osu! UCI Fall LAN

Last edited: 4 March 2016, 10:47AM


The osu! UCI Fall LAN was a one day LAN event that the osu! UCI club hosted on UCI campus. We had a total of 10 players compete for osu! supporter, osu! UCI merchandise, and Aaron's personally made two-key keypads. The tournament was split into two parts: one part Swiss and one part single elimination for the top 8.


To follow our string of quarterly tournaments, we started making plans for this fall tournament as soon as OSL ended. I didn't realize it at the time, but this would be the final tournament I'd work on with osu! UCI as I would be graduating after this quarter. Graduating was a decision I made midquarter when I thought about my options and decided to finish undergrad as soon as possible.

I won't go into too much more detail about my decision, but I talked with some peers and teachers about CS grad school and it seemed like the overall opinion from people urged me to not go. Some of the big points raised to me were that it cost a lot of time and money, and unless if I'm going specifically for research or into a very specific/obscure field, it's not worth it.

If I wanted to build my knowledge and experience, I'm probably better off investing my personal time learning something or, who would have thought, getting a job. In the end, I figured that if I really wanted to go to grad school for some reason down the line, the option's not going anywhere, so I didn't necessarily need to go now if I had other interests.

When I realized this was my last quarter, I wanted to slowly relieve my various responsibilities from the osu! club. I felt like at the time I was extremely anal about managing everything, from the website to meetings to events to all the projects. If I wanted the club to do well in the future, I needed all the officers to step up and take bigger roles so that we could continue to thrive.

For this tournament, it meant shifting over more obligations to other officers. I wanted Jimmy and Aaron to take over all the outreach and planning, and I wanted Justin to work on the website. In the end, I still had to pick up a lot of the slack, and we didn't do a great job on all fronts. The most obvious sign of our problems was having a really low entrant count, of just 10, which was much fewer than we expected and wanted.

I suppose I'll talk about planning first. This tournament was a little unusual for us because we were responsible for booking the event on UCI grounds. Press Start, our spring tournament, was part of a bigger LAN that already had a set location, and OSL, our summer league, was set online and had no physical events in person. We were looking for LAN opportunities around UCI during the fall, but after not finding any, we decided to host our own LAN.

I remember very clearly that it was Jimmy's responsibility to talk with UCI's room services about finding a room. But on the actual meeting day when we were supposed to consult with them, somehow he fell asleep, so I showed up an hour late to the meeting in his place. Apparently a signer of the club needed to be there in person or else we couldn't get the room.

I'm not sure how I feel about my meeting with the office. The guy I talked to was kind of fishy, and I'm pretty sure he tried to throw in a bunch of random expenses into our contract. Maybe he was just aggravated that we showed up late, but next time we should probably do a better job exploring all our options before committing to anything, perhaps even have multiple meetings.

The total cost for our room came down to around $50, which actually came as a surprise to me since I thought we wouldn't need to pay anything. There possibly could have been some ways to reduce this down, but we didn't want to go through the hassle of going over contracts again. It would also be difficult to arrange a meeting again, and personally I didn't want to handle this matter any further since it wasn't really my responsibility. $50 wasn't too much for a room anyways.

We could pay off some of the room costs by asking for tournament entrance fees. We'd break even with about 10 sign ups and any more money we could consider adding to the prize pool. We probably should have kept some of the funds for ourselves, but I was trying to run osu! UCI as non-profit as possible, even if I screwed myself in the end because of it.

I've made a lot of mistakes with handling money for osu! UCI in the past, everything from forgetting shipping costs for t-shirts, paying out a lot of money sending packages, buying supporter prizing, among other random expenditures. It's very safe to say I'm and some of the other members were definitely in the red under my leadership. This wasn't financially crippling for me, by all means, but we obviously could have done better managing money.

For this tournament we definitely didn't break even. We had too few entrants sign up, and I think Jimmy bought like a $30 extension cable or something, so our wallets were pretty screwed. In fact, I'd say we were way too generous on the prizing, but since we already set the prizes in stone, we didn't want to change that at the last moment.

