How to find a style in a battle?

General discussion about the games and the scene.

Moderator: Moporators

Post Reply
User avatar
kuchitsu
Kuski
Posts: 1423
Joined: 13 Aug 2010, 20:31

How to find a style in a battle?

Post by kuchitsu »

I'm in a battle. Just about everyone has 12 or 13, and F7 is constantly updating with their new finishes. Meanwhile, I'm stuck at the bottom repeating same shitty moves that give me some 18 once in a while, I have no idea how to improve or what I should change, and I can't even get a top10 by the end of the battle.

How do all these people know what to do? Do they try random stuff all the time until they unexpectedly stumble upon something good? Do they have some awesome analytical skills I don't possess? What is the secret?
User avatar
pawq
38mins club
Posts: 6547
Joined: 24 Aug 2008, 19:56
Team: TR
Location: Southampton, UK

Re: How to find a style in a battle?

Post by pawq »

plsfindstyle.exe
User avatar
Hosp
38mins club
Posts: 2126
Joined: 30 Aug 2009, 20:55
Team: MiE
Location: up yours
Contact:

Re: How to find a style in a battle?

Post by Hosp »

secret is play as many dff kinds of levs as possible until you just do stuff automatically because familiar with basicly everything that can be thrown at you, takes a few years tho
Image
ToMaT
User avatar
danitah
Kuski
Posts: 818
Joined: 24 May 2004, 16:22
Contact:

Re: How to find a style in a battle?

Post by danitah »

You analyze after the battle, by watching winrec etc, don't just watch it but try to play the style also. Sure there's some analyzing that can be done during the battle also, but it mostly comes down to intuition, which there is no quick secret trick to obtain.
User avatar
milagros
Cheatless
Posts: 4560
Joined: 19 May 2002, 17:05

Re: How to find a style in a battle?

Post by milagros »

don't play, watch the winrec and say to yourself - yeah, I would use the same style
[carebox]
User avatar
Zero
Kuski
Posts: 548
Joined: 19 Feb 2010, 14:16

Re: How to find a style in a battle?

Post by Zero »

Intuition plays a big role like dani said.. It could be also called game sense which takes a lot time to learn in any game.

What helps a lot is to learn various tricks by hoyling tricky levels competitively and later you can adjust those moves to any battle. When it comes to the actual stylefinding process it's important to start by trying to understand the big picture of the level

Start by looking at apples that maybe could be taken with a shortcut of some kind. If you start seeing possibilities you can work around that area and try to create a working style around it. There can be several loops and routes in a level but try to avoid as many of them as you can. You don't need to know any details immediately, just look for patterns of apples or floating ones that maybe could be taken without going through the full route. Point being that you start limiting your focus to the essential part of the level and try to create a working style in that one crucial area. Styles in other spots can be polished without too much effort usually. It's not always the funniest approach when it comes to the playability but it's effective to ignore any pointless styles from the very beginning.

Speed also plays a big factor. Some route might be shorter than another one but if it's very slow in nature and the other one can be cruised full speed all around then it's clever to at least try both of them because it can be too hard to tell by eye which will be better.

To sum it up to some extent:

1. Finding the main route and trying to understand the big picture

2. Understanding what is the most essential part for improving/losing time and trying your best to find a good style there.

3. Polishing the run by trying different mini styles all around that will cut your time little by little.

4. Checking through the map again once you have a working style! It's easy to forget but never feel too confident about your
route. Everybody misses something occasionally.
Last edited by Zero on 27 Jun 2018, 09:22, edited 1 time in total.
Rank #1 battler and the champion of BOSS cup :gaa:
36 club since 02/2024
User avatar
Zero
Kuski
Posts: 548
Joined: 19 Feb 2010, 14:16

Re: How to find a style in a battle?

Post by Zero »

It takes a while to start finding the essential areas but it will happen eventually.
Rank #1 battler and the champion of BOSS cup :gaa:
36 club since 02/2024
User avatar
pawq
38mins club
Posts: 6547
Joined: 24 Aug 2008, 19:56
Team: TR
Location: Southampton, UK

Re: How to find a style in a battle?

Post by pawq »

It often happens that I wonder if I'm playing the right lev when I see Zero time, but then I usually find the style within a few minutes.

