Archived discussions: 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012

Request for a publication[edit]

Hello, Ryan. Doom II is quality stuff, so I have to ask a Reviewer to get my change published. With this as your guide, could you give your judgement about the updated Compet-N table in the article? I added a separate table to my subpage; the content there is identical to the one in Doom II but the listing order is from the Compet-N site. If you make comparisons between the source site and the table, it would happen more conveniently via the subpage.

I've checked the data once myself and tested that the links work. There were no problems with those, but there's one questionable thing: according to the point policy for game movies (MAP01 - MAP30/32, "ALL"), some of the categories have two or three current records instead of one (they give a varying amount of points). However, I thought to only add the fastest records of the categories; that is the way things have been done in the wiki so far, and I'm pretty sure that the majority of readers are more interested in seeing fast playthroughs than knowing what kind of awards the speedrunners have had. Still, it's just an opinion among others.

Feel free to comment on your decision. --Jartapran 20:27, 10 May 2013 (UTC)

It's been years since I looked at the records, but I think all your ideas are workable.  Unfortunately, I can't help directly, because I'm not a Reviewer (see Special:Listusers).    Ryan W 00:02, 11 May 2013 (UTC)
AFAIK, Administrators are Reviewers by default. How else could you have done that to Unmaker's edit? :) Now, I cannot tell for sure, but I guess revisions turn into quality ones when the review grades (or some of them) reach a higher level in comparison with the Editor grades (Sighted/Basic/Moderate/Acceptable/Good/Concise). For instance, the last quality revision in Doom II was by Quasar with the grades Well sourced, High and Concise. [1]
If you're still unable to help, fine. Thanks anyway. --Jartapran 04:33, 11 May 2013 (UTC)
AFAIK, Administrators are Reviewers by default. -- not by default, I had to whine about it to get it to change so that admins be granted reviewer rights. --Gez 14:12, 11 May 2013 (UTC)
You are right; I totally forgot about that thread.  Apologies.  I'll take a look when I get home.    Ryan W 14:17, 11 May 2013 (UTC)


If you're on board with what appears to be a tiny consensus in favor of keeping the Eidolon article, I'd like to make some enhancements to it to de-emphasize the protracted description of Hexen II battle tactics, which are fairly off-topic here, in favor of more descriptive prose on his appearance, behaviors, place amongst the Serpent Riders, actions on the world of Thyrion, etc. Some kind of pic would be a nice touch too, IMO. --Quasar 20:16, 24 June 2013 (UTC)

Hi Quasar.  If we must discuss him, that's the way to do it.  Frankly our standards for inclusion still seem nebulous to me.  In a thread like that, however, with low turnout and at least one "disinterested" keep post, and not a copyvio or anything, we've generally kept the article assuming potential for improvement.  So feel free to close if you want.    Ryan W 02:00, 26 June 2013 (UTC)


I noticed you added the telefrag screenshot back to the article. Unmaker suggested that the picture does not depict a telefrag, and I concurred after making a test (see the screenshot below). I expected Axdoomer to reply to the matter, but he never did. Thus, the matter was left aside, and the validity of the screenshot has remained uncertain since. Now that you put it back to the page, I think it should be settled.


The zombieman§ stands behind an invisible wall in the modified E2M1 (so he was stationary when the screenshot was taken). When I surprised the monster by teleporting behind it, he didn't become telefragged. And considering that the gibbed player stood further away from the teleporter pad than in my picture, and telefragging doesn't include momentum, I agree that the phenomenon in the picture isn't a telefrag, but something that Axdoomer inadvertently thought to be one. I believe that what actually happened in the picture was what Unmaker said: the gibbed player flew through the teleporter and still had momentum left when he arrived to its destination.

