|
Introduction
+ Show Spoiler +A lot of players struggle with raising APM, and often try to mimic pro’s speed by spamming. This leads to a lot of poor mechanics and inconsistent results. Watching replays shows extreme variation in APM at all levels of play. Simple click speed is not the deciding factor.
I’d like to propose the introduction of a new term into the conversation about APM and play speed that I think more accurately reflects the difference between amateurs and pros:
RPM - Rotations per minute.
The Problem
+ Show Spoiler +Even most gold+ players have started to develop a sense of rhythm, and a "rotation" they cycle through.
For example a simple terran rotation might be: mule > build scv > build units > build supply depots> add tech/upgrades. Micro with any extra time, and repeat.
The problem is, even as a players apm goes up, their rpm-- rotations per minute-- stays low. I often see replays of plat or gold players where they micro a fight for 30 seconds, then switch to macro 30 seconds, and repeat. This leads to missed production cycles, supply blocks, lack of micro and awareness at key moments in fights, etc.
Learning from the Pros
+ Show Spoiler +What high level players have actually mastered is not necessarily clicking any faster, but cycling through their mental checklist very quickly. This is why you can see masters players with apm of only 70-80, but when you watch their games they move their screen quite rapidly, keep up with macro, micro fights effectively, etc. Their rotations per minute is high.
High rpm is accomplished through apm efficiency. Our silver terran player has 60 apm, but 1 rpm; he Micros for 30 sec and clicks 30 times, then macros for 30 sec and again uses 30 actions. His Apr ( actions per rotation) is 60: he is spamming, wasting time and actions, missing key timings.
A masters player with 60 apm and 10 rpm spends an average of just 6 seconds per cycle. In a typical cycle he might mule, check scv and unit production, build a supply depot, move a scout to a tower, and add some rallied units to a control group. Each step on the cycle takes just 1 action. He never sits waiting for a building or upgrade to finish, or spams 20 clicks for 1 move order. He uses each action efficiently, and moves on.
Accelerating your RPM - cycling through your mental checklist more frequently-- will increase your effective APM, awareness, and timing dramatically.
Techniques for Improvement
+ Show Spoiler +The foundation for speeding up rpm is good “tapping.” (see day9 daily #252 “secreats of hotkeys, apm, and mouse movement http://blip.tv/day9tv/day-9-daily-252-secrets-of-hotkeys-apm-and-mouse-movement-4730506 ) To work on your RPM, hotkey everything important--structures, upgrades, scouts-- and force yourself to watch the bottom of the screen while cycling through control groups. Try to make yourself perform just 1 action per rotation for each control group, and move on to the next. As you become comfortable with the order of your rotation, begin accelerating the cycle rate. Work on tapping through your rotation constantly, especially while in the middle of a fight! Learning to squeeze more rotations in between giving unit orders is the most important skill for increasing average RPM.
Conclusions and Feedback
+ Show Spoiler +
So, what does TL think of RPM as a tool for discussion and improvement of APM? Is this a more accurate picture of the difference between leagues?
Most importantly, what techniques do YOU use to work on speeding up your play?
|
There is no clear way to judge someone's "rpm". Not going to work.
|
I like the idea, but how can you talk about it without any way of measuring it objectively?
|
The hard part is going to be how you know what a cylce is. Each player has a different style of play and I just do not see any way of doing this in the game.
However, that said, I think that playing with the idea in your own head that it is rythem and cycles that matter, and not just 1a1a11a1a111a1a1a1a1a1a1a111.....
Good read all in all, too bad there is no way to measure this 
|
Wouldn't RPM just be for macro and not include micro (these groups wouldn't be in your rotation). Thats how I read this post, I could be completely wrong.
|
|
|
The hard part is going to be how you know what a cylce is. Each player has a different style of play and I just do not see any way of doing this in the game.
However, that said, I think that playing with the idea in your own head that it is rythem and cycles that matter, and not just 1a1a11a1a111a1a1a1a1a1a1a111.....
Good read all in all, too bad there is no way to measure this
Thanks for the response.
It's true there's no built in measurement within the game, but I don't think this needs to be a measurable number to be a valuable concept. You can see in a replay pretty quickly what a player's mental cycle is,and how fast they are tapping through their control groups.
