|
Disclaimer, please read:
This thread is obviously inspired by this week's SOTG discussion about mech in tvp. I personally believe that mech in tvp is not viable simply due to how the two races are designed and some of the nerfs that have been applied to terran over the lifespan of SC2. That being said, I think trying to make mech work regardless is still an interesting exercise – it's fun, and you can actually win games with mech.
I'm sure no one will care, but I'd really like it if this thread was free from the typical "mech won't work because of this, this and this (and chargelots!)" kind of deal. I know. We all should know by now that mech in the long run isn't viable. But most of us aren't tournament players, and some of us play for the fun of coming up with strategies and because it's fun to do unorthodox stuff regardless of our level of play. Most of us low/mid level players (for the record, I'm currently plat with #1 diamond as my highest ever rating) will win games simply from having a game plan and better macro than our opponent anyway, even if that game plan entails 50 minute games with 20 planetary fortresses.
Also, given that I'm just a platinum leaguer, I will for sure be saying things that seem obvious to a lot of people. I'm sorry for that in advance and please correct me if I say anything blatantly false.
Please feel free to share any replay you feel is relevant. I am personally very interested in replays where the terran loses.
+ Show Spoiler [note] +
I plan on changing this post considerably if anything productive comes out of an eventual discussion, adding on replays and other comments, etc. These are just lose ideas and observations to maybe get a discussion going, nothing more.
Openings:
There are many ways to get to a mech composition, but I've personally have had waaay more success with openings that don't go straight for factory units – although a 1/1/1 with siege tanks that somehow allows for a reasonably timed expo can for sure be very effective.
I'd love some input here.
+ Show Spoiler [How I open] +On a map like Shattered Temple, Shakuras or Metalopolis I'll always open with a reactored barracks and a proxy factory. The factory is primarily for a nicely timed scout, but it also functions as a kind of catch-as-catch-can harassment tactic: If there's a spot in the protoss main where they don't have vision, I will for sure try and get out a hellion or three and kill some probes. It's cheesy, but all is fair in tvp and mech. The crazy thing is, I've noticed that even players like Huk will often have a blind spot somewhere like next to the gas vents on Metalopolis. But primarily the proxied factory or the hellions built from it are for scouting. All the while, I'm making marines and bunkers. Taking workers off gas as soon as you have reach 150 lets you take your natural expo faster (duh) but there's a lot to be said for getting a faster starport and expanding after a raven as well (very safe against hidden DT tech, lets you get vikings against void rays, etc) Also, something kind of like Debo's marine/hellion all-in can be good against a fast expanding protoss, especially if you bunker up for a bit at the protoss' natural and all-in "less." I've had some bad losses doing this where I obviously over-committed to the to maintaining the position while expanding behind it. It's better to go home early than to die or let the game become more even because of a proxy pylon warp-in counter. ![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/proxyfacreturn.jpg) My first factory returning as I float my OC out to my natural. From here, something that looks a bit like Yoshi Kirishima's pure air build is quite good, although I don't build a PF at my natural – bunkers and repair works fine for the most part. The amount of pressure you can apply with just a few banshees is often enough to allow you to sneak a quick 3rd on a distant part of the map. Another good thing about going reasonably heavy air for a while is that it augments the protoss' composition – they'll make way more stalkers than what is good for them. I'm starting to think that getting up 2-3 starports while still on one factory might be a good idea. You should always have a reasonably sized air fleet of vikings, banshees and a raven or two flying around and doing damage. This air fleet will also do reasonably well against blink stalker harass – 2 PDDs, landed vikings and invisible banshees can take out a larger army of stalkers quite easily if there are no other protoss units around. ![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/harass.jpg) all this damage is the result of 3 proxied hellions and three active cloaked banshees By this point in the game, you'll start floating minerals, so it's time to start creeping forward with PFs, tanks, turrets and sensor towers.
The Chessboard:
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/thechessboard.jpg)
This is what interests me the most about tvp mech and it's where I want to start the discussion and share some simple observations. My main thesis on tvp mech is that it will not work unless you push forward with planetary fortresses - you don't just cut the map, you keep aggressively pushing with them. This limits the map selection. Unlike Artosis, I don't think a map like Antiga Shipyard is good for tvp mech, because it's so easy to get outflanked on that map – you can't get a good focues position with planetaries set up unless you somehow managed to set up just outside the ramp to his natural or something. I really can't see that happening.
I'm using Shattered Temple as a case study, since it has a big open center that allows for a lot of different tank and planetary configurations and it has some very distinct counterattack paths.
