|
On July 23 2012 17:10 Nightmarjoo wrote:Since nobody seemed interested in my stream vods, here are some recent reps. twitch.tv/lyrathegreat has my regular stream and vods if anyone changes their mind ~~ http://drop.sc/226582http://drop.sc/226589http://drop.sc/226590http://drop.sc/226594http://drop.sc/226595http://drop.sc/226583 This game is vs NSH Sage, in playhem daily tourney. I do lose it, but I think it should be fairly evident that I at least could have kept the game going on longer, if not won, if I had scouted his gate count (and thus his aggression/composition intent) early enough to construct a PF in front of the ramp in front of my nat. Without that PF I lost to his double aggression at my nat and 4th being unable to efficiently split up my army. I'd say this replay shows that the style can hold its own at "high level". Certainly I couldn't have lasted 28 minutes vs Sage with bio. Here are some scrappier games-- which mostly means I made some big blunders. http://drop.sc/226587http://drop.sc/226588http://drop.sc/226591So, I've picked up a few new ideas from practicing this. In the past I'd followed the original Lyyna upgrades of just mech armour before air ups. Then I was trying out double mech ups. Then recently I was getting armour then just attack. Now I think it's better to get +3 asap. I need to practice more vs my teammate, but he found a nasty 4base immortal/archon attack that hit right before or during initial bc production that was destroying me. When trying out a myriad of variations, minor and extreme, the scenario where I did the best vs it (excluding scenarios where I just had phenomenal unit positioning, which usually isn't possible since p isn't forced to engage you at any single location) was one where I had +3 instead of 2-1. Furthermore a higher ghost count than 1rax can produce is probably necessary. I tended to get a relatively late 2nd rax (and even later 3rd), but I think a 2nd rax on 3base is probably necessary. Though you can't constantly produce 2 tanks, 1 banshee, ups and 2 ghosts (you can do all of that minus the 2nd ghost) on 3base, you still get a few extra ghosts out of it, and 4base supports it fully. In addition to these changes stopping such an attack probably requires scouting that composition in advanced alerting you to make extra banshees and tanks instead of starting bc production. The Sage game taught me the importance of scouting gate count (or main in general for 3rd base transition). A pf that defends at least your nat is probably necessary vs high gate count aggression. I dunno the timing yet, but I'm guessing it's either a cc made on 3base simultaneous with extra oc production, or it's the cc made right after 4th base cc (with no extra ocs yet). It's hard to defend multi-pronged aggression without it. An annoying scenario I encountered was one where protoss made almost no gate units, using phoenix to fend off harass, making a lot of immortals into fast (3base I think, 4 at max) carriers. Since he wasn't wasting money on gate units his carriers were crazy fast and when he attacked with a mothership I didn't have the bc count to hold. I'm pretty sure the solution is identifying the strategy/composition early on and hitting with a thor/ghost/hellion timing (0 tank), with 2 reactor facts making hellion and 250mm cannon. I dunno for sure, but I'm guessing this is a 2base timing attack (not that you can't make a 3rd cc, but it's before that's fully up and running I think). I've been downplaying ravens a lot in recent games. I tend to only make extra ravens when I absolutely can't afford additional bcs (and it's late enough in game that I need some support for my existing bcs, or I just lost a battle/barely won and am prioritizing money on ghost/thor). Still not sure about port timing. I've been getting it + a raven pretty fast, but I don't know if that's ideal or if there's something better to make instead (don't know what either; suspect a faster 2nd fact is a bad idea given how mineral limited you are opening the way I am (with gas)). I'm also not getting cloak asap anymore (as I've seen lyyna do also in his most recent games). I don't know when the best timing for it is (or if it's even useful anymore) yet. So beyond questions of scouting and reacting what I'm working on is finding ways to harass more. The maps are all pretty bad for harassing these days (probably the reason for the relative recent terran decline), and I'm finding that most protoss are pretty well-prepared for basic hellion/banshee harass. While I don't need to do any damage, not being able to even make attempts at it is greatly limiting my scouting (which is significant: needing to blow scans on the off-change of getting useful and accurate information is certainly unideal). Gonna try to find out when the most appropriate time for a medic to add drops to that mix is. Don't know if I should try to autoturret harass early on either, or if preserving pdd energy is too important. I suspect the latter is true though. As always, appreciate any feedback on my reps. Please bring any patterns of mistakes or areas of improvement to my attention that you find.
Night,
Good post.
You said, "An annoying scenario I encountered was one where protoss made almost no gate units, using phoenix to fend off harass, making a lot of immortals into fast (3base I think, 4 at max) carriers. Since he wasn't wasting money on gate units his carriers were crazy fast and when he attacked with a mothership I didn't have the bc count to hold"
I've played against this before as high masters. It is a hard counter to mech which few know. My mistake is making too few starports. If you see this, my commentary to the community is have a minimun of 3 reactor starports and start all air upgrades ASAP.
I had to suicide tanks to try and kill his 3rd and 4th but it was too late. Make sure you scan and identify this build before it gets out of control. It was my fault for not scouting his army composition, but it is diffucult focusing on composition with this very "tanky" build. I'm only TvP mech (regardelss of map) but my builds certainly are dictated by the map, so it is best to try and scan in key places (too bad no cloaked obs or super speed ols) 
|
Do you disagree with the viability of the theroetical solution I posited for that scenario?
|
Long-time reader and player, first-time poster (unless you count my practice partner post), so please bear with me if the format or point-of-discussion is not up to snuff.
I have been closely following this thread and FilterSC's Bronze to Masters threads and videos, watching dozens of replays and tutorials while I try to get my head around the nuances of Terran macro, and have a few questions, not about the build itself, but its difficulty in execution.
