Hacker News Reader

Comment by someone_here | Hacker News Reader

You may notice that the player repeatedly strikes the same key in quick succession. Usually this is what some call "Spam" and count as actual actions for this metric but are not discrete actions that affect gameplay.

View on HN

A friend of mine started getting involved in the competitive Starcraft scene, and I noticed this tendency while watching him play. I think that any given player's APM is non-trivially influenced by the numerical "pissing contest" effect; every time I see a video like this it makes me want to see someone get really skilled while maintaining a reasonable APM and mop the floor with everyone else, just to show that it's a silly measurement.

EDIT: Full disclosure; I have played Starcraft since it came out in 1996, I possess a so-called "reasonable APM," and my spamming friend with an APM in the triple-digits could wreck me probably without even looking at the screen. I will never forget watching my entire fully-upgraded army somehow crumble impotently against a small, carefully-controlled contingent of foot soldiers at the perimeter of his base.

"ur micro is terrible dude, lol"

When did I hit the button that turned my guys into papier-mâché?!

a player's APM isn't directly related to their skill, but all the Korean professionals are quite fast (300 and above). there are so many things going on in the game, you really do need that much speed to play professionally and compete

I felt this way until I tried it. The rationalization, which I agree with as well, is that spamming in the early game and keeping up a steady rhythm "warms you up" in a way that makes it easier to maintain that speed when you really need it for micro-intensive portions of the game. So I don't believe it's really a pissing contest for very many players. (Although it's a poor metric.)

Also worth considering is that high APM is an unnatural requirement imposed by some limitations particular to SC:BW -- no automining, huge armies but a control group size limit, queued units consume minerals immediately (thus queueing is suboptimal.) In other RTSes without limitations like this, it's much more practical to have a lower APM. For example, in WC3, which has less units on the field at once and less base management, some players have been successful at a near-pro level with only around 60 APM. In SC2, with automining and large control groups, it seems like the best current players are frequently only between 100-150.

Fascinating! I never thought about using spamming as a "warm up" for late-game encounters which actually require triple-digit APMs to play at full effectiveness. Thanks for this post.

Note that it's 100-150 apm on SC2's fast setting even though you're actually playing at faster, so the apm is actually higher.

I forgot about that, thank you.

Starcraft came out in 1998 ;)

alpha tester.

"Spam" is done as part of warming-up during pre-game chat and during approximately the first 10s of most builds.

These players aren't spamming. They're selecting individual unit abilities within multiple control groups and switching off the membership of said groups in real-time.

That documentary clip is somewhat misleading. Some very strategic top-level Starcraft players with excellent micro-control can have as low as 200-250 APM. Some macro-oriented players such as Bisu ( http://en.wikipedia.org/wiki/Kim_Taek-Yong ) have ~420 APM.

Though APM is a datum which has grown during the evolution of the game, Flash ( http://en.wikipedia.org/wiki/Lee_Young-Ho ) ,the current highest ranked Starcraft player in the world, only has "average" APM for a pro gamer.

At the pro level, success is less about "action volume" and more about which actions one chooses to take.

For those that wish to learn more about the Korean pro Starcraft scene, I urge you to watch the English YouTube commentaries at SC2GG ( http://starcraft.homelinux.org/VODTracker/#v ).

You're talking about things like selecting groups of units, or what?

the metric came originally from a program called BWChart (http://bwchart.teamliquid.net/), and calculated APM based on all keystrokes and mouse clicks. So for the first 30 seconds of a game, it's trivial for anyone to have an APM of above 300. The pros, though, are able to maintain an insane 300+ APM over long games while maintaining meticulous control over their units.

That, also right clicking to move a unit, or setting a rally point, or mashing S or H to make a unit stop. There are a lot of actions that you can do that count in the metric and are spammed a lot.

Right, but what a lot of people don't realize is that what looks like "spam" is actually careful micro. In order to get the most effective unit control, it's actually necessary to order your units to hold position quite frequently while they're being positioned. It gives you some advantage, for example, if you just send your unit somewhere, it's possible for an enemy player to get close to your units and draw them to a disadvantageous place and kill them, where if you tell them to hold position this isn't possible.

So what you are seeing that looks like spam is actually individually grouping large amounts of units and positioning and holding position at just the proper point.

Reply on news.ycombinator.com