Castling:
The most important thing when building your big complex of fortresses is that you need to have pockets of space between your buildings for your tanks to move and re-siege, and there needs to be a path out of their safe pocket that can't easily be used as a counterattack path for the enemy.
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/fortress_sandwich.jpg)
A fortress sandwich lets your tanks be reasonably safe against chargelots and blink stalkers, and it's a position the tanks can easily move in and out of when your army needs to reposition. A possible improvement on this position would be to move the two tanks one or two hexes inward, depending on where it looks like the protoss player might attack. That way, chargelots will funnel into a narrow choke and the two fortresses will quickly kill them.
CC Production area:
Never, every build command centers in front of your army, especially if you've pushed so far up the map that the protoss is getting nervous and will attack anything on sight, regardless of the tanks behind the building CCs. You'll be forced to cancel buildings or worse. Have a designated area to build CCs and to float them out from – somewhere like the area in front of your gold base on Shattered Temple, since it's also protected by the narrow pool of water and the building CCs will serve as an emergency road-block in case a small attack is heading down that path.
Flank protecton:
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/unprotectedflankfail.jpg)
A line of depots is never enough defense and won't delay the protoss army for long enough that you can reposition your tank army.
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/reasonablysafeagainstcounters.jpg)
Here's a reasonably safe flank defense. Also note how there is plenty of space behind the PF to position tanks in case your sensor towers show a counter attack moving down the lane. You don't actually want a tank right behind the PF at all times though, it won't be able to participate in most fights.
A good idea is to make sure that typical counterattack paths are sealed tight. That way, a protoss player sending something like a DT down that path by just clicking the minimap will lead to the DT actually running into your PF/turret position closer to your designated choke and army. Dead DT.
Over-extention:
A typical mistake of getting too anxious to push the planetary fortresses forward, resulting in obvious weakness and possibilities for the protoss player to attack the building command centers or to counterattack somewhere else.
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/position010.jpg)
Here my position is starting to take shape and is very difficult to attack head-on for the protoss player with the army he currently has.
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/extendfail.jpg)
but here I'm pushing forward too fast
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/moreextendfail.jpg)
Ideally, I should have “boxed in” my initial position by making a row of planetaries close to the tanks on the west flank, kind of like this:
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/hardendflank.jpg)
Reinforcement flow:
![[image loading]](http://i1203.photobucket.com/albums/bb398/Broodthaers/reinforcement_flow_01.jpg)
At this point of the game, the protoss has started switching to void rays, carriers and a mothership. That doesn't mean you stop making tanks or adding on more factories. Here, I am adding on factories to the right, in a forward position, and the turrets create a nice flowing path for the reinforcements to take. The turrets will also funnel any attack trying to intersect the turret “path,” making the enemy units bunch up so that tanks do more damage to them. The important part of the factory positioning and the direction of the flow is that it leads directly to the part of the map I want to isolate and attack – the protoss' natural.
---
OK, so it's not much yet, but it's a start. Strictly speaking what I'm talking here isn't “mech” but rather a very constrictive, slow way of playing that reminds a bit of Thorzain's goofy “Spoonzain” style he does in tvz on his stream. Straight up mech with nothing to augment it will lose virtually every time to a moderately aware protoss player – I think what I've realized so far is that mech tvp hinges just as much (or maybe even more) on air units and static defenses as it does on factory units.
I won't be posting any replays yet, because my opponents frankly speaking haven't been very good. I think this could end up being a more theoretical discussion anyway, but please feel free to post your own.
|
To call it viable it should be working at Grandmaster level at the very least (and yes mech does work on grandmaster level at times but you have to work your ass off) GoOdy has shown us that, but with the most recent patches it has become really hard to beat an equally skilled protoss player with mech.
Ive beaten GM protosses with mech(mostly because they've been caught offguard) but i dont think it will be viable in Code S in WoL anyhow which is the true testament to any strategy.
Goodluck anyhow
|
Cool post.
Looked like a really funny game  too bad you didn't post the replay .
|
The thing I don't understand about mech play (Protoss POV here) is the lack of mobility. What happens when Protoss wants to go DTs and just fairies them in via WP? How does Terran respond to that?
Also, can't Protoss just take the map like crazy similar to BW? I recall a game White Ra vs Idra in BW where WhiteRa took a double expansion around the 9 minute mark and always had 1 base over Idra the entire game.
