Hi everybody! :)

Codeforces Round #168 will start in hours. haas and I (havaliza) are the authors of today's match. I'd like to thank Gerald and Delinur for helping us to prepare this contest and MikeMirzayanov for this great platform.

Hope you enjoy solving the problems as much as we enjoyed preparing them. ^.^

Good luck and have fun. :)

UPD1.

Score distribution will be:

Div2 = standard

Div1 = 500-1000-1500-2000-2000

UPD2.

Congratulations to the winners of both divisions. It's nice that after the contest all five winners of first division are red coders and all winners of the second division are first division coders!

Div1:

1. rng_58
2. Petr
3. Komaki
4. msg555
5. maksay

Div2:

1. qph-Jeremy
2. dojiboy9
3. FreezingCool
4. Hellis
5. yplusplusplusplus

Editorial is out and will be completed soon. :)

Before leaving Tehran I asked SeMeKh a couple of questions on how the training for the competitions works in Iran.

There are local and regional Olympiads in Informatics in Iran which take place across the whole country. Top 40 winners will get together in so-called Golden Camp and train the whole summer, with a break of 2-3 weeks in Ramadan, so the camp duration exceeds 2 months. Part of the participants with lowest scores leave the camp earlier. The rating of participants is classified as either gold, silver, or bronze. "Golden" participants get the right to enter any university in the country without Concur. The 4 participants of IOI are also selected from the "gold" category. Please note, that there is no national Olympiad in Informatics. — see note below.

Iranian competitors achieved following results in the International Olympiad in Informatics:

• 2012, Italy — two gold (mR.ilchi, havaliza), two silver (LGM, mruxim) medals.
• 2011, Thailand — three silver (including SeMeKh and sajad22), one bronze (kasnavi) medal.
• 2010, Canada — gold (a70babat), two silver (Mehrdad, SeMeKh), one bronze medal.
• 2009, Bulgaria — gold (a70babat) and three silver (including pooya_, Solej) medals.
• 2008 — unfortunately, Egypt didn't grant visas to the team of Iran.
• 2007, Croatia — gold and three silver (including Saeed_Reza) medals.

As you can see, results are very high, all 4 pupils normally get medals, there is often a gold one. Many organizations support and organize high school students for comptetion, which is unfortunately not the case with university students. I have not heard of a well-organized system of training for ACM ICPC in the country. Hopefully, the situation with support from universities, government and business will change, and I suspect it is a local business who will initiate the change.

There were not enough time to discover more details, so I will be happy to here rich comments from Iranians on the topic as well as tough questions from everyone interested.

UPD. Please see important comment from A.K.Goharshady below.

UPD 2. Here are some numbers from inoi.ir. There are about 10000 participants in the first stage. About 1000-2000 qualify to the second stage, which is a two-day event which includes multiple-choice test and a kind of examination. Some 60-70 people go to the stage where they solve problems using a computer, 35-40 summer camp participants will be selected here. There are 8 gold medalists at the camp, 4 of which will then take part in IOI.

Hi everybody, it's 11:10 local time at Tehran, I'll tell you how the Bayan Programming contest 2013 (1391) onsite Finals is progressing.

22:17. Peter50216, cerealguy and Egor are top-three! Congratulations!

22:05. Closing ceremony. The rejudge took place for one problem.

18:30. Not everything of Funkoders has been judged yet, so the results of this additional fun contest will come later.

18:29. It's over!

18:27. Announcement: two minutes to the end of the contest. It should have 5 though.

18:10. It's a fun competition, so there are two guys entertaining competitors and spectators:

18:04. Every couple of minutes one of the baloons exploid, making a person next to it jumping up.

17:54. Only the first submission will count. There are 15 problems. Competitors hope that the problems are not harder than in the main round.

17:51. Let's begin!

17:50. Funkoders challenge will last 42 minutes and will contains lots of problems.

16:26. The contestants are having lunch now, afterwards the Funkoders Challenge is scheduled — a contest for fun. While you are guessing about the winner's name, I'll announce the winner of our "guess-the-wireless-password" contest. Actually, nobody had it 100% correct. The password was "bayan", so MikeMirzayanov who suggested "Bayan" is winning a special prize: he won the right to contribute to Codeforces completely free of charge the whole day tomorrow, on Sunday! Congratulations!

