Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
ars4326 wrote:
Other than that, I haven't utilized RAM Watch or anything to track their HP (I'm admittedly not very savvy in this area).
Ah, that's too bad. You could have asked, and I or someone would surely have helped out.
I understand that attack power is exponential somehow (although more specifics would be interesting) and you also confirmed that there is a random element. How about sword type? How much stronger is the fire/thunder/ice/gaia sword than the previous? How much stronger is a charged attack? (Double?)
I guess it will be more relevant for an any% run when deciding which life stocks to get, but there are instances in this run where it could have helped as well. For example, is it worth going into the cave behind the sacred trees at once? At the 19 minute mark there are lizards which take some time to beat. Perhaps the single life stock here does not save any time, and beating the lizards later with higher attack would save more time?
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Nice to hear that you managed to resync it! I'm looking forward to seeing it on the workbench. Two questions out of curiosity:
- I understand that monster AI is different when redoing, but does the game also have subpixels which carry over between screens? Otherwise I don't understand why the screens without monsters wouldn't sync.
- Have you figured out how is damage is calculated? Is there any random element to it? Are you monitoring monster HP?
(Also BTW: it's どうも.)
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
feos: Thanks! The shoot em up stage works like a charm now. I am wondering, could there be any variable like the X scroll lock (0xFFDFC0) for vertical scrolling stages like the current airship hangar? I haven't found anything. (0xFFB79E seemed promising but only changes on he first scroll lock.) It's very helpful for timing purposes to know when you reach/leave locked screens.
netwizard: Yes, it is possible to hack your weapon level. In fact, this is what I tried first to see how much Green 7 would save. Unfortunately, the movie desyncs halfway through the boss fight when you increase Batman to Green 7, but it syncs all the way if you lower his weapon level. I suspect this is because Green 7 has a larger projectile hitbox than the other levels, which are all the same size. Because of this I had to use some calculations to estimate the result instead.
Archanfel: Heh, thanks man. I appreciate the support.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
feos: I tried to get in touch with you on IRC, because I think it's probably easier to discuss it there. Something is off with the script, because I get a number of end mismatch. I think it's around the end of HandleMsgTable, here:
if (MsgTable[ i ].timer_ <gens>= num1) and (var <= num2)
then return true
netwizard: I am looking into the skip of the weapon powerup, because at the moment I can't remember why I made that choice. (I made the first part of stage 1-4 nine months ago, and gave up on the balloons for a long while.)
Anyway, even though the projectile is a lot bigger, the damage increase isn't that much from Green 6 to Green 7. A test I made yesterday showed that having both at Green 7 saves about 55 frames at the boss. I will have to test how much picking up the powerup costs. My guess is that it's similar or less. (This is mostly out of curiosity. I'm not remaking this stage.)
About charging up Green 6 instead of Green 7:
You are correct that DPS of charged Green 6 is less than charged Green 7. (This actually isn't true for all weapons - the damage of charged Blue doesn't increase at all for some levels and for charged Red 1 to 7 goes 64 - 128 - 192 - 80 - 240 - 160 - 240, meaning higher weapon levels are sometimes significantly weaker.)
However, charging up Green 6 gives a larger percent damage increase than charging up Green 7. This means that
DPS of Green 7 + charged Green 6
is greater than
DPS of Green 6 + charged Green 7
Hope that makes sense.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Here's a small present for anyone following this topic:
http://tasvideos.org/userfiles/info/22158252089463586
Stage 1-4 (and all of level 1) complete! It's pretty intense.
Reminder to feos and r57shell: the script needs to be updated for the upcoming shoot em up stages. The hitboxes go a bit crazy when scrolling the screen up and down and don't match the objects on screen.
Also, now taking applications for people who like playing around in autoscrollers. Send your resume to me.
netwizard wrote:
What does vB mean?
ALAKTORN was right, down+B. I use v^<> when writing directions.
You can see the pattern in question on the first boss.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Clumsydoomer wrote:
1. It can be calculated easily because vanilla Doom's tickrate and framerate is 35. So, one frame is 1/35 (~0.028) of a second. There are some timing tools, but they all work in the same way as PrBoom-Plus (they never work on intermission screens though).
EDIT: I forgot to mention the levelstat function in PrBoom-Plus. It can instantly generate all statistics just by examining the demo for a few seconds.
Here is what it generated:
I used these commands: prboom-plus.exe -iwad plutonia -warp 31 -playdemo 30pl1241 -levelstat
-warp 31 isn't required, but it makes the process instant. Otherwise it will play the whole demo.
This is exactly what I was looking for, thank you. It took me a while to figure out that the result is dumped to levelstat.txt and not output to the terminal where you launch the command.
