Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Sign in to follow this  
xit-vono

Test and research demos

Recommended Posts

xit-vono said:

Here's a demo on plutonia map11 where one of the archviles is killed in 3 shots

Cool, we have proof. :)
So will the Doom wiki be changed? (and why does it say minimum 4 ssg shots to kill a vile?)

Btw, here's a map for controlled testing: 3ssgvile.zip

SSG is on all skills except UV. Play with the included 3ssgvile.bex dehacked file if you want a faster ssg and archviles that don't move or attack (so it's more controlled and doesn't take as long). I tried killing 800 viles in that map with the dehacked file and didn't kill any with 3 shots heh.

So, what are the odds? Does holding down fire or releasing and waiting until your shot is fired, then firing again, have any affect on the ssg damage? (I think it affects aim? or am I thinking about a different weapon like the chaingun?)

Share this post


Link to post

Do you know that Method has already killed a vile with 3 shots of SSG in 2004?
The evidence is this demo, though I noticed it only about one year ago when I tried af05.

By the way, I have done it only one time but I didn't keep the demo. This is because I thought everyone have already known it after watching Method's one.

Share this post


Link to post
TimeOfDeath said:

Cool, we have proof. :)
So will the Doom wiki be changed? (and why does it say minimum 4 ssg shots to kill a vile?)

Btw, here's a map for controlled testing: 3ssgvile.zip

SSG is on all skills except UV. Play with the included 3ssgvile.bex dehacked file if you want a faster ssg and archviles that don't move or attack (so it's more controlled and doesn't take as long). I tried killing 800 viles in that map with the dehacked file and didn't kill any with 3 shots heh.


Did you read notes under he table?

Doom Wiki, note under the table:
This table assumes that all calls to P_Random for damage, pain chance, blood splats, and pellet dispersal are consecutive. In real play, this is never the case: counterattacks and AI pathfinding must be handled, and of course the map may contain additional moving monsters and other randomized phenomena (such as flickering lights)

Probably your map contains no "other randomized phenomena", hence the above assumption holds and you indeed can't kill an arch-vile with 3 shots on your map.

Share this post


Link to post

I think I somehow killed a reverent with only 1 rocket in timeofdeath's reality map 01 demo.

Share this post


Link to post
gggmork said:

I think I somehow killed a reverent with only 1 rocket in timeofdeath's reality map 01 demo.

I think max rocket damage is 128 + 20 * 8 = 288. You probably hurt the revenant with your previous rocket.
BTW I checked af05-310 and it's real 3ssg kill (AV is thing #95)

Share this post


Link to post

TimeOfDeath said:
So will the Doom wiki be changed? (and why does it say minimum 4 ssg shots to kill a vile?)

My involvement in Doom Wiki is minimal and I haven't created a single word in the SSG article, but I suppose (like tempun) that the phrase "This table assumes that all calls to P_Random for damage, pain chance, blood splats, and pellet dispersal are consecutive" is the key. Open the "Fist page". Here you may find that the minimum berserk punches to kill a cacodemon is 2 (it should be 200HP damage+200 HP). But for a mancubus it is 4, not 3 (so they assume 200+200+200 is impossible; in fact, it is very unlikely, but possible).

So, what are the odds?

Ask Maes for estimated odds. Real odds should be insanely difficult to calculate. And pausing, shooting another thing in between, etc. should have impact on the random numbers responsible for damage, and maybe on the odds, too.

Share this post


Link to post

tatsurd - My mistake, I haven't seen that demo, thanks.
tempun - I did read under the table, but I'm not very smrt. :)

I had thought the minimum shots was 3 from when I played on ZDaemon a few years ago, but I guess that's different (ZDoom-based port). I *think* I finally killed one in 3 shots on the map I posted (with dehacked), after 825 tries (30 minutes).

I dunno, it just seems kinda weird to me to have that "minimum shots" info on the wiki when it's based on math, instead of being based on real play.

Share this post


Link to post

I just made a wad for this, and I tried to kill a vile with 3 shots while displaying their health and to record this to a demo. Although I spent 2 hours, the attempt ended in failure. I thought vile's body becomes a little wide when he starts to attack, so I can give the vile more damage. I don't yet know if it is true or not. The maximum damage which I gave a vile in 3 shots was 690. (3ssgfail.lmp, thing #22) Also, the maximum damage which I gave a vile in 1 shots was 245. The damage was not very rare, but it never happened continuously.

vdgg said:

Here you may find that the minimum berserk punches to kill a cacodemon is 2 (it should be 200HP damage+200 HP).

I did it twice recently. Here (Thing #97) and here (Thing #111)

Share this post


Link to post

TimeOfDeath said:
I dunno, it just seems kinda weird to me to have that "minimum shots" info on the wiki when it's based on math, instead of being based on real play.

Outside the "controlled lab case" method given in the wiki, actual cases exceeding those ranges can only be backed by demos or extended research. I'm guessing the editors haven't had any evidence till now and thus hadn't dared to guess 3 is possible. Someone should change the minimum to 3 and point to one of the demos, maybe xit's, in a clarifying note. (I would and I will if I get a bit of time to start editing the wiki again but can't guarantee it.)

