Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
<- 1  -   of 53 ->
--
--
List results:
Search options:
Use \ before commas in usernames
a real 5:05 wouldn't replace it? you're kidding, right?
shit, sorry, i read that as 5:06 for some reason. yes, of course it would.
Visit my profile to see my runs!
I'm just glad sdkess is back.  When he left, it was like seeing one of my favorite tv shows cancelled.

....that might not sound particularly endearing, but it is...  Wink
18319 frames.

5:05.622 seconds.

congrats!
bläää
Hey, Dr. Nate, How do you calculate frames from a tape?

It seems like your calculations isn´t the same as TGs.

Hope Scott pull off some great runs on other games soon. what about Ninja gaiden?
Edit history:
nate: 2006-10-19 06:59:17 pm
open file in vdub, trim off frames when player doesn't have control. but i actually timed it wrong just now according to radix -- it's actually about two frames closer to 5:06. but still easily 5:05.

the first frame is the frame *before* mario shows movement. the last frame is the frame *after* he no longer has control.
Haters gonna hate
Wow, just wow.

Way to go after a run many thought to be already optimal and beat it.  Congrats man!
Yep yep yep!
Quote:
18319 frames.

5:05.622 seconds.
Why, when I calculate that out plus those two extra frames, am I getting 5:05.35?

Anyway, this is exraordinarily awesome and I can't wait to see it.  I wonder what the lowest time is for a no-glitch run of SMB.  Would 5:04 be possible?
Edit history:
andrewg: 2006-10-19 07:49:50 pm
Hi! I'm andrewg!
so scott's is a 5:05? That doesn't make sense, considering by TG his time was 5:08.7. trevor's run is 5:09.6 by TG.

I almost nailed a perefect run again today. that would really stink if i made a perfect glitchless run and It was not accepted because it wasn't 1 second faster than scott's.

"I lost about 0.3 seconds there. In other words, this run is only 0.3 slower than if I had done it segmented. "

you sure it isn't 0.6 seconds? this doesn't make any sense to me... I have the same number of game units lost in my run as you did in yours, yet your run is a lot faster?

your 309 in 8-4 counted as basically a 310, and your 357 counted basically as a 358 in 4-2?

I got a 199 in 8-1, did that maybe count as a 200?

ok, so when the video is up, anyone that wants to can compare his time to mine.

Master-88
Extreme nice job ;). Can i wait video near future?

I think his time must be insane, because Trevor run looks insane/perfect.

I think 5.05 is lowest possible time with human skills.
Hi! I'm andrewg!
a 5:01 with glitches i think would be possible.
bläää
Nice video andrew. I´m not an expert when we talk about counting frames and game units in this game but there was a few places in andrews youtube speedrun that maybe can be improved.

in 4-2 when you hit the plant thing. it seems like your jump from the floor up to the first coin block can be more perfect. maybe even the second one? Is it possible to grab the plant a little higher too?

8-4 It seems like you can hit the coin block earlier/more to the left. (the one to reach the pipe wich leads to the water part

annyway, smooth runs like that is awsome. good luck in your attempts Andrew.
Master-88
Quote:
a 5:01 with glitches i think would be possible.


Yeah, but i think SDA rules forbid all glitches.

5.00 is best glitch time on TAS.
bläää
SDA will allow it but not TG
Edit history:
Trevor: 2006-10-19 08:41:47 pm
hi Andrew, this is Trevor (in case you didn't know). yeah, it's me, hi everybody.

your 5:09 run might actually be a very high 5:08. i'm serious. with virtualdub, your run is 9266 frames which, divided by 30, is 5:08.867. i timed it manually with a stopwatch and got 5:08.9.

are you sure it's 5:09.3 like you said before?

another thing: you didn't actually finish 1-1 in 20 seconds according to my timings. even though you finished with 370 timeunits remaining, you probably didn't go under the 21 frame rule.
Jungle Rat
Wow, I'm really glad that my run made it into the 5:05 range. I wasn't sure that it would, but I'm happy that it did.

Andrew, it is possible that I made a mistake when I was timing my practice runs, but I know for a fact that a fast 309 is exactly the same as any 310 that I got on 8-4. If you have a 363 going into the floating pipe, you are ok as long as it was almost a 364. I'm pretty sure that the 311 I had on tape was not any faster than my 310. To get a 311, you are only just a few frames faster than a solid 309, but these frames aren't enough to put you into the next 21 frame threshold. The 21 frame thresholds are completely independent from time units. I've never proven this, but it is probably possible to get a 310 (and maybe a 311) and actually be slower in real time than a 309!

