The DooMed Speed Demos Archive returns!

[Andy Olivera]

55 demos this week.

[Andy Olivera]

55 demos this week. Merry Christmas, all!

[Ryback]

Merry Christmas to our tireless DSDA admin!

[rehelekretep]

merry christmas to the legend AO!

[Andy Olivera]

53 demos this week.

[GoneAway]

Hello @mArt1And00m3r11339 👋

 

This thread is for Andy's updates and for discussing the archive itself, including notes about errors, etc.

Demos are posted in specific threads. If a thread exists for a wad already, it shows up in the list here: https://www.doomworld.com/forum/topic/18410-welcome-to-the-demos-forum-prboom-plus-usage-wad-and-irc-links-general-info/

 

For wads that don't have their own thread, you can create a new one (if you expect a lot of demos) or post in the miscellaneous thread here: https://www.doomworld.com/forum/topic/70300-miscellaneous-demos-part-3/

 

Typically you'll want to include information about the demo in your post and in the file name, including map, category, and time, so people get an idea of what it is.

[mArt1And00m3r11339]

8 hours ago, kraflab said:

Hello @mArt1And00m3r11339 👋

 

This thread is for Andy's updates and for discussing the archive itself, including notes about errors, etc.

Demos are posted in specific threads. If a thread exists for a wad already, it shows up in the list here: https://www.doomworld.com/forum/topic/18410-welcome-to-the-demos-forum-prboom-plus-usage-wad-and-irc-links-general-info/

 

For wads that don't have their own thread, you can create a new one (if you expect a lot of demos) or post in the miscellaneous thread here: https://www.doomworld.com/forum/topic/70300-miscellaneous-demos-part-3/

 

Typically you'll want to include information about the demo in your post and in the file name, including map, category, and time, so people get an idea of what it is.

Thanks for the information. I updated my previous comment on this thread. When it comes to file name, if I change the name of the file, it would not load in Boom.

[Andy Olivera]

47 demos this week.

 

mArt1: You need to include a text files with your demos. A template is here.

[Andy Olivera]

42 demos this week.

[Poncho1]

12 minutes ago, Andy Olivera said:

42 demos this week.

Oh, something's come to my attention: my DVII Map29 UV-Speed is NOT for the secret exit. It's the normal exit. 

[4shockblast]

Arsene1412's nomo for Doom E1M3 uses the secret exit, not the normal exit.

[SAV88]

Azuruish' 30hb6138 TAS movie plays back with the old (rather than new) version of Hellbound.

[Andy Olivera]

60 demos this week.

 

Poncho: DV2-29 has three exits, but the DSDA can only differentiate between two. Since two exits don't involve actually playing the level I'm calling both of those 'secret'.

 

4shock & SAV88: Fixed. Thanks!

[Poncho1]

Okay, understood.

[Ancalagon]

On 5/14/2012 at 11:54 AM, Andy Olivera said:

j4rio: I wasn't sure which of Uwe's demos were considered TASes, so I didn't note them as such.

 

So I sat through all of Winterfeldt demos to give a list of the runs that are 100% TAS, these are just the demos that are TAS without a doubt, I can't tell if his doom2 demos or the episode runs are TAS but I wouldn't be surprised if they are and we can't tell because by this time he got better at TASing, I'll watch Uwe's demos later.

 

E1M9-127
E1M7-157
E1M6-155 - The movement around the yellow key is a big red flag
E1M5-206 - Most obvious area is when doomguy is having a stroke while opening the chainsaw secret.
E1M4-028
E1M4-024 - This one doesn't scream TAS like the rest of the demos, but if you watch carefully doors and switches are opened and pressed perfectly, for a .97 exit, let's remember this is a 1995 demo.
E1M3-105
E1M2-036
30NM6520 - While there are no big red flags in the maps I watched at least, I don't believe someone can complete doom2 on nightmare while playing this bad, also quoting some of the map descriptions.

"Map 29: A tough one. Just stay alive." - No, just no, if you play this bad you can't beat map29 on nightmare without a clear strategy that you can put in words, this was simply save scummed and he doesn't know how to stay alive in 1 go.

"Map 22: As you see, I don't have a good solution for this one, I just hope for the best." - Describes most of the run (or walk), apparently this works?, also starts the map by facerocketing twice against spectres (shows the complete lack of practice, anyone with serious practice would have routed against this and used bfg instead) and then proceeds to do the rest of the map without taking much damage.

 

Please mark all of these runs as TAS.

Edited January 23, 2019 by Ancalagon

[Grazza]

1 hour ago, Ancalagon said:

I'll watch Uwe's demos later.

This post by AdamW is worth noting. That is, the absolute clincher was not the movement so much as the PRNG manipulation.

[Ancalagon]

24 minutes ago, Grazza said:

This post by AdamW is worth noting. That is, the absolute clincher was not the movement so much as the PRNG manipulation.

Killing all the map07 spiders in 2 rockets is clear segmenting, but e1m5-059 is harder to judge. By today's standards that run is nothing spectacular and I wasn't running doom at the time and I haven't watched enough 1996 demos to know how the standard demo looked at the time, the only thing that caught my attention is how perfectly he activates the line that raises the bridge at the start of the run, even today's runners run further to make sure the line is activated. 