§ I also tested with Our hero, and had the same results as with the zombieman. --Jartapran 19:15, 13 March 2014 (UTC)

(edit conflict x 2) That's excellent reasoning, and this may be one of those topics (few) which really needs a video rather than a single frame to be demonstrated.  Possibly I have been grasping at straws to avoid nominating images for deletion, because people find that unpleasant.  :7     Ryan W 19:30, 13 March 2014 (UTC)
I have to agree. --Quasar 19:29, 13 March 2014 (UTC)
Hi Quasar.  I saw your first reply, and it is probably correct but IMO not relevant.  :>   Ideally an article should be self-contained, which means the first screen shot should be a direct, immediate depiction without requiring the reader to visualize unseen events or ponder technical details.    Ryan W 19:38, 13 March 2014 (UTC)
I didn't read Jartapran's post fully enough to realize how well he'd tested it, and was trying to rationalize how it might still be a telefrag, so I was most likely wrong :) It remains vaguely possible that the shot is a telefrag in a source port that has wider players for some reason, but, we wouldn't want to use that as our example even if it were the case. --Quasar 19:43, 13 March 2014 (UTC)
If I were a newbie who had only DMed and was reading up on the classic maps, I might even think the dead player was gibbed by a rocket and flew in from the left after the teleportation.  DM moves pretty fast after all.  (No, there aren't multiplayer weapons on E2M1 but you get the idea.)    Ryan W 19:53, 13 March 2014 (UTC)
Hmm. Now I'm not that sceptical about the authenticity of the picture in question. I had done some inquiries after it was uploaded. The telefrag logic in the code checks if the colliding entities' boundaries overlap, but they should not necessarily occupy the exact same spot. If so, then we have a nonzero vector along which the damage will be inflicted (and momentum add is directed, according to P_DamageMobj). I'm not sure if the engine will permit the actual movement (it may consider the entities being stuck in each other at this moment), so probably my assumption is wrong.
But there is one more thing. Just at the time of the debacle I was playing Memento Mori 2, and something uncannily similar happened to me on MAP16. I took the teleporter after the blue door and arrived on top of the column at (480, 160), only to telefrag a hapless imp (who happened to teleport up there before me). Astoundingly, his remains were flung quite a distance away, clear off the pedestal. I was playing under Prboom+, and complevel was set to 9, I think.
So I am kind of at loss now, but I agree that the screenshot in question doesn't exactly portraits the subject, too much leeway and ambiguity. Unmaker 20:55, 13 March 2014 (UTC)
It'd be smart to say that "most of the time" the object should not have momentum inflicted on it, due to being inside the inflictor. Doom is excellent about not making things absolute, though. For every law it tries to enforce, it creates accidental exceptions. --Quasar 22:06, 13 March 2014 (UTC)
Sorry I haven't seen the discussion about that. I know it's been months. Yes, this is a real screenshot. I was deathmatching in E2M1 and recorded a demo. Player 3 (brown) stomped on Player 2 and his shredded corpse was thrown away. If you look correctly, you'll see a pixel from the brown player's gun at the right of the teleporter's green wave. I couldn't take a better screenshot where we could see player 3 because player 2 is thrown away so fast, he goes out of the screen so there's no point. If we could see player 3, I guess it would be much clearer that there is a momentum. As far as I remember, there are no rocket launchers in this level, so it's impossible that he was killed by a rocket. I'll do a video right now. --AXDOOMER 6:01, 16 September 2014 (UTC-05:00)
Me again. Some explanation I wrote somewhere else: Teleportation set thing's momentum to zero, so any momentum caused by a rocket's impact or a rocket's explosion applied to a player would be zeroed after teleportation.
Here's a piece of code from Doom: thing->momx = thing->momy = thing->momz = 0;
I uploaded a video to Youtube regarding this issue a few minutes ago (Link: I only have this explanation for the player being thrown away (, but it's only based on testing and not on the source code. I could reupload the image, but I see it says it was removed for "Copyright Violation", so I'll just wait.
--AXDOOMER 7:09, 16 september 2014 (UTC-05:00)
I had a severe flaw in the test I only now noticed, so I owe everyone an apology. I admit your screenshot was correct. At this point, I don't oppose re-uploading the screenshot and using it in the article, but I cannot comment on the copyright violation marking. --Jartapran (talk) 00:49, 17 September 2014 (UTC)
I have opened a discussion for undeletion of the screenshot. It was not removed as a copyright violation, but because screenshots are only fair use if they're being used to illustrate something in an encyclopedic context. Otherwise, we have no real legal reason to have them available. Now that it is accepted that the illustrated phenomenon is genuine, we can put it back in its rightful place. --Quasar (talk) 04:54, 17 September 2014 (UTC)
Per that thread, the Telefrag article has since had a video added, which strikes me as a much better way to demonstrate (a still image would look like any other gibbing).  It seems therefore that resolution has finally arrived.    Ryan W (talk) 01:24, 18 September 2014 (UTC)