Aaron had a few two-key keypads remaining that he wanted to gift for the tournament, and these weren't exactly cheap to make. When we sold them earlier, they were running for something like $20 each. The top 3 of this tournament, basically a third of the entire entrants, received them.

Additionally, the top 8 received supporter. Some of the supporter was paid out of pocket, while some was provided by osu! staff. And finally, top 8 received some charms that we've been holding onto for a while now. For an entrant, all these prizes were super awesome because pretty much everyone who entered got enough to immediately refund the $5 entrance fee we asked for, to our demise.

Actually, I didn't really tell the full story yet, because in reality we only had 8 entrants that came to the event. Royce and I, who didn't plan on competing, spontaneously joined at the last minute to fill some more spots. It wasn't much of a Swiss round since almost everyone made it to top 8. All Swiss helped determined was really seeding. We could have rearranged the brackets and just had a double elimination or something, but we already had map pools and everything planned out, so we tried to stick to our original plan.

I think we did a really awful job marketing and outreaching to get exposure for this event. This mostly fell on Jimmy and Aaron's shoulders who were responsible for these tasks, though the whole team shares the fault. I was working on a bunch of other stuff, and I wanted to step away from the outreach side of things.

We made I think only a single reddit post pretty close to the event date, and that was for the most part unsuccessful and got buried under other content. What I'm most annoyed about though was the fact that we delayed it for so long, because we requested the post to be made a lot of times a long time ago. But we waited, however, for the perfect timing at the peak hours which just resulted in the post never getting made until very, very late, and we didn't get much traction after that.

I guess we were hoping the post to be as popular as our OSL and Press Start posts that were quite a bit more successful. Maybe it was just unlucky, but at the very least we should have made multiple submissions. We also posted on the osu! forums and on some Facebook groups, but I don't think these reached out to a lot of people either. It could be something with the content within the posts. I don't know.

For these local events, I think the best way to get people is to ask and pressure people directly. We've built up a network now of many local players, and I think talking to each person individually about the tournament is the best way to spread word and gather interest. On this tournament we worked too indirectly and didn't catch enough attention because of that.

Also, using the nearby filter in osu! is a surprisingly great way of getting new people. I found some players I never knew through the filter very late into the quarter, but by then all of them already had plans. For this tournament, we had a relatively small target of about 16-32 people, and I don't think it's unreasonable to reach that number by asking people one by one.

Some of the lack of interest I guess was out of our control. We chose kind of a random weekend to hold the event, but this was unfortunately a time slot forced onto us. Apparently, when we were trying to book the room, the places we wanted were already reserved way, way in advance, something ridiculous like a year before. We had very limited times to pick from, but we did the best we could.

Originally, we wanted an event spread over two days, but we couldn't get that because there were just no free times. On the good sides though, the room we did get fit us all very well. We had room not only for all the entrants snugly but for whoever else came by to casually play with us.

Maybe just the idea of an osu! LAN wasn't very appealing. Press Start had a lot of additional games and attractions going on. OSL was very low committal with its online nature and even then a lot of people left the league as it continued on. Overall though, I don't think these reasons dragged down interest of this LAN too much. We just did a poor job reaching out and looking for people.

Most of my time was focused on our tournament website. I was fortunate to have quite a bit of resources to already work with from all the work I did in Press Start and OSL. The format of the OFL website followed the OSL one very closely, having one central page and shortcut buttons on the side to navigate it.

I did need to create some new assets. The biggest art piece was just the OFL logo. I really like the design and color scheme I eventually made for it. The leaf inside the O is actually just a close copy of the iconic MapleStory leaf. MapleStory is a game I hate with a burning passion, but at the time it was had a revival among my friends, and it was the first things I thought of when I wanted to draw a maple leaf.

Following on the colors, I edited the background and the side buttons to all have the fading pink to orange gradient. These were pretty simple changes that I think gave the tournament some extra flair. Pretty good effort to result ratio. It's kind of funny actually but I've been reusing this osu! background I made for a really long time now. You can even see it on this site (I hope at least), just colored pink now.