The same 21 frame deal goes for 8-1. I have one practice run on tape where I got a 199 (missed 200 by a single VCR frame), and it is exactly as fast as a typical 200 finish. All of my other 199 finishes are 0.35 slower than the 200 finishes.

Time units are a decent indicator of how well you're doing, but they are not precise. I based all of my play on actual stopwatch timed VCR recordings. I only took extra risks for speed when I knew that it made a difference.

I think SDA divides by 59.94/29.97 rather than 60/30 when calculating time. That could explain some discrepancies that people are having.
Edit history:
andrewg: 2006-10-19 11:50:08 pm
Hi! I'm andrewg!
I have the original VHS tape in my room, and I have timed it many times. The video online perhaps could be a little messed up, as the fact that I am a little new to putting the video online.

i understand what you mean that a 309 could be faster than a 311 due to getting slow times going down the pipe.

I thought that 1-1 was 20.9 seconds because it was faster than trevor's when I compared them side by side.

smb1 is kind of making me mad, why couldn't it have just gone by game units! Then I would be exactly tied with scott. i am so close to a 5:08 too.


I just realized something funny... check my post count.
Jungle Rat
Before I forget, I do appreciate the fact that some of you missed me being around. Cheesy Thanks!

As far as being "back" goes, I am but I'm not. My interest in speed runs began declining in mid-2004 until it finally bottomed out late last year. My interest level is still low, but low interest does not equal no interest. I'll still work on something from time to time if I feel inspired. The SMB run is actually my 3rd speed run of 2006. I might try to crank out another run before the year ends if I find the time.

Sometimes I run a game purely for fun/nostalgia and give up once it starts to become annoying. These runs, although not bad, are not that impressive to watch and will probably never see the light of day. Anything that I consider to be really good will be submitted to SDA.
Quote:
I think SDA divides by 59.94/29.97 rather than 60/30 when calculating time. That could explain some discrepancies that people are having.

absolutely. this is ntsc after all.
Edit history:
DRybes: 2006-10-20 01:53:41 pm
O Zlda?
It seems like the problem here, is that we are more precise in our mathematical calculations and timing than the very people who made the game. If only they had known they were making the most important video game ever created, maybe they'd have timed it a little better ingame :p

Question though, as far as game time and realtime not matching up, like all these situations you guys are talking about where a 309 and a 311 can take the same amount of real time... where is time gained or lost? Is it slowdown and thus random what happens?
our timing should be real time as far as i know.
The clock doesn't move while you're going down a pipe.  That can create a difference.
Edit history:
sdkess: 2006-10-23 02:34:54 am
Jungle Rat
Quote:
all these situations you guys are talking about where a 309 and a 311 can take the same amount of real time... where is time gained or lost? Is it slowdown and thus random what happens?

It isn't random or due to slowdown. Think of it like a window at the end of a level or pipe room that is only open every 21 frames. Say you have two separate runs where you reach the window on the 50th time it opens. If you arrive just after it closes the 49th time then you have to wait 20 frames for it to open the 50th time. If you get there just as it opens the 50th time then you can enter without waiting even though you arrived up to 20 frames later. Generally speaking, you can only save time in multiples of 21 frames (0.35 seconds).

This 21 frame window is completely independent of in-game time units. Level 8-4 can get weird because you are going down so many pipes. In some cases you might go down a pipe (like the floating pipe) just a couple of frames faster and think you gained time because you barely got a 364 instead of a high 363. Based on my tests, you are in the same 21 frame threshold either way so no real time is gained even though the game shows you being a time unit faster.

I've known about the SMB timing mechanism since I did my first speed runs. I discovered it by manually timing the first level (1-1) on various runs and observing no real difference.

Anyway, I hope that my explanation has cleared things up a little.
same principle applies to metroid fusion's random elements if anyone is familiar with that.
Edit history:
DRybes: 2006-10-22 08:29:37 am
O Zlda?
Whoa. That was an incredibly detailed explanation. I understand now, thanks  Cheesy

Edit/offtopic: nate has a quote from one of my favorite albums in his sig... awesome