I feel Blink Stalkers would prove to be extremely annoying as well. Since all you have to contest them is Hellions, Stalkers really have free reign all over the map, as opposed to worrying about Marauders+Marines. Blinking into your base, preventing you from slowly pushing forward by killing off SCVs building anything.
|
Great Read! Looking forward to more analysis as you play :D
Replay would be nice, but it's probably really long
|
United States2186 Posts
I've tried this exact idea about half a year ago before the blue flame nerf which obviously makes mech a lot worse; didn't work then either after people had seen it more than once or twice. Sad but, mech isn't used for a reason 
You will crush any ground force if you are slow, patient, and make enough planetaries. However, the immobility of mech is so easy to take advantage of with blink and especially nowadays when protoss actually realize the warp prism exists.
Another problem is cost efficiency. I've lost in even base shattered games versus pure ground armies. Protoss armies are much more cost effective than terran armies in general and when you are spending thousands of minerals on planetaries, it's far worse. 3-3 Immortals basically kill planetaries for free if you have nothing there and when you're stretched thin from blink/wp harass, it's hard to cover everything even on the smallest maps.
However, all of the above isn't that important. The real problem is carriers. If you let protoss get enough carriers (5-6+), you lose unless they are broke and you have many more bases. No way around it. Their air upgrades will finish faster, their cost efficiency is absurdly superior, and they will win almost all battles. In low carrier counts vikings can overwhelm them by sheer force but vikings suck once they get it enough; they also become paper planes if you cannot emp all of their templars. Marines beat carriers handily but they will automatically die to colossus/storm. Battlecruisers also can work but only if they don't get feedbacked, which is not something to bet on especially since protosses know that they can hide templars in wp's to dodge emps. If by some miracle you can hold a position that they can't break, you still just can't keep up with the mobility of carriers. Mass turrets can help hold one spot but you can't afford them in multiple areas. If you want to see this problem in action there's some replay of grubby vs sarens on shattered/lt from awhile back from some euro lan which lasted an hourish. Grubby played really poorly but it didn't matter.
|
It's an interesting idea, but there are a few things I think you have to consider if you want to continue using this strategy (especially in higher leagues).
First of all, what about drops? It's impractical to completely defend your side of the map with turrets, and the nature of a warp-prism means that the Protoss can bring in as many units as he wants. Since mech is so immobile, and you've invested so much into planetary fortresses, you can't really respond quickly (if at all). Even if you get planetaries at all your expansions, which severely hurts your economy, he can always go for your main/natural (which will always be orbitals) and kill whatever production buildings he can find. Even if you have some of your factories in the middle of the map, that's a big issue. Blink stalker harass is a similar issue -- until you manage to push almost all the way to his base, he can blink enough stalkers up into your main to kill everything almost instantly, or kill one of your mining bases (even if it's a planetary).
Also, keep in mind that at no point shown in those screenshots were you less than two bases ahead of your opponent. At that point, basically any strategy would have won you the game. I understand that your planetaries allowed you to secure that many bases, but I don't see why you couldn't have done the same thing with a planetary fortress/bio push. Also, keep in mind that higher-level Protoss players will expand as aggressively as you, since they know that you are committed to a very very slow push. If he had had that economy, his air switch might have worked out better for him (I'm assuming that he lost).
I understand that this is more of an idea you're trying out for fun, but if you do want to refine this style, these are some points you might want to consider.
EDIT: Looks like people already pointed all of this out while I was typing. Oh well.
|
On December 10 2011 10:01 Ver wrote:I've tried this exact idea about half a year ago before the blue flame nerf which obviously makes mech a lot worse; didn't work then either after people had seen it more than once or twice. Sad but, mech isn't used for a reason You will crush any ground force if you are slow, patient, and make enough planetaries. However, the immobility of mech is so easy to take advantage of with blink and especially nowadays when protoss actually realize the warp prism exists. Another problem is cost efficiency. I've lost in even base shattered games versus pure ground armies. Protoss armies are much more cost effective than terran armies in general and when you are spending thousands of minerals on planetaries, it's far worse. 3-3 Immortals basically kill planetaries for free if you have nothing there and when you're stretched thin from blink/wp harass, it's hard to cover everything even on the smallest maps. However, all of the above isn't that important. The real problem is carriers. If you let protoss get enough carriers (5-6+), you lose unless they are broke and you have many more bases. No way around it. Their air upgrades will finish faster, their cost efficiency is absurdly superior, and they will win almost all battles. In low carrier counts vikings can overwhelm them by sheer force but vikings suck once they get it enough; they also become paper planes if you cannot emp all of their templars. Marines beat carriers handily but they will automatically die to colossus/storm. Battlecruisers also can work but only if they don't get feedbacked, which is not something to bet on especially since protosses know that they can hide templars in wp's to dodge emps. If by some miracle you can hold a position that they can't break, you still just can't keep up with the mobility of carriers. Mass turrets can help hold one spot but you can't afford them in multiple areas. If you want to see this problem in action there's some replay of grubby vs sarens on shattered/lt from awhile back from some euro lan which lasted an hourish. Grubby played really poorly but it didn't matter.
