EPTUsing BWAPI to take spectating to the next level! - Page 12
| Forum Index > BW General |
|
YPang
United States4024 Posts
| ||
|
AssuredVacancy
United States1167 Posts
A new worker stat is added: Total number of idle worker seconds as well as the total amount of minerals that could've been harvested by the idle workers. The idle worker seconds work like this: if there are 4 idling workers idling for 1 second each, that's 4 worker seconds, it's the same if there's one worker idling for 4 seconds. The number of minerals "missed" from idling is calculated by this: if the worker is idling for that frame(1/24th of a second), the number of minerals missed is equaled to the instantaneous mining rate per frame of that player's workers divided by the number of workers. E.g. If the player is on one base with 50 workers mining 8 patches, an idling worker will not increase the "amount of minerals missed" as much if the player is on one base with 4 workers mining 8 patches, simply because one added worker in the oversaturated mineral field is not going to be that big of a deal where as one added worker in an undersaturated mineral field makes a bigger impact. This is a fairly accurate measurement of the amount of minerals missed from idling workers. This will help you refine your builds as you can see how much more minerals you could've gotten if you had better mechanics. ![[image loading]](http://imgur.com/yJPzr.png) This was fairly late into the game, but you can see that the amount of minerals missed by both players is a pretty big amount. You can also see that the ratio of minerals:minutes is not the same. This is because the mining rates of each player is different when their workers are idling. Units making in production facilities, research/upgrades, unfinished buildings are all labeled as you can't tell what they are unless you click on them. This just saves a bit of time: ![[image loading]](http://imgur.com/xbcit.png) The option to turn them off is in the settings file under the field "showHiddenNames". The bug of not reading options correctly is also fixed. | ||
|
AssuredVacancy
United States1167 Posts
On February 08 2010 09:53 YPang wrote: what is the "windows folder"? sry im confused by that xD It should be either C:/windows, or C:/winnt | ||
|
Nevuk
United States16280 Posts
| ||
|
AssuredVacancy
United States1167 Posts
On February 08 2010 10:00 Nevuk wrote: It's kind of difficult to tell whether it's speed up or not- sometimes I hit + / - and nothing happens. It increases in kind of a weird fashion, usually you have to hold + or - to see a difference. | ||
|
mizore
40 Posts
| ||
|
Funnytoss
Taiwan1471 Posts
On February 08 2010 10:04 AssuredVacancy wrote: It increases in kind of a weird fashion, usually you have to hold + or - to see a difference. Is it possible to change the command currently used for changing speed? I have a laptop keyboard so I don't have keypad numbers... | ||
|
AssuredVacancy
United States1167 Posts
On February 08 2010 13:53 Funnytoss wrote: Is it possible to change the command currently used for changing speed? I have a laptop keyboard so I don't have keypad numbers... Hmm good idea, laptops usually have a numpad though(if you have a numlock, then you have a numpad, it's just overlapped with the normal keyboard) | ||
|
Bajadulce
United States322 Posts
| ||
|
Funnytoss
Taiwan1471 Posts
On February 08 2010 14:19 AssuredVacancy wrote: Hmm good idea, laptops usually have a numpad though(if you have a numlock, then you have a numpad, it's just overlapped with the normal keyboard) That's true, but for some reason it hasn't been working very consistently - speed toggling isn't very precise at the moment. The best solution I've found is to watch replays in multiplayer mode so at the very least "fastest" game speed will always be an option. | ||
|
YPang
United States4024 Posts
| ||
|
AssuredVacancy
United States1167 Posts
On February 08 2010 15:37 YPang wrote: I think ihave a good macro formula . I think u should calculate the amount of time the building is t making units . And you'll obviously want that time to be low. And for zergs u can do he same thing but for he time a larvae is idle. These should be applied 5 mins after the game cuz in he beginning players sometimes purposely leave them idle. I'm not really sure what you meant by macro formula or if his is what your seeking Posted this a bit earlier in the thread:
This number is calculated every frame, so naturally if you have idle buildings, you'll get penalized every frame and so you'll end up with a lower macro rating. This seems somewhat accurate right now as my replays never get above 9000, while flash/best gets up to 15k easily. | ||
|
yann
119 Posts
I recall I might run the chaos launcher as an admin... just in case. It's a great tool! | ||
|
pyrogenetix
China5098 Posts
| ||
|
NiteKat
United States151 Posts
very very nice tool though. Must have taken a lot of work great job and keep up the good work. Oh yeah on a note something I noticed. When I used your tool all of the menus in the game sped up. For example the animation that makes the menuse move in and out of the screen as you navigate to single player/lan etc. all moved faster. Kind of an odd thing to have happen and not super important but just weird. lol | ||
|
Shadowfury333
Canada314 Posts
On February 08 2010 15:46 AssuredVacancy wrote: This number is calculated every frame, so naturally if you have idle buildings, you'll get penalized every frame and so you'll end up with a lower macro rating. This seems somewhat accurate right now as my replays never get above 9000, while flash/best gets up to 15k easily. Have you taken into account that certain MU's have idle buildings as a matter of course? For example, in TvP, the Terran isn't using their barracks for production after the first few minutes, but it still counts as idle here. | ||
|
Durak
Canada3685 Posts
On February 09 2010 02:51 Shadowfury333 wrote: Have you taken into account that certain MU's have idle buildings as a matter of course? For example, in TvP, the Terran isn't using their barracks for production after the first few minutes, but it still counts as idle here. Good question. Another common example I can think of is Zerg saving larvae for mutas. | ||
|
AssuredVacancy
United States1167 Posts
On February 09 2010 02:51 Shadowfury333 wrote: Have you taken into account that certain MU's have idle buildings as a matter of course? For example, in TvP, the Terran isn't using their barracks for production after the first few minutes, but it still counts as idle here. Well then you can choose to ignore the provided macro rating for that period of time. The rating was never meant to be an end all be all estimate of one's skill, but rather provide a rough estimate of your current macro progress. But something like you suggested is already implemented, as idle buildings/supply blocks are not counted for supply numbers lower than 50, because I realize there are certain timings that require you to either be stuck at 17/17 supply(10/15 ranged goon for example) or for you to have idle larvae, like mutas. | ||
|
NiteKat
United States151 Posts
Now what would be really awesome would be if someone like Day[9] or the Korean professional scene decided to use the data this program is able to grab from replays and come up with their own macro formula. edit: the one thing I noticed is that command centers and nexus are counted as production facilities and that you do eventually stop producing workers in long games. So those will count against this "macro score" produced. However if you don't try to compare a "terran" to a "zerg" and just compare terrans to terrans etc. then that isn't an issue as all players will have the same handicap there. | ||
|
emucxg
Finland4559 Posts
| ||
| ||