id logo removal[edit]

Hello Ryan, I'm concerned for the removal of the id logo in this edit of yours. I think the inclusion of that logo there is covered by the definition of fair use and, if that's not the case, the logo should be completely removed from the wiki. --Kyano 23:45, 15 March 2014 (UTC)

Hi Kyano.  Fair use is not a property of an image, but of a particular situation where the image appears.  For example, a newspaper might place the logo of a university next to a sports score, but not in the margins of articles every time an alumnus is mentioned, as that would almost always be unrelated to the content.  A more topical example is using a screen shot to illustrate a bug, on a page about the bug.  Using the same screen shot as the background of my user page would not be fair use, being entirely decorative and irrelevant to the wiki's purpose, but that doesn't invalidate the first case; they are considered separately (see the second bullet under "Common misunderstandings" in your link).
Regarding the id logo, I believe that copyrighted images should not be used when we can make essentially the same point with text alone.  Using the logo in id Software serves the main purpose of the article by helping to identify the subject, similar to the sports score example (which is not to say we couldn't be sued, but logos are routinely used this way on Wikipedia with its far stricter image policies).  Putting the logo in the template does not serve the main purpose of the template by helping the reader find the desired article, any more than text alone would do, so it seems unnecessary to assume the additional risk.
Obviously, unless we end up in court someday, all we have on this issue are opinions, including my own.  If you or StoneFrog or anyone else disagrees, I'm not sure what the next step is.  I have researched copyrights myself only because I can't afford to hire an attorney on the site's behalf, and I think the Doom community is generally appallingly sloppy about copyrights, based on (apparently) the idea that we will never have a major incident because we have never had one before.  Maybe I'm wrong though... if we could manage to hold a thorough discussion about fair use on the wiki, including some of the port maintainers who have to deal with it all the time during development, I'd certainly abide by whatever guidelines came from that.  To date, such discussions have always fizzled out.  Therefore, each of us (meaning everybody, not admins) just takes their best guess as to what will help the project.
I am not a lawyer, and this is not legal advice.    Ryan W 01:51, 16 March 2014 (UTC)
I preferred having the logo myself, it looks naked now. I don't think it was a problem. --Quasar 03:56, 16 March 2014 (UTC)
Thank you Ryan for this explanation. I'm not in the position of choosing what should and what shouldn't be in this wiki, even though I personally think there was no problem, it's better to play safe. "I think the Doom community is generally appallingly sloppy about copyrights, based on (apparently) the idea that we will never have a major incident because we have never had one before." Yes, but we now have Bethesda. --Kyano 10:17, 16 March 2014 (UTC)
Well, I'm not in that position either; no single person is (sorry if sounded like I was overruling you).  The problem is, this is one of the few issues where our collective choices aren't the final word on the matter, because the real world can intervene.  We could have a unanimous vote that our current situation was OK, but that wouldn't change the case law, or Zenimax/Bethesda's plans (as far as I know, neither is at all interested in a relationship with the community, e.g. [2]).    Ryan W 15:23, 16 March 2014 (UTC)