How big of an air army were you getting when you were trying this?
I'm noticing I'm getting almost a 50/50 air/ground army split at some points of a game.
Edit: regarding drop defense, I think it's basically a case of "play better and watch the minimap more." You only need to split off a couple vikings to stop a WP that hasn't reached your base yet, and hellions typically deal with what a protoss player will chose to drop you with. One guy did a two archon drop, though.. which was pretty hairy. You have money to burn on turrets if you've managed to get to the planetary pushing part of the game, so most drops should not be getting through.
|
i believe the main problem is that there are nothing we can use other than PFs to zone out parts of the map and the damage of the seige tank is not being respected enough. I wish 4 seige tanks defending an expo with a narrow choke + wall can hold off a decent amount of army cost effective enough which was what BW was. You needed less tanks to get expos due to their dmg of 70. Therefore more expos means more tanks faster.
On December 10 2011 10:03 moocow2009 wrote:
It's an interesting idea, but there are a few things I think you have to consider if you want to continue using this strategy (especially in higher leagues).
First of all, what about drops? It's impractical to completely defend your side of the map with turrets, and the nature of a warp-prism means that the Protoss can bring in as many units as he wants. Since mech is so immobile, and you've invested so much into planetary fortresses, you can't really respond quickly (if at all). Even if you get planetaries at all your expansions, which severely hurts your economy, he can always go for your main/natural (which will always be orbitals) and kill whatever production buildings he can find. Even if you have some of your factories in the middle of the map, that's a big issue. Blink stalker harass is a similar issue -- until you manage to push almost all the way to his base, he can blink enough stalkers up into your main to kill everything almost instantly, or kill one of your mining bases (even if it's a planetary).
Also, keep in mind that at no point shown in those screenshots were you less than two bases ahead of your opponent. At that point, basically any strategy would have won you the game. I understand that your planetaries allowed you to secure that many bases, but I don't see why you couldn't have done the same thing with a planetary fortress/bio push. Also, keep in mind that higher-level Protoss players will expand as aggressively as you, since they know that you are committed to a very very slow push. If he had had that economy, his air switch might have worked out better for him (I'm assuming that he lost).
I understand that this is more of an idea you're trying out for fun, but if you do want to refine this style, these are some points you might want to consider.
EDIT: Looks like people already pointed all of this out while I was typing. Oh well.
Drops shouldn't be a problem cuz of missile turrets. Just like in BW if you allow one arbiter into your main he can recall his army there. Terrans in BW just figured out the best position to place turrets that cover the largest area with min amount of turrets.
|
On December 10 2011 10:13 Quotidian wrote:Show nested quote +On December 10 2011 10:01 Ver wrote:I've tried this exact idea about half a year ago before the blue flame nerf which obviously makes mech a lot worse; didn't work then either after people had seen it more than once or twice. Sad but, mech isn't used for a reason You will crush any ground force if you are slow, patient, and make enough planetaries. However, the immobility of mech is so easy to take advantage of with blink and especially nowadays when protoss actually realize the warp prism exists. Another problem is cost efficiency. I've lost in even base shattered games versus pure ground armies. Protoss armies are much more cost effective than terran armies in general and when you are spending thousands of minerals on planetaries, it's far worse. 3-3 Immortals basically kill planetaries for free if you have nothing there and when you're stretched thin from blink/wp harass, it's hard to cover everything even on the smallest maps. However, all of the above isn't that important. The real problem is carriers. If you let protoss get enough carriers (5-6+), you lose unless they are broke and you have many more bases. No way around it. Their air upgrades will finish faster, their cost efficiency is absurdly superior, and they will win almost all battles. In low carrier counts vikings can overwhelm them by sheer force but vikings suck once they get it enough; they also become paper planes if you cannot emp all of their templars. Marines beat carriers handily but they will automatically die to colossus/storm. Battlecruisers also can work but only if they don't get feedbacked, which is not something to bet on especially since protosses know that they can hide templars in wp's to dodge emps. If by some miracle you can hold a position that they can't break, you still just can't keep up with the mobility of carriers. Mass turrets can help hold one spot but you can't afford them in multiple areas. If you want to see this problem in action there's some replay of grubby vs sarens on shattered/lt from awhile back from some euro lan which lasted an hourish. Grubby played really poorly but it didn't matter. How big of an air army were you getting when you were trying this? I'm noticing I'm getting almost a 50/50 air/ground army split at some points of a game. Edit: regarding drop defense, I think it's basically a case of "play better and watch the minimap more." You only need to split off a couple vikings to stop a WP that hasn't reached your base yet, and hellions typically deal with what a protoss player will chose to drop you with. One guy did a two archon drop, though.. which was pretty hairy. You have money to burn on turrets if you've managed to get to the planetary pushing part of the game, so most drops should not be getting through.