I noticed that there is a slightly faster non-TAS MAP12 on DSDA - I assume this is because that movie uses death, and death would slow down the following maps more than it would gain for you?
http://doomedsda.us/wad946m212.html
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Two questions to anyone who knows (probably Clumsydoomer):
- How do you extract/show fractions of a second for the levels? Does it have to be calculated by inspecting the movie file, or is there an option in PrBoom, or what?
- Could you elaborate on the method of constructing a Doom TAS? It seems to be quite different from how the rest of our tools work.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
feos wrote:
14256 is the frame where the ending song first loop ends.
But here rises a question: should we use that for all Doom submissions? It's not easy for the author to figure out the exact length, I did it by trimming the dump at different frames and comparing.
Neither way is consistent:
1) Cutting the movie at the last input doesn't allow the demo viewers to see the ending, and forces the publisher to extent the demo just to encode the ending. But it's the best practice regarding the site's time calculations.
2) Including the entire ending might be hard for the runners.
3) Including only a part of it just feels wrong.
(I numbered the options.)
I think option 1 is the best. We want to make it as easy as possible for players. Most of them are not regulars at TASvideos, and we already have a wall of rules which makes it hard for newcomers to join. It is better to make extra work for encoders (or, realistically, me, since I am the only one who answered the call to edit the movie).
In the publication description, put a note along the lines of "Note: The encode has been extended to include the full ending music. This is not present if you play back the movie in Doom."
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
hegyak wrote:
1) Final Doom (Added Episode 4 to Doom 1), Eviloution and Plutonia as well. That's 4 major games (Final Doom supersedes Doom 1 since it has All of Doom 1) for one "system."
(...)
2) What about Quake? We haven't gotten there yet, but what about that game? Does it get it's own section as well? Because there's options there too. We have the main game (Quake 1), Hipnotic and Rogue. That's three games that use the same engine.
(...)
3) ID games made Doom II for Windows 95. It is Doom on Windows 95 (and above, including NT). Not sure if that's a "Port" or not. It does run Doom 1 and 2 and the other two WADs from Ultimate Doom's bundle.
1) Just wanted to clear this up: it was Ultimate Doom which added the fourth episode (Thy Flesh Consumed) to Doom 1. Final Doom is the expansion to Doom 2, which contains Evilution and Plutonia.
2) The Quake movie format does not record keypresses (like most of our emulators) or player commands (like Doom), and can therefore never be accepted at TASvideos. (DeHackEd can probably explain the Quake movie format better.) Any Quake games will have to be TASed through JPC-RR or Hourglass (if WinQuake is allowed), and will be listed in the same place as other DOS/Windows TASes.
3) Doom 2 was released for DOS. A later port, Doom95, made Doom 1 and 2 playable on Windows. It does not have demo playback compatibility with movies made in DOS Doom 2.
EDIT: Sorry, had to run yesterday, didn't finish my thought>
To a casual viewer, it will not make much sense having Doom listed separately, while other similar games such as Quake are listed on their platform. Under published movies, Doom should be listed with other MS-DOS games. If that should be further merged with Windows, I don't know.
In the submission list however, I think it could be a good idea to still list it as Doom (and not MS-DOS). It gives relevant info to judges and encoders. The site code will have to treat it differently in any case, since it's a different file format.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Warp wrote:
The three episodes were all part of the one and same game. Just because the shareware version had only part of the game included doesn't change that fact. The shareware version was an incomplete version of the game.
Allowing only part of a game to be completed, and considered a full TAS on its own right, would be rather unusual and exceptional. Are there any other examples of this?
Some examples were already posted by other people. I would like to add that it is not technically possible to record an LMP demo which spans over multiple episodes. Once you have completed an episode, you need to start a new game to select the next episode, and the LMP format does not support starting new games. Therefore, Doom has to be recorded and submitted in several LMP files.
(Source: I ran through episode 1 and tested.)
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Looking good as always.
One thing I noticed is that you are very low on health, and without Eke-eke. Are you planning on dying for some reason? I seem to remember that glitching through some doors without a key required dying next to it and being revived by Eke-eke...
It is slower because this movie does not stop input until the ending has been shown. The final stage actually ends a bit earlier in this movie, than in the movie you linked.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
This was great! I vote yes and wouldn't mind this having a star.
Since this is a run which aims for score, wouldn't it be a good idea to at least put the score you achieve in the submission? Perhaps per level, and a comparison with the current WR would also be interesting.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
>(I'm assuming that when a player makes a match, their turn continues.)
Correct.