16:05. The contest has finished!

16:00. Five minutes until the finish of the contest.

15:35. Half an hour before the contest end (again). Here are some details about "rejudge". The score for each problem depends on submission time, number of wrong tries and number of people successfully solved that problem. So, after the contest end everything will be rejudged to double-check and make sure the final results are correct. Discrepancies, if any at all, will be investigated manually.

15:19. Announcement: 45 minutes until the end of the contest. Ranklist will not be updated anymore. Some rejudges may take place.

15:13. Current standings:

• cerealguy C, F, G
• peter50216 A, F, G
• Egor A, F, G
• watashi A, F, G
• Solej A, F
• Dmitry_Egorov A, F
• havaliza A
• _M_ A
• poopi A
• LGM F
• anrieff F
• ali_fiujnomali F
• a70babat F
• liympanda G
• m.sabouri G
• kelvin F
• haas F

15:05. Announcement: according to the jury team decision, the contest duration will be extended by 30 minutes.

15:02. Half an hour till the end. I see a small chance that the contest will be prolonged.

14:45. With baloons it looks like this (larger photo here):

14:36. There are few more people who solved one problem: havaliza, Nick name, liympanda and cheshme.

14:34. Baloons are distributed now for every solved problem: green is assigned to the most popular task F.

14:20. Announcement: please check clarifications, the constraint for interactive problem is published now. The interactive problem will not be removed from the problemset, although this possibility has been considered before.

14:18. The problems are hard, there are not many submissions.

14:16. Nicknames of Galapagos, fmwviormv and Haas do not match those on TopCoder. Also, cerealguy uses a nickname mikhail in the contest today.

14:11. We have a new leader: peter50216 with A, F, G. Then cerealguy — F and G, mruxim, Galapagos with A, watashi with G. Then come anrieff, fmwviormv, kelvin, Haas, Egor and Dmitry_Egorov with F.

13:59. watashi is on the second place at the moment with problem G. He is followed by mruxim, peter50216 and Galapagos who solved problem A. Three more people with problem F follow: kelvin, Egor, Dmitry_Egorov.

13:55. Almost the half of the contest has passed, and we have just 8 people who solved at least one task. Just recently cerealguy has got his second accepted and he dominates the ranklist with problems F and G.

13:53. I hope you missed me. I missed Iranian food so much, so I had to have a lunch break. Standings follow.

13:01. Announcement: please check clarifications. (One of the constraints has been fixed.)

12:57. It is very quite in the contest area. Egor is the only one with his headphones on, he codes something thoroughly, looking at the problem statements frequently.

12:43. There are 7 problems for three hours. Unfortunately, there is no standings table available on the web.

12:40. cerealguy seems to be the only one who brought Mac to compete on. He seems to write a code for at least 3 minutes now.

12:33. Contestants have printed versions of problems.

12:32. THE CONTEST HAS STARTED!

12:31. My laptop shows Swiss time, which is 10:01 now, and I'm getting very good in adding 2.5 hours to it to put a timestamp in Iranian timezone to each update.

12:28. One problem may be removed from the problemset, because the are some difficulties with it. We hope to begin soon.

12:26. By now everyone should be logged in with their username and password.

12:18. Four executables are available: {32-bit, 64-bit}x{Windows, Linux}. SeMeKh: "I hope nobody has any other OS."

12:14. Contestants can test their solution to the interactive task locally, providing the simulator with N (size of task) and a random seed. The simulator will generate a hashed result ("secret key"), which can be submitted to the judge.

12:11. Mehran apologises for the delay and encourages the participants to make themselves familiar with interactive problem. Every announcement is made in English and in Farsi.

12:05. Here is the contest area:

12:02. There 56 contestants here, representing Iran, Russia, China, Taiwan, Bulgaria.

11:52. All of the password guesses in comments correctly spotted the main keyword of the contest :-) But there is no 100% correct answer yet.

11:49. Network issue is resolved. SeMeKh announces, that one of the problems will be interactive, the binary executable will be available.

11:38. It has been anounced, that one of the contestants has problem with network. We'll try to fix it and start as soon as possible.

11:27. All contestants have entered the competition area. Mehran, one of the problemsetters from Bayan, has announced the Wi-fi network name and password where contestants have to connect. Can you guess the password?