My point is that that isn't really enough. By the time a warp prism is actually visible on your mini-map, it's too late to send vikings to defend it. You can stop them warping in endless reinforcements, but you can't do anything about the first wave, and vikings aren't very good against ground units. Turrets are helpful, but all it takes is one little chink in your defenses turret, and they're all useless. In fact, I'm pretty sure (I'd have to check to make sure) that as long as there's an area which is only covered by a single turret, he can still drop and get one warp-in before the prism dies. By the time you can afford to ring half the map with closely-placed turrets, you've already won, whatever strategy you're going for. You could try sensor towers + vikings, but since your sensor towers will be seeing his base, it's hard to tell which dot is a warp prism until, once again, it's too late for you to send vikings and prevent the drop.
Once he gets a drop down, as you said, hellions will deal with it if it's a zealot drop. If it's anything else (stalkers, archons, high templar, immortals) you're in a bit of trouble. I'm not saying that a drop will automatically kill you -- I'm saying that a properly executed drop could easily do a lot of damage even if you respond well, and if you're not already way ahead (like you were in this game), might make the difference between a win and a loss.
On December 10 2011 10:22 gfever wrote:i believe the main problem is that there are nothing we can use other than PFs to zone out parts of the map and the damage of the seige tank is not being respected enough. I wish 4 seige tanks defending an expo with a narrow choke + wall can hold off a decent amount of army cost effective enough which was what BW was. You needed less tanks to get expos due to their dmg of 70. Therefore more expos means more tanks faster. Show nested quote +On December 10 2011 10:03 moocow2009 wrote:
It's an interesting idea, but there are a few things I think you have to consider if you want to continue using this strategy (especially in higher leagues).
First of all, what about drops? It's impractical to completely defend your side of the map with turrets, and the nature of a warp-prism means that the Protoss can bring in as many units as he wants. Since mech is so immobile, and you've invested so much into planetary fortresses, you can't really respond quickly (if at all). Even if you get planetaries at all your expansions, which severely hurts your economy, he can always go for your main/natural (which will always be orbitals) and kill whatever production buildings he can find. Even if you have some of your factories in the middle of the map, that's a big issue. Blink stalker harass is a similar issue -- until you manage to push almost all the way to his base, he can blink enough stalkers up into your main to kill everything almost instantly, or kill one of your mining bases (even if it's a planetary).
Also, keep in mind that at no point shown in those screenshots were you less than two bases ahead of your opponent. At that point, basically any strategy would have won you the game. I understand that your planetaries allowed you to secure that many bases, but I don't see why you couldn't have done the same thing with a planetary fortress/bio push. Also, keep in mind that higher-level Protoss players will expand as aggressively as you, since they know that you are committed to a very very slow push. If he had had that economy, his air switch might have worked out better for him (I'm assuming that he lost).
I understand that this is more of an idea you're trying out for fun, but if you do want to refine this style, these are some points you might want to consider.
EDIT: Looks like people already pointed all of this out while I was typing. Oh well. Drops shouldn't be a problem cuz of missile turrets. Just like in BW if you allow one arbiter into your main he can recall his army there. Terrans in BW just figured out the best position to place turrets that cover the largest area with min amount of turrets.
As I pointed out, it's basically impossible to completely ring your side of the map with missile turrets, especially given how much HP warp prisms had. Placing turrets in only strategic locations only really works if your opponent isn't paying attention -- if they are, they can just go around. Remember that unit control is easier in SC2 than BW -- when your warp prism is under attack, it responds more quickly when you tell it to fly the other way.