As far as fair use goes, the image is on the wiki anyway. Removing it from a template does not change this fact, so it doesn't mitigate the risk. The id template is semantically strongly associated to id Software, so it isn't a misleading or gratuitous use of the image. Finally, even in the extremely unlikely case where id or zenimax objected to the presence of this image in the template, they'd start with a cease & desist rather than going directly to a lawsuit. To be honest, that change seemed very unnecessary to me. But I am also not sure about the rest of the change (removing {{prettynavbar}} and nesting a table inside the table); none of the other navboxes look like this. --Gez 15:40, 16 March 2014 (UTC)

My research suggests the inclusion of the logo falls far within the bounds of trademark (not copyright) fair use for corporate logos. If there is consensus, I am in favor of restoring it. --Quasar 15:56, 16 March 2014 (UTC)
Yeah, the trademark issue should be OK either way, because we haven't modified the logo and we only use it as part of documenting the real company and its products, not in promoting our site or someone's personal project.  I suppose Zenimax could theoretically ask us to use the current logo instead.    Ryan W 16:30, 16 March 2014 (UTC)
Gez, I was hoping you'd see this discussion since it takes me forever to figure out template formatting.  :>   I'd have rewritten it as an HTML table, only that seems to slow MediaWiki down (IIRC you and others have also objected to those as being hard to maintain).
the image is on the wiki anyway  — I'm sorry, but I think this is oversimplifying.  We have screenshots of monsters to show what they look like.  That doesn't automatically clear the image for anything, so that I could crop it, draw a crown and a goatee, and smack it on my user page with the caption "i am the imperor lololol".  Granted, the usage I removed was not totally frivolous like that and related peripherally to the encyclopedia content, but that was the principle.  Strictly speaking, if it came to a legal filing, the hosting and embedding could be contested individually.  If I make 20 copies of a CD and put them in my closet and don't tell anyone about them, that's still a violation.  Following Wikipedia, we generally ignore that point and host images locally if they have at least one agreed-upon mainspace use, because external images would be a thousand times more troublesome.    Ryan W 16:30, 16 March 2014 (UTC)


I believe, regarding interwikis, that the only one I ever disabled was one that would link to the old incarnation of this site. The community one, and ones to specific wikia wikis which had no known external alternative, I left in place deliberately, in part to avoid broken links. One thing I have to stress is that, for referential integrity purposes, I cannot delete rows from the table. I can only insert new ones, or change the URLs of existing ones. The best I could do to disable ones with no alternative URL would be to make them generate an invalid URL, and I don't think that would be good for the wiki. --Quasar 21:58, 25 March 2014 (UTC)

I don't think anyone ever proposed losing every Wikia row, only those for the "community hub" subdomains which would have no conceivable uses in articles.  At the time, you were also willing to accept invalid URLs [3], although I can definitely see the other side of that — the result is awkward and ugly, and we should just dig up a more appropriate link in each case.    Ryan W 22:41, 25 March 2014 (UTC)

Discussing deletion off target article's talk page[edit]

Please try to keep relevant discussion on deletion proposals confined to the appropriate article's talk page, as otherwise I'm chasing the discussion all over the place. --Quasar (talk) 16:40, 27 February 2015 (UTC)

Obviously.  This particular case meets speedy deletion criteria 4 and 15, however, so I was posting to Jartapran's talk as a courtesy (in case he remembered something I didn't).  That said, it's not "uncontroversial" if someone objects, so I won't delete.    Ryan W (talk) 18:07, 27 February 2015 (UTC)

Youtube'ing walkthrough LMPs[edit]

I admire your dedication towards making "encyclopedic" .LMP files for Doom's levels. It would be nice if they were more easily viewable by ordinary web users of the website as there's obviously a barrier to entry with downloading a WAD, learning how to play it back, etc. There are also disadvantages to the LMP format - eg. inability to rewind or fast forward. So it would be nice if we could make them available as embedded videos. Putting them on Youtube would be the most obvious solution.