11:13. While you wait for updates from me, you can read the first part of my nano-research on how it works (can also be read as "IT works") in Iran. Or just look at the pictures from the first two days in Tehran.

11:11. With a delay of about one hour contestants started to enter the competition area.

There is less than one hour till the contest start, participants are waiting in the lobby while the last preparations are being made in the second floor (Mezzanine).

UPD. The contest start delays, for now we expect about half an hour delay.

Internet connection in the hotel has been repaired at about 4:00 in the morning, so we have a bit more pictures from the past 50 hours:

https://get.google.com/albumarchive/pwa/112972602033679009172/Bayan?authuser=0&feat=directlink

Today, with all international participants of the Bayan Programming contest finals arrived in Tehran, we have visited some places in Tehran. You can see on the following picture (from right to left, because this is how things are being read in Farsi): video-op making documentary about the contest, watashi, liympanda, anrieff, Egor, cerealguy, Sadegh with Dmitry_Egorov hiding behind him, kelvin and peter50216.

We started at Niavaran Cultural Historic Complex which has several palaces where Iranian Shahs (Kings) and their families used to live before the Islamic Revolution in 1979.

After visiting museums with impressionist arts from Shagal, Dali, Picasso, vehicle museum with Rollce-Royses, Palaces with persian carpets, mirror halls und even mountain skis of former palace owners, we went to the bazar in the northern part of Tehran. I hope to post more pictures as soon as possible, because Bazar is must-see: at least on photos or even better to experience it personally.

We had an impressive lunch after the sightseeing. Sitting on the rugs, we tasted bread with cream, olives in a sauce made from bean and fruits, rice and Bakhtiari-kebab made from large piece of beef covered with mix of chicked with eggs. Our café is situated right across the street from Shahid Beheshti University, recommended to us in the comments to my previous post.

Later, at the hotel, we met some of the Iranian participants of the contest. Some of them represent a Sharif University team, which will go to the ACM ICPC Finals in Saint-Petersburg in summer. You can see havaliza in the red t-shirt and I hope you'll help me to identify other Codeforces members in the comments.

As the contest starts in less than 7 hours, it's time to wish a good luck to all of the participants of a Bayan Programming contest 2013 finals: good luck!

I arrived to Iran last night. Almost simultaneously with me two Bayan Programming Contest participants have also arrived to the Tehran Imam Khomeini International Airport.

Mr. Sadegh Hajsamadi from Bayan warmly welcomed us: peter50216, kelvin and myself at the airport and hired a taxi which took us directly to the hotel. Thank you, Sadegh!

The hotel arranged by Bayan for the contest is very modern, nice and has a conference room that will fit more than 60 contestants.

I have not met other contestants yet, but I got a chance to visit the Bayan office. Mr. Ali Ghadiri, company's CEO, and Mr. Mostafa Rokooie, the contest manager, have invited me to a very traditional Persian lunch with local specialities. The food was delicious, thanks to Ali and to Mostafa!

Unfortunately, I cannot show you the inside of the office because of the problem statements all over the place and algorithm ideas that are written on whiteboards, but here is at least a competition banner near the entrance that you can see:

And if you feel like starting learning Persian, here is the first word — Bayan:

Hello everyone!

Codeforces Round #167 will take place on Wednesday, February 13th at 19:30 MSK. This is my fourth Codeforces round and I hope not the last.

I'd like to thank Seyaua, sdya and Gerald for helping me to prepare the round. Special thanks to Delinur for translation of all problem statements into English.

Problems’ point values will be standard in both divisions.

I strongly recommend you to read ALL the problems.

Gl & hf ! :)

Contest is over. Congratulations to div1 winners:
1). tmt514
2). tourist
3). scott_wu
4). rng_58
5). dreamoon_love_AA

and div2 winners:
1). yefllower
2). Harlos
3). pseudopodia

You can view tutorial here.

Good day, friends)

Soon is coming regular Codeforces round #166 for Div.2 participants. Traditionally the others can take part out of the competition.

