Quarterfinal
Deadline: Saturday, November 12, 2005, 9pm CET
The Rules
Ok, after a quite hard second round I decided to make it a bit smoother using a well known parameter
set. This time it's under the 94nop hill rule except the max. warrior length is just 50. P-space is
allowed this time, but you have to share the same p-space than your opponent using PIN 13. To force
you for agressive strategies the scoring scheme will be Win: 1 Point; Tie: 0 Points, Lose: 0 Points.
The parameter setting are:
Coresize | 8000 |
max. Cycles | 80000 |
max. Processes | 8000 |
max. Warrior Length | 50 |
min. Distance | 50 |
p-space | allowed, but must share the same PIN 13 |
Rounds | 1000 |
The commands for pmars is:
pmars -s 8000 -c 80000 -p 8000 -l 50 -d 50 -r 1000 -= "(W-1)/S" |
Fixtures and Results
Well, I can't put you off the scent because you all see through the scoring scheme;-)
I thought I would see more p-spacer, but it looks that the max. warrior length of 50 wasn't
enough for you to consider it.
el kauka vs. Roy van Rijn |
Fizmo's Comment
Roy seems the clear favourite in this match. Let's see if el kauka can surprise him
with an unexpected entry
The Comments
el kauka:
Regardless what i'm sending - to face roy means you
are doomed. Victory needs no explanation, defeat allows none
The Results
Roy sent a modified Hullabaloo, while el kauka sent a p-warrior including a oneshot.
That sounds like a close battle, but the p-warrior's oneshot wasn't able to dominate over
the stone/paper. Overall, a clear victory for Roy van Rijn.
Battle QF1 | Result |
---|
el kauka | 143 |
Roy van Rijn | 458 |
Fizmo's Comment
A very interesting match, because both surprises with very good entries so far.
The Comments
inverses:
This time I'm not well prepared, I just had no time. And the rules are
too typical (still can't understand why Fizmo alternated scoring formula,
this doesn't affects 1-on-1 results).
So I took my best warrior PhatPaper, added a stone, trimmed qscan
and here it is; no time to optimize it. It only loses to some oneshots.
I've tried to make a P-switcher, but it exceeded maxsize by few instructions.
Probably I'll lose out but anyway participating in CSEC was a great fun.
The Results
inversed sent a qscan into stone/paper, while Fernandes paper/imps. The slight
advantage of inversed's qscan seems lost by the slower booting of his stone/paper. The
match was running nearly in a tie but S. Fernandes' entry finally achieved some few wins more.
Battle QF2 | Result |
---|
S. Fernandes | 26 |
inversed | 12 |
John Metcalf vs. Chip Wendell |
Fizmo's Comment
Two corewar giants clashes in this match, but only one can survive. Who will it be?
The Comments
Chip Wendell:
1. First off: use p-space, or not? The short warrior length
would make it difficult, plus I don't have much experience
building p-warriors. So I won't use it, but I may brainwash
if it doesn't hurt me too badly.
2. The "aggressive" scoring scheme would be a factor to
consider if this was a round robin, but for a one-on-one
battle, it's completely irrelevant. As with the usual scoring
formula, the only important thing is to win more than I lose,
regardless of ties.
3. So, what strategy should I pick? Stone/papers seem to be
the most robust warriors, vulnerable mostly to scanners. John
is a scanner guy, but he may avoid those for that reason.
Besides, there's a chance (however slim) that John will be
fooled by the scoring scheme into thinking that I won't enter
a paper.
4. Hullabaloo seems to be about the best recent S/P, so I'll
start with that. The qscan section is too long, but I don't
want to enter without one, since that would make me even
more vulnerable to scanners (no decoy). Reepicheep has a
nice short one, so I'll borrow that. It fits very nicely.
5. I want to add brainwashing, to foil any of John's attempts
at p-brains. After wasting time determining that adding an
STP line to the paper wrecks the paper, I added a line to
the stone. It drops from .5c to .4c, but it doesn't seem to
hurt its score much, so it's probably worth it.
6. Send it in, and cross fingers...
The Results
John sent a anti-imp stone hoping for a scanner or stone/imps, while Chip expected a
scanner or a p-spacer choosing a stone/paper with brainwash. Finally it was a clear
victory for Chip Wendell.