Ideally I'd quite like to come up with a script that can translate LMPs into videos and upload them to Youtube automatically. Chocolate Doom has an experimental branch that will dump videos, though it may be that something like PrBoom+ would be more appropriate (widescreen fits better for Youtube, and it's nice to have the higher res).

Anyway, curious what you think about it. Do we have any numbers for how many LMPs are on the site at present? Fraggle (talk) 01:12, 25 May 2015 (CDT)

Hi Fraggle.  It's occurred to me, yes, and you make good points about practicality.  Some platforms don't support LMPs at all, or the user may find it helpful to watch on a separate screen (kids these days).
The counterargument is that the barrier is OK because some content is copyrighted.  I for one don't consider my work important enough that the potential benefit outweighs the risk, given Google's past behavior.  However, most of our LMPs are freely licensed, so the uploader's permission isn't needed for reuse.
Counting demos is a bit slippery.  Category:Original LMPs has 74 items, but a semi-manual wildcard search gives 106.  14 of those are -nomonsters runs with max secrets, 12 are "encyclopedic" walkthroughs, and 46 are concept art bullshit.  :D   Nearly all the others are bug demonstrations.    Ryan W (talk) 13:05, 25 May 2015 (CDT)

Account creation captcha[edit]

Thank you very much! —The preceding unsigned comment was added by (talkcontribs) .

Er, no problem.  :>   Happy editing!    Ryan W (talk) 16:10, 29 May 2015 (CDT)


You can still find all the info about EmbedVideo at mw:Extension:EmbedVideo - it has changed quite a bit because a new maintainer took it over and rewrote the entire extension, pretty much. We were in fact severely behind (v1.0, vs latest v2.2.6). I had to upgrade the extension due to the version we had not properly supporting the urlargs parameter, which I've added to Template:Youtube so that annotations are disabled. The video on Barrel suicide was 90% covered by large text boxes due to User:Fraggle's well-meant attempts to ward off a hoard of idiots that were making outrageous claims about that video on youtube (it would have been better to just disable comments on the video and let them fuck off, IMNSHO) --Quasar (talk) 09:55, 21 July 2015 (CDT)

Thank you, but I can't even locate the original issue now (I think it was someone trying to use one of the obscure services), and honestly that's a good sign I wasted my time.  If and when we actually see a correctly formatted call that doesn't work, I'll take another look.    Ryan W (talk) 22:29, 21 July 2015 (CDT)

Old SEO page[edit]

The project is still "active" but that list is very out-of-date at this point as it was based on what were then - a couple years ago already - the most often searched phrases that returned us in the search results but at positions of 9.x, meaning we were at risk of falling off the first page of hits on Google. If I were to pull the list now, there'd invariably be a lot of differences. But also as we discussed on IRC, various things on that page have been misinterpreted or taken in less than useful directions too and I'd like to avoid that happening. Either way there are sure to still be things on that list that could be visited and improved which is why I went ahead and undeleted it. I personally never stopped working on the "project" there because it's ultimately part of improving the site overall. --Quasar (talk) 11:17, 27 September 2015 (CDT)

Hi Quasar.  Good to hear that work is ongoing, because it's quite an important task IMO.  Nothing I've said previously was meant to contradict that; I apologize if it sounded so.
My main question is how non-search-engine-experts can help.  By editing, or by staying out of the way??  You said on IRC that many improvements don't require a total rewrite, only a well-placed disambig or page move.  Great, less effort per article, but smaller edits must be hard to delegate, because in the 10 minutes it would take to offer advice you could just do it yourself...
During the prior discussions, I began revising Plutonia [4], which I think does need a rewrite because it's a core topic, the title isn't going anywhere, and almost all the 2011 content comes from one source.  I would have sandboxed it first, hoping for additions/corrections from people who were actually in the community back then.  Was I on the right track at all (at least in theory)?    Ryan W (talk) 19:12, 29 September 2015 (CDT)