And again the problems were prepared by the group of authors: Pavel Kholkin (HolkinPV), Nikolay Kuznetsov (NALP), Rakhov Artem (RAD) and Gerald Agapov (Gerald). Traditionally thanks to Michael Mirzayanov (MikeMirzayanov) for Codeforces and Polygon systems and Mary Belova (Delinur) for translating the problems.

UPD: Score distribution will be not standard a little bit — 500, 1000, 1500, 2000, 3000.

We wish everyone good luck, successful hacks and high rating)

UPD2: the contest is over, we hope you enjoy it)

Congratulations to winners:

1) xrvpud221
2) xyz111
3) nanoha
4) wyx528
5) GuyUpLion

UPD3: the editorial is published, you can find it here

Bayan has published the schedule of the upcoming finals of their contest: http://en.bayancontest.ir/post/Contest-schedule

The contest should be quite compact: just two days and everything in the same hotel. All participants has to register themselves on Friday, February 15. Actually, the schedule does not say "February 15", it just says "Friday". Participants of the onsite event have got email invitations to Tehran for February 14-17, so let's assume we are talking about 15th of February, 2013. Speaking about dates, I have to mention that it's right now the year 1391 in Iranian Solar Hijri calendar. The New year is celebrated on March equinox, that's why the Sun is mentioned in the calendar name.

The contest will be held next day, on Saturday. All events should take place in the Parsian Azadi Hotel, which is located at the very north of the city next to Alborz mountains. Looks quite fascinating:

As far as I know, the Bayan Contest team has not communicated any transportation details to foreign participants. I checked out how you can get from the IKA airport to the hotel. One can take a bus to the subway station, cross the city in metro, exit near the Embassy of Russian Federation (Swiss Embassy is also nearby) and take a taxi. Here are the details: https://maps.google.com/maps/ms?msid=203576579397348428033.0004d54f4678ed65db970&msa=0 — please beware, this is not the official information, it's just my idea on getting around in Tehran.

UPD. All the participants will be welcomed and picked up at Imam Khomeini Airport by Bayan staffs, and will be transported to the hotel afterwards.

Hi everybody!

Have you spent the January as efficiently as we have? Here's a small list of innovations the users can see:

• Now not only the red coders can become coaches. The coders with a yellow handle are now accepted as well, provided that they have taken part in an official Codeforces round no earlier than 6 months ago and have taken part in at least 30 rated Codeforces contests. This innovation roughly doubles the number of potential coaches. So we invite you to check out your new role in the Gym!
• Now it’s absolutely easy to add a Polygon contest to the Trainings. All you need to do is to copy the data from contest.xml from the Polygon system contest interface to contest.xml in the sandbox folder in the Gym (use FTP).
• We’ve added protection against losing data in the Codeforces and Polygon projects’ forms. Now the textareas have a semitransparent ‘Drafts’ button in the upper right corner. You can use it to view or choose some previous versions of the text. The text is saving to drafts automatically. Now if the browser crashes unexpectedly or if you lose the session, that’s not a problem as you will be able to recover your text from the drafts.
• We’ve done some work on introducing the testing message shift from the system to the browser. You can see it on the Status page.The work hasn’t been completed yet and still has some issues, but on the whole it’s a functional system.
• We’ve updated the testlib to 0.8.7: we've boosted the performance of the ensure() function, the readStrictDouble stopped classify more-than-20-digit numbers as invalid, we’ve fixed some compilation warnings and fixed the compilation for some exotic cases.
• The Polygon system has better performance for problems with large manual tests.
• The contest and problem statements in the Polygon system now have reasonable URLs.
• We’ve fixed some bugs caused by cases in some old problems when the standard checker updates automatically, but testlib doesn’t. As a result, the checker cannot be used in such a problem, as it does not compile with the outdated testlib.h. Just in case, we’ve introduced the autoupdate feature for testlib, now updating it is going to be really easy.
• One more Polygon improvement. If a solution is marked as Main or Correct in the Polygon system, but it gets OK on a test in more than half of the time limit, then such situation gets highlighted in the invocations. Similarly, the system highlights the situation when a TL-solution gets TL, but it gets an OK in the double time limit.
• The problem descriptor in the Polygon system got attribute for the sample (example) tests. Now test element has attribute sample="true" value, if it is the test from the statement. This attribute is optional, it its absence is equivalent to sample="false".
• The compilers on judge servers have been updated to modern versions.