No, it won't be an e-peen number players can tack on their wall. But is it a concept we can use to show growing players how to improve?
Wouldn't RPM just be for macro and not include micro (these groups wouldn't be in your rotation). Thats how I read this post, I could be completely wrong.
The problem with pulling micro out of the equation is that you get players switching between macro and micro mode and not macroing during fights or scouting actively while macroing. The goal is to keep these integrated so important things don't get lost.
You see good players tap through army control groups and re-arrange them regularly even when there is no fight impending, just as part of their habitual cyce.
|
though there is no way to measure "rpm", it should be deep within anyone that wants to improve their game. its all about efficiency, doing what needs to be done within the shortest time available.
for example, some players will build 5 gateways, wait, then select all to add to group key.
but efficient way will be to build 5 gateways, do something else, then come back and group them all.
|
Yes, a human watching a replay can see what that cycle is, however, I still do not see how this would be something you can mesure with algorithms.
Also, a cycle for Zerg could be * Drone, inject, creep spread, scout, build ov* which is only 5 things, I would be able to cycle that many more times a min that I would be able to cycle *SCV, MULE, Build Units, depot, check build time on depot, scout* has 6 things in it and will take longer for that cycle to happen, its the same amount cycles... well scratch that... I guess if someone has a cycle of 10 things that they go through in 5 seconds, and someone has a cycle of 6 things they go through in 15 seconds, the better player is the one that can do.... fuck.. more actions per min.....
|
I've known this problem but didn't really put much thought into it. This is a good tip and make players focus on increasing their rpm. I have a problem of having really high apm up to 5 minutes in the game and drops back down to 100ish the rest of the game... spiking during engagements.
I don't think it's important to have a really accurate/computerized way of recording rpm. Watching your own replays and having a clear goal of increasing your rpm is a very useful way of improving.
|
I am a mid-high level masters player and I have always prided myself in my low APM, granted just out of practice it has risen and now averages around 80 but when i first got into masters i had an average of around 60. I like the way you describe making a conscious effort to avoid spamming however i do not necessarily like the idea of having set rotations built into your mechanics. I think that it will allow for fast improvement up to high diamond maybe but after that because of your set rotations you will be preforming actions because they are the next action on your list, not because they are the most important action that needs to be dealt with immediately. I believe you should make a conscious effort to slow down, avoid spamming and just think, what is the most important thing right now? do it then ask yourself the same question again. It allows for adaptive thought and flexibility rather than set patterns. I think your way of thinking will allow for quicker growth at lower levels but you will hit a block as you get higher that will be hard to break, because you have set habits that give you good fundamentals but are not ideal.
|
I really like the idea's you mentioned, you really do see all Pro's rotating in the same cycle. It was a nice read like some others said. Cool Stuff!!! : )
|
Yes, a human watching a replay can see what that cycle is, however, I still do not see how this would be something you can mesure with algorithms.
Also, a cycle for Zerg could be * Drone, inject, creep spread, scout, build ov* which is only 5 things, I would be able to cycle that many more times a min that I would be able to cycle *SCV, MULE, Build Units, depot, check build time on depot, scout* has 6 things in it and will take longer for that cycle to happen, its the same amount cycles... well scratch that... I guess if someone has a cycle of 10 things that they go through in 5 seconds, and someone has a cycle of 6 things they go through in 15 seconds, the better player is the one that can do.... fuck.. more actions per min.....
Correct, comparisons between races and players aren't really relevant. It's not about the number, it's about looking at your own play and improving the rate at which you move through your mental checklist, whatever it is. The exact structure of the cycle will vary by style, but the *principle* of practicing it and increasing how quickly you move through it applies across the board.
I am a mid-high level masters player and I have always prided myself in my low APM, granted just out of practice it has risen and now averages around 80 but when i first got into masters i had an average of around 60. I like the way you describe making a conscious effort to avoid spamming however i do not necessarily like the idea of having set rotations built into your mechanics. I think that it will allow for fast improvement up to high diamond maybe but after that because of your set rotations you will be preforming actions because they are the next action on your list, not because they are the most important action that needs to be dealt with immediately. I believe you should make a conscious effort to slow down, avoid spamming and just think, what is the most important thing right now? do it then ask yourself the same question again. It allows for adaptive thought and flexibility rather than set patterns. I think your way of thinking will allow for quicker growth at lower levels but you will hit a block as you get higher that will be hard to break, because you have set habits that give you good fundamentals but are not ideal.