I didn't really follow BW competitively, but it seems like that turrets wouldn't even really work there. Did people really do that? With Recall on a relatively fast unit (and one you can make multiple of), and corsairs with disruption web, it seems like a turret defense wouldn't work that well. Were corsairs not really in the meta-game (it seems like they'd be good against the mech people keep saying Terrans used)? Pardon my ignorance.
|
Might as well wait till the expo before giving mech a shot again in TvP. There are just a million reasons mech doesn't work right now in the matchup.. it's slow, can't deal with blink stalker harass, can't expand easily and can't deal with many P lategame stuff. Mech suffered splash damage from almost every patch as some mech units were too strong in TvZ.
Battlehellion, warhound and shredder is basically changing the entire mech game plan which might make it viable and change PvT completely. Would probably be good as PvT is probably the least interesting of the 3 non-mirrors because the options are too much set in stone.
|
On December 10 2011 10:49 moocow2009 wrote:Show nested quote +On December 10 2011 10:13 Quotidian wrote:On December 10 2011 10:01 Ver wrote:I've tried this exact idea about half a year ago before the blue flame nerf which obviously makes mech a lot worse; didn't work then either after people had seen it more than once or twice. Sad but, mech isn't used for a reason You will crush any ground force if you are slow, patient, and make enough planetaries. However, the immobility of mech is so easy to take advantage of with blink and especially nowadays when protoss actually realize the warp prism exists. Another problem is cost efficiency. I've lost in even base shattered games versus pure ground armies. Protoss armies are much more cost effective than terran armies in general and when you are spending thousands of minerals on planetaries, it's far worse. 3-3 Immortals basically kill planetaries for free if you have nothing there and when you're stretched thin from blink/wp harass, it's hard to cover everything even on the smallest maps. However, all of the above isn't that important. The real problem is carriers. If you let protoss get enough carriers (5-6+), you lose unless they are broke and you have many more bases. No way around it. Their air upgrades will finish faster, their cost efficiency is absurdly superior, and they will win almost all battles. In low carrier counts vikings can overwhelm them by sheer force but vikings suck once they get it enough; they also become paper planes if you cannot emp all of their templars. Marines beat carriers handily but they will automatically die to colossus/storm. Battlecruisers also can work but only if they don't get feedbacked, which is not something to bet on especially since protosses know that they can hide templars in wp's to dodge emps. If by some miracle you can hold a position that they can't break, you still just can't keep up with the mobility of carriers. Mass turrets can help hold one spot but you can't afford them in multiple areas. If you want to see this problem in action there's some replay of grubby vs sarens on shattered/lt from awhile back from some euro lan which lasted an hourish. Grubby played really poorly but it didn't matter. How big of an air army were you getting when you were trying this? I'm noticing I'm getting almost a 50/50 air/ground army split at some points of a game. Edit: regarding drop defense, I think it's basically a case of "play better and watch the minimap more." You only need to split off a couple vikings to stop a WP that hasn't reached your base yet, and hellions typically deal with what a protoss player will chose to drop you with. One guy did a two archon drop, though.. which was pretty hairy. You have money to burn on turrets if you've managed to get to the planetary pushing part of the game, so most drops should not be getting through. My point is that that isn't really enough. By the time a warp prism is actually visible on your mini-map, it's too late to send vikings to defend it.
It all depends on what stage of the game we're talking about, but you should have pretty good map coverage with sensor towers when you start PF creeping your way across the map
---
Since some people here were asking for a replay, I guess I can upload one. I don't want to put it in the OP because frankly both my opponent and I played like garbage. That said, I would probably have lost this game if I was playing mmm, since my army control is so bad
http://drop.sc/71465
|
Planetary's cost way to much and die to easy focus fire.
I had a similair idea last february, but instead of Planetarys, I used bunkers like this
![[image loading]](http://i911.photobucket.com/albums/ac313/XXXSmoke_2010/Screenshot2011-02-1719_17_04.jpg)
Bunkers were nice because they made a wall, they were free(back then salvage was 100%) they built fast, and with the armor upgrade they were still pretty good at taking some dmg.
With some good tank cover behind them the toss would take a few tank volleys before they could attack my tanks, which often made a big difference in battles.
It kind of worked, it threw a kink in the toss at least. Still no where near as good as bio which is unfortunate cuz bio has made me hate T.
Now a days, 75% salvage and BFH nerf hurt this alot as well.
|
wow u guys got quite a lot of shit out at only 19-20 minutes xD
Anyways i'm interested at the idea of building your factories closer, especially when they're that close. You can reinforce much faster but at the same time it can be riskier. But you don't need to put them at the front as long as you put them next to some PFs, it should be a lot more safe than building them at your main or such.
