Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Aoi]

It'd be kinda awesome to see something as being worth 5 hours and 29 minutes of time, even better if it's 5 hours and 29 minutes of Aristocrat time (equivalent to maybe 2-3 times that much 'commoner' time).  Hell, you could even intentionally fudge the numbers a bit by random, with the severity decreasing as and when you develop better methods of timekeeping.

That reminds me of a project I put together in one of my Econ classes a few years back-- to illustrate a point, I converted the typical/average/approximate pay for labor in different positions and around the world, normalized to US minimum wage, and converted it to time.

It's actually rather depressing: the current life expectancy of somebody born now in the US approximately 79 years, or 700,000 hours. Federal minimum wage is 7.25, so, by this count, a life is worth a baseline of ~5,000,000$. According to Forbes, Tiger Woods made 78m$ in 2012. That entire life is worth 3 weeks of Tiger Woods.

Having said that... I like that idea, but would that that skill levels should be able to "downshift". Skilled labor can replace unskilled labor, but at the same cost, but unskilled cannot replace skilled, regardless of the case. In the above case, an aristocrat can probably do okay digging a ditch, but he sure isn't going to be turning out masterwork knit socks.

Number fudging can be graft. Improve policing and auditing processes to minimize!

[MoLAoS]

The teacher didn't have a problem with the fact that your hypothetical worker worked literally every hour of his life? According to my back of the envelope calculation where you actually only work a full time job .4*hours of your life, its really only 2million$ for a minimum wage worker.

Tiger Woods is an interesting choice. Entertainers occupy a pretty weird space. Their activity would be worth less than minimum wage in a pre-recording society. But suddenly they are catapulted to the top of the food chain. Also interesting is say, musicians compared to industrialists. A musician may generate 10million song sales, at a $1 a pop. But the music player company will generate effectively 3billion sales, if we use the iPod Touch. And the musician again will only make 1million, since the publisher and various other things like equipment and such will eat up 90% of the money. It would take 3000 hit songs per generation of music players to cause musicians to generate as much profit as Apple. And the publisher of course would have 10x as much as either of them in that case. Well, technically the various operating costs would limit the publisher and the hardware vendor, but still compared to the musician, they make a killing.

[Puzzlemaker]

Todays block post excites me!  Combat!

Will terrain be deformable for, like, explosions or something?

As for that trench problem:  Why not side-step the issue entirely by making the trench building-like?

Like this (or something:

Code: [Select]

#\$$   $$/#
##\$   $/##
###\___/###

# = Dirt
\_/ = dirt edge according to marching cubes
$ = space filled by trench object

It's sort of hack, but it would let you put a trench in without causing too much pain.

[lordcooper]

Or maybe barricades instead, if trenches turn out to be too much of a PAIN?

[Puzzlemaker]

Or maybe barricades instead, if trenches turn out to be too much of a PAIN?

No man, you have to have trenches.  I mean, what's a steampunk colonialism game without trenches!?

[Urist McScoopbeard]

Ya, how will we defend out encampments from zulu warriors cultists!?

[Citizen Daniel]

I'm pretty sure we'll have trenches.  We just have to, uh, create an algorithm to solve a really big, stupid linear algebra problem using our fixed point math system in a way that doesn't take 10 minutes.  This is what we do to bring you the muddiest of subterranean combats!

(EDIT: I guess that isn't really subterranean is it.  Well, they're trenches, whatever.)

Terrain is deformable, technically.  We have some limits on just how deformable, though, because the perspective of the camera is isometric, and if height differentials are too big, bits of the terrain get hidden behind higher structures, and that's really aggravating.

The thought of using objects to "fill in" trenches has been discussed, it's an option, but it would be a bit cleaner if it weren't necessary.

[Puzzlemaker]

I'm pretty sure we'll have trenches.  We just have to, uh, create an algorithm to solve a really big, stupid linear algebra problem using our fixed point math system in a way that doesn't take 10 minutes.  This is what we do to bring you the muddiest of subterranean combats!

(EDIT: I guess that isn't really subterranean is it.  Well, they're trenches, whatever.)

Terrain is deformable, technically.  We have some limits on just how deformable, though, because the perspective of the camera is isometric, and if height differentials are too big, bits of the terrain get hidden behind higher structures, and that's really aggravating.

The thought of using objects to "fill in" trenches has been discussed, it's an option, but it would be a bit cleaner if it weren't necessary.

Sounds fun, don't let the bugs bite!

[Boksi]

The marching cubes sound like some sort of eldritch abomination, really. I wonder what relation they have to the invisible geometers?

[Citizen Daniel]

Oh!  For the people who asked for links to the blog posts, here's today's.  As previously mentioned, it's about trenches.

http://www.gaslampgames.com/2013/11/13/combat-2-son-of-combat/

(Update: turns out we don't actually need to solve a "really horrible" math problem!  Our previous solution actually looks really good if we tweak it a bit.)

[Farmerbob]

Oh!  For the people who asked for links to the blog posts, here's today's.  As previously mentioned, it's about trenches.

http://www.gaslampgames.com/2013/11/13/combat-2-son-of-combat/

(Update: turns out we don't actually need to solve a "really horrible" math problem!  Our previous solution actually looks really good if we tweak it a bit.)

If you are having difficulties with performing calculations for trenches, perhaps it might be a good idea to simply maintain a lookup table of some sort if the number of equations you would potentially have to solve is small to modest?

You are doing a top down simulation, based on tiles, so it's conceivable that the number of equations you would have to solve might be small.

Even if the number of equations you have to solve is large, perhaps the more rigorous math is actually duplicated a lot amongst all the potential equations, allowing you to create the "North Trench Wall" constant, the "Northeast Trench Corner" constant, etc.

Sortof like fluid dynamics in real life

[ahappydude]

Dungeons of dredmor gave me so many hours joy so will back this with my lsat money later on, it will totally be worth it.. Cheers for thses blogposts, fun reading!

[Citizen Daniel]

The issue was with the fact that what we have is basically a height map that's generated by a voxel map, and the algorithm which generates that height map from the cube representation can't do sheer walls easily because, well, we didn't think there would be situations where we'd need a really stark angle, so we chose the fastest one we had lying around.  It'll still do cliffs, but it has some rounding behaviors at the top and bottom.  If we tweak the parameters a little though, it gets close enough to look good.  I mean, real trenches aren't perfectly angular anyway.

Edit: @ahappydude - glad you're enjoying them!

[PTTG??]

You guys seem to do a lot of "original research" in your design project. Have you found anything like the Fast Inverse Square Root property (Good ol' 0x5f3759df)? Possibly in your elaborate mathematical structures defining the angles of roofs or some chicken-related pathing optimizations? If you can't say, just blink it to me in Morse Code.

[Citizen Daniel]

No   There's honestly not a lot of motivation to chase those sorts of things to improve performance, not at this point in the project, and not unless you're obsessing over speeds the way that really competitive games do.  And those guys hoard their secrets these days. Nicholas and Micah spend a lot of time mangling the C++ compiler which is interesting though.  Really we have Canada to thank for giving us tax credits for doing weird experimental stuff to make simulations run faster on newer architectures.  Honestly, I'm somewhat convinced that you'll be seeing a lot of these methods in other games in the next 5-10 years.

Just got characters in-game to recognize each other as made of meat.  Good day.