Thanks, I appreciate this feedback and high level perspective.
I agree that adaptive thinking is far more powerful as you progress. However, I think about this like learning an instrument.
When you first pick up a guitar, you can't just play improv jazz by thinking about which note comes next and how to play it. You have to first learn the structure and practice the basics until they become autonomic, muscle memory. Then, when you don't have to think about the basics anymore, your mind is free to think creatively and explore outside the box.
I think you are correct in saying this is an effective way to practice up to diamond, but that to really break through players will need to think flexibly and adapt to the game.
|
In my honest opinion APM is an irrelevant variable. I think it comes down to decision making and micro. Although my average APM is around 80-120 when I'm microing my combat units, my micro peaks at 350-400.
I'm mid master zerg and I often time find that relaxing, getting good map control, scouting and taking advantage of your "timing windows (expo/attack)" often time wins you the game despite how well you micro your units.
When I see players button smashing to keep 300 apm constantly where 250 is wasted... I wonder if they're trying to punish their keyboard/mouse. 
|
So, the OP is more so saying that this should be a tool that each person uses on their own, rather than a stat inside of starcraft. I was under the impression that you were advocating to replace the APM with RPM in game.
As a training tool and used on your own while watching replays, yes. awesome.
|
This isnt World of Warcraft man, i dont think theres a set way to identify rotations per minute(im assuming this is what this means)
Theres alot of fluctuations that happen that could throw anything you do normally off.
Theres quite literally zero way to figure out "RPM"
|
On September 17 2011 03:47 arb wrote:
Theres quite literally zero way to figure out "RPM"
He's not advocating some quantifiable number for players to focus on to determine RPM. Rather, he's saying that lots of players focus on useless actions to improve APM and think they are getting better and instead they should focus on a set cycle of actions such as building workers -> army rallies -> scouting -> creep spread/inject/mule/chrono -> etc and by perfecting and speeding up a rotation, you will in turn increase your effective APM.
Of course, there are going to be times when rotations may get messed up such as 200/200 micro battles/harass, but as you get used to the rotations you've setup you can more effectively keep them up with production/re-supplying/etc during all parts of the game.
For some reason most people need a quantifiable number to see how they compare with others (see APM x.x) whereas this technique doesn't provide that, but does, arguably, provide a much better avenue of improving.
Good read and good concepts.
|
It's something players should learn to do (I guess?) but it really serves no purpose. People who brag about their 300 APM but are still curious as to why they're stuck in Gold League will continue to brag about their 300 APM. You can't set any sort of numerical value that is supposed to be representative of a person's skill because there's so many components that go into being a skilled player. What I think is interesting is when a player like  GoOdy or  White-Ra manages to play with such low APM compared to the higher APM that players like meRz play with.
|
The OP is right in stressing the importance of RPM. I think he is suggesting that "rotation" is just a guideline. It's a structured way to view the game in order to multitask well and stay productive in non-critical moments (Although you really still need to rotate back to econ/base management during battles in order to expand behind your pressure, etc).
It's true that making good decisions is more important than APM, but people who say APM doesn't matter are wrong! When you're faster, you just give yourself the opportunity to make more good decisions per minute. Off the top of my head: defending multi-prong attacks - You know exactly what you should do at both locations of aggression, but you're simply not fast to perform optimally.
I think APM/RPM is something you should worry about once you're at mid-master. Anything before that is knowing your own builds, recognizing your opponent's builds, and basic economy management such as when to cut probes, how many to transfer, etc. You can't rotate between 3 things when you don't even know how to play optimally while concentrating on 1 thing.
|
I measure my performance in wins and losses
|
I like the idea, the question is how to implement it in practice. Use of macro mechanics would perhaps give an indirect measures (e.g., regularity of mules/chronoboosts/injects).
|
I think APM is fine. Much less vague than "RPM". Rotations... lol....
