Super Mario Bros.

Edit history:

Blubbler: 2015-02-24 01:09:08 pm

2015-02-24 01:08:00 pm

I made some changes to the Knowledge Base entry: https://kb.speeddemosarchive.com/Super_Mario_Bros.

Ideally I want there to be animated gifs for every trick, where that makes sense, and videos containing both easy and optimal strats for every level in each category.

I have some trouble with early runs in the "Notable Runs" page. I used this Twin Galaxies news article (http://www.twingalaxies.com/content.php/2577-Scott-Kessler-Attains) for everything prior to 5:14, but I have some issues with this, which somebody who was actually involved with it, when it happened, might be able to help me with:
- It says 6:22 happened in November 2003 and the 5:28 happened afterwards, but on the rankings it says the 5:28 was verified in June 2002.
- It says Scott Kessler improved his 5:17 to 5:13, but he also had a 5:14 before his 5:13, which is ignored in the article. Other articles on their page confirm that it indeed went 5:20 -> 5:17 -> 5:14 -> 5:13. I fear more records may have been ignored between 5:20 and 6:22.

I made some changes to the Knowledge Base entry: https://kb.speeddemosarchive.com/Super_Mario_Bros.

Ideally I want there to be animated gifs for every trick, where that makes sense, and videos containing both easy and optimal strats for every level in each category.

I have some trouble with early runs in the "Notable Runs" page. I used this Twin Galaxies news article (http://www.twingalaxies.com/content.php/2577-Scott-Kessler-Attains) for everything prior to 5:14, but I have some issues with this, which somebody who was actually involved with it, when it happened, might be able to help me with:
- It says 6:22 happened in November 2003 and the 5:28 happened afterwards, but on the rankings it says the 5:28 was verified in June 2002.
- It says Scott Kessler improved his 5:17 to 5:13, but he also had a 5:14 before his 5:13, which is ignored in the article. Other articles on their page confirm that it indeed went 5:20 -> 5:17 -> 5:14 -> 5:13. I fear more records may have been ignored between 5:20 and 6:22.

Blubbler

Edit history:

Blubbler: 2015-03-04 04:49:03 pm

2015-03-04 04:44:50 pm

I did some tests with framerules on the All-Stars version and I think we need to change something (as in the community, idc about SDA rules, but of course the same logic applies).

SMB1: The framerule runs during the title screen and stage intros, but does not delay the entry into 1-1 as in the original.
This means that you can save up to 20 frames over another run just by starting the game and skipping the intro of 1-1 at a different time.
If you start timing at character control or pressing start doesn't really matter, the randomness is there either way.
One way around that would be letting the 1-1 intro play out, but that would invalidate every run so far and just sound ridiculous.
Another solution would be to start timing in the main menu, which would negatively affect some runs.
Or just accept that timing/luck on the title screen plays a factor.
I would prefer starting time from file select at the main menu. It solves this issue and the current best run by andrewg (5:05) mashes through the titlescreen anyway, so would not lose time unfairly (on the speedrun.com leaderboard a 5:35 and 5:46 would be negatively affected).

Lost Levels: Here the framerule does not run during the title screen, but it does run during the stage intro and the whole random 20 frame advantage applies again, when starting from character control.
I would definitely support switching to timing from pressing start. Solves the issue, does not negatively affect any run that I know of. Many people do it this way anyway.

I did some tests with framerules on the All-Stars version and I think we need to change something (as in the community, idc about SDA rules, but of course the same logic applies).

SMB1: The framerule runs during the title screen and stage intros, but does not delay the entry into 1-1 as in the original.
This means that you can save up to 20 frames over another run just by starting the game and skipping the intro of 1-1 at a different time.
If you start timing at character control or pressing start doesn't really matter, the randomness is there either way.
One way around that would be letting the 1-1 intro play out, but that would invalidate every run so far and just sound ridiculous.
Another solution would be to start timing in the main menu, which would negatively affect some runs.
Or just accept that timing/luck on the title screen plays a factor.
I would prefer starting time from file select at the main menu. It solves this issue and the current best run by andrewg (5:05) mashes through the titlescreen anyway, so would not lose time unfairly (on the speedrun.com leaderboard a 5:35 and 5:46 would be negatively affected).

Lost Levels: Here the framerule does not run during the title screen, but it does run during the stage intro and the whole random 20 frame advantage applies again, when starting from character control.
I would definitely support switching to timing from pressing start. Solves the issue, does not negatively affect any run that I know of. Many people do it this way anyway.

Alko

2015-03-06 03:23:19 pm

Caution: This user contains Kana ^_^

Both suggestions sound reasonable to me (being the kid who watches the cool kids from behind the fence and wants to be part of them Wink

).

