| Forum Index > SC2 General |
|
SarcasmMonster
3136 Posts
On June 18 2012 07:43 Existor wrote: Why you need global ladder? Who need it? You're best in your region and thats enought, isn't it? If you want to be best in Korea, then switch region in game settings and go to korean ladder! What the problem? Isn't that exactly what they are doing? | ||
|
KonohaFlash
Canada1590 Posts
On June 18 2012 07:15 FlamingForce wrote: Sorry to rain on your parade here bro but the A.I. is very well capable of making MKP's micro look like bronze league  + Show Spoiler + http://www.youtube.com/watch?v=IKVFZ28ybQs Again, I'm not saying that the A.I can't micro well. What I am saying and what Chris Sigaty said in the interview, is that making the A.I Imitate MKP's micro 1:1 will be difficult. Of course playing against A.I with better marine micro than MKP will allow you to practice your baneling and infestor micro very well, but it's unrealistic setting that you will find in an average ladder game. However, there are plenty of great and smart people in the Starcraft Community, and anything is possible. In my opinion, this tool will really help with the metagame, I can see myself spending hours practicing against shit I lose to all the time on ladder :D | ||
|
Alexstrasas
302 Posts
On June 18 2012 06:10 Shikyo wrote: I think that it could be programmed to be significantly better though. Maybe people are lazy, maybe they don't know how, but the technology is there. For instance scouting and being able to guess from that what the opponent is doing, giving each alternative a chance (Say DT rush 50% Blink stalkers 35% chargelots 13% citadel fake into something else 2%) upon scouting citadel, and then making it figure out the best possible response to that, etc. I really think that it'd be possible to program a good AI but it'd be very difficult and it'd require the programmer to have extensive knowledge about pro level play as well That is a slippery slope my friend. Yes you can work on the AI so it does better strategic options, however if you do that + machine micro reflexes 9001 apm you can end up with an AI that is actualy better then the players themselves thus opening a window for other problems. Let´s not forget that the game is not that complex, you can, with enough time, just "teach" any and all builds, army compostions,etc on a certain map, and the AI will always play flawlessly. | ||
|
Existor
Russian Federation4295 Posts
Isn't that exactly what they are doing? Global play =\= Global ladder. Understand? Launch Diablo 3 and see what is Global play function. You can simply change your server and play everywhere you want, but you will have separate statistics, like in Diablo 3 - separate auction and separate heroes with separate single progress | ||
|
StarcraftWonders
United States59 Posts
| ||
|
paralleluniverse
4065 Posts
On June 18 2012 08:10 Existor wrote: Global play =\= Global ladder. Understand? Launch Diablo 3 and see what is Global play function. You can simply change your server and play everywhere you want, but you will have separate statistics, like in Diablo 3 - separate auction and separate heroes with separate single progress Yes, global play and global ladder are different. So the description in the OP should have mentioned global play, because that's what's happening. A (region/server-wide) global ladder sadly isn't coming, nor was it discussed, so it should not have been deceptively placed in the description of the interview. | ||
|
SarcasmMonster
3136 Posts
On June 18 2012 08:10 Existor wrote: Global play =\= Global ladder. Understand? Launch Diablo 3 and see what is Global play function. You can simply change your server and play everywhere you want, but you will have separate statistics, like in Diablo 3 - separate auction and separate heroes with separate single progress Why you need global ladder? Who need it? You're best in your region and thats enought, isn't it? If you want to be best in Korea, then switch region in game settings and go to korean ladder! What the problem? I'm confused. I thought Global Play (which is what we are getting) was what you wanted from your original post. It seemed like you had an issue with Global Ladder. | ||
|
Alchemind
Albania142 Posts
| ||
|
malaan
365 Posts