Players who "focus" on raising their APM are simply focusing on a useless number. Once you're familiar with your openings and econ timings (eg:MULE), it frees up more brain time for controlling units, etc. Spamming APM and missing your timings, just to see a higher number, while still losing the game is working on something useless. Getting better at doing more things at once by optimizing your available real APM will gradually increase your overall avg. APM over more and more time.
If you've gone from spamming @300 APM in the first minute of the game to spamming @500 APM in the first minute of the game, you've built up something completely stupid and absolutely useless.
|
The idea is good but i dont really think there is a proper way to measure rpm as of now. Honestly i rather just look where those "apm" is being used and see how effective it is rather than cycling
|
A high apm is usually an indication of good mechanics. I think most people know about how seriously to take that measurement. The RPM you described is pretty consistent with the way OGsTOP plays. He never misses a depot or an scv, but that doesn't make him the best player in the world (obviously). APM is spent on unit micro, scouting, and idle cycling too. All of which are important.
Ultimately the only good benchmark for how good a player is, is how well that player plays. A high APM is usually a good indication of good mechanics, but that is it. Your RPM concept could just as easily be described when talking generally about a players "macro".
EDIT: I used to think spamming was a bad thing, and not really productive for developing good play. As I have become a better player, and I have realized that spamming is a productive way to practice. You should always be pushing yourself to play beyond your comfort zone, more so in mechanics then anything. Forcing yourself to play to fast it the only good way to develop better mechanics. Ideally every action should be useful sure, but that takes time and practice. And at a certain level of play, having more then 100 apm really, really helps.
|
This is useless. How is knowing your rpm going to help your game? If you were practicing your mechanics correctly already you wouldnt need to care about how many cycles you perform in a game. Apm helps because it is a judgement of your activity level, which needs to be high in order to have good mechanics because there are so many actions you need to perform simultaneously. Spamming increases your activity level because it keeps you going all game. If you have good mechanics, spamming will only help since you already know wtf you are doing. Players who just spam for the triple digit number without knowing wtf they are doing will get nowhere.
Best example: top koreans all have high apm. It is possible to play with sub 100 apm and be good like whitera and sheth etc. But it doesnt change the fact that apm matters. Pay attention and you will see their apm jump beyond 200s when required, and they arent even microing yet, just controlling units and macroing.
|
IMO it all comes down to what you are doing per screen at any given time in your game. I could say I build 5 racks and do something else and come back then claim this is more efficient than waiting there for the racks to finish building. But if that action of "something else" is just watching the river flow then how is that more efficient than waiting your racks build to completion.
|
Nice idea, I think the new patch will give us a better sense of our own apm.
And to anyone who hasn't already, definitely watch the day9 daily mentioned in the post. It is one of the most helpful mechanics videos out there.
|
rpm is too dificult to measure, simply because you don't have the exact correlation between the actions and whats really happening... Only if you could analyse the replay and the actions at the same time you could determine the RPM of a player... There is yet no tool for this, other than watching the replay in slow mo and actually observing the APM, and whats happening.
In other words... it's STILL impossible to measure RPM. Just to clarify that it doesn't make it an unvalid concept. Like others, a definition doesn't need any proof, or measuring technic to exist..
|
People are saying RPM is "so" hard to measure, yet do you go through someone's game to find which actions they make are effective and which aren't to determine effective APM? No, you don't. Nowhere did he state that RPM should be quantifiable, he said focusing on the idea of RPM versus APM will lead to quicker improvement.
|
There was a great thread on a topic with the same point as this one:
http://www.teamliquid.net/forum/viewmessage.php?topic_id=250059
The crux of the thread was something like:
- Since APM is a measure of the number of clicks one performs in a task, it doesn't maximize speed or time, and so it's not necessarily a valid measure.
- APM is probably the necessary condition for rapid task completion, but having high APM doesn't necessarily mean anything.
- If you want to improve your APM, you should instead focus on picking some aspect of your play and drilling it over and over in that specific situation so you can automate and complete that task faster in a real game. (for example, roach stutter step, moving workers to gas when taking a new expansion, taking xel naga towers with lings and unbinding them from your control groups)
I feel like the linked OP accomplishes what you're trying to do only a bit more articulately and completely (no offense to you of course).
