The DooMed Speed Demos Archive returns!

22 hours ago, Opulent said:

A Cyberdemon531 maxdemo is just as good as a Cyberdemon531 maxdemo that happens to be of shorter length.  This isn't my opinion, this is a fact.  It is a crime to remove either demo and goes against the main purpose of the DSDA to me.

Dude, thanks for the shoutout! Glad to know someone watches my runs :D

On 25.10.2017 at 6:13 AM, Grazza said:

Yeah, some people have some growing-up to do. Once it's released, it's released and that's that. I'd be happy to see Andy refusing all such requests out of hand. It's pointless extra work on top of everything else he does for the community. And deleting history, even if only in a small way, loses something of value. If compet-n demos got deleted each time they were beaten, we'd have very little idea whom to credit years later for route and trick ideas.

I will wholeheartedly second this.

EDIT: On the same subject, I don't think merging demos is any good either, because it makes it less likely to see how a run came to be over time.

2 hours ago, Cyberdemon531 said:

Dude, thanks for the shoutout! Glad to know someone watches my runs :D

How can no one watch your sexy runs ;P

5 hours ago, Nine Inch Heels said:

I don't think merging demos is any good either, because it makes it less likely to see how a run came to be over time.

Yeah, I agree. I'll know to just keep my demos up as they are from now on. My mistake.

f1b1-153.zip (UV Max for 01fava E1M1) seems to compressed with a rarely used compression algorithm which results in the zip being recognized as corrupted by many utilities dealing with zip archives. Could it be repackaged?

idgames has newer version of nmdu.wad than the one that is on DSDA. These demos desync with the idgames version:
UV Max: E3M3 (both), E3M5, E3M6, D1EP3
UV Speed: E3M3

Staying with nmdu, Skepticist's D1EP2 UV Max in 39:36 desyncs with both DSDA and idgames version, using nmd2_old.wad doesn't help either.

23 minutes ago, Keyboard_Doomer said:

a rarely used compression algorithm which results in the zip being recognized as corrupted

Seems like it's a .7z with the wrong extension.

23 minutes ago, Keyboard_Doomer said:

idgames has newer version of nmdu.wad than the one that is on DSDA. These demos desync with the idgames version:

Maybe this ruling applies? Which reminds me, I never did reply to that post...

On 23/04/2017 at 3:29 PM, Andy Olivera said:

RjY: Since I started separating incompatible versions (no more 'old' WADs) I no longer host those that don't have demos for them.

Fair enough. But I think it needs to be made clear when the wad a demo is recorded on is a different version from the "common" one.

It's reasonable to assume most players who watch demos get wads from /idgames or doomworld, then go to dsda to look for demos for them; I doubt that many use dsda itself as their primary wad source.

Now, if a demo obtained this way then fails to play, it seems to me that "the demo was recorded on a different version of the wad" is pretty far down the list of suspects. I may have a skewed perspective, but I think it's far more likely to first suspect an engine bug. So I ended up wasting quite some time looking for a bug that didn't exist. :/

If there was some indication or hint on your page that the wad was different, a date in the title, or anything really, more than just the wad name/author, then I would likely have suspected the wad much earlier and not spent so much time chasing wild geese. :)

Not expecting you to go through your whole archive and compare it with /idgames, just hoping you'll consider requests to mark the pages of such wads as and when differences are found.

I see my assumption about the zip was wrong. It indeed appears to be 7z.

I agree that most people probably have the wad they want to play from idgames or Doomworld but I think it happens fairly often that if a demo caught your interest and you don't have the wad yet, DSDA might be the place from where you download it.

I also remember someone mention that when they're about to record a demo for a wad, they download the wad from DSDA rather than idgames.

Sure, splitting the wad table for a version that doesn't have any demos yet and is unlikely to have anytime soon is annoying but considering there's a chance someone will consider the wad hosted on DSDA as up-to-date I don't think DSDA should host only the outdated version just because the new one doesn't have any demos yet.

1 hour ago, RjY said:

If there was some indication or hint on your page that the wad was different, a date in the title, or anything really, more than just the wad name/author, then I would likely have suspected the wad much earlier and not spent so much time chasing wild geese. :)

Wads being updated (even multiple times) after initial idgames release is a common occurrence nowadays. I really wish the idgames provided a checksum/hash for the wads included in the zip so you can compare you have the newest version without having to download anything. I hope at least the new DSDA that's in the works will add this feature. 

1 hour ago, Keyboard_Doomer said:

(version problem stuff)

(Probably totally off-topic)

I remember that when I was playing through Disjunction, floatRand updated the WAD on Google Drive, but not on /idgames... This means I am potentially risking my time for desync demos or such because I thought /idgames one is the updated one. Gladly that there's no entry on DSDA just yet, and floatRand updated the /idgames WAD file just in time to avoid confusion. To be honest, version problems increase unnecessary workload, but it's very difficult to avoid... It's difficult for Andy to do all the stuff if an author suddenly wanted to update his/her WAD.

BTW: The Doom 2 D2All UV Max in 1:10:57 by Xit Vono and Vincent Catalaa, should it be a coop run, not being in this section?

45 demos this week, and now over 3,000 WADs!

Keyboard: 01fava demo and nmdu tables fixed. Thanks!

