|
Make sure you read the OP before asking a question. Asking a question already addressed in the OP will result in moderation action. Also, please put some effort into your questions. Do not just ask for build orders. |
United States7483 Posts
On February 22 2015 02:21 imabigboi wrote:
From the point building a new nexus, when will the investment fully paid back?
That depends on a lot of factors, ranging from whether you have probes to saturate the base already and transfer them or have to build them still, whether it's harassed and you lose workers or mining time there, etc.
Each probe up to 2 probes per patch returns around 40 minerals a minute, so with optimal saturation (16 probes) immediately, you're looking at almost precisely 37.5 seconds on average of mining to return the cost of the nexus.
If you're mining gas, it complicates the issue because you have to ask the value of the gas compared to the investment of the nexus and assimilators.
If you didn't have enough probes to saturate, it takes quite a bit longer. In general, when you have no more than 16 probes mining at a base, each probe at that base takes around 75 seconds of mining for a probe to pay for itself (not counting supply costs and pylon costs, which adds an extra 100 minerals every 8 probes, and therefore increases it to 93.75 seconds per probe on average).
There are also opportunity costs to take into account as well, such as the delay in tech or units, which must be started later.
So this question is more complicated than it looks at first glance.
|
On February 22 2015 06:56 Whitewing wrote:Show nested quote +On February 22 2015 02:21 imabigboi wrote:
From the point building a new nexus, when will the investment fully paid back? That depends on a lot of factors, ranging from whether you have probes to saturate the base already and transfer them or have to build them still, whether it's harassed and you lose workers or mining time there, etc. Each probe up to 2 probes per patch returns around 40 minerals a minute, so with optimal saturation (16 probes) immediately, you're looking at almost precisely 37.5 seconds on average of mining to return the cost of the nexus. If you're mining gas, it complicates the issue because you have to ask the value of the gas compared to the investment of the nexus and assimilators. If you didn't have enough probes to saturate, it takes quite a bit longer. In general, when you have no more than 16 probes mining at a base, each probe at that base takes around 75 seconds of mining for a probe to pay for itself (not counting supply costs and pylon costs, which adds an extra 100 minerals every 8 probes, and therefore increases it to 93.75 seconds per probe on average). There are also opportunity costs to take into account as well, such as the delay in tech or units, which must be started later. So this question is more complicated than it looks at first glance.
Hmm... Can u give the shortest time needed to get the investment back. 400(nexus)+16*50（probes). Ignore the gas part and opportunity cost, also assume non-stop probe production.
|
United States7483 Posts
So you're asking at what point, after starting the construction of the nexus, do you make back 1200 minerals?
Since you asked shortest time needed, I'm going to assume 2 nexus production of probes, with constant chrono boost. That means every 11 1/3 seconds, 2 probes will finish and start mining. Probes return minerals at a rate of 2/3 of a mineral per second, on average. The nexus has a build time of 100 seconds, which cannot be chrono boosted.
This requires some calculus, since the first probe built will be mining far longer than the last probe mining.
Doing this, you finish 16 probes 190 2/3 seconds after starting the nexus. Disregarding transfer time from the other nexus to the new base, they'll have mined ~423 minerals at this point. It will take just under another 73 seconds to collect the rest of the 1200 minerals at this base, assuming no interruption or worker loss.
Thus, you're looking at around 263-264 seconds (or 4:23-4:24) from the instant you start the nexus until you've fully paid for the nexus and the workers to mine the mineral patches, disregarding travel time.
Taking travel time makes the final answer a little more complicated, because it differs from map to map. However, the way you would do the increased time is fairly simple: a probe mines 2/3 a mineral per second, so take the amount of travel time, multiply by 2/3, and multiply that by 8 (8 probes travel), and add that result to the final time. So if it took 8 seconds to get from the main nexus to the natural mineral lines, you'd be looking at an extra 42 2/3 seconds. Average is around 15 seconds, so taking that, you'd be looking at an extra 80 seconds due to travel time on a map with 15 second travel time from main to natural, making the final result around 5:43-5:44 seconds.
Not using constant chronoboost on your nexii will increase this time.
