Talk:Skill level

Clockwork Orange

 * The name of the fourth skill level, Ultra-Violence, comes from the 1971 film A Clockwork Orange. The word itself did not originate in the movie, however; it was first used a decade earlier in the Anthony Burgess novel (also named A Clockwork Orange) upon which Stanley Kubrick based the movie.

Is it not enough to say:


 * The name of the fourth skill level, Ultra-Violence, comes from the 1962 novel A Clockwork Orange, by Anthony Burgess.

?


 * I guess you'd have to ask the id programmers that one. :>


 * In other words, if the storm of sewage in "Mr. Hankey - the Christmas Poo" was reminiscent of the Mickey Mouse sketch in Fantasia, do you credit Fantasia or Der Zauberlehrling? There's something to be said for a particular, striking interpretation, so in principle it could be either. Ryan W 21:50, 22 Jun 2005 (UTC)


 * ok, how about


 * The name of the fourth skill level, Ultra-Violence, comes from the 1971 film A Clockwork Orange (itself based on the 1962 novel by Anthony Burgess).

Respawn delays
if (!respawnmonsters) return;

mobj->movecount++;

if (mobj->movecount < 12*35) return;

if ( leveltime&31 ) return;

if (P_Random > 4) return;

I am too fried from finals to figure it out exactly but it looks like it takes at least 12 seconds for monsters to respawn, then after that each tic it first checks if the leveltime is a multiple of 32 (so the bitwise-and is zero) and then it calls P_Random which goes from 0 to 255 and returns if the value is greater than 4

so i guess that means that after 12 seconds, the probability of a respawn in any given tic is 1/32 * 5/255 or 6.13e-4 which suggests that on average it should take 1632 + (12*35) tics or 58.6 seconds for a monster to respawn 08:41, 11 December 2007 (UTC)


 * This is interesting. I'd like to see a graph that shows the probability (as a percentage) of respawn on a second-by-second basis. To note, however: The probability of a pseudorandom number being less than 5 is not 5/255; it would be x/256, where x is the number of bytes less than 5 among the 256 pseudorandom numbers. Zack 19:22, 11 December 2007 (UTC)


 * Looked at the code again, the pseudorandom table is unsigned char rndtable[256], I sorted the values and got 6 numbers <= 4 (0, 0, 2, 2, 3, 4 - why the hell didn't they use all the possible values? 145 appears 5 times in the table).  That changes it to an average time of 51 seconds to respawn instead of 59...


 * Thanks - you beat me to it. :) I got the same result. 51.0 seconds (plus one third of a tic, which is negligible). By the way, you can sign your comments on talk pages with four tildes ("~") :)


 * EDIT: If it interests anybody or is deemed worthy enough for this article, 51.0s is the mean average respawn time, while 38.72s (38s + ~24 tics) is the median average (where 50% of all demons respawn in 12-38.72s and 50% respawn after 38.72s). This is determined by calculating T = 32 log(x) / log(250/256) + 12(35), where x is the percent of respawns that occur after T tics. Also, the mode average respawn time is ~12.5s. Zack 04:51, 12 December 2007 (UTC)


 * US$0.02: Careful there.  Any NM player will tell you that the averages are much less important than the fluctuations, because that's what determines how quickly you can backtrack through a level.  That said, if the article currently says the minimum is 8 seconds, and you think it's obvious from the source that it's 12 seconds, then I would suspect every number of originating from a pre-source-release walkthrough, wherefore you are absolutely correct that a rewrite is in order.    Ryan W 05:15, 12 December 2007 (UTC)

NM vs. UV -fast -respawn
Graf Zahl states:

I found one piece of code which is specific to the nightmare skill: if (gameskill != sk_nightmare) mobj->reactiontime = info->reactiontime; (This is in the function P_SpawnMobj) This is certainly demo critical but in a running game you will most likely not notice the difference. Aside from that there are only very few places where nightmare is checked but in   all of them the fast paramteter is also checked.

I looked through the code for other uses of the variable gameskill. No, I don't really know what I'm doing. :>    However, it seemed to confirm the final sentence of the above (either NM and -fast are checked together, or it's something very predictable, like initializing a demo, or checking difficulty class before placing a Thing). Thus I suggest that the first table in this article is incomplete without mention of this reaction time discrepancy... except that I don't know what info->reactiontime  means, so I can't tell the reader how gameplay would be affected. Ryan W 02:02, 29 January 2008 (UTC)