The website was up pretty early, but there were still things we needed to wait on before it was fully complete. Of course, we couldn't put up the brackets just yet since that would be done the day of. To make my life easier, however, I decided not to make any HTML tables that I would have to manually update on the server. Instead, after looking at Challonge more closely, I decided to use their convenient embedded features.

After testing some Challonge brackets out, it turned out that in each of the matches, you could specify some extra information like time of match and lobby name. This saved us a lot of time since we could edit brackets and bracket information directly on the Challonge site rather than messing with HTML. It saved admins a lot of work, so it was definitely worth it.

And finally I was also waiting on maps by Jimmy. This time we would only be using two map pools, one for Swiss and one for top 8. Having fewer pools was something we learned from our previous tournaments. It was a lot less work on Jimmy's side, and it gave players more time to practice maps.

We actually had quite a few arguments about mods. The first point of contention was removing one of our past rules of restricting the mods someone could choose. Basically, it said that the winner of a map could no longer pick that mod later on. So for example, if someone won on hard rock, he couldn't then pick another hard rock song in the future.

This was to prevent abuse cases where someone could just spam a certain mod to their advantage, and this promoted more well rounded play. We wanted to remove this rule because it was kind of difficult to explain easily, and we had a lot of accidents in the past dealing with that. We didn't think it would run into too many issues. On the contrary, I was thinking it was good to reward players for investing into a certain mod.

The rest of the officer team agreed, and this was one of the things we tested during the tournament. The results were kind of strange. For the most part, I think it didn't matter too much. The good sides were that some people were just able to play mods that they liked, and having that player freedom was definitely good. Some abuse did come up, however, though not as one may expect. These abuse cases happened in the opposite direction.

The best example was Jason/leluffy vs Ryan/[Neetwork] in the grandfinals. It turned out that Jason had a really apparent weakness playing hard rock maps because he heavily preferred playing HDHR vs just HR. He remarked that the hidden mod made the screen less cluttered and easier to read.

So taking advantage of that, Ryan pretty much just steamrolled him on consecutive HR picks. Was this necessarily a bad thing? I don't really know. It seems more like a personal problem on Jason's side than a problem with the rules themselves, and I guess the decision we have to make is, do want to protect or hurt players with mod weaknesses.

Technically though, if you looked the other direction, I suppose it did favor Ryan because he could abuse HR maps, but I think in this case it's more of a problem with just Jason himself. Overall though, I'd say I'm a little more in favor of keeping the mod rule. It shows more variety of maps and prevents abuse in both directions, both to protect weaknesses and to prevent strengths.

There was another area of heavy contention that we argued for a while about which was the idea of 100% free mod. The idea was that we would make our entire map pool free mod, and players could attach any mods they wanted on top of it. I just want to note by the way that this idea was brought up long before anyone knew about Jason's deficiencies, so we weren't covering for him or anything.

I wanted to try this out because I thought it'd be cool to explore the catering-to-people's-strengths idea some more. Looking at the reality of a tournament, it's really different from a player's normal practice and play session. Players generally gravitated towards a specific mod because they preferred one over the other. Someone may like the faster double time over the more punishing HR or vice versa.

So having a completely free mod pool helped cater towards that. Unfortunately, the idea starts to break down the more you think about it. First of all, choosing to add a mod was already a bit of a stretch. Would you really take the chance to put on a mod when you could just play a song no mod? It doesn't seem like it's worth the risk.

Also, double time had some problems. This mod, first of all, worked differently from other mods since it's forced onto both players. You can't just double time anything either since the song gets considerably harder. It's also very difficult to add HR onto double time. HD and DT might be still be okay, but HR was out the question for most songs. With all these problems, we needed to readjust a lot of things. We considered something like maybe a no mod, double time, and free mod pool.

Jimmy for some reason, and I don't quite remember, was extremely adamant about not doing this format. He really preferred having HR and HD separate and was very against combining the two down to one pool. The biggest stretch he would compromise was making a separate free mod pool on top of HR and HD to allow these kinds of shenanigans to occur.

Okay, I'm not a big mapper, map analyzer, expert tournament format maker, or even a big mod player, so I can't comment too much about mods. Fine, I'll let Jimmy's points slide. The next iteration I proposed and definitely something I still want to try, was what if we just let people add mods whenever they wanted, though not encouraged to, on top of a normal pool. So we would still have our typical no mod, double time, HR, HD map sections, but a player could add any mod they wanted on top of it.