Still, just as other posters have pointed out, the high APM itself is not the important part, but being active and able to process tasks more quickly and accurately is (which will naturally raise your APM anyway).
|
The OP makes perfect sense but i dont see a new word like RPM being used by many. Everything you described with RPM falls under a word that already exists: "multitasking". And as others have pointed out measuring someones ability to multitask in a number is nearly impossible thats why RPM wont be needed. But you pointed out well that multitask ability doesnt have anything to do with apm. People need to understand that
|
On September 17 2011 06:42 michaelhasanalias wrote:There was a great thread on a topic with the same point as this one: http://www.teamliquid.net/forum/viewmessage.php?topic_id=250059The crux of the thread was something like: - Since APM is a measure of the number of clicks one performs in a task, it doesn't maximize speed or time, and so it's not necessarily a valid measure. - APM is probably the necessary condition for rapid task completion, but having high APM doesn't necessarily mean anything. - If you want to improve your APM, you should instead focus on picking some aspect of your play and drilling it over and over in that specific situation so you can automate and complete that task faster in a real game. (for example, roach stutter step, moving workers to gas when taking a new expansion, taking xel naga towers with lings and unbinding them from your control groups) I feel like the linked OP accomplishes what you're trying to do only a bit more articulately and completely (no offense to you of course). Still, just as other posters have pointed out, the high APM itself is not the important part, but being active and able to process tasks more quickly and accurately is (which will naturally raise your APM anyway).
Great link, though I think we are talking about slightly different things.
His article is (I gather) a highly technical way of saying "practice one action at a time to make it faster, rather than trying to do everything faster at once."
RPM doesn't focus on the specific actions one takes in microing a unit, etc, but rather on the rhythm of moving from one action to the next. Put another way, ACT is about actually doing the task faster, RPM is about stringing these tasks together more efficiently.
I will definitely use some of his thoughts in working on my own play, though, thanks for the link.
|
I feel that APM comparisons are next to worthless.
|
As a concept, I think this is actually quite clever. Unfortunately, I do think it will remain as such - just an abstract concept. It's a good thing to think about, but I don't think that there is really any practical way to apply it to games. As other people have said, it's just not something that is easily measured.
|
United States360 Posts
There would be a LOT of factors you'd need to cover for a calculation like this.
For one, yo'd have to analyze the replay first to find out what the player's cycle "order" is, as well as determining how many steps they have in their cycle, such as a terran with no OC won't have mules yet, a Zerg with no queens yet, etc. So, as the game goes on, your "steps per cycle" will increase as you get upgrade structures, macro mechanics, expansions, units to micro, etc.
Then you'll need to actually gauge how quickly they are cycling through all these steps, including such things as certain steps not being included in EVERY cycle, perhaps there will be activities, such as upgrades that only go every 2-3 cycles, or something.
You could possibly avoid this by simply grouping similar activities into one "grouped" step, but that weakens the rating overall, I think.
Additionally, if a players missteps and accidentally misses a step, like missing an inject, or missing a production cycle - how would you add that to the calculation? Do you just skip that step, or lengthen that cycle until they do it again?
This WOULD be a great rating to be able to use - but it's simply far to complex to be of any real use in practice.
|
this is very interesting but as others have said it would be difficult to measure
|
I like this. I think a lot of the people responding have the wrong idea (WE NEED A FORMULA AND IT WOULD BE IMPOSSIBLE TO MAKE ONE) - it struck me as more of a good way to measure your own progress. For example, my cycle would be production-injects-creep-upgrades-army, and my RPM is probably currently at about 2 or 3. If I could boost that up to 7 or 8, it would mean actually spreading creep at a reasonable rate, getting upgrades out more promptly, sacrificing macro while microing less often, and generally playing at a higher level all around. I think this is a really cool idea and I'll be implementing it in my practice!
|
I see holes in this idea because everyone's rotation is different depending on build, race, micro ability, etc. Saying that your rpm doesn't go up when your apm goes up is just dumb unless you are spamming to get apm up... If you have more apm, naturally you will get more ACTIONS done in a MINUTE so your rpm will go up.
|
On September 17 2011 08:23 envisioN . wrote:
I see holes in this idea because everyone's rotation is different depending on build, race, micro ability, etc. Saying that your rpm doesn't go up when your apm goes up is just dumb unless you are spamming to get apm up... If you have more apm, naturally you will get more ACTIONS done in a MINUTE so your rpm will go up.