Battle QF3 | Result |
---|
John Metcalf | 120 |
Chip Wendell | 628 |
German Labarga vs. Zul Nadzri |
Fizmo's Comment
Zul Nadzri sent the last rounds always papers, will he continue? Will German Labarga
surpises Zul with something unexpected?
The Comments
German Labarga:
Zul will surely submit a p-spacer with some kind of brainwashig.
If i submit a single warrior it will be probably beaten by
any of his components, so i have to submit a p^3 with some
brainwash detection cappability.
I first wanted to use a scanner+stone/paper or similar using a version
of Gihegruoerg, ;but can't fit something like that in 50 instructions.
So finally try a p-spaced paper/imp, with and without stone.
I know that there will be many ties and the scores will be low,
but taking a few poitns from my opponent would be enough to win.
The p/i is very strong and i think it can only be beaten by a
fast anti-imp scanner, so then, the stone pops up.
Zul Nadzri:
Again (and again) victory mostly depends on good anticipation
of opponent's strategy. Usually, it is better not to anticipate
anything and put the passion into a warrior. This time, I am going to
resurrect another off-hill warrior. May be can submit to the hill after
this, after all the effort spent for this round.
I picked Dawn 2, Roy's warrior because the core is definitely less than
50 lines.
Why Dawn? I admit it has disadvantages. So do other warriors. Remember
the stone/scissor/paper analogy?
Why paper? I am sure G.Labarga loves to destroy papers with his scanner...
...and I am going to give him a challenge that he is looking for! You
better win dude!!
More tips: this warrior can't beat Recon2
First, I removed Dawn's Qscan to meet the 50 lines standard.
Second, I removed all empty codes, making the warrior compact to attract
scanners to the 'decoy'.
Third, I filled gaps with stp.ab >0, #0 just in case it can be executed.
Put PIN 13 on top.
Fourth, Gave a good name "Peaceful Jawn". Ain't is beautiful?
Fifth, Jawn = John + Dawn. Acousticly, 'Jawn' is 'John'...almost the same
sound and frequency.
Sixth, send early to Fizmo and enjoy my holiday.
Seventh, if your name is John, this is not related to you ;)
The Results
Zul didn't sent a p-warrior as German expected but paper/imps. There was no scanner who could
be defeated by Labarga's p-warrior to get some wins. Zul get a close victory with only
a few (early round) wins because he hadn't spent cycles for a p-switcher before launching his
code.
Battle QF4 | Result |
---|
German Labarga | 0 |
Zul Nadzri | 60 |
The Second Chance Tournament Tree Fixtures
There is a second chance for all players who wants to continue the CSEC2005:
Here are the fixtures for the second round:
The Comments
datagram:
I don't know who my opponent is, but assuming they don't make
a scanner that is better than my scanner, I should have pretty good
chances to win. The warrior is a p-switcher that is 2 parts zooom
scanner and 1 part kline paper. The scanner uses a SJD clear,
SPL-JMP-DAT; works better against imps than SSD. Kline paper does
well against most Stone/Imp. Should be immune to brainwashing. Not
optimized, but scores decently against most mid-level warriors.
The Results
# | %W | %L | %T | Name |
Author | Score | % |
1 | 12.2 | 10.7 | 77.1 |
Planter | datagram | 122 | 100.0 |
2 | 10.7 | 12.2 | 77.1 |
Mowgli | LAchi | 107 | 87.7 |
Battle | Result |
---|
datagram | 122 |
LAchi | 107 |
The Comments
Mizcu:
Why should i spend time trying to make
something myself? Fooling around with
pspace would drive me crazy, and
i doubt that Nenad will use it
The Results
# | %W | %L | %T | Name |
Author | Score | % |
1 | 7.3 | 4.7 | 88.0 |
QHack1 | Nenad Tomasev | 73 | 100.0 |
2 | 4.7 | 7.3 | 88.0 |
CSEC | Miz | 47 | 64.4 |
Battle | Result |
---|
Nenad Tomasev | 73 |
Mizcu | 47 |
Jens Gutzeit vs. Janeczek |
The Comments
Michal Janeczek:
The new scoring scheme doesn't change anything in 'one must
fall' tournament.. So given my lack of time and decent performance
of Microvenator in RF4, I'll submit just this. Hopefully the
new name will be enough to stop Jens :)
Battle | Result |
---|
Jens Gutzeit | 384 |
Michal Janeczek | 403 |
The Off-Topic Round-Robin Results
Just keep in mind that only here the scoring scheme will affect the results.
|