Also yes, definitely, air units compliment mech very well. A fleet of banshee viking raven can be very good at harassing and deal with harsass from protoss, so you don't have to make those food heavy hellions in the lategame and so you can have more tanks, and having instant splash damage (SMs) will help a lot; the fight becomes more in terran's favor exponentially every time protoss loses a unit before fully engaging due to splash and SMs etc at a long range.
aside from nuking, i wonder what's better to get, ghosts or air (given u get 3/3 on mech and 3/3 on air)
|
On December 10 2011 12:22 XXXSmOke wrote:Planetary's cost way to much and die to easy focus fire. I had a similair idea last february, but instead of Planetarys, I used bunkers like this Bunkers were nice because they made a wall, they were free(back then salvage was 100%) they built fast, and with the armor upgrade they were still pretty good at taking some dmg. With some good tank cover behind them the toss would take a few tank volleys before they could attack my tanks, which often made a big difference in battles. It kind of worked, it threw a kink in the toss at least. Still no where near as good as bio which is unfortunate cuz bio has made me hate T. Now a days, 75% salvage and BFH nerf hurt this alot as well.
The problem with bunkers vs PFs is that if you lose a big chunk of your army and all you have is bunkers, the wave of gateway reinforcements will push you all the way back to your natural pretty quickly. If you've been slow and defensive and creeped forward with way more PFs than you'd ever think is reasonable, the protoss army won't be able to push you back as quickly. And if you have a lot of turrets around your PFs, a protoss air army will take a lot of casualties as well.
But as Thorzain said on his stream yesterday, mech can work if you've already won the game... I guess that's the big problem.
On December 10 2011 12:26 Yoshi Kirishima wrote:
wow u guys got quite a lot of shit out at only 19-20 minutes xD
Anyways i'm interested at the idea of building your factories closer, especially when they're that close. You can reinforce much faster but at the same time it can be riskier. But you don't need to put them at the front as long as you put them next to some PFs, it should be a lot more safe than building them at your main or such.
Also yes, definitely, air units compliment mech very well. A fleet of banshee viking raven can be very good at harassing and deal with harsass from protoss, so you don't have to make those food heavy hellions in the lategame and so you can have more tanks, and having instant splash damage (SMs) will help a lot; the fight becomes more in terran's favor exponentially every time protoss loses a unit before fully engaging due to splash and SMs etc at a long range.
aside from nuking, i wonder what's better to get, ghosts or air (given u get 3/3 on mech and 3/3 on air)
I personally think that a good flock of ravens is easily just as strong as ghosts - maybe even stronger.. in the replay I posted, I was killing off colossi with seeker missiles. You can't do that with emp ;p
but yeah, if there are HTs out, ravens are basically pointless.
Although, I kind of wish that all the people who at one point were heavily experimenting with mech would give it a go again, just this time try and go for a lot of ravens at the end of the game. I think people will be surprised how good they are when you have a fortified position to fall back on. (again, given that there are few HTs on the map)
|
On December 10 2011 12:39 Quotidian wrote:Show nested quote +On December 10 2011 12:22 XXXSmOke wrote:Planetary's cost way to much and die to easy focus fire. I had a similair idea last february, but instead of Planetarys, I used bunkers like this Bunkers were nice because they made a wall, they were free(back then salvage was 100%) they built fast, and with the armor upgrade they were still pretty good at taking some dmg. With some good tank cover behind them the toss would take a few tank volleys before they could attack my tanks, which often made a big difference in battles. It kind of worked, it threw a kink in the toss at least. Still no where near as good as bio which is unfortunate cuz bio has made me hate T. Now a days, 75% salvage and BFH nerf hurt this alot as well. The problem with bunkers vs PFs is that if you lose a big chunk of your army and all you have is bunkers, the wave of gateway reinforcements will push you all the way back to your natural pretty quickly. If you've been slow and defensive and creeped forward with way more PFs than you'd ever think is reasonable, the protoss army won't be able to push you back as quickly. And if you have a lot of turrets around your PFs, a protoss air army will take a lot of casualties as well. But as Thorzain said on his stream yesterday, mech can work if you've already won the game... I guess that's the big problem. Show nested quote +On December 10 2011 12:26 Yoshi Kirishima wrote:
wow u guys got quite a lot of shit out at only 19-20 minutes xD
Anyways i'm interested at the idea of building your factories closer, especially when they're that close. You can reinforce much faster but at the same time it can be riskier. But you don't need to put them at the front as long as you put them next to some PFs, it should be a lot more safe than building them at your main or such.
