### hmehta's blog

By hmehta, history, 10 days ago, ,

The TCO20 Round 3A and Parallel Round are scheduled to be held on Saturday, August 1 at 12:00 UTC -4.

The match will be rated

Please note that you must register for this round in the Arena. Registration is now open for the round in the Arena or Applet and will close 5 minutes before the match begins, so make sure that you are all ready to go. Click here to what time it starts in your area.

Members eligible to compete in the Parallel Round include:
- Members who did not qualify for Round 3
Qualified for Round 4 or TCO20 Algorithm Finals from Online Stages 1,2 and 3

Don’t know how to compete in Topcoder Algorithm Competitions?

Check out this guide to successfully compete in an algorithm match.

You can compete using:

• Topcoder Java Applet - You can refer to this guide here to set up the applet. (Note that those who have Java 8 installed on their machine will see a security issue — You will have to add Topcoder in security exceptions in Java Control Panel. Please refer to the details in the guide.)
• Topcoder Web Arena(Beta) - Please watch this video for step by step guide

Best of luck to you in the Arena!

• +25

 » 9 days ago, # |   0 Is this round is div-1 or div-2(or above) ? Sorry for asking silly question, this is my first topcoder contest. I will participate if it's div.2 or above.
•  » » 9 days ago, # ^ |   0 This is close to Div I in difficulty :) Open to everyone!
•  » » 9 days ago, # ^ |   0 I'd say it's somewhere in between as far as problem difficulty goes, i.e. d2 C-E or d1 A-C are both comparable, but the time format is shorter.
 » 9 days ago, # |   0 Gentle Reminder: Round begins in 35 mins
 » 9 days ago, # | ← Rev. 2 →   +6 Where am I supposed to register for the contest in this stupid Web arena? I've opened it 20 seconds before the registration deadline and haven't found anything except for "enter" button.UPD ok, it seems that now registration ends 10 minutes before the contest. When did it change?And there is no ending time in the match schedule in Java applet.
•  » » 9 days ago, # ^ |   0 The same. but I don't know why I can't open a problem if I'm not registered
•  » » » 9 days ago, # ^ |   +16 This makes sense because it decreases opportunities for cheating.
•  » » 9 days ago, # ^ |   +3 Everywhere (calendars, blogs etc.) they advertise match begins at XX:00 or XX:30. Registration ends at XX:55 or XX:25 (5 mins before their advertised match start time.) But recently they started coding phase at XX:05 or XX:35 instead of their advertised match start time.These days they need 10 mins for room assignment. 5 mins were not enough in round 1 (or sometime around that I don't remember exact match). They were forced to add 5 mins delay in those matches (just like CF does).Sometimes they add "Please note that the coding phase will begin at 07:05 AM UTC -4 but the registration will still close at 06:55 AM UTC -4." in cf announcement. Sometimes they don't. Based on which blog they copied for making an announcement.
•  » » » 9 days ago, # ^ | ← Rev. 2 →   +10 These days they need 10 mins for room assignment. Wut? Is room assignment done by hand? I'd have thought xx:x5 is done for psychological reasons like on CF.
•  » » » » 9 days ago, # ^ | ← Rev. 2 →   +29 No, I'm sure it's not done manually. They just didn't change the registration time, which isn't surprising to me since they haven't switched to a good contest environment.
•  » » » 9 days ago, # ^ |   +3 That’s completely my fault — I will make sure it is explicitly mentioned everywhere. 10 mins for room assignments to give us time to fix if there are any issues
 » 9 days ago, # |   0 The idea that $L/3$ edges can be different was too much for me in 500 :) .
 » 9 days ago, # |   +3 A is painful.
•  » » 9 days ago, # ^ | ← Rev. 2 →   +10 Not really? Take the answer without an obstacle and check each $x_1 \le x \le x_2$. Find the highest coordinate reachable without the obstacle, then with it when you go to the left of the obstacle first, and then when you go to the right of the obstacle first, take the maximum of the latter two. Compare them with $y_1$, $y_2$, subtract the cells above $y_1$ that are reachable without the obstacle, add the cells above $y_2$ that are reachable with.You can mess up and implement it in an overly complicated way or mess up a formula, but this is short and clean.
 » 9 days ago, # | ← Rev. 2 →   +67 Easy stinks, med is again not some cool construction but rather some riddle to solve on the paper and hard is the easiest because $O(T \cdot P^2 \cdot N^2)$ passes.Cool round :/
 » 9 days ago, # |   +48 I like Easy. I'm poor at this kind of Medium.In Hard, why did you have such constraints and time limit? I see $O(T P^2 N^2)$ passes (even in Java), and FFT passes (which is kind of on the border, so ok), while the intended $O(T P N (P + N))$ is rather simple and fast enough.
 » 9 days ago, # |   +15 Thanks for the round!My screencast: https://youtu.be/rPRXifb4K-U
 » 6 days ago, # |   +21 It's been four days after the match, but the practice rooms for old SRMs are not open yet. hmehta, Could you please open the Practice Room?