Sonic 2 Emerald Hill Act 1 in 0:23

[LucaPM]

New Record! 0:21

Alright.In a Sonic forum that I often visit (mostly because there are lots of fans of my own fangame) there was a topic where everyone could post their own records in any Sonic game so I decided to post my own =P

http://www.youtube.com/watch?v=wUgQ0F5aeAs

Please comment and rate

Edited October 14, 2009 by LucaPM

[azoo]

0:23 isn't bad, if I do say so myself. However, I think I could do Emerald Hill Act 2 in about less than 0:40 (thanks to that fantastically-placed Speedshoes monitor). I'll post up a video soon if I can get one to load.

[DarkOverord]

I've done EHZ1 in 0:35 once before (I don't normally due to lol emeralds). It doesn't surprise me that someone's done it that quick.

[Super Lombax]

According to the UK edition of 'Guiness world records:gamers edition 2009' The world record for emerald hill is 0:20, so your records pretty close. Mine is 0:33. I've been wanting to break the world record for fun since I've read about it. I still try from time to time...

Super Lombax

[LucaPM]

Thank you all guys!

Acutally,all this was born from a challenge with my friend called Alessandro where him challenged me to beat his own record (0:26)

Still,It's an huge satisfaction to have done that run

[bmn]

The Speed Demos Archive record is 19 seconds, by Joe Stanski. The route is almost identical to the one that got the 20 seconds WR at SoS '08, but gets over the first main junction quicker using a perfectly-timed spindash.

Because of the nature of video game records, Guinness is less stringent about them than they are about general records. As well as the odd adjudicated record as with EH1, they generally source their records from Twin Galaxies and to a lesser extent SDA, but they're very hit-and-miss about it

TG, which calls itself the official WR site due to its Guinness affiliation, is a very straight-laced records site; emulators only for certain systems, no glitch abuse allowed, one-session only, and some very strange arbitrary rules in places - e.g. you're only allowed to use the default pistols in Tomb Raider 1. SDA focuses on publishing videos of speedruns rather than compiling records; no emulators, glitch abuse allowed except for game-breaking ones. Both sites use real time when the game timer doesn't do its job properly (as opposed to e.g. the Sonic Center, which allows timer-manipulation tricks such as time stopping).

That said, neither TG or SDA are anywhere close to being a comprehensive source of records. Most of SDA's records that have equivalents on TG are quicker than their "world records", even discounting the differences in rules, but SDA's stringent video quality and peer verification procedures mean it takes a lot of work to get a run on that site (and thus there's often stuff on Youtube etc that is quicker). In general, if you're reading the GWR Gamer's Edition book, you should take everything you read with a grain of salt.

Back on topic...

Personally, I can't remember my best time on EH1. I never really put any effort into trying to go fast on it (I prefer GH1 >_>), but IIRC it's between 28 and 30 seconds.

[Tornado]

0:23 is pretty good. The best I've ever managed to do is 0:24 I believe, and I'm not sure where one could get another 5 seconds out of it (probably why I haven't been able to do any better).

you're only allowed to use the default pistols in Tomb Raider 1.

That is odd. It sounds a bit like a Tyson Doom run, actually. Is there a separate record for other types of times when the games have such strict rules sets?

Edited October 13, 2009 by Tornado

[Frozen Nitrogen]

The Speed Demos Archive record is 19 seconds, by Joe Stanski. The route is almost identical to the one that got the 20 seconds WR at SoS '08, but gets over the first main junction quicker using a perfectly-timed spindash.

Oh.

My.

God.

The bouncing! The bouncing!!

[bmn]

0:23 is pretty good. The best I've ever managed to do is 0:24 I believe, and I'm not sure where one could get another 5 seconds out of it (probably why I haven't been able to do any better).

Video linked in the above post if you didn't see it earlier.

That is odd. It sounds a bit like a Tyson Doom run, actually. Is there a separate record for other types of times when the games have such strict rules sets?

TG's rules in full for Tomb Raider

Nope, that's a hard and fast rule. Justification: TR1 is easier than the other games, thus you can't use anything but the pistols.

There's not a lot of those kind of weird rules, tbh. However, I'm affiliated with SDA which likes TG about as much as a hardcore Sonic fan liked Mario back in the day - you can't expect me not to pick on something like that ;p

Oh.

My.

God.

The bouncing! The bouncing!! :blink:

You should see GH3 in the TAS of Sonic 1. 30 freaking seconds...

[LucaPM]

I've made a new record!

Emerald Hill Act 1 0:21

http://www.youtube.com/watch?v=wUgQ0F5aeAs

Getting so close to the WR...

[Ashley Marsh]

You should see GH3 in the TAS of Sonic 1. 30 freaking seconds...

