View Page Source

Revision (current)
Last Updated by adelikat on 1/17/2022 7:21 PM
Back to Page

[https://files.tasvideos.org/bisqwit/hloop.gif|left]
!!! Interview of Bisqwit for [http://www.gameinformer.com/|Game Informer]
----
This interview was answered by [user:Bisqwit] for Game Informer
in November 2008. The interviewer is John Ondrey. It was
conducted by e-mail.%%%
Links and formatting were added afterwards.
----
> What do you do?  Where do you live?
I live in Finland,
and I'm a computer software engineer by trade -- been
doing that professionally for about 15 years.

> What does it take to become a tool-assisted speedrun champion? Can anybody do it?
It takes dedication and a bit of luck. Anyone can do it, provided they
take their try at a game that hasn't been beaten to perfection yet.
As with regular speedrunning, a TAS player must be keenly focused at
anything that may cost time in the game, such as bumping into obstacles,
non-obvious shortcuts. Even collecting some bonuses in the game may end
up costing time, when the game may spend time tallying your score later
in the game.

> How many TAS records do you own?  What is your most impressive run?  Why?
Currently I possess only two TAS records, though I have had a few more.
My most impressive run was that of the original Mega Man for NES.
It abused many glitches of that game that required extremely high
precision and understanding precisely how the game works, to pull it off.
Lately someone else improved the record with a few clever discoveries,
though most of the movie looks exactly the same.

> How long did it take to achieve your most impressive run?
> How many hours a day did you play?

About a year, but naturally I did not play every day.
I'd say it took perhaps around 300 hours to produce it. Possibly more.
A lot of the time was spent in making graphs, discussing plans with other
experts of that game, disassembling the game code to figure out how it
works, etc. Such work is carried out very rarely for TASes. Maybe only
for the top five or ten, of all the thousand or so TASes that exist.

> Are TAS's more difficult than 'regular' speedruns?  Why or why  not?

They are not easier and not more difficult. With unassisted, i.e. "regular"
speedruns, a significant portion of time is spent trying to develop one's
reflexes to be able to carry out some stunt in the game. With TASes, an
equivalent portion of time is spent running the game at extremely slow  
motion, perfecting the tiniest movements in the game. Both types of
speedrunners still usually spend the majority of time just ''planning''
the speedrun and probing the game for its weaknesses. To figure out
''how'' they could pass through a wall, for example.


> What is it about tool-assisted speedruns that you enjoy so much?

I enjoy the insight to game programming that it gives; the subverting
of the games' rules by finding out glitches and abusing them; and the
visual performance that comes from movement without a single flaw.


> What would you say to those who disregard tool-assisted speedruns
> altogether, saying it's cheating?

I would say: you are missing a lot, beginning with the point,
but I cannot blame you.


> Do you think it's possible for humans to perfect a speed run for any
> game using tools? Why or why not?

This question is not as interesting, as is the question of whether it is
possible for humans to perfect a speedrun for a game _without_ using tools.
The answer to that question is, in my opinion: rarely, except for the
simplest games. For games that require reflex or luck, you must have
incredible dedication to come close to the TAS records. In my opinion,
that dedication is much better spent by using TAS tools. Much less life
wasted :) But when the unassisted speedrun movies are ''not'' perfect
(though still trying hard to be), they are most interesting to watch.


> How do you go about figuring out new techniques for faster times?
> Do you experiment random things? How do you know what to experiment?

A lot of it comes from experience and insight, some from luck.
There are typical pitfalls in game programming that cause
similar bugs/traits in many games, and learning about them
helps developing techniques. Some people just try random things.
Having a keen eye for anything out of ordinary helps a lot.
As in, "wait, how did that happen? Can I do it again? Is it useful?".


> To you, what is the most impressive TAS you've ever seen? Why?

I don't have a personal favourite, but many of the Super Mario World
movies are quite entertaining to watch.