I got TLE in 1000 milliseconds (1 sec)
when I tried to submit my 929.
Is the default time limit 1 second now for
all problems?
Thanks.
New Time Limit is 1 second?
Moderator: Board moderators
-
- Experienced poster
- Posts: 122
- Joined: Sun Nov 13, 2005 10:25 am
- Location: Taiwan
The test data used on the contest is different from which used on the old system, and usually, the test data used on the contest is smaller.
I think that new system using the datas which is used on the contest, because on 11266, my code run 1.908(s) on old system but 0.330(s) on new system. 11267, my code run 0.219(s) on old system but only 0.070(s) on new system. (So some problem's time limit only 1 second)
However, the running time on the ranking compare those submissions which use old system as which use new system. I think it is... strange. And it's not fair for those using old system.
I know there are a lot of things wait for the system designers to do now, I think that everyone can wait one or two month(even longer) to see this bug be fixed.
Finally thanks all who dedication themselves to new system. The system can not be found without them, thank you
I think that new system using the datas which is used on the contest, because on 11266, my code run 1.908(s) on old system but 0.330(s) on new system. 11267, my code run 0.219(s) on old system but only 0.070(s) on new system. (So some problem's time limit only 1 second)
However, the running time on the ranking compare those submissions which use old system as which use new system. I think it is... strange. And it's not fair for those using old system.
I know there are a lot of things wait for the system designers to do now, I think that everyone can wait one or two month(even longer) to see this bug be fixed.
Finally thanks all who dedication themselves to new system. The system can not be found without them, thank you

Well, someone set a rather arbitrary time limit of 1 sec for all problems. I really doubt that the new server is 10x faster than the old one. I just tried a last few accepted submitions I had and they were 3-5x faster... or 10x slower(!). These slower ones are the ones with low times on the old UVa - I think the startup overhead of Java is counted against the running time of the solution, I don't know...
I tried some of the ones that barely passed time limits on the old UVa and they get TLE on the new UVa (both C and Java)
I tried some of the ones that barely passed time limits on the old UVa and they get TLE on the new UVa (both C and Java)
Time Limit now 3seconds
Just a few bugs now ...
1) The default time limit has been changed to 3 secs, but
it's still the same for all problems (3secs for all problems) ...
making some impossible.
2) There is also no memory information for some reason.
3) Many special character do not appear in the HTML descriptions,
as they did on the old server.
Otherwise the new server is great !!! Everything else
seem to run smoothly!! Thanks to the UVA team.
1) The default time limit has been changed to 3 secs, but
it's still the same for all problems (3secs for all problems) ...
making some impossible.
2) There is also no memory information for some reason.
3) Many special character do not appear in the HTML descriptions,
as they did on the old server.
Otherwise the new server is great !!! Everything else
seem to run smoothly!! Thanks to the UVA team.
I have posted similar things in topic http://online-judge.uva.es/board/viewtopic.php?t=21466 .
I see that contest ID 172 has 5 ME (memory limit exceed) verdicts. So memory limit still exist?
I also think that Problem ID should be shown in "Last 50 Submissions" and "My Submissions" page, because there are some problems with same name.
I see that contest ID 172 has 5 ME (memory limit exceed) verdicts. So memory limit still exist?
I also think that Problem ID should be shown in "Last 50 Submissions" and "My Submissions" page, because there are some problems with same name.
Some other flaws:
1.
I think it should be a submission error.
2. I submitted the same code twice, and get two very different runtimes 0.000s and 0.010s! Such big difference do not happen in old judge system.
1.
Code: Select all
5902457 SRX Time limit exceeded C++ 0.000 2007-09-08 10:06:56
2. I submitted the same code twice, and get two very different runtimes 0.000s and 0.010s! Such big difference do not happen in old judge system.
When I want to submit the problem , the system asked me to relogin totobby wrote:Some other flaws:
1.I think it should be a submission error.Code: Select all
5902457 SRX Time limit exceeded C++ 0.000 2007-09-08 10:06:56
2. I submitted the same code twice, and get two very different runtimes 0.000s and 0.010s! Such big difference do not happen in old judge system.
submit , then I relogined it , but it asked me to relogin again .... ,
this time it runs correctly , but I got TLE with 0.000s and no problem name
displayed.
studying @ ntu csie
-
- System administrator
- Posts: 1286
- Joined: Sat Oct 13, 2001 2:00 am
- Location: Valladolid, Spain
- Contact:
Time limits have been changed. They will be adjusted in the future depending on the problem.
We'll check those RE submissions.
Differences of 0.010 always happened in the old system. In fact, 0.010 used to be the precision of the clock.
Sure memory limit still exists.
html descriptions will be revised in the future
I hope I didn't miss anything....
We'll check those RE submissions.
Differences of 0.010 always happened in the old system. In fact, 0.010 used to be the precision of the clock.
Sure memory limit still exists.
html descriptions will be revised in the future
I hope I didn't miss anything....
DON'T PM ME --> For any doubt, suggestion or error reporting, please use the "Contact us" form in the web.