I think it says a lot about our stylefinding abilities/tactics that seeing a time several sec faster often makes us find that style very fast. I mean, just look at Hooked. 18 years without a major breakthrough, then a few days after another guy figures it out too, after just seeing the time :D
User avatar
Ramone
50mins club
Posts: 1978
Joined: 20 May 2002, 15:42

Re: How to find a style in a battle?

Post by Ramone »

Hey!

My approach:

After doing one cruiseride to test/see level I go to editor to get full view. Even if its a quite small map it helps me alot to see full map on screen. Then I eyetest all possibilities. To get a full view of level helps understand distances/angles and really allows you to see what is really possible/best.

When have some idea you go back to level and test diffetent styles/moves/routes. I always save several recs during battle (0.rec, 00.rec, 000.rec etc. 0 cause top of rec folder for fast merge. If 00 best of the first merge of 2,3,4 or more ideas remember 00 and then keep pkaying/saving/merging. Its also good to not esc/hoyl toomu h earlier in the battle. By keep cruising testing whatever styles/routes/tricks you might find something better/easier/faster.

It is always a balance of best style to most plausable style/trick. If no uberpros playing I might go for safer/easier/slower style if best seems too hard. Then if have lead can try harder. If uberpro playing its straight to best style. All or nothing.

Keep merging. If you randomly do a weird trick/move save and merge with your current best time. Dont need perfect attempt of move/style to get an idea from merge.

Often earlier in battle some trick/move/sc is hard af, often you can find new moves/ways to achieve same thing, keep trying new stuff while hoyling.

When/if clueless go back to editor. Dont be stressed while spying map during battle, 30 secs in editor is very very much. And equals only a few failed attempts, so its worth it. Also, I realized many players dont realize you can hide all pics/masking/grass in normal editor to get clean viev, helps with making trainlevs fast too.

If there is some maybe possible sc/megatrick later in level make a trainlev (0.lev) so you can test it more efficient. Sometimes a move/sc needs totally dif approach than your other style so no need waste time hoyling moves/routes you may not use if trick/sc not possible. And again, here is the balance of plausability to make trick/sc. Sometime a few tries is enough to know it is possible, but hard, and a bit lucky, then go back to real lev. Sometimes trainlev is great to find setup moves and hence make you less nervous once get to that one crucial area with good ride. Usually I always play trainlev in editor, but if its a longer lev I might go to the actual lev and merge recs from trainlev to see what style/moves are better later in lev.

I spend a lot of time in editor just looking at map (Eyetesting is good skill you know)

Its never too late to change style.

If its 10 mins left of a 15 min battle and you feel you can msybe beat leader if get a great/lucky ride. If its around the same time, change style. Its esrly in battle and leadtime will improve. Aim higher, there are always better styles. All battles are very short (max 60 mins) and best moves are never found that fast. Detours can be faster, some scs requires detours (usually for speed/dif angle).

People often get stuck looking for scs or routes, where there is often this one (or two, or three...) trick on the same route all use that cuts alot of time.

In high speed levels moves usually means less gain. In low speed areas/levels much more to find/invent, ways to get speed or reach areas ppl dont think about. If its a long(er) level with mixed parts the ones where you change directions, such as going to a drop from high speed or the slower areas are usually where you need to spend your time.

Getting two apples at once or stuff like reaching apple by stretching, can save alit alot of time, test in editor, move start, maybe need move speed. Often good to add poly so can test with more speed. If still cant do it, leave it. If can get it, do you think you can get that much speed or needed angle in real lev?

Its always a balance, best style you can pull off in given time.

And again, dont fucking start hoyling right away, get a feel for the level. Maybe you wont even start same direction as you first saw, or have any need for the moves you initially thought.

Many ppl, Bjenn, but usually new pkayers, think its cool to build a level around a sc, and adds a detour, more often than not detour is less polished/fun to play than rest of lev, idiot levelmaker didnt even bother to test it/make it a good part of the level.

Know your designer, some levelmakers more than others have set ways of creating a map. Quantz, always hid built in scs. Blaztek always have an intended route etc. When I testplayd Blazteks levels he says I ruined his lev by finding faster style than his intended, so he edits it. Some players make more open/raw stuff where you may need to look for more unobvoius stuff.

Also, I often spend full battle searching for styles. So often havent even decided style when battle is over. So maybe dont listen to me...