We seem to interpret the problem the same, but still your answer for a 3-pair game is different than mine. We agree this far at least:
1 pair:
p1: 1/1
draw: 0/1
p2: 0/1
2 pairs:
p1: 1/3
draw: 0/3
p2: 2/3
Let's see where one of us goes wrong...
3 pairs:
p1: 1/5 for finding one pair (AA known, BBCC left). After this, the situation is identical to a 2 pair game:
-p1: 1/3 for finding two pairs to win 3-0
-p2: 2/3 for finding two pairs to win 2-1
total chance to win in this scenario:
p1 = 1/5*1/3 = 1/15
p2 = 1/5*2/3 = 2/15
p1: 4/5 for no pair (AB known, ABCC left)
-p2: 2/4 for 1 pair directly (B known, BCC left)
--p2: 1/3 for B, 2 more pairs to win 3-0
--p2: 2/3 for C
---p2: 1/2 to win 3-0
---p1: 1/2 to win 3-0
-p2: 2/4 for turning up C
--p2 1/3 to find the other C, 3 pairs to win 3-0
--p2 2/3 to find A or B, which means p1 wins 3-0
p2 chance to win is then
4/5*2/4*1/3 = 8/60 = 2/15
4/5*2/4*2/3*1/2 = 16/120 = 2/15
4/5*2/4*1/3 = 8/60 = 2/15
p1 chance to win is
4/5*2/4*2/3*1/2 = 16/120 = 2/15
4/5*2/4*2/3 = 16/60 = 4/15
In total
p2:
2/15 to win 2-1
6/15 to win 3-0
p1:
7/15 to win 3-0
So my original calculation of 9/15 for p1 was wrong, and 7/15 like you had it was right.
This branches into a lot of possibilities pretty quickly. Maybe I'll work out 4 pairs by hand too, if one player gets a pair directly (1/7 chance of this happening) the game basically becomes a 3 pair game which we already know the solution to. In that case 2-1 above becomes a draw.
There should be a better way than doing all possibilities by hand though...
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
GoddessMaria15 wrote:
Just Skate? Why no love for Blaze? o-o
Blaze and Axel are fun to play, but don't offer that many advantages in terms of speedrunning. They have no way to move around quickly and do moderate damage. I'd still watch if someone made a speed run with them, though...
TheReflexWonder wrote:
Well, on Normal difficulty, the Stage 3 alien area has many enemies survive with a sliver of health after getting sworded once; maybe they have 40 HP?
I think the important thing is to land while moving away (either by drifting backward halfway or going really deep toward an enemy) so that you can forward-forward at a slight distance.
Yeah, they have 40 HP. Actually, all enemies have starting health which is divisible by 8.
Maybe it's possible to get this combo working regularly, but I'm skeptic. Feel free to prove me wrong. :)
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Who wins at Memory?
While playing Memory today, I thought about 1) the best strategy to win and 2) who wins statistically, the player to go first or the player to go second?
Let's assume that both players have perfect memory, so that all that matters is statistics. Let's also assume that you must reveal at least one new card on your turn, so we don't get stuck with both players only choosing known cards because revealing a new card decreases your chances of winning.
1) I have a good idea about the optimal strategy. When there are lots of cards left, it's is better to only reveal one new card, and then pick a card which is already known.
Reasoning: if you reveal a card which you don't know the match for, you should not reveal another card, because that increases the risk that your opponent can match that card with an already known card.
If there are only a few cards left, it can be better to switch to reveal two new cards, if the first one isn't a match to a known card. For example: 6 cards left (AABBCC), 2 cards known (AB). You reveal a card which is C. If you guess on one of the three remaining cards, you have 1/3 chance of winning. If you instead pick one of the known cards, you have 0 chance of winning, because all of the new cards which the opponent turns up will match a known card.
2) For a small number of pairs, calculating player 1's chance of winning is easy. For example, for 2 pairs is 1/3. Reveal one card, A. You now have 1/3 chance of guessing the matching card. Whoever gets the first pair also gets the second.
For 3 pairs, it is 1/15 (win directly) + 8/15 (win on second turn), or 3/5.
How about x number of pairs? Is there a way to generalize this?
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Got 30/35 questions, or 133 as reported by the test (but you shouldn't believe IQ scores you get online). Got stumped by the same question as everyone else it seems.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
TheReflexWonder wrote:
1) a decent bit in the arcade of Stage 3 and the Jungle area of Stage 6, where you have to walk downward to proceed.
2) but the Dash Attack does more damage anyway (40 as compared to the sword's 32) with comparable time for opponents to hit the ground.