Garrett: What do you mean? It's tagged as Co-Op; the CN designation just displaces the CO one. You can test this by disabling Co-Op demos on the 'Settings' page; it should disappear. However, Co-Op demos aren't segregated, so it's right where it's supposed to be.

Regarding removing demos, I very rarely do this, as in cases like rehelek's last request where the demo serves no real purpose, historical or otherwise (or for compatiblity; see below). Merging I do on my own sometimes, such as when there are several improvements in quick (as in days) succession. I'm with pretty much everyone else in disliking the current trend of merging improvements with demos that are weeks or months old. That isn't going to continue.

Regarding multiple WAD versions, this is the curse that just keeps getting worse. I know in the past a WAD could have several releases, so it's possible the only thing that's really changed is the number of demos and how quickly players pick up new WADs for recording, but I'd wager it's actually becoming far more common. Nowadays, I'm far more likely to delete a few demos rather than add a second version of a WAD that's only been out for a week and which will probably have yet another version in the future. Trying to keep up with all the revisions is a futile endeavor, so I just wait until it becomes an issue for the DSDA, then handle it, either by splitting the versions or updating the current version and removing an old demo or two. Maybe the next DSDA iteration can use some sort of script to check for updates to idgames WADs to stay ahead of the curve, but for now, if there're no demos for a WAD, there's no need for it to be hosted. Like the name says, the DSDA is a demo archive, not a WAD archive. I know this can cause confusion, but I think the issue with nmdu, a WAD with over a hundred demos and compatibility issues that sat unnoticed for at least six years, is a good indicator that the problem isn't as bad as it might sound. Even keeping the WADs current and labeled doesn't prevent an old version from being used to record new demos, though it's uncommon, thankfully.

Andy: Thanks for the explanation because I thought coop demos have their own section.

On 10/29/2017 at 1:32 PM, Andy Olivera said:

Maybe the next DSDA iteration can use some sort of script to check for updates to idgames WADs to stay ahead of the curve

I do have some plans for wad page versioning to make it easier to organize things, but I probably won't be hosting wad versions with no demos either.  Also, re: merging / deleting demos, dsda3 is insert-only, so if anything I will be retroactively "unmerging" demos back into the table.  Everything has historical value.  But, I will mark demos anonymous if people don't want their "reputation" tarnished or something.

My Cchest map29 demo got mysteriously marked as a valid max some months ago but it misses a monster.

my chillax map 43 tas requires mapfix and noblock in order to actually play the demo

34 demos this week.

Ancalagon & A7MAD: Fixed. Thanks!

My Hellbound MAP14 NM100S time is actually 2:24. I mentioned that "2:30" as PAR :)

Also, Rayzik's MAP27 Max record is 20:53.

35 demos this week.

SAV88: Fixed. Thanks!

87 demos this week.

Oh, I thought I don't need to put the zero to hold the digit when the demo is less than 1 minute. Next time, I'll know what to do not to waste extra time from you.

do you mean like, 59 seconds as "059" or "59"?

@rehelekretepYes. I put "59", but they are changed to "059". Well, I just don't like to put extra work on others.

0xf00ba's Sunlust EP2 UV-Speed was labeled as a UV-Max.

8 hours ago, GarrettChan said:

@rehelekretepYes. I put "59", but they are changed to "059". Well, I just don't like to put extra work on others.

ah thats what i thought - ''59'' would be 59minutes to most people :)

62 demos this week.

Garrett: Your efforts to follow standards are appreciated. :-)

rdwpa: Fixed. Thanks!

My Going Down map02 max is for the 2015-08 version.

46 demos this week.

kmc: Fixed. Thanks!

67 demos this week.

Both of Community Chest MAP12 nomos by @doomdaniel95 desync. Haven't yet found which WAD version (and/or port setting) do they require.

Also, Doom 2 In Name Only MAP12 UV Max by @Revved plays back with d2ino_b6.wad.

Also, FC19's nochance NM100S is TAS.

Also, the Eternal collection table mostly lists gross nomonsters times instead of exiting times.

Level    Exiting time    Gross time
MAP01            1:48          1:56
MAP02            3:33          3:39
MAP03            4:48          4:57
MAP04            6:08          6:21
MAP05            5:30          5:39
MAP06            2:49          2:56
MAP08            1:52          2:03
MAP09            2:34          2:40
MAP10            2:57          3:06
MAP11            6:08          6:17
MAP12            9:01          9:09
MAP13            0:08          0:13
MAP14            3:15          3:19
MAP15            8:23          8:28
MAP16            2:19          2:23
MAP17            1:43          1:49
MAP18            2:41          3:02
MAP19            4:21          4:30
MAP20            8:51          9:00
MAP21            1:57          2:02
MAP22            2:46          2:53
MAP23            7:21          7:26
MAP24           10:05         10:14
MAP25           10:44         10:52
MAP29            5:24          5:35
MAP31            9:47         10:11
MAP32            4:53          4:59

Moonblood map31 uvspeed by @paymentplan, not by me

I downloaded Cchest from DSDA again, so I should have the up-to-date version. Hopefully whatever the problem was is solved. Would someone be kind enough to tell me if this desyncs?

Cchest (1) MAP12 nomo in 4:23.

cc12o423.zip

That one works fine for me @doomdaniel95