hacker.org

Hi folks,
I'm just trying around the Flash Flood challenges which are very interesting, but it seems kind of strange for me that the error message "failed to flood correctly" _always_ appears, even when the real reason is exceeding the cycle limit. My code definitely floods correctly because I used the same for the "Flash Flood warmup" challenge - the only difference in the sequel is the lower cycle limit.

Could it be that the challenge evaluation is kind of buggy? (maybe in connection with the server hack?) I'm asking that because the "Flash Flood warmup" challenge actually _does_ inform you when the cycle limit is exceeded, and the evaluation should exactly be the same, only with another cycle limit.

Want to try it out yourself?
Try to type in a simple endless loop "04-g" as solution in "Flash Flood warmup.". Result:

Array ( [result] => -1 [message] => too many cycles: 25002 (PC=1, STACK_SIZE=1) [output] => )
Cycle count: 25001

Submitting the same answer in "Flash Flood", however, results just in

'failed to flood correctly' is incorrect.

Strange, isn't it?

i also tried to solve it today and wondered that the error message is not that detailed like in the warmup challenge
i tried to solve the warmup challenge with my program and it took only about 17500 cycles to solve it
now i don't know whether there is a mistake in the program or the program is still too slow, so it would be usefull to know at least the number of cycles

I brought this up before - search for other "Flash Flood" threads. The lack of informative error messages is deliberate, to make the challenge harder in an obfuscated kind of way. It depends on the algorithm, but you do need to be well under the cycle limit on the warmup in order to pass Flash Flood, because the board is more complex. The other threads have warmup cycle counts for algorithms that do pass the main challenge.

@gfoot: Oh, you're right, thanks for the hint with the other thread (I should use the search function before posting in future ) - after reading the discussion I'm pretty sure now that my algorithm *does* flood correctly, but it simply takes too many cycles. Well, time to start happy optimizing