HYPOTHESIS
After my post in the Katedra thread I was curious what actually makes a good baseline ammo HP to monster HP ratio, so I cheesed some debug info into progs.dat and loaded all the stock maps with it, as well as an utterly unscientific list of random custom maps I had at the ready.

EXPERIMENT
My metric was to add up monster health, except for zombies, then add up all ammo HP, subtracting the number of zombies from the number of rockets before figuring them in.
- Multipliers: 25 per shell, 9 per nail, 30 per cell, 180 per rocket.
- Starting shells were not counted.
- Totals include all ammo in secrets (there's no way to programmatically filter them out).
- No rockets vs Shamblers math was done. We can assume players don't use rockets on shamblers anyway.
- Chthons, Shubs, and crucified zombies were excluded.
- Numbers for E1M7 are inaccurate anyway due to meat fireworks.
- All numbers are skill 2.

http://lunaran.com/quake_mapammo.txt

Obviously there's a lot of assumptions inherent in this, and how hectic any given map's combat is (hecticness? hecticity? hectitude? let's go with hectitude.) the hectitude of any given map's combat is going to affect how these numbers actually feel in practice, but this is a good learning exercise to establish some boundaries.

CONCLUSIONS
- the first map of an episode always has way more ammo than necessary. (if I had counted starting shells these numbers would be a little higher even.) e2m1 wins the award for most excessive ammo placement, with enough ammo to kill every monster in the map 7 times. however, we can consider these numbers flawed, because enforcers' cells are all dropped in the first map but can't be used until the player has a lightning gun.
- e1m3 wins for most rockets, with 101. (you can only carry 100.)
- ratios otherwise hover around the 2.0-2.5 mark. episode 1's are the highest, episode 4 the lowest, predictably, although the plethora of quads in e4 probably offsets that.
- even though players of one-off custom maps don't have the benefit of bringing all the previous map's ammo with them (because there was no previous map), the handful of custom maps I checked were still a little more scarce with ammo than the id maps.
- the ammo to monster ratio in Katedra is just under 3:1. 

CORRECTION
- e2m4 wins for most rockets with fucking 118 

smqe08d_lun.bsp 

Very interesting. 

for science! 

As fascinating to study these numbers are (and I admire this kind of academical approach to Quake's gameplay), ultimately they're not very useful as benchmarks, because of how much maps vary.

For instance, jam2_lunaran's ratio is 1.9, and could's is 2.7, which would suggest that the former is less "balanced" than the latter, and therefore harder. And in reality it most certainly isn't.

Perhaps this code could be expanded to calculate the maximum monster damage/health pickups ratio? 

If there are tarbabies, it needs more ammo. 

 

This is really interesting stuff. Definitely something the more studious mappers are likely to use 

You woke the troll again >___________<

OTOH,

- e2m4 wins for most rockets with fucking 118

Made me lol far too much. The indignation of it all! :) 

So, NarNar Katedral was not so unbalanced after all. Believing your metric, maybe ammo would just be better positioned. 

That's a good point. The raw math is fun but placement and pacing is also key. 

Katedra throws a dribbly stream of monsters at you in wide spaces up until the end, so it's not difficult to be leisurely and precise and wind up not needing much more than 1:1 on ammo. Again, the map's design matters, and for that map 3:1 was probably rather high. Everyone's experience with it seems to bear that out.

The big criticism of RPGSP1 right after it came out was that everyone ran out of ammo in the last battle, and it's one of the lowest on the list at 1.5. He then released COULD two years later and probably overcompensated with too much ammo. ikspq1-4 are also below the average, but on skill2 ikka tended to pack his maps with fucking wall to wall ogres so it was pretty hard to miss anything.

I would have said after looking at these numbers that the best target for a map that feels resource-strict and tense but is still fair and entirely beatable with a bit left over would be a little under 2:1, so, the ammo in jam2_lunaran is obviously perfect. Terra1-7 were very well balanced, and once you get out of the dimension of grunt backpacks, the episode squarely averages out at 2:1.

Perhaps this code could be expanded to calculate the maximum monster damage/health pickups ratio?

If the ammo thing is an inaccurate yardstick, this would be a complete crapshoot.

Upload The Progs.dat
for science!

I actually didn't put it in the id progs. I'd have to redo it in the stock messy item and monster code, so, I could do that later I guess, but not everyone seems terribly impressed with science. 

I suspect scampsp1 would've had a better ratio if the end arena was cut off... I gave way more ammo than you'd ever need in the last 2 rooms because by that point, conservation of ammo wasn't really the intended challenge anymore.

These are cool numbers though. I think you are right that a 2:1 ratio might be just about right, and I think that's roughly what I mentally shoot for even without seeing the actual numbers. 

I'd have to redo it in the stock messy item and monster code, so, I could do that later I guess, but not everyone seems terribly impressed with science.

Just finish lunsp2 faster, duh! 

<Daz_> czg fucking make id let machine games make the inevitable Quake remake ok?
<czg> no. 

Oh please, no. No Quake remake. Oh god ... 

 

with the stellar job they're doing with Doom we'll all be clamouring for them to remake Quake. 

guys, we're safe, read this fanfic to find out why:

doom4 comes out. as soon as people start reading reviews, sales tank just like what happened for rage. bethesda finally has to confront what a turd they purchased, and they see the negative fan reaction to doom4 as a narrow window in which they can close id software's doors without backlash. 

https://pbs.twimg.com/media/CVoMyrWWIAARpbX.png:orig 

Nothing wrong with the IP, it should be able to absolutely shit money. But you need to get people who actually know and care about the franchise to make the game.

Wolf TNO was an absolutely incredible game. Get those fuckers to make Doom. 

Wolfenstein TNO was great, I really enjoyed it. I plan to give it another play sometime soon.

I never thought Rage was bad. Actually I thought it was pretty good. It certainly could have been much better, but it was far from bad. 

I enjoyed Rage. It got a repetitive after a number of hours, the game ended very abruptly though. It's like they just shipped it without finishing the last 5 levels or something. 

While Wolf:TNO is in fact great, it's also very different from Wolf 3D. IMO there is no evidence that anyone currently sad about id's new-Doom would be happy with the new-Doom that machinegames would make.