| Forum Index > SC2 General |
|
Stow.Wif
France67 Posts
now I average 150 for T and 300 for Z. My guess is that discarded commands are also counted : whenever you try to build drones by holding the key, after morphing all your larvas, the following commands are also counted in the apm. That the only thing that I see that can make sense. Due to macro mechanics Z apm has always been the same, but very far from 2 times my T apm ! | ||
|
Cheerio
Ukraine3178 Posts
On February 22 2013 17:15 ETisME wrote: I heard it is measured in real time rather than in game time? Clearly this change alone can't explain all the differences with the APM before patch. | ||
Olli
Austria24422 Posts
| ||
|
fabiano
Brazil4644 Posts
| ||
|
iamho
United States3347 Posts
| ||
|
Thieving Magpie
United States6752 Posts
On February 23 2013 00:55 iamho wrote: My BW APM (aka true APM) is in the low 200s but I can easily do 400+ on SCII now, despite not knowing half the hotkeys. According to Blizzard we're all a bunch of low-self-esteem 15 year-olds who will cry and quit the game if we're not tricked into thinking we're better than we really are. I'm pretty sure this is the same logic that they used when they decided to hide wins and give everyone useless division ranks. Um... BW did not have APM counters in it. We as a community willed it into existence and Blizzard is simply keeping pace with what we in the community are already doing for ourselves. | ||
|
ApocAlypsE007
Israel1007 Posts
Well, i'm not DRG, and my APM before the patch sits with Zerg at around 130... Apparently this takes into account all the times I hold 'd' for example to build a round of drones, each sticky stroke it counts as action... | ||
|
TheRabidDeer
United States3806 Posts
I tried: Moving just the camera with hotkeys (no apm) Moving the camera by scrolling to the edge (no apm) Moving the camera by minimap (no apm) Holding production key down without enough money (no apm) Sending repeated actions for a unit, such as right clicking multiple locations for an overlord (apm) Selecting units (apm) Basically, every action that actually makes something happen is an action in APM. A minute of testing (ingame time) and it seemed to be pretty close to how many actions I had performed during the test. I suspect there is just a bug in some games causing apm to inflate, but under normal circumstances it seems accurate. | ||
|
Befree
695 Posts
Okay here was my first test. I used the select command and attack command to kill my own base, and then continued to switch through selecting different units of my own to get a certain number of actions per game: 20 actions in a 1:49 game Game time APM would be ~11 Adjusted real time APM ~15.4 Score screen apm was 12 (or ~22 actions in 1:49) -------------------------- 40 actions in a 1:38 game Game time APM would be ~24.5 Adjusted real time APM ~34.3 Score screen apm was 27 (or ~44 actions in 1:38) --------------------------- 59 actions in a 1:37 game Game time APM would be ~36.5 Adjusted real time APM ~51.1 Score screen APM was 35 (or ~56.6 actions in 1:37) ---------------------------- So it definitely does not seem as if the score screen average is using real time APM. Which makes plenty of sense considering the game timer still isn't in real time, and adjusting the speed of a game drastically changes your APM (since it always calculates based on fast speed, no matter what). I think it is very safe to rule out the idea of this being real time APM ------------------ I also played a game where I spammed camera control including mouse scrolling on map, edges, and with my middle mouse button, as well as creating saved camera locations and recalling them. These had no contribution to the APM. ------------------- What I have noticed over my games is that the longer the games go, the higher the average APM goes. When watching replays Current APM seems to look perfectly fine, but the way the average increases just does not make sense. My guess would be that there is something wrong with the average APM algorithm and that it has nothing to do with blizzard time, camera controls, or actions only being counted at the most intensive periods. While I can't disprove the 3rd one very easily, it seems extremely unlikely and a very odd idea. But 1 and 2 are certainly wrong If only we could just fix this and change the standard game time to fastest so the world would just make sense again... | ||
|
Disposition1989
Canada270 Posts
| ||
|
FancyCaTSC2
56 Posts
| ||
|
Silentenigma
Turkey2037 Posts
| ||
|
Thieving Magpie
United States6752 Posts
On February 23 2013 07:03 Befree wrote: Conclusion: APM is still counted in blizzard time, and camera movements do not contribute to APM Okay here was my first test. I used the select command and attack command to kill my own base, and then continued to switch through selecting different units of my own to get a certain number of actions per game: 20 actions in a 1:49 game Game time APM would be ~11 Adjusted real time APM ~15.4 Score screen apm was 12 (or ~22 actions in 1:49) -------------------------- 40 actions in a 1:38 game Game time APM would be ~24.5 Adjusted real time APM ~34.3 Score screen apm was 27 (or ~44 actions in 1:38) --------------------------- 59 actions in a 1:37 game Game time APM would be ~36.5 Adjusted real time APM ~51.1 Score screen APM was 35 (or ~56.6 actions in 1:37) ---------------------------- So