It definitely is an interesting fact to keep in the back of one's head in case there's any SMB marathon commentary to be done xD

presjpolk

Edit history:

presjpolk: 2015-03-08 08:33:26 am

presjpolk: 2015-03-08 08:32:45 am

2015-03-08 08:32:28 am

HELLO!

Other games do this: Press start at a different time and you change stuff. Legend of Zelda runners are controlling Start presses in order to manipulate bomb drops, for example.

Why is this a problem that what you do before control (and therefore timing) begins, affects the time of the run?

dxtr

2015-03-08 10:34:49 am

Speedrunner

^ Indeed. TMNT also, starting the game at a certain point for better rng isnt anything new. It's Any% right? Kappa

Blubbler

Edit history:

Blubbler: 2015-04-25 02:13:35 am

2015-04-24 07:48:54 am

darbian

2015-04-24 06:08:22 pm

Congrats Blubbler! Really amazing run. That second hammer bro you killed in 8-3 was terrifying.

Le Hulk

2015-04-25 12:55:58 am

SNES rules!

I was fortunate enough to see the end live. Congrats Blubbler!

Blubbler

2015-05-04 12:25:50 pm

I wanted to improve that run for a long time, but I didn't get around to it until now. It's pretty good except for the axe grab in -3. Can still be significantly improved with a Small Mario wallclip in 1-2, but I can't do it.

Alko

2015-05-06 05:03:10 pm

Caution: This user contains Kana ^_^

Real nice runs you had there ^__^

andrewg

Edit history:

andrewg: 2015-05-15 02:02:14 pm

andrewg: 2015-05-15 01:58:17 pm

2015-05-15 01:57:38 pm

Hi! I'm andrewg!

Blubbler, congrats on the warpless, very nice Smiley

I knew it was only a matter of time before someone topped it.

I wanted to pop in here to get very picky on timing schemes. I was reading over the SDA rules because of the timing for Super Mario Bros 2. I believe (I'm pretty certain anyway), that timing for smb2 starts upon being able to control the cursor to select a character, but SDA rules say from character control. My assumption had always been first input of what is considered gameplay to last input of gameplay:

"For games without timers, a simple real-time measure is used. When the player first gains control of the game's character, timing begins. At the end when control is lost, even if that's long after the final battle, the timing stops. "

So I was thinking about this game, and I think timing should start at the same moment, but end 1 frame earlier? I'm not sure. Anyway, just something I was thinking about. I'm more interested in adjusting the smb2 SDA timing, because I believe it is incorrect. Thoughts?

Blubbler, congrats on the warpless, very nice :) I knew it was only a matter of time before someone topped it.

I wanted to pop in here to get very picky on timing schemes. I was reading over the SDA rules because of the timing for Super Mario Bros 2. I believe (I'm pretty certain anyway), that timing for smb2 starts upon being able to control the cursor to select a character, but SDA rules say from character control. My assumption had always been first input of what is considered gameplay to last input of gameplay:

"For games without timers, a simple real-time measure is used. When the player first gains control of the game's character, timing begins. At the end when control is lost, even if that's long after the final battle, the timing stops. "

So I was thinking about this game, and I think timing should start at the same moment, but end 1 frame earlier? I'm not sure. Anyway, just something I was thinking about. I'm more interested in adjusting the smb2 SDA timing, because I believe it is incorrect. Thoughts?

darbian

Edit history:

darbian: 2015-05-15 09:30:06 pm

2015-05-15 09:27:49 pm

I had a good run yesterday:

http://www.twitch.tv/darbian/c/6691086

Using the frame count from my local recording, it comes out to a 4:57.88, however I know this isn't the actual time since I'm recording at a frame rate that differs from the NES. I remember with my last PB the timing from my recording ended up being 0.01s slower than the true time, so I think it is within +/- 0.02s from 4:57.88. I'll hold off of submitting to speedrun.com until I get an official timing (Blubbler?). Time to start grinding fast 4-2......maybe.

Mistakes:
Too high on the wall jump in 8-2 and also way too far to the right when entered the pipe.
Went 3 or 4 pixels too far in the turnaround room
Touched down briefly in the middle of the water section, and also again shortly after (I have NEVER touched down at that spot before, guess it was nerves)
No swag coins in 4-1 Sad

Water exit looked pretty perfect to me, also got the 1 frame jump to start room 2 in 8-4

darbian

2015-05-18 02:03:31 pm

Before I was able to get the exact time of the run above, I managed to best it:

4:57.709 http://www.twitch.tv/darbian/c/6698125

This run is identical time-wise until 8-4. My last 8-4 had a sloppy wall jump up to the pipe, a sloppy turn around room and multiple touchdowns in the water (although a perfect or near-perfect water exit).