In a real game, you can expect around 6 minutes for your natural expansion to fully kick in and pay off it's total investment. Transferring excess probes from your main to your natural as soon as your natural finishes will reduce this time a bit.
|
I'm a mid-high diamond protoss trying to make that push for masters! If anyone can answer these that would be fantastic.
I have a few general questions involving several matchups.
PvT
- Are the two general openings following a reaper opening the 3 rax, or widow mine drops?
- If you see a reactor with your probe, is it generally a reactor expand? What units come out of this, and what do I need to be afraid of?
- Around what time should storm be finished, assuming nothing goes wrong, and we did a colossus opening?
(Also, there are some midgame openings where protoss stops production of colossus at around 3, and go to templar, or keep colossus production. can someone elaborate more on this? I'm not very familiar with it.)
PvZ
- Before getting colossus, about how many immortals would you make? (while getting a third) I'm aware its pretty situational but any answer involving this is fine.
- What maps are worst for pvz in the current map pool? It seems like deadwing and overgrowth (and maybe vaani) are horrible because of swarm host play.
PvP
- If the opponent goes phoenixes (4+) and I go robo, is the correct answer to get blink asap and 2 cannons per base?
- What situation is needed in order for one to go tempests? Every time I try to transition to tempests, or my opponent transitions to tempests, they die doing so.
Again im looking to get most of them answered, and any answers are greatly appreciated!
|
On February 22 2015 09:53 Whitewing wrote:
So you're asking at what point, after starting the construction of the nexus, do you make back 1200 minerals?
Since you asked shortest time needed, I'm going to assume 2 nexus production of probes, with constant chrono boost. That means every 11 1/3 seconds, 2 probes will finish and start mining. Probes return minerals at a rate of 2/3 of a mineral per second, on average. The nexus has a build time of 100 seconds, which cannot be chrono boosted.
This requires some calculus, since the first probe built will be mining far longer than the last probe mining.
Doing this, you finish 16 probes 190 2/3 seconds after starting the nexus. Disregarding transfer time from the other nexus to the new base, they'll have mined ~423 minerals at this point. It will take just under another 73 seconds to collect the rest of the 1200 minerals at this base, assuming no interruption or worker loss.
Thus, you're looking at around 263-264 seconds (or 4:23-4:24) from the instant you start the nexus until you've fully paid for the nexus and the workers to mine the mineral patches, disregarding travel time.
Taking travel time makes the final answer a little more complicated, because it differs from map to map. However, the way you would do the increased time is fairly simple: a probe mines 2/3 a mineral per second, so take the amount of travel time, multiply by 2/3, and multiply that by 8 (8 probes travel), and add that result to the final time. So if it took 8 seconds to get from the main nexus to the natural mineral lines, you'd be looking at an extra 42 2/3 seconds. Average is around 15 seconds, so taking that, you'd be looking at an extra 80 seconds due to travel time on a map with 15 second travel time from main to natural, making the final result around 5:43-5:44 seconds.
Not using constant chronoboost on your nexii will increase this time.
In a real game, you can expect around 6 minutes for your natural expansion to fully kick in and pay off it's total investment. Transferring excess probes from your main to your natural as soon as your natural finishes will reduce this time a bit.
thx dude, nice math work 
|
These are some pretty general questions so sorry in advance if I'm not precise enough :
PvT :
- Idk what you mean, yes 3 rax and widow mine drops are both standart follow-up of the repear opening, I don't think I can tell you than one is particulary more popular than the other.
- You mean like no repear reactor ? That's weird, I can just tell you he's gonna make marine and that you shouldn't try to deviate from your original gameplan (It's generally a good advice to follow when you're in doubt )
- There's a lot of way to play the midgame, and definitly no set in stone answer to this, as a general rule. You want to tech to storm when you are 3/3 (if opening double forge) OR are 100% sure he's not gonna scv pull (all 6 gas taken/double engibay armory ) and that the game is going for the longer game. The number of collosus doesn't really matter that much, you just keep pumping them as long as you're not feeling safe to transition to templar.
PvZ :
- None if you're facing no roachs at all, so it's usually a prism straight into colo(full lings/fast hydras), One if it's a few units into some tech (Imo into prism into collossus), 2 if it's presure into tech (Since you're under presure you can't afford a prism and needs the second immortal), as much as possible if it's mass roachs.