it definitely does not seem as if the score screen average is using real time APM. Which makes plenty of sense considering the game timer still isn't in real time, and adjusting the speed of a game drastically changes your APM (since it always calculates based on fast speed, no matter what). I think it is very safe to rule out the idea of this being real time APM ------------------ I also played a game where I spammed camera control including mouse scrolling on map, edges, and with my middle mouse button, as well as creating saved camera locations and recalling them. These had no contribution to the APM. ------------------- What I have noticed over my games is that the longer the games go, the higher the average APM goes. When watching replays Current APM seems to look perfectly fine, but the way the average increases just does not make sense. My guess would be that there is something wrong with the average APM algorithm and that it has nothing to do with blizzard time, camera controls, or actions only being counted at the most intensive periods. While I can't disprove the 3rd one very easily, it seems extremely unlikely and a very odd idea. But 1 and 2 are certainly wrong If only we could just fix this and change the standard game time to fastest so the world would just make sense again... It might be possible that the calculations keeps track of it chunks which would reduce the amount of "downtime" that normally occurs in games. For example, lets say in a 10 minute game you had 100 APM. Now, you don't actually have 100 actions in Minute 1, 100 actions in minute two, etc... This is an average. In ten minutes you made 1000 actions and that averages to 100 APM. How its normally calculated is in current rate. Example--at minute X I did Y actions in Z time and so my APM is W. This means that lulls in the actions reduces your average. In otherwords, even if you make 1000 actions in 10 minutes, if the lulls are too big during certain sections of the game you'll still have less than 100apm. Unless APM is literally dividing the number of actions total in a game by the amount of time passed. Lulls would not care. If you make 1000 actions in the first two minutes then walked away from your keyboard (dying at the 10 minute mark) you'll still have 100apm. In the old system, the moment you stop moving for 3-5 seconds your APM begins to drop and 10-20 seconds later your APM goes all the way to zero. If you walk away for 8 minutes you'll be barely above 1-5 apm. But if Blizz calculates simply the total number of actions and divides it by total time played--then hell yeah your APM would spike. | ||
|
-HuShang-
Canada393 Posts
Edit: nop | ||
|
neverlast
Austria62 Posts
"Average APM" was 100 and "current APM" floating around 80 for a while and suddenly then my "avg. APM" went up to 120, even though I my "current APM" was never higher then 100!! How can my current APM never go higher than 100 but my average is going up?? Blizz there is a bug in your numbers.... | ||
|
Henk
Netherlands578 Posts
| ||
|
Kaitokid
Germany1327 Posts
| ||
|
Pursuit_
United States1330 Posts
On February 23 2013 00:55 iamho wrote: My BW APM (aka true APM) is in the low 200s but I can easily do 400+ on SCII now, despite not knowing half the hotkeys. According to Blizzard we're all a bunch of low-self-esteem 15 year-olds who will cry and quit the game if we're not tricked into thinking we're better than we really are. I'm pretty sure this is the same logic that they used when they decided to hide wins and give everyone useless division ranks. I'm really confused by this logic. As long as APM remains consistent (which admittedly it hasn't recently) and measures actions the players take, then faster players will continue to have higher APM than slower players. If IQ was suddenly changed to be more detailed and accurate but increased the rating by a magnitude of 10x, and your IQ "improved" from 110 to 1100, it wouldn't be a self esteem boost if your friend with 125 IQ improved to 1250 at the same time. That being said, they really need to just pick a measurement of APM and stick to it, changing it all the time is what makes it useless. | ||
|
RaZorwire
Sweden718 Posts
It just keeps getting more mysterious. | ||
Falling
Canada11369 Posts
On February 25 2013 05:51 Pursuit_ wrote: I'm really confused by this logic. As long as APM remains consistent (which admittedly it hasn't recently) and measures actions the players take, then faster players will continue to have higher APM than slower players. If IQ was suddenly changed to be more detailed and accurate but increased the rating by a magnitude of 10x, and your IQ "improved" from 110 to 1100, it wouldn't be a self esteem boost if your friend with 125 IQ improved to 1250 at the same time. That being said, they really need to just pick a measurement of APM and stick to it, changing it all the time is what makes it useless. I think real action over a real minute is pretty reasonable how about you? How hard can this truly be? It's not a made up, arbitrary number. Key strokes and mouse clicks/ time. Just like rpm is cycle over time. revolutions per minute. There's no reason to get tricky by including non-standard times and the like. | ||
| ||
[ Submit Event ] |
|
|
|
OSC
The PondCast
LAN Event
Replay Cast
OSC
LAN Event
Korean StarCraft League
CranKy Ducklings
WardiTV Korean Royale
LAN Event
[ Show More ] |
|
|
EPT