If you are playing for rank/stats you can ignore my tips. I play to win. 2nd or last doesnt matter. I dont give two fucks about ranks.

Good luck!

And, more important, the real winner is the one that has the most fun.

PS. Spy your own battles. Spy 100% of battletime. Dont only spy Zero/zweq/adi, spy everyone, noticing diffetent players approach to a battle will help you understand the game better, and probably make you realize what approach is best for you. Different players have so different approach, there is no right or wrong. Also its very interesting to see what players plays in which ways, who plays more like you than not. It is always easier to learn a move/trick/style by spying several attempts, both fails and makes, than watching a replay. (Unless you are Ilkka that watches rec frame by frame to try and copy exact. But by doing same, you get same, not better. Find your own ways.

Oh yeah, when trying a hard trick/sc that will for sure give you (a solid) lead, no need to try hard move thats 1/25 in start to gain 0,5 or even a sec. Go for safer. Unless much time and dont wanna reveal possible times, then go for max and if no success, change to easier late in battle if its still easy win.

In other words, use common sense :)

Also chat alot so other players get distracted.
Elasto Mania - ez better
User avatar
Zero
Kuski
Posts: 548
Joined: 19 Feb 2010, 14:16

Re: How to find a style in a battle?

Post by Zero »

Ramone wrote: 27 Jun 2018, 11:41Also chat alot so other players get distracted.
LOL! It was a plan all along :mrgreen:
Rank #1 battler and the champion of BOSS cup :gaa:
36 club since 02/2024
User avatar
Zweq
34mins club
Posts: 4055
Joined: 28 Nov 2002, 15:54
Location: suo mesta

Re: How to find a style in a battle?

Post by Zweq »

play 20 000h, apply accumulated intuition
Image
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: How to find a style in a battle?

Post by bene »

Think :bear:
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
anpdad
38mins club
Posts: 933
Joined: 6 Nov 2003, 06:04
Location: Russia

Re: How to find a style in a battle?

Post by anpdad »

Interesting that everyone's approach is so different.

My thought process is quite simple when battling, so i'll just go ahead and ramble about it for a little bit, even though it is slightly offtopic.

For simpler levs (non-apple harvests osv), I look at lev for 10-20 seconds with f1-enter, devise the most basic route that "should" work, and throw away the obvious bad ones, and also the ones that are too complex to evaluate without actually trying them out. Most of the time this is enough to come up with the general outline. After that it's just a matter of refining the details and adding ornamentation (which is limited by your personal experience, fundamental driving skills, and motivation). This approach applies for probably 95% of levs. The main bottleneck is in the amount of useful ornamentation you can spot, and your ability to add it to your style and pull it off within the given battletime. If you have trouble with that, then imo the best way to improve rapidly is to actively watch winner recs, actively analyze the whats and whys of their moves, and practicing afterwards. Using collective knowledge as a personal stepping stone is never a bad decision. Just dont slack and keep your mind open and greedy for new knowledge, and attempt to generalize and abstract your new knowledge so as to make it applicable to more general cases. I.e. the mindset is not the "oh so he indeed just did get inside the pipe fullspeed, meh", but the "oh so he got enough momentum to rotate in the desired direction prior to getting into pipe by utilizing the bump in the end of the polygon and then correcting the rotation with alo, kinda like in int29. What if there was no bump on that poly, would it still be possible? Seems like the only option would be to break heavily in that case, but i'll make sure to remember that for future in case i see a better solution. Or maybe i'll just start a battle with such trick and see what other people come up with".
For complex levs i usually stick to the first good route too. Rarely i'd just opt to save myself some time by booting up the good old sl and checking out couple of routes quickly (without going into details), and then go back to eol and refine it in the same way as described above.


TLDR: use prior experience (thoroughly generalized and categorized into broad ideas) in order to optimize the details of the style locally throughout the lev. Use common sense fueled by personal fundamental skill in order to throw away bad global routes. Refine the result by adding small specifics that work for the given lev. If the route does not yield good enough results, try the next one that looks fast, and again, attempt to go through your mental backlog in order to come up with moves that make it work. If this doesnt work, try random stuff (fueled by intuition) in an attempt to stumble upon something new. Repeat until you eventually lose. Proceed to ragequit and make sure to insult other participants to vent off the stress.
Post Reply