- That's pretty funny actually, since to me overgrowth and deadwing are two very good PvZ maps. I would say inferno is by far the worst PvZ map, followed by expedition lost (the damn rocks makes everything really akward). These 2 would be my veto into a Bo3 vs a good zerg.
PvP :
- This situation is pretty bad for you, you should try to get blink ASAP, 2 canon per base is overkill tho (unless he's going crazy on the phenix count). You should always try to survive with one canon and a lot of blink stalker/templar for feed back, so that you have a strong force for counter presure in the midgame.
- In collosus vs collosus, it's a risk often worth doing, but you're right saying that in an even situation, the player trying to tech to tempest can die to a well timed timing attack, but he will also die really easily if he doesn't react to the tempests in time, it's a delicate balance with no real definitive answer, sorry. In general the defender advantage is really strong in collosus vs collosus, so that's why you will often see player trying to tech up to tempest
|
On February 22 2015 10:53 PtitDrogo wrote:
These are some pretty general questions so sorry in advance if I'm not precise enough :
PvT :
- Idk what you mean, yes 3 rax and widow mine drops are both standart follow-up of the repear opening, I don't think I can tell you than one is particulary more popular than the other.
- You mean like no repear reactor ? That's weird, I can just tell you he's gonna make marine and that you shouldn't try to deviate from your original gameplan (It's generally a good advice to follow when you're in doubt )
- There's a lot of way to play the midgame, and definitly no set in stone answer to this, as a general rule. You want to tech to storm when you are 3/3 (if opening double forge) OR are 100% sure he's not gonna scv pull (all 6 gas taken/double engibay armory ) and that the game is going for the longer game. The number of collosus doesn't really matter that much, you just keep pumping them as long as you're not feeling safe to transition to templar.
PvZ :
- None if you're facing no roachs at all, so it's usually a prism straight into colo(full lings/fast hydras), One if it's a few units into some tech (Imo into prism into collossus), 2 if it's presure into tech (Since you're under presure you can't afford a prism and needs the second immortal), as much as possible if it's mass roachs.
- That's pretty funny actually, since to me overgrowth and deadwing are two very good PvZ maps. I would say inferno is by far the worst PvZ map, followed by expedition lost (the damn rocks makes everything really akward). These 2 would be my veto into a Bo3 vs a good zerg.
PvP :
- This situation is pretty bad for you, you should try to get blink ASAP, 2 canon per base is overkill tho (unless he's going crazy on the phenix count). You should always try to survive with one canon and a lot of blink stalker/templar for feed back, so that you have a strong force for counter presure in the midgame.
- In collosus vs collosus, it's a risk often worth doing, but you're right saying that in an even situation, the player trying to tech to tempest can die to a well timed timing attack, but he will also die really easily if he doesn't react to the tempests in time, it's a delicate balance with no real definitive answer, sorry. In general the defender advantage is really strong in collosus vs collosus, so that's why you will often see player trying to tech up to tempest
Wow im very very grateful for your response.
pvt: - So you generally get 3/3 first before teching into templar, unless he's not pulling the boys. I originally thought it was the other way around. Thanks!
pvz: - I'm curious as to why one would get the prism, or a prism so early in the game. Is it just to pressure while building up colossus and mining the third base?
pvp: - What would you do to prevent a phoenix player from getting such an advantage? Is going robo generally bad if a stargate player goes phoenix? Here is a replay of someone going phoenix and I scouted it late. http://ggtracker.com/matches/5816851
I'm not sure if you have time to watch (its ok if you dont haha) but I took lots of damage and it took several minutes for me to get blink and stalkers to ward off phoenix harassment. How do I negate a phoenix opening?
once again thank you thank you thank you
|
Austria24422 Posts
Thanks for the help PtitDrogo! Could you elaborate some more on SCV pulls? What are definite signs of it - 2 starports, ghost academy off 4 gas, 5 rax, single ebay? I'm really not sure, I keep forgetting 
What are key timings to scout for those signs?
|
After doing some gateway pressure, (cronoboosted MSC and cronoboosted Stalker attack) using MC's 2 stalker MSC opening, and following it up with an expand, with the intent to go into the templar tech route, which of the following is the best way to transition?
1) Get double forge then twilight into templar archives (this delays archives because upgrades like 1/1 and blink/charge cost alot of gas)
2) Go straight into high templar and then follow it up with double forge.