I think giving players this freedom was very rewarding. It lets people eek out a potential advantage in double time if they wanted to use HD. If they were extremely comfortable with a mod, they could throw it onto the no mod pools. I guess in Jason's case, he could actually play the HR maps now. This would also give more justice to the removal of the rule saying winners couldn't play a mod again. And oh yeah, flash light could make an appearance(!).

This idea, however, was still heavily rejected by Jimmy. I don't really remember what his oppositions were this time, just that the rest of us kind of gave up and ultimately threw out all our free mod ideas. At the end of the day, he was the map pool maker, so he had the final say. If he didn't want to do it, we weren't going to force him to.

I think his reasoning was probably something to do with not respecting the mod categories, maybe? That doesn't sound really convincing, but that's the best I can recall. I guess if you HDHR'd everything it defeats the purpose of having HR and HD categories. You can count no mod too in there too, although no mod maps are no mod for a reason because they're supposed to be significantly harder.

I talked to a friend about this issue, and he raised an interesting point that reading no mod's AR9 was an important skill to have, and spamming HDHR or HR on top of it kind of defeated the purpose. I don't know, I haven't discussed this with a lot of people, but I think it's an interesting idea. I'll let this sit and stew for now. In the future I may bring it out again.

It's time to move on from maps. Let's see, I guess I'll talk a little bit about the format of the tournament that we agreed on. The format used for OFL was a product of what we tried and tested earlier through Press Start and OSL. Once again, we would be doing a two phase tournament broken up into Swiss and single elimination.

Swiss was a good format to account for a wide variety of players. It helped deal with rank discrepancies and still ensured some competitive matches. The top 8 single elimination helped reach an exciting finale. With the combined two phases, we could finish everything within a day quite concisely, as they saved us a lot of time compared to other formats.

The biggest change we made compared to previous tournaments was bump up the Swiss rounds to best of 7, or, as I say, first to 4 on the OFL website to be a little clearer. This was a lesson learned from OSL. A lot of players there were unhappy about best of 3 finishing so fast. The question you might ask then is why did we not do best of 5 then? Best of 7 may seem a little excessive.

There were a couple reasons for this. The biggest were time slots. If we did best of 3, we could cram everything within half an hour. 1 warm up map + 3 other maps at most totaled around 20 minutes and with some extra setup time maybe 25 minutes. It's pushing the 30 minute time slot a bit, but it should fit.

Since we wanted to move away from that, the next step up was best of 5. 1 warm up map + 5 potential other maps + some buffer time summed up to around 35-40 minutes. We could probably afford to squeeze in another 2 maps and adjust that to best of 7. This way we more fully used an hour to our advantage, and if you happened to completely stomp your opponent early, this gave you more time to rest. Pretty good scenario for everyone I would say.

Also to note from above was that each Swiss round was an hour long. This could be a little troubling in some circumstances because having a low number of rounds can make it difficult to really find out who the top players of a Swiss group really are. Ideally, we would have liked two separate days for the tournament so that we could dedicate one entire day to more Swiss rounds. Since we squeezed everything within a day, however, we could only afford 4 Swiss rounds of 1 hour each.

Having best of 7 allowed us to try and get the best representation possible for the players. I'm pretty sure the Challonge backend calculations weigh match scores at least a little bit, so a 4-3 is better than a 3-2. Of course, last but not least, we just wanted players to play as much as possible. My worst experiences in tournaments all boil down to not doing anything and just waiting around dreading my next match. Swiss and single elimination promised everyone marathon matches to always play until the very end.

Top 8 was structured a little differently as well. If I could, I probably would have set something like best of 9 for top 8, best of 11 for semis, and then maybe even best of 13 for consolation and grandfinals. This was kind of what made sense if we started with best of 7 in Swiss. If you actually look at the numbers though for something like best of 13, playing for over an hour back to back in a marathon tournament seems extremely stressful in osu! and probably not the best setup to do.