Share this post


Link to post
tatsurd-cacocaco said:

I don't yet know if it is true or not.

Probably not true.

tatsurd-cacocaco said:

The maximum damage which I gave a vile in 3 shots was 690. (3ssgfail.lmp, thing #22) Also, the maximum damage which I gave a vile in 1 shots was 245. The damage was not very rare, but it never happened continuously.

af05-310: 210 + 245 + 245 (thing #95)

Share this post


Link to post
myk said:

Someone should change the minimum to 3 and point to one of the demos, maybe xit's, in a clarifying note. (I would and I will if I get a bit of time to start editing the wiki again but can't guarantee it.)

Don't forget to change the note "this table assumes...." then. Also "Min" and "max" columns in the table will become pointless: maxshots = ceil(HP/mindamage); minshots = ceil(HP/maxdamage)

Share this post


Link to post

Yeah, all that should be changed there is to add "unless specified otherwise".

Share this post


Link to post

whoa, how are you guys displaying the monster's health and weapon damage?? That sounds hella cool.

About the vile body: the sprites can be different sizes, but the radius and actual size of monsters never changes, right? So they can look bigger or smaller, but the size never changes?

Share this post


Link to post

Yeah, the engine size of the thing and the graphic are independent. The arch-vile is always 20x56 units to the engine.

As for seeing the hit points of monsters, in PrBoom+:

Additional command-line parameters for cheating/TAS functions:
"-trace_thingshealth ThingID [ThingID] [ThingID]" - displays the current health of the specified Things (in the HUD)
"-trace_thingspickup ThingID [ThingID] [ThingID]" - shows if the specified Things have been picked up (in the HUD)
"-trace_linescross   LineID  [LineID]  [LineID]" - shows if the specified Lines have been crossed (in the HUD)
"-trace_givendamage  ThingID [ThingID] [ThingID]" - shows the latest and total damage inflicted by any specific monsters or players (in the HUD)

Share this post


Link to post

Cool, thanks. Could you give an example on what exactly to write in the command line? I r dum.

This will also be useful for seeing how much health is taken away from a cyber in one bfg shot, with multiple attempts lols.

Share this post


Link to post

Where it says ThingID just place the number of the specific thing in the map. Like, if you want to follow the health of the shotgun guys in E1M1 that are behind the two blocks by the stairs you could use:

prboom+ -playdemo e1m1-029 -trace_thingshealth 102 103

Share this post


Link to post
myk said:

Yeah, the engine size of the thing and the graphic are independent. The arch-vile is always 20x56 units to the engine.

OK! I understand it now.

tempun said:

af05-310: 210 + 245 + 245 (thing #95)

So I tried many times to occur it. A lot of luck is really necessary.

Share this post


Link to post

Has anyone calculated the maximum damage of the rocket launcher as well? According to the Doom wiki, the Arch Vile takes 3.72 rockets with an SD of 0.55, but the minimum number of shots is actually 3 instead of 4. I've seen quite a few vanilla and boom demos where an arch vile is 3-shotted by rockets, versus the rarity of the 234-245 damage ssg shot (since the average needed is 233.3 to 3 shot an arch-vile).

So, the pseudo-rng for the launcher must have a greater average damage calculator than the ssg (not including the splash damage immune monsters). I assume this is because of the greater dispersion and damage range of the ssg, so even though it has a greater capacity for damage (300 vs 288), it has a lower average and lower maximum damage?

Edit: By "maximum" damage, I mean the maximum actual damage in a vanilla/boom engine, not the theoretical 300/288.

Common example of a 3 shotted arch with rockets: Henning's pl20-339.lmp - http://www.youtube.com/watch?v=1A29Tur5jT0...1 min mark.

Share this post


Link to post
Qaatar said:

So, the pseudo-rng for the launcher must have a greater average damage calculator than the ssg (not including the splash damage immune monsters). I assume this is because of the greater dispersion and damage range of the ssg, so even though it has a greater capacity for damage (300 vs 288), it has a lower average and lower maximum damage?

There are 20 pellets in one shot of SSG. Each pellet have 5-15 point damage of randam number. Therefore, lower or higher damage of SSG shot doesn't happen easily. Their damage is usually stable (around 200 point damage). However, RL has only 2 elements of damage. The one is direct hit damage, and the other is blast damage. Blast damage depend on the range, so the enemy which take a direct hit would receive near maximum balast damage. As a result, damage range is 148-288 point. It's not stable.

Share this post


Link to post
tatsurd-cacocaco said:

There are 20 pellets in one shot of SSG. Each pellet have 5-15 point damage of randam number. Therefore, lower or higher damage of SSG shot doesn't happen easily. Their damage is usually stable (around 200 point damage). However, RL has only 2 elements of damage. The one is direct hit damage, and the other is blast damage. Blast damage depend on the range, so the enemy which take a direct hit would receive near maximum balast damage. As a result, damage range is 148-288 point. It's not stable.