Preview button[edit]

That would be fair criticism, if it weren't for the fact that preview mode doesn't display the Categories bar. ;)
Now, if this omission could be remedied, then this would indeed help avoid mishaps there in the future. --Xymph (talk) 08:59, 5 March 2016 (CST)

It actually does, at the very bottom below the save button and list of templates.  I'll grant that probably seems counterintuitive at first.    Ryan W (talk) 09:05, 5 March 2016 (CST)
P.S.  Not one, but two open bugs to change this behavior [5] [6], so you're hardly alone.    Ryan W (talk) 09:11, 5 March 2016 (CST)
<looks almost below bottom edge of screen>... oh, there. Well, thanks for the pointer then. --Xymph (talk) 09:29, 5 March 2016 (CST)

Compet-N questions[edit]

Can we count on the UV Max category to be always present? Looking at the article for Doom E2M8, there are pretexts for not having UV Speed, UV Respawn, and Pacifist. In maps without secrets, there's no difference between NM Speed and NM 100S. Some maps are not possible in Tyson (Looper's weird trick to pistol-shoot the head in Icon of Sin hasn't been replicated yet for the Final Doom IWADs). I'm planning something but, to avoid headaches and inefficient code, I need at least one category that will never be missing. Same question for UV Fast. --Gez (talk) 11:44, 20 March 2016 (CDT)

Short answer: UV max is safest given that the current WADs all have it.

I think I see what you're doing; I didn't launch into that because in the current situation, extra links don't give 404s, only lists with zero entries each.  The speedrun community is generally very optimistic about future discoveries (starting with MAP07 pacifist, I believe), so IMO it doesn't hurt to acknowledge such by assuming table-fillers may be just over the horizon.  :>

Anyway, I've never seen documentation of which category is maintained when two are equivalent (the official instructions have contained errors in this regard for a while, and I even exchanged emails with AdamH about them), so probably no category is 100% guaranteed to exist.  This post is just thinking aloud.

E2M8 (perhaps also Heretic E5M8) may be the most limited case, where

  • no secrets are required;
  • 100% kills are necessary to exit (these two together mean UV speed or UV max must be open, but not both);
  • infighting can never result in 100% kills.

I'm sure some mapper has plugged the MAP30 Tyson loophole by simply burying the head a mile below ground.  :D   As you say, that leaves UV -fast... BUT that category is no longer accepted for episode runs using CnDoom [7].  That would make an exception, for instance, of the proposed Boom-compatible WADs which may eventually be added.  Or perhaps AV episodes 2 and 3 are so difficult that no one will go to the extra trouble of using a DOS emulator!

HTH, and thanks for your markup expertise as always.    Ryan W (talk) 13:40, 20 March 2016 (CDT)

