UVa OJ Board

I have sent corrected set to admins before.
Not sure if it's uploaded yet.

Some harder I/O to help you: 0 28 1999 11 18 1999 11 28 5 28 1999 11 18 1999 11 28 5 13 1999 1 1 1999 12 31 0 30 1976 6 30 1999 11 28 1 1 1 1 1 1 3 1 4 1 1 1 1 1 3 1 0 1 2001 3 1 2001 12 31 5 23 2005 9 23 2005 9 25 6 24 2005 9 24 2005 9 24 0 25 2005 9 23 2005 9 25 3 7 2005 9 10 2005 10 5 0 9 2005 9 ...

Hi,

I have created a set for this, and hopefully it will be set soon. Here is an illustration of ACD > B is not needed, since
it can be derived from:

ACD > ACG > AB > B

Since D>G, and CG > B

Can you now find the remaining unneeded one now?

Note the Sample output in 461 only demonstrate the formatting (not correct output), since party of 4 clearly takes <60 minutes, so we can schedule the part of 4 for bench 1 at time 0, and it will be still available at time=60 minutes to be scheduled again. The correct output to the sample input is t...

Are you sure your puzzles satisfy:
"Then, every (vertical or horizontal) word of S occurs only once, and the set of all (vertical or horizontal) words of S is equal to W."

All horizontal and vertical words appear in the initial list W?

Is the data updated yet?

What is the maximum cassette duration? it does not say in the problem.
"Standard" seem to imply maximum should be 120, is this the case?

Also what's maximum number of songs?

Could you post some critical I/O? I keep get WA on this one, but don't know why.
thanks!

Actually the problem limits are too loose. It should disallow O(V^2) type of solutions. It should be O(V). 1) The better solution is first construct SCC (strongly connected components), as well as the DAG after SCC decomposition. 2) Perform bellman ford negative cycle detection on only one node for ...

It would be great if you can post the dataset somewhere, so I can check my result (since the judge data files are likely wrong). Thanks!

Seems the problem in all problems with special judge ... also 11208

Nevermind

I do get 5838170, but it would hit tle for me (without precalculation and nasty uncoding in base 100 to make the 40k source limit, which seems not what's intended!!!). At any time the partial cumulative result cannot become negative, and no fractions are allowed. Conisder posting your code, if you s...

Can you post your code?

Actually, I got ac in 0.5 seconds, using just min cost max flow.
I use a very simple trick to conver min cost flow to min cost max flow
-- just connect source to sink with infinite capacity and 0 cost.
This allows X to be "thrown away" before been produced. Then at max flow it's fine.

Hi,

Could it be roundoff error issues?
How did you solve this problem? Use spatial partition (quad tree) ?

best,

Josh