Also yes, definitely, air units compliment mech very well. A fleet of banshee viking raven can be very good at harassing and deal with harsass from protoss, so you don't have to make those food heavy hellions in the lategame and so you can have more tanks, and having instant splash damage (SMs) will help a lot; the fight becomes more in terran's favor exponentially every time protoss loses a unit before fully engaging due to splash and SMs etc at a long range.
aside from nuking, i wonder what's better to get, ghosts or air (given u get 3/3 on mech and 3/3 on air) I personally think that a good flock of ravens is easily just as strong as ghosts - maybe even stronger.. in the replay I posted, I was killing off colossi with seeker missiles. You can't do that with emp ;p but yeah, if there are HTs out, ravens are basically pointless. Although, I kind of wish that all the people who at one point were heavily experimenting with mech would give it a go again, just this time try and go for a lot of ravens at the end of the game. I think people will be surprised how good they are when you have a fortified position to fall back on. (again, given that there are few HTs on the map)
but then you can get ghosts out as well, and the way he deals against that is to use colossi or zealots, and you already have ways to deal with those
interesting idea about the bunkers, too bad they're not 100% anymore, being able to salvage them and move them around sounds fun, though you can always do that with barracks which can also float and give you vision and even let you make some marines on the fly xD
Holy fuck barracks are 1000 hp bunkers are only 400 lol o.o
barracks ftw!
|
Every Terran player that used to mech TvP has stopped doing it because it doesn't work anymore. The hellion nerf was the nail in the coffin for mech. And even then it didn't work well at all versus good Protosses.
There was a time when mech was completely viable and could cost effectively compete with Protoss even into the lategame, and that time lasted up until the siege tank damage nerf.
Everything after that in terms of Terrans making mech work was us trying to find brute force solutions and things to make it work, and whenever it showed signs of becoming viable, blizzard nerfed it into the ground, with the hellion nerf being the most recent nail in the coffin.
The most potential right now for TvP transitions deals with vikings, banshees, and battlecrusers with ghosts. Mech is more of a dead end than bio is right now
|
can you post some tvp mech replays?
|
where's the replay for the game in the op?
|
On December 10 2011 12:57 avilo wrote:
Every Terran player that used to mech TvP has stopped doing it because it doesn't work anymore. The hellion nerf was the nail in the coffin for mech. And even then it didn't work well at all versus good Protosses.
Yeah, I know and I agree, but it also kind of isn't the point of the thread. Just the fact that you have to get an obscene amount of planetary fortresses to even hold a position with mech proves that mech is essentially futile. So why not try and come up with some things that can make the usage of the planetaries themselves more effective? Your perspective on the game is fundamentally different from mine though. I'm not playing in online tournaments and I have no ambitions or even desire to get to masters.. so I think for a player like me there's room to mess around with really futile builds. But I don't think this is every going to be my go-to tvp build either.
There was a time when mech was completely viable and could cost effectively compete with Protoss even into the lategame, and that time lasted up until the siege tank damage nerf.
yup :\
Everything after that in terms of Terrans making mech work was us trying to find brute force solutions and things to make it work, and whenever it showed signs of becoming viable, blizzard nerfed it into the ground, with the hellion nerf being the most recent nail in the coffin.
I've started to think that even with this kind of silly attempt at mech, that getting hellions to buffer really isn't a good idea. And I don't think it was that great back when hellions did full damage either. Hellions got chewed up by chargelots when they were out of room to micro, and chargelots don't bunch up like zerglings do either, so it's basically futile.
But I'm starting to think that if you never really attack with the tanks you make (and instead make fewer tanks and more air units) that you don't really need hellions anyway. Maybe have like 12-16 tanks and the rest of your army is air units. All the minerals goes to more CCs and more bases.
The most potential right now for TvP transitions deals with vikings, banshees, and battlecrusers with ghosts. Mech is more of a dead end than bio is right now
Yeah, I saw on your stream that you won some games with mass BCs.. it was pretty cool, but to me it almost seems like it's harder to get a ton of BCs than it is to spam Fortresses everywhere. I do appreciate that you have a very unorthodox style though. Tried ravens?
On December 10 2011 13:11 lightsentry wrote:
where's the replay for the game in the op?
the OP is like 2-3 different games actually.. but they're all bad and I don't want the discussion to be derailed by how bad they are. :o Also, this isn't a [H]elp thread or anything like that... but I did post a replay from a different game further down the thread.
|
|
|
|
|
|
EPT