Not necessarily, if a terran goes allin with marine/scv his APM could go much higher while his RPM drops to 0 (if a rotation consists of micro and macro and the player only micros 1 control group of units there is no rotation involved). It is entirely possible to increase your APM without spamming or increasing RPM, just focus entirely on micro - there really is no limit to the amount of non-spam APM micro can soak up, it's just a matter of deciding when the relative benefit of sinking more APM into micro is no longer worth it.
Many lower level players completely abandon macro while microing, thus creating a situation where their APM goes up while their RPM goes down
|
One of the patch notes for 1.4 is that spamming will no longer count towards your in game APM. If spamming no longer artificially increases your APM that means that any increase in APM that a player gains through the course of practice will mean that they are able to complete more useful actions per minute than they did before. This means that after 1.4 goes live you can now watch a pro level replay and measure your apm vs a pros and get a good idea of how many more things you should be doing per minute.
|
On September 17 2011 09:33 HTODethklok wrote:
One of the patch notes for 1.4 is that spamming will no longer count towards your in game APM. If spamming no longer artificially increases your APM that means that any increase in APM that a player gains through the course of practice will mean that they are able to complete more useful actions per minute than they did before. This means that after 1.4 goes live you can now watch a pro level replay and measure your apm vs a pros and get a good idea of how many more things you should be doing per minute.
I thought they were just changing command group spamming by not making it contribute to apm? I believe you can still select units and cursor box them to spam APM, which is what a lot of pros do anyways during the lulls in the game.
|
You might be right actually I forgot about box spamming I dont do it so I cant comment on if its changed in the new patch as well.
|
The number of "rotations" that you define can vary on different stages of the game and would not make for a useful measurement.
If you don't have many minerals, you're not going to be making very much and have lower rotations. A Zerg waiting for larva to pop out would have low rotations. Protoss waiting for warp cooldown to finish would have lower rotations - but if he warps in a bunch of units, rotations would sky rocket. RPM wouldn't make much sense to base a players skill on, assuming it could even be measured.
I also agree with the poster that said APM is a good indicator of good mechanics. I don't even know why Blizz bothered with tweaking the APM calculation, it's like the bunker nerf... unneeded.
|
On September 17 2011 09:44 Hypnotic42 wrote:+ Show Spoiler +On September 17 2011 09:33 HTODethklok wrote:
One of the patch notes for 1.4 is that spamming will no longer count towards your in game APM. If spamming no longer artificially increases your APM that means that any increase in APM that a player gains through the course of practice will mean that they are able to complete more useful actions per minute than they did before. This means that after 1.4 goes live you can now watch a pro level replay and measure your apm vs a pros and get a good idea of how many more things you should be doing per minute. I thought they were just changing command group spamming by not making it contribute to apm? I believe you can still select units and cursor box them to spam APM, which is what a lot of pros do anyways during the lulls in the game.
Repeated control group and selection commands will no longer count as unique actions for APM calculation purposes.
|
I guess RPM is another way of looking at it...
I like to tell all the newbies I tutor, that instead of APM to use UAPM (useful actions per minute).
I figure if you refined all your spam into the actual action being done is your actual UAPM. A player with 60 UAPM can outplay someone with 300 APM... Especially if the 60UAPM is consistant.
|
Now that I think some more about it, I've noticed that if I get very flustered early in the game by losing a battle I sometimes start rotating too quickly. As a result I start to queue units (I guess because I'm desperate to get them out), which in turn messes up my build.
|
people overvalue apm so heavily... just play the game
|
Don't focus on RPM or APM, they may both correlate with skill and winning, but why not just focus on the other side of things i.e. figure out why you are losing, different build orders, how to scout. With more games and more skill comes APM and RPM. I doubt the opposite is true at all (The only way I can get better is if I increase my APM/RPM.... yeah right)
|
tldr version:
RPM is not something you will try to increase as a measure for improvement, RPM is something you will use as a conceptual framework to help identify the most efficient use of your time with the components of the rotation and the time required to perform each part of the rotation and the rotation as a whole forming your boundaries.