I saw that Blizzard said they wanted a unit that could break siege lines from the ground and give zerg some 'map control' capabilities. Sounds good, but what happens when this unit is in it's 'siege' mode and terran decides to drop marauders on them with a scan? the DPS of marauders is insane compared to what I see in the DPS / unit release of the swarm host. I just have a vision in my head of 3-4 marauders derping 2-3 swarm hosts in 5 seconds flat. Also - the protoss base raider unit. I just can not see how protoss will have the APM time free to bother with it's 'base raider' abilities... seems completely stupid to me. Also, saw a video somewhere of someone at anaheim using the 'tempest' unit. It has absolutely insane range, and he did some kind of 1 base tempest rush, sieging the zerg natural and there seemed like zerg could do nothing about it. | ||
|
KimJongChill
United States6429 Posts
On June 18 2012 09:13 malaan wrote: So about the swarm host.... I saw that Blizzard said they wanted a unit that could break siege lines from the ground and give zerg some 'map control' capabilities. Sounds good, but what happens when this unit is in it's 'siege' mode and terran decides to drop marauders on them with a scan? the DPS of marauders is insane compared to what I see in the DPS / unit release of the swarm host. I just have a vision in my head of 3-4 marauders derping 2-3 swarm hosts in 5 seconds flat. Also - the protoss base raider unit. I just can not see how protoss will have the APM time free to bother with it's 'base raider' abilities... seems completely stupid to me. yeah, also considering that enough dps on the part of p/t makes swarm host useless :/ | ||
|
SarcasmMonster
3136 Posts
On June 18 2012 09:13 malaan wrote: So about the swarm host.... I saw that Blizzard said they wanted a unit that could break siege lines from the ground and give zerg some 'map control' capabilities. Sounds good, but what happens when this unit is in it's 'siege' mode and terran decides to drop marauders on them with a scan? the DPS of marauders is insane compared to what I see in the DPS / unit release of the swarm host. I just have a vision in my head of 3-4 marauders derping 2-3 swarm hosts in 5 seconds flat. Also - the protoss base raider unit. I just can not see how protoss will have the APM time free to bother with it's 'base raider' abilities... seems completely stupid to me. Swarm hosts will die to Marauders just as fast as Siege tanks will. At least the Locusts can meat shield if the Marauders approach by ground, and attack Medivacs if they approach by air. | ||
|
malaan
365 Posts
On June 18 2012 09:16 SarcasmMonster wrote: Swarm hosts will die to Marauders just as fast as Siege tanks will. At least the Locusts can meat shield if the Marauders approach by ground, and attack Medivacs if they approach by air. The point was how slowly they spawn compared to how quickly stimmed marauders can fire. Also, zerg AA on ground units in it's current form sucks (say they use queen, slow as shit off creep. Say they use hydra, too expensive to use hydra / swarm host and in theory this composition would be useless anyway.) Don't forget zerg doesn't have marines to protect siege tanks ;-) The whole weakness of broodlords against terran is the fact that they are slow - and very easily picked off by vikings. Imagine how hard it is going to be to protect your swarm hosts from siege tanks and scans, you cant exactly run infestors into siege lines to fungal them can you? It kind of gives me a 'rock paper scissors' worry for TvZ right now but we will see. I think it was IdrA who was saying how all the units just feel stupidly overpowered and completely counter each other in every matchup, so we will have to see what happens in beta. I really don't know how I feel about HOTS atm... | ||
|
SarcasmMonster
3136 Posts
