Race to Fusion

[Paul Stevens]

We better fix the web-site problem rather than try
to mail you the game.

When you go to dianneandpaul.net what do you see?
You should see a 'CSBwin' link. When you select the
CSBwin link, what do you see? You should see a list of
files. One of them should be MovieDM36.rar.

[Antman]

I fixed the problem, all i had to do was hit the refresh button once i got into the list of CSB files, sorry

[PaulH]

Another new record, using Stamm, reincarnated, (ie zero starting levels) and a time of 34m and 46s.

It was an up and down game, I am finding new ways to shave a second here, shave a second there.

I suppose this shows the importance of mana and wizardry of the game...

[Paul Stevens]

The 34 minute 45 second record is posted at
dianneandpaul.net
File MovieDM3445.rar

I figured I better start putting the seconds into the name.

(Fourth try....the charm)

[Paul Stevens]

Here we go with a sub-34-minute record. 33 minutes and
58 seconds by Paul Hayes playing reincarnated STAMM.

dianneandpaul.net
MovieDM3358.rar

[Second attempt]

[PaulH]

I set a new record using a ressurected character of 31 minutes and 45 seconds using Stamm. This shall be my last attempt, the mouse can't take any more!

(6th attempt)

[Paul Stevens]

The 31 minute 45 second movie is at dianneandpaul.net.
File MovieDM3145.rar.

So how much time can be saved by using four champions?

[Fifth attempt....too many]

[PaulH]

Would it save any though? More characters to heal and rest, longer to get the BOS, and weaker characters at the end. Two people may shave off some time though I doubt it.

[Antman]

I think none, that makes the boots of speed useless, unless you have four pairs.

5th

[DR]

Furthermore, using only one champion gives you the interesting possibility to dodge fireballs and other attack spells. ;)
And of course you are likely to never run out of food, as you need supplies for one instead of four.

[Max Cannaday]

I just spend hours reading this whole thing. I've been looking through the dmweb.free.fr site for a utility that will let me edit my players LUCK to maximum. dmute won't do it. Just want to play an otherwise normal FULL game from beginning to end, to see how mega-high luck changes the gameplay..

And it brings me here :)

Yes, using 1 char makes you immortal to all single-barrel projectiles. Just move his positional icon from left to right and they all wizz past you.

I often use the Mirror-Death bug (with gothmog) to grab the vorpal blade on level 6, and all the gold keys for level 2, in ghost-mode, before I even choose my champions :) (Vorpal blade so Hulk is a journeyman wizzard by the end of level 1)

I'm goign to install all of this and see the game won in 34 minutes. I find this so hard to believe!! It's so funny because the posts at the top of this forum talk about how hard it is (and impossible) to beat 1 hour :)

I want you all to know I share your obsession for this game. This world is long overdue for an official Windows Online Multiplayer version.

I'm a senior-level windows/network/protocol programmer specializing in C++... if anybody wants to team up for a fun project :) I'm also a GUI/ActiveX/ATL master to boot.
- Max

[Zyx]

Max, you can try Paul's editor: CSBuild.
Current version is 788.
http://dianneandpaul.net/CSBwin

[Antman]

I just finished watching Paul Haye's 31 minute record and decided inflict some more attempts apon myself. I doubt i could actually beat it, but i might aswell see if i can get below 35 minutes at least, after playing alot of conflux 2, i'm sure i can move much faster than in my previous DM/CSB records. So now i can drive myself mad with some more attempts.

[Tom Hatfield]

Guys, I would love to watch these movies, but I have absolutely no idea how to do it. I downloaded CSBWin and subsequently one of the movie archives — only to find that the archives already contain the correct version of CSBWin — but I can't figure out how to play back the log. Help is very much appreciated.

You might want to include a readme in the archives so other people don't run into this problem.

[beowuuf]

the movie needs to be called 'playfile.log' and be in the root directory with the correct csbwin version and dungeon.dat/graphics.dat, i believe (config.txt file is optional, right?)

then you go to the dungeon entrance, click on the 'misc' pull down menu and click 'playback' - the movie should run.

[Tom Hatfield]

I was just coming back to say I figured it out. Thanks anyway.

[fara]

the rules need to be accurate. make a non-save, non-reload category and a category with save builds. only the non-save,non-reload categories should be considered the real records. else it's too easy to build a fake run using multiple reload and the hics edited out.

also, the linux port of CSB should be equally admitted for runs and also it should be possible to submit avi files of demos played on the atari st while filming the atari monitor using a video camera.

[sucinum]

>>it should be possible to submit avi files of demos played on the atari st while filming the atari monitor using a video camera.

i think, we can decide this if anyone should try

[zoom]

i have an idea :
how about, at the end of the savegames from personal computers,
get a list of what hidden attributes have been used and how often.

not so special like how many fireballs and magic torches, how many swings and chops and climb downs, and warcries and vi potions ; but something more general.