Or is there a third better way of doing so?
|
On February 22 2015 17:21 AkashSky wrote:
After doing some gateway pressure, (cronoboosted MSC and cronoboosted Stalker attack) using MC's 2 stalker MSC opening, and following it up with an expand, with the intent to go into the templar tech route, which of the following is the best way to transition?
1) Get double forge then twilight into templar archives (this delays archives because upgrades like 1/1 and blink/charge cost alot of gas)
2) Go straight into high templar and then follow it up with double forge.
Or is there a third better way of doing so?
If you really want to go HT then you're still gonna need some form of detection. With your pressure you should be able to identify if your opponent is going for a 1-1-1. However, he can still randomly drop widow mines at around 9-10 minutes off of a standard 3 rax opening. Back when templar openings were popular people used to open up with an oracle for detection, scouting and harassing. Lately, the only thing we've seen was oracle-phoenixes into templars. Phoenixes help out at dealing with medivac drops (which are otherwise really strong since you won't have blink stalkers) and lifting widow mines.
Otherwise, you can go for a robo first and then switch to the templar tech. You can build a few observers for map vision/detection and even a warp prism to delay any attack. You can also go into storm drops from there. Also immortals complement pretty well a chargelot/archon army, since they have a slightly better range to pick off widow mines and marauders without being kited to death.
Regarding the forge timings, it's pretty hard to survive and take/defend your third base with templars, but then it should get easier once you have plenty of high templar energy available all across your bases. So, I would suggest to stay on a single (early) forge until you're safe from immediate attacks. Armor upgrades are actually much more important than weapon upgrades when your army is mostly made of melee units, and when your main source of DPS are storms which aren't affected by upgrades.
|
Italy12247 Posts
Keep in mind that going straight into templar is INFINITELY worse than going colosus, so try that at your own risk.
|
Answering to Imrising and DarklordOlli :
PvZ : In my current PvZ build I get the robo after taking the third, so really the prism is not that early (To me Robo before third is an outdated build, but if you're doing it just add 1 immortal to the number I said above ) The job of the prism is just to harass when you have the money, nothing fancy.
PvP : Yeah robo expand vs phenix is rly rly rly bad, im gonna copy paste my response to a similar question that I answered some times ago (sorry no time to watch the replay ^^)
[ Yeah it's pretty much a build order loss, the less sentry you have, the better it is for you to eventually come back (I usually get 3 in my robo expand build so I cry everytime tt ), you need to try to make as much stalker as possible (constently produce out of the gateways even if it means cutting probes) so that you can defend both main and natural ASAP. You need to slowly tech blink or else the guy can just win by never stopping phenix production, make one obs to see his movements is a good idea too. Only after You have a good amount of stalker you can start a forge and try to go into a macro game. I usually take a third and try to do a timing with blink stalker chargelot/archon before they get too much colossi (making immortals is like a really bad idea they just lift them up ). It still is possible to win in this situation even if it's hard ( a commun mistake from the phenix players perpective is to tech to collosus too late and being overly safe vs blink stalker all-ins with too many immortals, meaning that a bunch of zealot archon becomes very effective in the midgame) ]
DarkLordOlli : Yeah It's kinda hard sometimes, A terran that wants to go for the late game is always gonna double gas at his third pretty quickly. Sometime it's really hard to scout the terran so my own plan is to take my gas when the terran take his, that way I always keep up in the units count (sometimes scv pull with ghosts comes with 5 gas, So don't try to overthink it if it happens and just take one gas yourself ^^)
|
Hi Guys! Maybe a stupid question.. But I'm new to Starcraft 2 (played Brood War before back in da dayz), and I've started with watching replays to get familiar with all the new units.
So now my question: Units have these special abilities, like the Phoenix that can do a Graviton Beam. When you want to use this, and you have like 10 Phoenixes selected, can you then just press the hotkey to beam a unit up, and will the system automatically select 1 phoenix to beam... Or will this action cause all 10 Phoenixes to do a beam (and all lose energy).
Or do you have to select 1 specific Phoenix and then use the beam?
Hope this question makes any sense :p..