That's what I was wondering. There must be a much higher probability of a rocket doing near maximum blast damage. Do rockets always do 160 points of damage if it's a direct hit? If they do, that means that a direct hit with a rocket will nearly always do 160 (direct hit dmg in multiples of 20) + 96-128 (blast damage in multiples of 8)...which means a direct hit will nearly always do over 250 damage? That doesn't sound right.

I'm sure there has been calculations of this sort before.

Edit: I mean in vanilla as well. I realize in boom and beyond the pseudo-rng was corrected. Also, the doom2.net weapon damage calculations by George Bell only provide a few ratios regarding a couple of scenarios, and no exact calculations of percentages and damage. For example, the 1 in 5 million for the ssg is completely wrong in vanilla, as the ssg can never do 300 damage in the pseudo-rng.

Share this post


Link to post

Qaatar:

http://www.gamers.org/pub/idgames/lmps/tyson/weapons.html (basics)

http://www.doomworld.com/vb/doom-general/50365-rocket-berserk-damage/ (quite a recent topic)

Oops I see you already mentioned Bell's page. So to bring something constructive in my post... I think the odds are the same in Boom and in Vanilla but this is based merely on observation. Also, in case of rockets unlike SSG I think if we assume total randomness, we aren't very far from "real play" conditions, so the phrase "There must be a much higher probability of a rocket doing near maximum blast damage" is false.

Share this post


Link to post
Qaatar said:

Do rockets always do 160 points of damage if it's a direct hit?

No, rockets always do near 128 points damage if it's direct hit because the monster receive the blast damage in zero range. In addition, the monster take 20 - 160 point damage randomly.

Qaatar said:

Also, the doom2.net weapon damage calculations by George Bell only provide a few ratios regarding a couple of scenarios, and no exact calculations of percentages and damage. For example, the 1 in 5 million for the ssg is completely wrong in vanilla, as the ssg can never do 300 damage in the pseudo-rng.

I just calculate a theoretical value. About SSG, one pellet has 5, 10, or 15 point damage, so the maximum damage is 15 and the probability is 1/3. The probability that all pellets have 15 point damage is the 20th power of 1/3. The answer is 1/3486784401. However, the probability of one rocket has the maximum direct hit damage is only 1/8. Like I said, SSG's damage is stable but RL's isn't stable. For example, even if 250 over damage happen continuously and you can kill the vile with 3 rockets, it is not a very rare case, unlike SSG. Also, it means that the probability that one rocket has the minimum direct hit damage is 1/8. The damage is 20 (+ near 128 point blast damage), so you might be able not to kill a demon. If you doubt it, you should try this wad.

Qaatar said:

the pseudo-rng for the launcher must have a greater average damage calculator than the ssg

As a result, it is wrong. The damage of RL is only uncertain.

Share this post


Link to post
tatsurd-cacocaco said:

No, rockets always do 128 points damage if it's direct hit because the monster receive the blast damage in zero range. In addition, the monster take 20 - 160 point damage randomly. Therefore, mean damage is 218. this wad.

As a result, it is wrong. The damage of RL is only uncertain.


I see. I also never used the cheating/TAS functions before, so I couldn't trace the damage. So the 20-160 is completely randomized instead of 20-160 in intervals of 20 that I thought...someone should change the doom wiki then. However, the only problem is I can't know for sure how much blast damage was inflicted, and only calculated based on the 20-160 direct damage variable. Are you sure that it always does 128 points of blast damage in a direct hit?

Edit: for example, the last 2 demons in your demo, the rocket inflicted 188 and 193 points of damage. They were both direct hits, so assuming 128 points of blast damage, the direct hit would be 60 and 65. The 60 would fit into the 20-160 in multiples of 20, but the 65 wouldn't. However, that's only assuming it was 128 points of blast damage, which we don't know for sure?

Share this post


Link to post
Qaatar said:

I see. I also never used the cheating/TAS functions before, so I couldn't trace the damage. So the 20-160 is completely randomized instead of 20-160 in intervals of 20 that I thought...someone should change the doom wiki then. However, the only problem is I can't know for sure how much blast damage was inflicted, and only calculated based on the 20-160 direct damage variable. Are you sure that it always does 128 points of blast damage in a direct hit?

According to doom wiki and Bell's page, 128 point damage always do if it's a direct hit. I believed that but I know that it's wrong after checking RL's damdge in my wad, so I wrote "*near* 128 point damdge". I don't know the detail.

Share this post


Link to post
Qaatar said:

I see. I also never used the cheating/TAS functions before, so I couldn't trace the damage. So the 20-160 is completely randomized instead of 20-160 in intervals of 20 that I thought...someone should change the doom wiki then.

I'm not sure exactly how blast damage works, but I am sure that the direct hit damage is in increments of 20. If you want to confirm this for yourself try firing a few rockets at a Cyberdemon or Spider Mastermind.

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  
×