|
3340 Posts
In the Broodwar forums there was a thread posted recently called Training Exercise for Starcraft. Since my slow hands bug me I decided it would be a good idea to try out the exercises BlueIris suggested. The hope is that it'll eventually speed me up in useful ways even if it's pointless spam to start off with. I'm gonna try and fit in 5 games a day and report the results here. Here's today's report:
All my games were ZvP on Luna. I chose ZvP because I'm Z, and I think the protoss comp's zealot push is arguably the most difficult to deal with out of the comp's arsenal of attacks. I chose Luna because it's boring and familiar and I'm pretty sure the comps don't get confused on Luna.
The first 3 games I opened with 3 hatch lurkerling, expanded twice to gas expos, then switched to ultraling. The last 2 games I opened with 4 hatch hydra (4th hatch at a min-only), and then teched to lurkers and expo'd to 2 gas expos... both 4 hatch games ended before my gas expos were up and running.
I noticed in the 4 hatch hydra games, where I had the min only, I spent my minerals very quickly, and needed to tech lurker to spend all my gas. The gas probably could've been used for faster hive tech instead. Its something I should try.
In the 3 hatch games almost all my unspent resources were minerals, and the extra minerals remained high even after putting down 2-3 extra hatcheries in my main for pumping lings.
Both 4 hatch hydra games I met the macro goals, though I had one less mineral line and 2 less gas than in the 3 hatch lurker games. I may need to place a few more hatcheries to make the macro goals in the 3 hatch lurker games.
As far as making the APM goal, I'm very far away from that for the time being. In fact, my "APM Max" only broke the 300 mark for a few moments in just one of my games... and it was 100% hotkey spam while I was containing the Comp and waiting for more larva. This may be an indication that in order to meet this goal I'll need to hotkey spam constantly, and push myself to fit actions inbetween the lightning fast spam.
Other notes: Game 1 the comp Toss FE'd... Comps are so bad at that. The rest of the games were comp's standard 4 gate zealot. I was never in any danger of losing a game. That's probably the result of not enough useless spam to meet the exercise's goals.
Goals = 300 apm, less than 600 resources.
Comp Spam Practice
Game01: 130 APM; 1349 unspent resources; 14:42 time.
Game02: 113 APM; 1380 unspent resources; 18:36 time.
Game03: 153 APM; 3590 unspent resources; 16:46 time.
Game04: 153 APM; 355 unspent resources; 14:39 time.
Game05: 149 APM; 237 unspent resources; 13:19 time.
Game01-05 Averages
140 APM; 1382 unspent resources; 15:36 time.
|
So you did improve as the day went on. Nice!
|
go there
http://www.teamliquid.net/forum/viewmessage.php?topic_id=50034
search for
BW AI Project
Homepage
Download
Authors : Entropy and Bajadulce
Version : 3.1
Description : players have always complained about the original AI. Therefore, Entropy (aka Physician on most SC forums) tried to enhance it using some scripts of his own. He was helped in this task by Bajadulce and it appears that the AI is now quite competitive... give it a try !
get it.
since you'll be playing vs pc that AI it's way better then blizzard AI
|
On July 10 2007 13:08 xM(Z wrote:go there http://www.teamliquid.net/forum/viewmessage.php?topic_id=50034search for BW AI Project Homepage Download Authors : Entropy and Bajadulce Version : 3.1 Description : players have always complained about the original AI. Therefore, Entropy (aka Physician on most SC forums) tried to enhance it using some scripts of his own. He was helped in this task by Bajadulce and it appears that the AI is now quite competitive... give it a try ! get it. since you'll be playing vs pc that AI it's way better then blizzard AI
the point of the exercise isn't for a challenging opponent though... then he could just play real people
the goal is to raise his APM and overall speed which is easier to focus on with worse opponents, hence using the computers
in fact tadzio i'd argue that your practice session was a failure because you did not meet your goal of 300 APM and under 600 resources (continually not just at the end of the game)
it's mostly irrelevant that you beat the computer, don't worry if the computer beats you with zealot rushes but rather that you're making your APM goals, that is after all how i felt blueiris explained it
anyhow that thread also interested me and made me wonder if i should go about artificially boosting my APM with more spam like that, idk though in my experience i've reached my highest APMs when i was challenged the most (playing someone much better than me)
|
3340 Posts
I agree that I didn't meet my goals. The last two games I think I met my macro goals... but I didn't have a huge economy like I did the first three games so it's hard to tell. Anyway, the reason I'm doing this is so I can meet the goals. I don't think I'd bother with it, and I certainly wouldn't blog about it, if I could already do it.
I'm impressed that your APM climbs when you're most challenged, Locked. The more difficult my opponent, the less I do.
|
nice :o
Maybe we should play a few games someday.
|
I actually tried BlueIris's technique and I saw immediate results (although staying <300 minerals was a big problem).
It jumped from 140s to about 190s after several games; I have so much trouble with Pylon timing that it's ridiculous  If I didn't have this problem I'd be in the mid-250s by now T_T
|
they should make the BW-AI works without a cd ..
|
This is stupid.
You are going to make it a habit to press 5 times every time you could just as well press once. Of course, then your APM will be higher even if you aren't spamming consciously. But you are still spamming. The result will be that you won't know if you improved or not.
And it also means that the time spend on every spamming action cannot be spend on real actions. So 150 APM with spam is probably worse than 100 APM without spamming.
Of course in the end playing 5 games a day while macroing like hell will make your real APM go up.
|
APM is not really that important. It will increase with the skill. Just go play real people, you'll learn faster that way. But having low apm does not make you a noob. Look a Oldy,he's top10 Iccup and I probably have twice his apm, but not even a quarter of his skill and experience.
|
would you recommend trying this vs the brood war ai project?
|
|
|
|
|
|
EPT