As a quick thought exercise, consider the cool-down on inject, the recharge time on nexus/orbitals, and the production time of units. With these things in mind it is clear you cannot complete a full rotation more than 1.something times a minute for any race, so the key is to identify the ideal RPM for YOU, and then examine your game play for times when you deviate from this RPM and figure out why.
The long version:
+ Show Spoiler +
This is a useful lens which I have seen discussed before, the idea of setting up a task rotation is discussed in various learning guides (And of course by Day 9), and adding in the layer of "RPM" is definitely a valuable learning tool. APM really is an outgrowth of increased efficiency of action, as you develop muscle memory for a task such as stutter stepping you are able to perform the same mouse movement and keystrokes faster and more accurately, this is the nature of muscle memory. As many have said, trying to increase APM directly is a very futile task, you may be able to spoof the calculations but it wont translate into wins unless those actions actually accomplish something for you.
RPM on the other hand IS something you can directly focus on to improve your game. The first step is to establish an action cycle which ensures that you are not missing critical actions such as mules, injects, or unit construction. Then you should figure out how often each component of that cycle takes to "reset" and ensure that you are able to perform that action and rotate through the rest of your cycle before that element of your rotation is active again.
Now with this rational foundation in place you can start working on making those actions completely automatic. They NEED to happen at that time, every time, and as you drill that cycle this will begin to happen. As you settle in you will start to see that you have time in between certain actions in your cycle, these little gaps are where you will start to fit in extra actions like pre-splitting your marines or setting camera hotkeys, and they are also the gaps in which you will execute micro during combat.
Now as many have said RPM itself is unmeasurable and is actually pretty useless as a measure if you think of it as a number to increase, as I explained above it is important to identify the "hard" limits inherent in your rotation, so you will be inherently capped at how frequently you can cycle through your full rotation in any given time period. It makes very little sense to return to the Mule portion of your rotation every 10 seconds just so you can hit 6 RPM instead of 1.3. It is also worth noting that the RPM potential will increase as a game goes on since you will end up with multiple orbitals/queens/nexus/units/production facilities that all need to be checked, so the rotation you set up should leave room for more actions within it as the game goes on.
What the idea of RPM really does for us is provide a tool for examining our own gameplay after the fact and refining our use of our actions. When you are watching your replay you can watch your own actions and see whether you are missing actions due to distractions during the game, which in turn provides a lens through which you can observe the impact of these distractions on your overall macro timing, as each delayed action pushes the entire remainder of the cycle back AND every cycle that follows. By adding the "per minute" to the idea of setting up a task rotation you add the idea of putting this rotation into the overall time based framework of the game.
The biggest difference between APM and RPM is that APM will always be seen as a "more is better" measurement, even if you can win with lower APM, having a higher APM (or more accurately uAPM as one of the other posters puts it) would be handy. With RPM you WILL hit a wall with how many rotations are actually useful. That being the case, RPM acts as a tool for both improving the efficiency of your rotation to help fit more actions in, and also as a tool for REDUCING the number of actions you are taking which serve no purpose. Why select your production when it still has 10 seconds before it even COULD be useful to do so? Place a creep tumor, build a supply depot or give a scout another queued order instead.
|
I see what you are saying OP and I totally agree with everything said.
RPM is a solid concept to keep in the back of your mind, even if you cannot actually measure it.