 

In the Winterfeldt post I only mentioned the demos that are obvious TAS, all of these were recorded during the second week of August 1995, except e1m4-024, dated november 6th 1995, I don't think slowmo is enough to produce this run, the exit time is .97 and the run gets progresively more perfect towards the end meaning this is probably segmented, but I can't just call the suspicious runs like e1m5-059 TAS all by myself, those would need a community judgement.

Edited January 23, 2019 by Ancalagon

[Ancalagon]

And now for Uwe's demos, again, I only included the demos that are blatant cheating, many demos are very suspicious but I can't say they are tas with 100% certainty.

 

E1M1-040 - This run is so tas it's hard to believe one day it was accepted
E1M3-102 - Blatant slowmo
E1M4-029 - what is this? to produce this mouse movement at the time you had to play the game at 10% speed or so.
E1M5-205 - Blatant slowmo
E1M6-149 - Blatant slowmo

0602UV01 - This demo has many "???" areas, but map07 rings all the alarm bells, killing all the spiders in 2 rockets is not possible, simple as that. and... did he just read the entire map07 intermission text?, yes he did!

0911UV21 - Okay, this one doesn't have an "alarm!" moment like the episode 1 demo and the evidence is more situational than "blatant tas", but as the note the admin left in the demo "Seven whole Minutes faster than the previous best... And with Every Level done faster than ever before", so, a player we known for cheating does a ground breaking demo, 3 months later records an episode 1 run that we know for sure is cheated, does things like map29 jump and map30 1 cycle without any setup and protagonizes a funny moment when an ellastic collision against a rev pushes him back into the tiny curved ledge stairs in map29 and reacts fast enough to not fall... I guess that is a small alarm moment. Overall, I don't believe the legitimacy of this run.

[Looper]

I'm sure Winterfeldt didn't manage to run legit doom2 map02 uv-speed in 43 seconds. Just look what a crappy mess the whole demo is, killing monsters behind him and having very tiny suspicious stops for save/loading. To be honest, I am not sure he had ANY non-cheated runs because why would he? Already over 50% of his demos are proved to be TAS, so why would ANYONE trust a guy like that? :D I would mark all Winter's and Uwe's demos as TAS btw, even though there might be a couple extremely shitty runs non-TAS.

 

In my opinion, in the case of these two dudes, it should go the other way around: prove that they did NOT use TAS, otherwise mark the runs as TAS.

[Grazza]

2 hours ago, Looper said:

In my opinion, in the case of these two dudes, it should go the other way around: prove that they did NOT use TAS, otherwise mark the runs as TAS.

I agree. Marking just some of them as TAS would give an unjustified impression of legitimacy to the remaining ones. And in the case of Girlich's it would suggest that the only thing you need to do to avoid a run being considered suspicious is to avoid jerky movement and practically-impossible PRNG sequences.

 

Of course, this is all a part of Dooming history, and any outrage is a thing of the past.

 

BTW, there was discussion here, including the question of "did these guys do any legit runs?".

 

Edited January 23, 2019 by Grazza

[Ancalagon]

I also think they all should be labeled tas, or maybe label them as cheated instead to differenciate from legitimate tases, there is one special run from Uwe, e1m8-125. This run was recorded during may 1996 and the gameplay is really bad, turns are really slow, he can't even turn fast enough to walk around one of the barons and shoot him continuosly, this demo looks legitimate. If you watch this demo, then watch e1m1-033 recorded 3 weeks before, is this the same player? e1m1-033 is full of fast turns and it looks like a decent demo, nothing like e1m8-125. Looking at e1m1-033 again that secret at the end is activated with some extreme precision, 4 days before he recorded e1m1-035 that also has pretty fast turning.

[4shockblast]

Yeah, I agree that TAS isn't exactly the correct label here (same regarding Kimo's demos). There needs to be a comment or maybe some other tag indicating that they were not submitted as TAS demos because having them default to TAS as-is makes them mixed up with actual TASes IMO.

[GoneAway]

Cacatou's plutonia 22 nomo in 0:51, which is the current record, appears to not be cl4 but instead mbf.

I also looked at his scythe and scythe2 nomo packs and many of these demos are mbf as well...

Not sure it is worth enumerating everything, but probably the plutonia run should be marked since it's an iwad record.

 

(Might be cl -1 instead of mbf, but mbf is in the header, not sure what the format is for cl -1)

Edited January 26, 2019 by kraflab

[Andy Olivera]

25 demos this week.

 

Regarding Steffan and Uwe's demos: they've all been tagged 'TAS'.

 

Kraflab: Fixed. Thanks!

[Ancalagon]

Thanks for tagging all those demos, there's another player with very fishy demos that might get a post soon, depends on how much evidence can be found.

 

In Back to nirvana my 6:15 map26 max is actually for map11, sorry for the error in the txt file.

 

And I hope you're prepared for a massive demopack soon to be released ;)

Edited January 30, 2019 by Ancalagon

[Andy Olivera]

38 demos this week.

 

Ancalagon: Fixed. Thanks!

[Andy Olivera]

40 demos this week.

[Andy Olivera]

93 demos this week.

[GoneAway]

Bob9001's NoMo100 demos for plutonia map 2, 16, and 21 are actually just NoMo (those maps have no secrets).

[4shockblast]

For hdsped01, I'd like my map 9 pacifist recategorized as UV-Speed as it has an unintentional telefrag, but one that is avoidable, so I don't think it makes sense to call it pacifist. :D

 

Also my swtchrm e1m9 nomo is actually in 0:23 (the DSDA page lists it as 0:26).