Thanks in advance!
|
Austria24422 Posts
You can have all of them selected and only one will use the spell. Unless you click on two units, then two phoenixes will use it. And so on...
|
On February 23 2015 04:44 DarkLordOlli wrote:
You can have all of them selected and only one will use the spell. Unless you click on two units, then two phoenixes will use it. And so on...
Ok! Thanks 
|
Sorry i have a few more silly questions that I have floating in my head.
- If i 13 scout a zerg, and see a hatch first, I can skip the zealot right? Around what time do i need a unit to plug the wall?
- If I scout a double hatch before pool, is the best reaction simply to play standard, and not get the zealot?
- If a zerg is getting a third at the gold (of vaani or expedition) what is the best way to punish this? A quick 2 base all in?
- I played a pvp with a gm, and I did naniwa's safe expand, which goes around:
13 gateway
15 gas
16 pylon
18 cyber, 2nd gas,
21 pylon, warp gate, stalker
and gets a sentry, expand, gateway, 2nd sentry, and robo in that order. He proceeded to go straight to phoenix, deal some damage to my economy while im teching straight to blink, and he has an economical and tech advantage.
After the game he said that after seeing 1 gas and no 2nd gateway, he knew I was going to fast expand, so he went phoenixes. I'm curious to know, how good of an indicator is 15 gas, 18 gas and 1 gateway? Because if I keep facing gms and they know I go robo fe I'm always going to be at an disadvantage since the opponent will always have the build order win. How do I prevent this? Is there a good solid build that doesn't have such a glaring problem?
|
On February 23 2015 04:36 Lucchese wrote:
Hi Guys! Maybe a stupid question.. But I'm new to Starcraft 2 (played Brood War before back in da dayz), and I've started with watching replays to get familiar with all the new units.
So now my question: Units have these special abilities, like the Phoenix that can do a Graviton Beam. When you want to use this, and you have like 10 Phoenixes selected, can you then just press the hotkey to beam a unit up, and will the system automatically select 1 phoenix to beam... Or will this action cause all 10 Phoenixes to do a beam (and all lose energy).
Or do you have to select 1 specific Phoenix and then use the beam?
Hope this question makes any sense :p..
Thanks in advance!
It depends on the spell. Specifically for Graviton Beam, only one phoenix will lift a unit. However, something like Blink or Stimpack will make every unit selected use it. In the editor there's a checkbox for each ability (called "Best Unit" I think?) that chooses between those two behaviours. If it's selected, only the closest capable (enough energy, not on cooldown) unit will use the ability. If it's not, every capable unit selected will use it.
They did a good job applying it correctly to spells, so whichever behaviour makes the most is sense is generally what will happen.
|
Questions about 2 base blink style in PvT (pressure with 10 stalkers while teching)
1- shall I start gate exp or 15 double gas? Double gas let me do faster units and a poke to terran, while 2nd is delayed. Is this risky not having units in main for mine drops? Standard gate exp give me better eco with units in base. What are your thoughts?
2- how to delay terran with 10 stalkers? frontal attack is good trading shields for scv mining time? does the over reaction with bunkers worth it? When I blink to main, shall i kill army, scv or tech labs?
3- best follow up? 2 forge+templar archive for chargelot archon?
4- shall I take 3rd or finish him on 2 base?
|
Italy12247 Posts
1) There's no difference between the two. Im not sure why you think 15/15 allows you to pressure the terran, it just gives you more early game gas at the cost of a slightly later nexus
2) Just kill whatever you can and take map control. Depending on how hard you are going into pressure, you may or may not want to commit to a straight up fight. Generally you want to just harass and pick off whatever you can unless you are cutting probes for extra stalkers. Check out the vods on the Blink guide i wrote a while ago: http://www.teamliquid.net/forum/sc2-strategy/467544-partings-pvt-the-big-boy-build
3) Always colossus, templar builds are dead
4) Take a third and go colossus. If you commit to a hard blink pressure your robo tech is too delayed to be able to hit a reasonable colossus timing, unless you completely destroy him with blink (which is unlikely with this style, you want to take map control more than anything).
|
I need a PvP build, I'm tired of just cheesing every single game.
I want it to be safe, no cheese and with some sort of timing attack. Any ideas? All of the ones in the main help me section seem outdated. I'm Masters.
|
|
|
|
|
|
EPT