It would be a solid idea to drill into the head of anyone who is new to the game.
|
I don't think the objective of this post was to say that we have to start MEASURING rpm, but it should instead be what players strive for. I agree completely, especially since day9 goes over this in one of his mechanic dailies. Check your 4,5,6 (ex. rax, upgrades, cc)
Good post!
|
Well, I'm a quantitative psychologist, M.S., working on PhD in it (to the leyman, a statistician), and I know a thing or two about Theory of Measurement. First, there are two things to understand, the abstract construct (the thing you want to measure) and the concrete operational definition (OD) of said construct (how you measure it). E.g., intelligence is a construct, and the IQ test (pick any version, doesn't really matter for this demonstration) is its OD. Now, not all OD's of any given construct incapsulate the whole of the construct, and many OD's have "construct irrelevant variance". For example, imagine I want to measure SC2 expertise, and my OD is APM. First, does this OD incapsulate all of the construct of SC2 expertise? No, there's more to it, e.g. knowing your build order, proper timing sense/scouting, knowing how and when to react, etc. So, APM isn't the whole story. Also, does APM have construct irrelevant variance? Yes! By this I mean, you can be in the bronze league, select a worker and your first building (nexus/cc/hatch), assign hotkeys 1 and 2 to them, respectively, and spam 12121212... all game long till it's over (you'll likely lose unless they dc  ). So, APM may not measure SC2 expertise at all, and in more tangible cases, maybe someone is addicted to spamming, and plays as normal, but with an extra 121212... in between thoughtful actions.
Now to consider your idea, RPM... This is in theory an OD of "SC2 expertise". This seems to me to be an attempt to take APM, but polish it in a way that attempts to remove some of this construct irrelevant variance. I like it, though unfortunately I can't think of a way using the current built-in measures Blizzard provides, to measure this, short of going through each replay and hand-coding.
Despite this huge issue of work-intensive measurement (and who knows, maybe some 3rd-party add-on could achieve this, though I wouldn't have it running during a ladder game, but rather during replay, in order to avoid being flagged as a potential cheater, or worse, wrongly labeled one), I do feel that this is a important conversation to consider, and I think this idea is a great way to measure one's own improvement, given the stamina to measure it and the ability to put the data garnered to good use.
|
Because it is impossible to measure, this is more of something that is good to just keep in your head as a mental reminder that you need to again cycle through the things you need to do. More of a self development thing more than anything I would say. Just keeps you conscious of these things.
|
What really needs to happen is that the idea that APM isn't really a measure of skill needs to be implemented. What happens is that if you forget about your APM and try to spend time getting better, your APM will go up naturally and just naturally you will play faster than before. If you start to concentrate on raising your APM so you can "play like a pro", nothing will happen. Most people who say they "don't have the APM" to do something actually do, it's just that they don't have enough practice for their APM to go up since they are way to occupied with thinking, when the pros who have hours and hours of practice don't need to think as often, but instead most of what they do is second nature. It was a lot like what was done on Xellos, where they analyzed his brain while he was playing and saw that most of his decisions were second nature.
|
RPM is roach per minute and it's usually higher than APM*
|
If only we had something that took everything into account- APM, RPM, decision making, preparation, micro, macro, etc. Something maybe where you would match 2 players together and compare them by letting them express the different avenues of skill and intelligence they have with respect to the game. O yea, WINS.
|
On September 17 2011 02:16 PolSC2 wrote:
There is no clear way to judge someone's "rpm". Not going to work.
True, there is no way to measure RPM as there is with APM, but it is still helpful as it highlights the need for a mental checklist for lower leveel players. You can't measure it but if you are aware of it and you can increase the rate at which you go through your checklist then you are obviously going to improve as you will not get supply blocked, forget to make works, forget to make units etc.
Doing this will also keep your money low and allow you to see what you can afford off X number of bases. I played a game against a protoss yesterday who went 5 gate chargelot blink stalker off 1 base, then went for 7 gate collosus when he expanded after a failed push. If he had better RPM then he would know he could not afford those 5 gates and research off 1 base but because he did not keep up with his check list he had extra resouces.
On a side note, in the OP there is a typical RPM for Terran. For me Protoss RPM checklist would would go something like this:
Probes > Pylones > Warpin > Robo/Stargate > Upgrades/Tech > Micro
and for Zerg it goes:
Inject > Spread creep > Am i about to die? No, make a drone. Yes, make a unit > Overlord > Scout > React > upgrades/tech > Micro
I add more into Zerg because i main as Zerg and thats my checklist. Scouting could be put in with 'Micro' but it's so important for Zerg i feel it needs it's own point
|
|
|
|
|
|
EPT