The reason we ended up going with best of 7 for top 8 and best of 9 for all the rest was that we were actually limited by Challonge's notation system. The highest best of value that you could set in the bracket was best of 9. You couldn't do anything higher, so we just said alright, let's go with 9 then. I think things worked out the best this way. Everyone was pretty exhausted by the end, and best of 9 was a fairly good limit.

That covers most of the prep for the tournament. Let's jump forward to the actual events of the LAN. Come the day of, we got there very early to set up and test. Jimmy, Aaron, and I brought our entire computer setups. I also brought a lot more like stuff in case we needed it like an extra laptop, my router, and some power strips.

Jimmy and Aaron handled the main casting and stream broadcast. We sat around 20 to 30 viewers average over the tournament. Not exactly the biggest turnout, but I can see that OFL wasn't exactly the biggest prestigious osu! tournament to stay and watch. Most of our viewers were either ourselves or friends.

Casting was pretty fun though. Jimmy and Aaron handled most of the Swiss round casting, but once people got eliminated from the tourney, all the losers joined together, and we had an interesting half a dozen person cast. Victor brought his TV to the event for us, so we crowded around that and had a good time watching people play.

The entrant turnout was pretty awful as mentioned before, so I ended up joining the tournament at the last moment and playing. Most of the other players were several folds better than me, so I didn't expect to place well. I'd still like to say I had some pretty competitive games though.

I might not be the best player, but I probably am one of the most unusual ones. I played full area using a really weird thumbs up grip, and my accuracy was notoriously bad. Despite that, my strengths were just being able to hold combo well and taking some random games here and because of that. I didn't make it pass the round of 8, so obviously my true colors showed through, but I still had a fun time playing a lot.

Being put into the player role showed me the good aspects we planned for the tourney. The best of 7 format was pretty nerve wracking and tiring, but it was rewarding as well. It felt good to play a ton of rounds, and I liked having more competitive games. Format-wise, I'd say we got it down pat, and I'll adopt a lot of it for the future.

I think it's good to switch up small changes each tournament and build on what we learned before. We took the best parts of Press Start and OSL but also played around with some of the existing rules as well, namely removing the mod rule restriction. Doing things like this spices up each tournament so each isn't to repetitive. Next time maybe the club'll do some more drastic changes, who knows.

Welp, my run went pretty south, but I'm not exactly the guy to thrive under competitive pressure, so dipping out and relaxing early wasn't too bad. Now out of the tournament, I could keep a closer eye on the games to see if anyone needed anything. I was actually a little worried early on whether the rest of the admins could handle things without me.

Usually I was the central figurehead where all the match results were relayed through. Since we had a low turnout and did everything directly via the Challonge website, the other admins took over all my responsibilities without a problem. I think Justin by himself handled pretty much everything, so the workload didn't seem that bad.

That's pretty much it. We saw some pretty exciting matches in the tournament. There were quite a few games that went the full stretch. The grandfinals weren't that particularly exciting since Jason got bodied on all the hard rock picks, but regardless he still put a decent fight. Even with a low turnout, I think we still held a fun and successful tournament. I'm sure everyone went away feeling like they played quite enough osu! for the day, and mostly everyone was happy to receive prizing in some form or the other.

I'd like to say thanks everyone for coming, whether they were here for the tournament or not. We had a lot of random friends come by just to hangout with us. Big special thanks to the admin team for putting on a good show. In this tournament in particular, there was definitely more background discussion and planning involved since we had to book the room ourselves. Hopefully with organizing so many tournaments now, the officers will be fit and ready to hold more in the future. Good luck, good luck.

One last note for a funny story. There was actually a heavy debate among the officers about what to call this LAN. I think the story went something like me wanting to try and make the acronym NFL somehow, to fit Fall for the second letter and LAN for the last one. So we needed a N word for the first. I couldn't think of a word for it, at least one appropriate.

Then Aaron suggested Nightfall LAN, and I really liked this name, like really, really loved it. Unfortunately though, it was rejected by the rest of the officers. I guess it may have been too cheeky or something. I didn't see any problems with it. I thought it was probably one the best things we could have thought of, but oh well, one day we'll use it, one day.