;___;

I think I died a little inside watching that...

[Cortez]

You can't beat the 20 second world record, I'm good friends with the person who achieved it (Louis Tsiatallou baby ) and hes since edited the EHZ Act 1 stage to be completely flat and even then it can only be done in 20 seconds.

[bmn]

You can't beat the 20 second world record, I'm good friends with the person who achieved it (Louis Tsiatallou baby ) and hes since edited the EHZ Act 1 stage to be completely flat and even then it can only be done in 20 seconds.

The Speed Demos Archive record is 19 seconds, by Joe Stanski. The route is almost identical to the one that got the 20 seconds WR at SoS '08, but gets over the first main junction quicker using a perfectly-timed spindash.

Admittedly, the above is something like 19 seconds and 57 frames, but it's still under 20. TSC lists 12 people with 19s (sure, most of them will be emulator runs and you should take TSC's times with a grain of salt sometimes, but 12 people is a lot).

The current quickest TAS uses a crazier bouncing route to get 17, but even if a human player could do it, they'd have a hard time getting 18.

[Cortez]

Admittedly, the above is something like 19 seconds and 57 frames, but it's still under 20. TSC lists 12 people with 19s (sure, most of them will be emulator runs and you should take TSC's times with a grain of salt sometimes, but 12 people is a lot).

The current quickest TAS uses a crazier bouncing route to get 17, but even if a human player could do it, they'd have a hard time getting 18.

Heh, The Sonic Center, it'll take more than a grain of salt for me to believe anything they do. Though surely there'd be a slight button delay on the Megadrive version compared an emulated version? I'm talking litterally milliseconds but it must be enough to take into consideration?

[bmn]

Heh, The Sonic Center, it'll take more than a grain of salt for me to believe anything they do. Though surely there'd be a slight button delay on the Megadrive version compared an emulated version? I'm talking litterally milliseconds but it must be enough to take into consideration?

Input delay's a weird thing. I read a tech article not long ago that stated that all consoles (not sure if this includes PC or not) take at least three frames (at 16.6.. or 20ms of time per frame) to process your input. Even if PCs via keyboard don't have that inherent delay (as I say, I don't know), most PC controllers and adapters today also have a delay due to USB auditing being scheduled and not handled by interrupts as the controllers of yore did.

I'm more than willing to believe that there's a heavier input delay on the real thing, even if it's only, as you say, in the timeframe of 10ms or so. I recall playing Sonic 1 on MD a while back and having a harder time of it than playing on emulator - at the time I put it down to the PAL/NTSC speed difference (using PAL as a benchmark, the early games run 20% faster on an NTSC console).

That said though, either stanski's run was on a Genesis console, or he's fooled a lot of people for nearly 3 years (SDA's peer verification includes the weeding out of emulator runs, and it's not usually that difficult to distinguish between the two). Not counting the faster play - a game second is 1.001s in NTSC, 1.2s in PAL - UK and US Sonic 2 are exactly the same game, so anything an American player can do, a British player could theoretically do. But, as I say, that's a slightly faster strategy than the WR, and even then it only just makes it.

[Phos]

I'm pretty sure the Genesis poles each button multiple times per frame, I don't think there's going to be any lag from that, and if there is, it's probably less than you'd get from a PC controller.

On the PC front, some of the higher end mice and I think keyboards as well can run in some kind of 1000Hz mode to eliminate delays like that.

[bmn]

Found it.

http://www.eurogamer...-factor-article

The meat:

• The lowest latencies a video game can have is 50ms (three frames) - the PS3 XMB runs at this rate, but few games reach it.
• Most 60FPS games have a 66.67ms latency - Ridge Racer 7, for example.
• 30FPS games have a minimum potential lag of 100ms, but many exceed this.
• Game developers should test their own games using the camera technique in order to weed out bugs - West says that Heavenly Sword's response slows down to 300ms just by turning the character, and reckons it's a technical issue that should have been resolved before going gold with the game.
• Citing GTAIV as an example, West suggests that a 166ms response is where gamers notice controller lag, which could also explain the Killzone 2 furore too.
• Game reviewers should accurately measure latency for their reviews where controller lag is an issue, in the hope that sloppy game response times come under far more scrutiny.

Multiply all those times by 6/5 for a 50/25fps game.

I don't have numbers so this is all guesstimation, but as a stock Windows audits USB every 8ms, a typical response time for keyboard/mouse is likely somewhere around that, plus the time it takes the OS to send it to the emulator or whatever, and the emu to handle the input (typically two frames or so - it can't just stop drawing a frame because you've done something). So, at a guess, 30-50ms, assuming it's handled well.