IceCTF 2016 guide

IceCTF 2016: Format, Tasks and Final Results

IceCTF 2016 was a long online jeopardy built for beginners and veterans alike. This guide covers the organizer, the schedule, the scoring, the published top of the board and what a first-time team can expect.

Hexfield editorial

OrganizercRUcible
Dates12-26 Aug 2016
FormatOnline jeopardy
Teams1,671
Event
Byte histogram of plain English text, with a tall cluster of lowercase ASCII letters and a spike at the space character
Plain text leaves a very uneven byte histogram, which is the first thing a beginner-friendly CTF teaches players to read.

Organizer and place in the calendar

IceCTF is organized by cRUcible, an Icelandic security team, and the 2016 edition ran online from 12 to 26 August 2016. CTFtime lists it as event 319. The scoreboard there counts 1,671 participating teams, which makes it one of the largest open contests of that summer by head count. Most of the big names in the CTF calendar play events with a few hundred teams; a contest in the four-digit range is large enough that the long tail of the board is made of beginners, school groups and one-person entries, and that mix shapes how the tasks are written.

The event has no rating weight on CTFtime. The platform assigns weight from the quality of the field, but it does not do so for competitions that stay open for several days, and a window of this length falls into that group. For a team that cares about its yearly rating the contest therefore counts for nothing; for a team that wants practice it is close to ideal, because the schedule leaves room for sleep and for reading.

Format and scoring

The contest is a classic jeopardy. A board of tasks sits in front of every team, each task carries a point value, and a solved task returns a flag in the form IceCTF{...} that is submitted for the points. There is no server of one’s own to defend and no direct interaction between teams, so the ranking is decided only by how many points a team collects and when it collected them. When two teams finish on the same total, the usual rule applies: the team that reached the score first ranks higher.

Because the board stays open for two weeks, the organizers can publish tasks that need a real amount of work, for example multi-step web challenges or reverse engineering tasks with a custom checker, without having them fall unsolved on the final day.

Categories and the kind of task to expect

The task list follows the usual jeopardy split. Web tasks cover injection, broken session handling and source-code leaks. Binary exploitation starts with stack overflows that overwrite a saved return address and moves up to format-string bugs. Reverse engineering tasks hand over a small ELF or Windows binary with a hard-coded comparison routine. Cryptography ranges from classical substitution and Vigenere ciphers to textbook RSA with small exponents or shared factors. Forensics and steganography tasks pass around a packet capture, a disk image or a picture with a payload appended after the end-of-file marker.

For anyone training with binary data in mind, the forensics and stego tasks are the most useful. A file that looks like a PNG but carries a ZIP after the last chunk, or a capture where one stream is base64 inside HTTP, rewards a habit of looking at the bytes before opening the file in a viewer. The binary data visualization guide describes the byte histogram and digram views that make such anomalies visible in seconds.

One feature of the 2016 edition was a light-hearted task that asked teams to send in a photograph of the members posing at a landmark, a nice reminder that the organizers treat the event as a community occasion as well as a ranking.

The published top of the board

The scoreboard on CTFtime shows a very tight head of the field. Two entries share first place on 4,506 points, and Whitehatters Academy UK is one of them. Behind them three teams are level on 4,306: DeliciousHorse, Pwnium and dephault. That is a gap of 200 points between the leading pair and the next group, which in a board with a typical task worth 50 to 200 points amounts to one or two solved tasks.

The shared totals show how a long jeopardy ends. Once the strongest teams have cleared everything that can be solved, the order is set by the timestamp of the last accepted flag, so a few minutes of speed in the first days can outweigh hours of work at the end. Teams that are tied on the final screen are often separated by a margin that nobody would call a skill difference.

What it is like for a new team

A first-time team gets a gentle ramp. The early tasks in each category can be solved with a handful of tools: strings, file, xxd, Wireshark, a hex editor, binwalk, a Python interpreter and gdb with a pattern generator. The sensible way to work is to pick one category per person, keep a shared notes file with the tools tried on each task, and rotate when a task stalls for more than an hour. Writing down what failed is as valuable as writing down what worked, because the next task in the same category frequently reuses the idea.

With 1,671 teams on the board, public solution write-ups appear within days. Reading three or four of them for the tasks a team could not solve is the fastest way to learn, and comparing the intended solution with the path actually taken often shows a shortcut that was missed. The team page and the blog collect further notes on how contests of this kind are approached.

Where the series went next

IceCTF returned in later years under the same name and the same organizers, again as an open online jeopardy aimed at a broad audience; the CTFtime event page for each year lists the dates and the number of teams. The 2016 edition remains a handy reference for how a beginner-oriented CTF can sit alongside the heavyweight events: it asks for the same skills, spreads them over a longer time and leaves the rating race to other contests.

Questions readers ask

Who organized IceCTF 2016?

The competition was run by cRUcible, a team from Iceland, and announced on CTFtime as event 319. It was open to anyone, played entirely online, and aimed at a mixed audience, from students taking their first CTF to experienced teams that play the whole calendar.

Why does IceCTF 2016 carry no rating weight?

CTFtime weights an event by the strength of the teams that play it, but it does not assign a weight to contests that stay open for a long stretch. IceCTF 2016 ran for several days, so it counted as a standalone result and did not feed the yearly team rating.

Is it a good first CTF for a new team?

The format suits it well. A long window removes the time pressure of a 24-hour sprint, tasks are spread over many difficulty levels, and the number of participating teams means plenty of public discussion afterward. Bring a notes file and expect to learn the standard toolchain along the way.

New to capture the flag? Formats, teams and a first-month plan in one place.

Read the CTF guide