On June 18 2012 09:21 malaan wrote: The point was how slowly they spawn compared to how quickly stimmed marauders can fire. Also, zerg AA on ground units in it's current form sucks (say they use queen, slow as shit off creep. Say they use hydra, too expensive to use hydra / swarm host and in theory this composition would be useless anyway.) Don't forget zerg doesn't have marines to protect siege tanks ;-) The whole weakness of broodlords against terran is the fact that they are slow - and very easily picked off by vikings. Imagine how hard it is going to be to protect your swarm hosts from siege tanks and scans, you cant exactly run infestors into siege lines to fungal them can you? It kind of gives me a 'rock paper scissors' worry for TvZ right now but we will see. I think it was IdrA who was saying how all the units just feel stupidly overpowered and completely counter each other in every matchup, so we will have to see what happens in beta. I really don't know how I feel about HOTS atm... It's pure speculation but I think Swam hosts will be decent at protecting themselves since Locusts can last up to 25 seconds and they spawn every 25 seconds. Even if you spawn them, and camp them over your Swarm hosts. Say they use hydra, too expensive to use hydra / swarm host and in theory this composition would be useless anyway. I think you may be getting ahead of yourself here. Swarm host + hydralisks looked damn strong in the Battle reports. It might be a great composition in HOTS. Time will tell. | ||
|
Existor
Russian Federation4295 Posts
the DPS of marauders is insane compared to what I see in the DPS / unit release of the swarm host. I just have a vision in my head of 3-4 marauders derping 2-3 swarm hosts in 5 seconds flat. Use hydralisks, fungal and blind to marauders. Also Locusts can crush medivac at any time. Their stats are similar with roach, but shorter range and less hp, plus anti-air atack included. Also, saw a video somewhere of someone at anaheim using the 'tempest' unit. It has absolutely insane range, and he did some kind of 1 base tempest rush, sieging the zerg natural and there seemed like zerg could do nothing about it. Tempest is a "long range siege support", not a "long range siege flying tank without deployable mode". Also tempest speed is 2.25, that is slower, than mutalisk and corruptor You can kill it with: * vipers + hydralisks * fungal + locusts (cheaper in terms of vespene and mostly at tier2) -> stun and locust them * mutalisks and or corruptors. There are no splash Don't forget zerg doesn't have marines to protect siege tanks They have hydralisks and Queens. Queens can heal Swarm Hosts. Also, locusts can do small support vs air Current problem is how zergs will deal vs Oracle. | ||
|
AndAgain
United States2621 Posts
| ||
|
synapse
China13814 Posts
| ||
|
GohgamX
Canada1096 Posts
 | ||
|
GohgamX
Canada1096 Posts
| ||
|
Existor
Russian Federation4295 Posts
So global play will finally allow us to have a separate MMR for each race, right? Right, on different servers. | ||
|
Bombadil819
United States45 Posts
On June 18 2012 06:41 KonohaFlash wrote: I'm just repeating what Chris Sigaty himself said. I think he knows better than anyone else. I'm sure he's aware of the level of AI that's actually in the game, and I wouldn't be at all surprised if he followed stuff like the BWAPI competitions, but a) You didn't specify the "AI that's currently in game", you said AI in a very general sense. If you meant the former, that's entirely reasonable, but I still felt like the mistake needed to be pointed out so no-one would be mislead by it. (Please don't take this the wrong way, my post is intended to be educational, rather than combative). b) While I'm sure that what he meant was the "AI that's currently in game" (and it definitely can't micro like MKP  ), I'm certain that he doesn't know that AI as well the software engineers who wrote it, and he's definitely not an expert on the field of AI in general. It really wouldn't be that difficult to replace the existing AI with something EMAPF (Evolutionary Multi-Agent Potential Field) based, at least for micro. The difficult part would be reactive macro, but an evolutionary algorithm based on the responses they've already given it would probably also be an improvement. The in-game AI could become an actually useful training tool in that case. | ||
| ||
The PiG Daily
Best Games of SC
BASILISK vs T1
Maru vs Clem
Serral vs herO
Clem vs SHIN
[ Submit Event ] |
|
|
|
CranKy Ducklings
Online Event
FEL
AI Arena Tournament
BSL: Ladder Tournament
Sparkling Tuna Cup
INu's Battles
herO vs Classic
Percival vs Rogue
BSL Open Qualifier
BSL Open Qualifier
Replay Cast
[ Show More ] |
|
|
EPT