3) Skate's jump-in attack -> Dash Attack is quite fast when you have a single high-HP enemy in a loop on wake-up compared to Max's Atomic Drop, so maybe the superior speed per knockdown is worth the lower damage there (that Skate combo does 64 while Max's Atomic Drop does 70).
4) Anyway, I think there's some potential here to consider, and if nothing else, it'd be a nice change of pace from all the creeping slowness and Atomic Drop spam of Max when he's not Thunder Tackling (though Skate trades the latter for Dash Attack spam, so... :p ).
5) Does anyone happen to have HP values for enemies so I don't have to gather them manually? Otherwise I guess I'll just play through normally and make my own notes.
1) Also stage 1!
2) The sword actually does 38 when used by the player, so it's not that big of a difference. It's 32 when used by the enemies, though.
3) Skate's jump down+attack into dash attack is possible, but really hard to pull off, at least for me. Most of the time you end up grabbing the opponent instead.
4) I agree! It will almost definitely not be faster than a run with Max, but some variation is nice.
5) What, all of them? :) I think it might be unique per enemy, but if it's not, it varies at least by enemy variant (different color Galsias have different health) and difficulty level.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Personman wrote:
Am I understanding correctly that only three teams used TAS Editor?
It is mindblowing to me that anyone would choose not to use it when it is available. I understand that people who've been doing this for a long time might be less comfortable with the interface, but it seems like there must be some other huge advantage to doing it the traditional way to make comfort outweigh what seems like the enormous advantage in convenience.
legendarymudkip already explained how we did it in team TTT. I can add that in the first team I was on (team 9), hegyak used it.
I experimented some with it, but noticed I was just not nearly as productive recording new material as with old-school save-state TASing. I've used that method for like 10 years now, so I have quite some experience with it, and learning a completely new method (branches? tags? where are my savestates? hotkeys?) takes time.
For changing earlier material, when an improvement was found, it was obvious that TASeditor was easier. But mudkip handled that for our team, so I was never forced to deal with it.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Thanks Moth for all the effort you put into organizing this. It was really well done. The ghost encode in particular is hilarious, like a little army of metal forces doing the conga line.
Prophesy which turned out to be true:
<Truncated> god, it will feel like such a waste improving this 3 frames when it turns out we are behind by 200 frames when the results are posted
Yeah, not finding the stupid pause glitch annoyed me. I started watching team MMM's movie directly when it was posted, and right there in the first screen is a huge trick we missed. But eh, it is what it is. All teams missed out on some tricks it seems, anyway. We did our best. Thanks to my final and previous teammates for their effort!
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Aqfaq wrote:
I would like to offer 100 player points to anyone who makes a good run for one of these games:
* Dune II (Genesis or DOS version)
* Monster World IV
Can I do this? Can I give my own player points in exchange for runs that I would like to see?
Like Warp said, player points are not like a currency you gain, but more of a rating. If your runs are obsoleted or voted down your player points go down too.
Hopefully the player points a player gets for the run itself is enough. Otherwise, you could always offer some other sort of currency, like money. (It has been done before.) In that case, you should probably define "good run" in an objective way to avoid conflict later.
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Twice now when I have left BizHawk 1.9.0 SVN 8231 running while I left the computer for some time, it had crashed when I return.
- NES game (Metal Force) was loaded.
- A movie was in playback mode.
- Emulator was paused.
- RAM watch was on.
- No lua script was running.
Problem signature:
Problem Event Name: APPCRASH
Application Name: EmuHawk.exe
Application Version: 1.9.0.8231
Application Timestamp: 5470b314
Fault Module Name: ig4icd32.dll
Fault Module Version: 8.15.10.2291
Fault Module Timestamp: 4d419d9f
Exception Code: c0000005
Exception Offset: 00736e69
OS Version: 6.1.7601.2.1.0.768.3
Locale ID: 1053
Additional Information 1: 0a9e
Additional Information 2: 0a9e372d3b4ad19135b953a78882e789
Additional Information 3: 0a9e
Additional Information 4: 0a9e372d3b4ad19135b953a78882e789
Editor, Experienced Forum User, Published Author, Reviewer, Experienced player
(969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Samsara wrote:
Spikestuff wrote:
[23:36:07] <ais523> RGamma: Nach said outright he planned to rig DTC5 in #tasvideos
[23:36:58] <ais523> RGamma: I should note that this sort of behaviour (not the gambling specifically, admins hinting at shady goings on at the top) is one of the main reasons that I leave #tasvideos for months at a time
Yeah, I wish I didn't have the ability to understand jokes, either. It would make the world much more simple to live in.
From the context given above, it seems that ais523 did understand that it was a joke.
That doesn't stop it from being tired old patently unfunny joke which potentially alienates outsiders and we should stop doing, though.