There are the main classes Fighter ninja wizard priest and each is divided into 4 subcategories, hidden skills . Just show these hidden skills and the main skill and the steps walked and a monster toll as well as the levels of the characters.. i think that would be good, and maybe it is even easy to do..
For me, single player mode is far too difficult; but there should be a way of making these movies interesting, not only by a short duration. Maybe an extra race to fusion mod made with csbuild would be it....

[Gambit37]

I was just watching Pauls' record-breaking video on Youtube. It's completely misleading, because it's only 20 minutes of real-time and goes insanely fast. Why is it so fast? Speed-runs can only be viewed legitimately if they truly reflect the actual time spent, but these playbacks are way too fast and it's simply not possible to play the game like that.

I imagine CSBWin is doing something strange with timers/ticks to create playbacks this quick?

[Paul Stevens]

I imagine CSBWin is doing something strange with timers/ticks to create playbacks this quick?

Nothing that I know of. One-hundred percent certain
that it is nothing intentional. Perhaps the
Youtube encoding or playback?

It is certainly possible for the player to record
at low speed and playback at high speed.
(used to be called 'over-cranking') I don't
think there is any way to know this except to
ask the people who did the deed.

I'll record a twenty-minute session tonight and
keep a stopwatch on it. Then play it back and
see what happens. I'll report.

[Gambit37]

No need to record and time it -- you see very easily from the speed-run video that it's impossibly fast and no-one can actually play the game that fast.

I suspect you're right about the slow record versus fast playback. If that's the case, then the YouTube videos are very misleading (I did notice a few comments from people saying they couldn't play the game that fast and was he cheating)

[Gambit37]

Here's an example, speed run at 8 mins:
http://www.youtube.com/watch?v=CBmRWMx47Yg&NR=1

The DM Encyclopedia says that the record is currently 10m 37secs. http://dmweb.free.fr/?q=node/856

[Paul Stevens]

The time we use for the competition is the
number of clock ticks divided by 6.

If you run the game at a very slow speed
then it certainly becomes easier to keep
up with the Gigglers. This may or may not
be called cheating. I will appoint myself
the final arbiter in this matter but am willing
to relinquish the post. My decision....play
the game at any speed you like. The best
time will be determined by the number of
clock ticks, just as before. You may play it
back at a fast rate to impress your friends.
This allows for technical skills to compete
with physical dexterity and eye-hand
coordination to some extent.
Those are the rules. (Subject to change
when everyone cries "foul".)

As for the YouTube video. Mr. Bracegirdle
did not choose to display the time at the
end of the video. Perhaps he can submit
the recording in order to claim the new record.

[Gambit37]

Right OK, I understand now. I know it's only a bit of fun, but it is a tad misleading.

Just to check, I screen-captured the record breaking 10mins 37 secs at normal gameplay speed with no extra ticks, and it's actually 13mins 48 secs. Still insanely impressive though!!!

[beowuuf]

Youtube cuts off anythign above 10 minutes for normal uploads, right?

[Gambit37]

15 mins these days. Probably was 10 mins when that speedrun was uploaded.

[Antman]

Hi there, been lurking a bit recently again.... I didn't upload that video but it is the record holding 10:37 video, it's run without extra ticks so it plays back slightly too fast, %20 it seems.

That is the speed it always plays back at without extra ticks. But with extra ticks the playback it too long, the real time the game was played in is somewhere in the middle. I didn't slow down the game while I was playing. I thought because the game's internal maximum speed is 6 moves per second, but with the boots of speed you can move faster, but it's an illusion. You move faster real time, but the game time increases to keep up with you!

So when these games are played back they're not truly acurate of the speed you actually played the game in as the recording gets played back at strict 6 moves per second, which is slower than you originally played. I think that's what's going on anyway... could you verify Paul?

I didn't play as fast as the recording without extra plays, and I didn't play as slow as the game played back with extra ticks. I think you can demonstrate the effect by starting a game and watching how fast the ticks are going by on the clock, then holding down the spin button, you spin faster than 6 times per second, and you can see the rate at which the ticks were ticking over increases to keep up, so if you recorded yourself doing this, when you play it back you'll spin either too fast with without extra ticks and too slow with them.

I would guess that the actually time I played the game in was around 9 minutes maybe, I think it's impossible to tell.... the ticks are the benchmark for records, but unfortunately they nullify the boots of speed and how fast you are at turning a bit.

[Jan]

I think I'll need more coffee to understand this. *runs into the kitchen*

[Paul Stevens]

could you verify Paul?

I doubt that you are correct.
The 1/6th second ticks are generated at
a **very** low level in the code and could
not possibly be aware of what is taking place
in the game.

But......Since you seem quite sure....I will
investigate. Stranger things have happened.