can see how context is required though: would "kmps" be "km * ps" or "km/s"? And without the contextual understanding that "k"="km", we have your initial interpretation of kilopound-seconds!

When I (personally) go to do unit analysis though, I always write it out longhand. Too many hoops to jump through to get my computer to show nested divisions in a way that is clear to me!
(I first learned unit analysis in high school using Imperial measures as examples--converting miles-per-hour to feet-per-second types of things, before we moved on to actual science-and-engineering.)
Conversion factors are less relevant under SI, but IMO they still apply to time: days and hours (or minutes) are more intuitive than kiloseconds.

I don't agree with you on all points, but I appreciate the discussion! I've seen that TI has a widely international audience, so I hope that this will help the devs better understand your perspective.

[icomeinpeace]

Regarding antimatter production: yes the prefix 'a' is used (among others), but the unit of space resources is elsewhere noted to be decatonnes, seemingly based on 1 boost = "1 decatonne into LEO".
This makes antimatter prefixes confusing for everyone

Used consistently, i.e. if space resource production was written as "210.3 dat + 2.4 dat", and the antimatter as "5.4 ag + 0.11 ag" then, although those prefixes and units are perhaps... well, alien to many, at least there would be some way to figure out what is meant. ("It's all SI prefixes and units, look them up on Wikipedia.") Edit: I looked them up on Wikipedia, and realised I'd mixed up d (deci) and da (deca). Fixed now. Unless there's a reason for displaying resource production in decatonnes specifically, perhaps better to use tonnes instead. Or why not Mg?

I can see how context is required though: would "kmps" be "km * ps" or "km/s"?

At least in this case it couldn't be the former since prefixes are never stacked. I feel the reasons to use / instead of p are mostly 1) it's by definition the correct way to do it according to the standards (see e.g USMA, linked to by NIST), and 2) it stands out more than "just another letter" which differentiates it nicely from prefixes and units. So again, it helps with unit analysis, for example.

When I (personally) go to do unit analysis though, I always write it out longhand. Too many hoops to jump through to get my computer to show nested divisions in a way that is clear to me!

Qalculate is a great tool for "everyday unit analysis"! (And just a great calculator in general.) It can do things like this (calculation of time to heat up 1 kg of water with an electric heater under ideal circumstances):

Qalculate.png (25.11 KiB) Viewed 3428 times

Conversion factors are less relevant under SI, but IMO they still apply to time: days and hours (or minutes) are more intuitive than kiloseconds.

True. The dozen is just too practical since it's so easily divisible by integers! By 2, 3, 4, and 6. Great when speaking of time periods. In general, for display (and input) prefixes and conversions are required (but remember that the unit for minute is "min", not "m").

Another example is the degree (angles). While I'm happy with radians, I'm probably in a very small minority.