This 8-4 was the opposite: Nearly flawless wall jump/pipe entry, great turnaround room, no touchdowns, but a horrible water exit. I also think I lost a frame on the axe.

After achieving this time I finally decided to give the faster 4-2 some attempts, and I actually got it once after a few tries. Unfortunately the run died in 8-1. I tried many more times but was unable to get it again, so to date I've only had one faster 4-2. Hitting that one specific pixel on the pipe is just ridiculously difficult. I'll still attempt the faster 4-2 on and off, but I honestly think at this point I have a better shot of further improvement by getting a better 8-4 with the easier 4-2. The odds of getting the faster 4-2, not getting screwed over by 8-1 framerule RNG, getting the optimal bullet bill shot + successful flagpole glitch and a great 8-4 just seem so astronomically low.

Before I was able to get the exact time of the run above, I managed to best it:

4:57.709 http://www.twitch.tv/darbian/c/6698125

This run is identical time-wise until 8-4.  My last 8-4 had a sloppy wall jump up to the pipe, a sloppy turn around room and multiple touchdowns in the water (although a perfect or near-perfect water exit).

This 8-4 was the opposite:  Nearly flawless wall jump/pipe entry, great turnaround room, no touchdowns, but a horrible water exit.  I also think I lost a frame on the axe.

After achieving this time I finally decided to give the faster 4-2 some attempts, and I actually got it once after a few tries.  Unfortunately the run died in 8-1.  I tried many more times but was unable to get it again, so to date I've only had one faster 4-2.  Hitting that one specific pixel on the pipe is just ridiculously difficult.  I'll still attempt the faster 4-2 on and off, but I honestly think at this point I have a better shot of further improvement by getting a better 8-4 with the easier 4-2.  The odds of getting the faster 4-2, not getting screwed over by 8-1 framerule RNG, getting the optimal bullet bill shot + successful flagpole glitch and a great 8-4 just seem so astronomically low.

Alko

2015-05-18 04:44:48 pm

Caution: This user contains Kana ^_^

Wow, congrats².

Sorry for ignorance and stupidity from my part, but does that make you the first player to achieve 4:57 on console? (At least you're the only one I see on speedrun.com …)

andrewg

2015-05-18 04:56:06 pm

Hi! I'm andrewg!

Darbian, huge congrats! I knew it was only a matter of time. Smiley

Great work!

darbian

2015-05-18 08:25:40 pm

Quote from Alko:

Wow, congrats².

Sorry for ignorance and stupidity from my part, but does that make you the first player to achieve 4:57 on console? (At least you're the only one I see on speedrun.com …)

It does. The first 4:57 didn't surprise me quite as much because I was on pace for it so many times in 8-4 in the days before I actually got it. I knew it would happen sometime. The second 4:57 happened on the next run that made it to 8-4 on the same pace, that was stunning to me.

Quote from andrewg:

Darbian, huge congrats! I knew it was only a matter of time. Smiley

Great work!

Thank you! Your dedication to this game is incredibly inspiring, and I'm really excited for your return to attempts.

Blubbler

2015-06-06 03:28:33 pm

http://www.twitch.tv/i_o_l/c/6799462

Blubbler

2015-06-07 06:42:38 pm

http://www.twitch.tv/i_o_l/c/6805250

Blubbler

2015-06-23 01:38:23 am

and another one:

Blubbler

2015-08-03 12:03:02 pm

http://www.speedrun.com/smb1/thread/dsmuv
some questions about leaderboard rules, linking here for visibility

also I got 19:09 SMB1 warpless and 8:06 SMB2j any% since my last post

Tompa

2015-10-19 03:27:46 am

Darbian got a new any% WR! A time of 4:57.62!!

All info and links can be found here: https://www.reddit.com/r/speedrun/comments/3pawbn/wr_super_mario_bros_any_in_45762_by_darbian/

PJ

2015-10-19 10:42:29 am

Is PJ

DARBIANNNNNNNNNN

sinister1

2015-10-19 02:10:22 pm

sinister1

so pro u don't even know

Simply amazing!

AdHoc

2015-12-23 10:53:14 am

I cannot believe andrewg wasn't the one breaking 4:57.
Are you still alive?

cak

2015-12-25 02:52:28 am

twitch.tv/ccaakk

Lol...in fact, andrewg hit 4:57.9 this week, using every trick possible, but sadly messed up a bit in 8-4. Otherwise, could have been low 4:57 and a new record.