Alright, done, what do you think of E2M8: Tower of Babel (Doom) and Template:Competnmap‎ now? (Also, are we really supposed to be using divs instead of colons in talk pages now?) --Gez (talk) 16:57, 20 March 2016 (CDT)
The template appears completely shipshape now, and the "parser data" of E2M8 hasn't exploded.  Fantastic!  Of course I will test further as I run across other applicable maps.
(About the divs, I don't claim to comprehend the underlying issue — something about MediaWiki generating non-standard HTML?  When we reach a supported MediaWiki version I suppose a bug could be filed, assuming manual formatting is still officially sanctioned in the first place.  :>    Ryan W (talk) 15:20, 21 March 2016 (CDT)
P.S.  Yikes, never mind, the issue was noticed in 2006: [8].    Ryan W (talk) 16:01, 21 March 2016 (CDT)
<dl> is a semantic HTML tag which, as per the HTML 5 standard, must be followed by one or more <dt> / <dd> sequences before the closing tag for the dl - it is called a definition list. The abuse of ":", which produces an orphaned dd tag (which is invalid HTML), to produce indentation on MediaWiki talk pages is an extremely unfortunate historical accident caused by lack of anything else suitable. I am neither suggesting nor requiring that we use the divs with class dw-div-dl on talk pages because it's impractical - it IS useful however for indenting your comments if they're going to contain extensive markup like ul lists, because those do not cleanly cooperate with a dl indentation. However, dw-div-dl is required for use in all other namespaces. Invalid HTML is a negative signal for page ranking and can also cause problems for tools that work with the data in an automated fashion. This is why I went to the trouble recently of expunging its use from all templates. --Quasar (talk) 00:47, 22 March 2016 (CDT)
Yeah, that's about what I remembered from prior discussion.  So it's a nonzero (but hopefully small) search rank hit, because user talk pages are indexed.    Ryan W (talk) 16:28, 23 March 2016 (CDT)


Do you not consider our custom text being replaced with text copied from an unreleased game an issue? -- 10:33, 28 April 2016 (CDT)

A loaded question.  In a vacuum, of course I do.  In this specific case:
  • the newbie will likely get blocked (persistent refusal to discuss), giving others ample time to review and salvage any legit additions;
  • I know nothing about Doom (2016) other than its title and publisher, so I couldn't fill gaps left by undoing the recent edits/uploads;
  • in two weeks we'll be flooded with INFORMED input anyway, so it will come out in the wash; and
  • the current content presumably improves search rank in the meantime.
When Doom 3 was released on Xbox, we saw very similar newbie editing.  The site didn't explode (although as you've noted, some of that material remains unpolished to this day).  You'll have to forgive me for acknowledging community consensus, i.e. copyright concerns are of marginal priority, very rarely grounds for swift action.    Ryan W (talk) 17:54, 28 April 2016 (CDT)
All fair enough points (except the consensus about rights issues one, I have to still fight against that as much as it sucks). However as I mentioned on IRC this was prompted by a misunderstanding/bit of a rash "oh shit!" moment and I have my regrets so I apologize here as well for jumping in your direction; I won't assume you're active just because I saw an edit 10 minutes ago, I tend to forget not everybody goes on the 6-hour editing binges like I do ;) --Quasar (talk) 19:56, 28 April 2016 (CDT)


I've been trying to replicate the damage data behind the histograms such as those used in the SSG article, but I can't replicate the same distribution you've got, so there's probably something I'm overlooking somewhere. My current assumption is that between two consecutive pellets, there are 9 extra P_Random calls (two from horizontal spread, two from vertical spread, three from bloodsplat-or-puff spawning, and the last two from P_DamageMobj. Meaning that if the first pellet uses the RNG value at index 0, then the second uses RNG value at index 10, the third at index 20, and so on until the 20th at index 190. That however gives me a distribution with a result peak at 205 total damage (48/256 probability) whereas the histogram shows a peak at 185 damage with about 18%. If you still have whatever code you've used to get the values I'd like to compare with my monstrous spreadsheet. --Gez (talk) 12:23, 23 February 2017 (CST)

Wow, it's been a while since someone took those seriously.  :>   I have the code, but it is some kind of encoded Mathematica archive, so actually I can't read it myself anymore!  If anyone desperately wants to import it, I'll post it somewhere, but you're probably better off starting from scratch.
What you say makes complete sense, and sounds consistent with my goals at the time.  I do remember getting stuck on the SSG computation: I couldn't determine how many P_Random calls were in the vertical spread.  Perhaps I never actually solved it, and the current version still assumes zero additional calls among the damage (try that first).  I'm no longer surprised when data mistakes linger for a decade, because it is so much work to double-check, and I'd be the first to congratulate you if you replaced my image.    Ryan W (usually gone) 17:31, 23 February 2017 (CST)