The reason this build in particular appeals to me, is that it's so UNLIKE the heavy bio play that FilterSC teaches in order to help players grasp macro fundamentals. To lower league players especially, does the macro (with all the macro-OCs, SCV exchanges, and expansion tactics, seem at all counter-intutive or clumsy? As a lower league player, I find it difficult to benchmark the quality of a player's economy when performing these kinds of builds compared to a much more standard format of play, and would be curious as to whether there were opinions on the economic backbone of this kind of play in general.
In addition, is there any reason to not use the lessons and tactics brought to bear in this thread in every matchup? TvT it seems that the player with the right mixture of mech/bio/air and map control with a decent economy behind it will always come out ahead. TvZ anymore seems to be driven more and more to the late game (disregarding all-ins and busts) as people are hungry for Infestor/Broodlord heavy armies which I would loooove to Mech the hell out of. (Neural parasite still kinda scares me...but I'll get over it)
I suppose my reason for posting, is that again, I really like the way this build looks, and thematically the only thing better than marines are giant robots, so if the wise ones active in this thread could lend some insight into whether I would be taking a step forward or backward by training on this kind of play while also training on macro/bio it would be greatly appreciated.
Again, sorry if maybe this was the wrong place to post this. Please don't ban me, lol.
|
I'm not familiar with what he advises, but I'm inclined to suggest improving your fundamentals with bio before trying to learn mech. Bionic play is very mechanical in nature (player-mechanics) With bio it almost doesn't matter if you're doing the right or wrong thing as long as you're doing it fast. Your units are powerful but flimsy, requiring good attention and control to get the most out of them. Succeeding in gaining any level of mastery over bio will inevitably help your core mechanics as a whole. And you can get by in bio just by focusing on having efficient macro and active units.
In sharp contrast mech is much more decision-based than bio is. In tvz mech you'll have 3 techlab facts for a large percent of the game. The difference between making three thors and two thors and one tank with a production cycle could be the difference between losing horribly and winning decisively. When you complete a new CC you have to decide whether you can safely expand with it, giving you a new place to drop mules with the added requirement of more space to defend (more opportunities for your opponent to damage you), make an extra OC out of it for scanning and mule-dropping, or making a defensive PF out of it helping secure what you already have. Expanding when you can't hold that expansion is a huge waste of money, making an extra OC when you have no place to mule is a huge waste of money, and making an extra PF when you're running out of income is going to lose the game. Efficient macro isn't inherently good here if you aren't making the right thing. And often the bulk of your units aren't active, which makes getting the most out of the units which can be active far more imperative.
However, just getting to that point in the game where you have to make a decision, a choice based on your experience and the information you have specific to the game's scenario, requires good mechanics. You can't focus on your mechanics playing mech if you're losing by making poor choices, but you can't survive to learn how to make those decisions without decent mechanics. My standard tvz opening is cc first or 1rax cc into hellion/banshee. I use the hellion/banshee to scout my opponent, limit their creep spread (which effectively gives me time to react to anything), slow their economic and military growth by forcing them to use larvae replacing drones and lings instead of making new things to hurt me with, and to give me a safety net of force and distraction behind which I can secure my 3rd base allowing me to produce what I need to stay in the game. If you don't have the micro, macro, and multi-task to do all those tasks simultaneously, you're just going to lose to a single roach attack. And while you could try to develop those skills playing mech, you're better off doing so with bio imo. Being forced to multi-task like that with repetition and time will improve your multi-task, but only if you get there alive by knowing to make your bunker next to the cc instead of in part of your wall because you scouted his fast gas making a mass ling runby a possibility, and knowing to scan again at 6:30 to rule out the possibility of a roach all-in which would require you to make marauders with the rax-techlab and make the port-techlab with the port itself instead of floating to the rax techlab, etc etc.
I have no idea how viable it is to try to climb the ladder with mech. I started sc2 with the skill and experience playing bw competetively for 6 years yields and have been the equivalent of high masters since the beginning of beta. I played very nearly exclusively bio in sc2 until about last spring (except tvt, which I started meching in after mlg anaheim last year or so). Switching to mech was relatively easy for me once I came to understand the elementary decision making behind each matchup. I can't possibly say how easy or difficult that switch can be for someone at all relatively lacking in experience and skill.
Also with bio you have the huge advantage of being able to copy and mimic what progamers do. When you aren't sure how to handle a situation you can find replay or vod showing you the solution, or giving you a different build altogether to try out to avoid the scenario entirely, etc. With mech you mostly have to make up the answers for yourself since it isn't popular. Thus what games there are out there (and at whatever level) additionally include a great number of idiosyncrasies that you can't always copy for success. So if you choose mech, know that you do so at your own win% peril.
As for benchmarking economy, just compare income to necessary spending. I have games where my economy is just in shambles because I had to sack SCVs early to defend an attack, or to free up supply to adjust my composition safely to be able to stay in-game, etc, while having 10 OCs. You think, 10 orbitals? That's pretty good! But 10 OCs isn't good if I actually need 13 to support the constant production of everything I need, and thus I have no money and everything I choose to make is a potential game-ending hazard if I budgetted for the wrong thing (chose a BC when I needed a thor, etc).
Alllllllllll that being said, I think mech is wonderful style in each matchup, and that every terran should learn what they can about it to increase terran's strategic diversity as a whole across the world. If it turns out you can improve while learning mech, great! But if you choose this route, you may hit barriers forcing you to pause playing mech to improve your skill with bio.
|
On July 31 2012 03:36 Nightmarjoo wrote:
I'm not familiar with what he advises, but I'm inclined to suggest improving your fundamentals with bio before trying to learn mech. Bionic play is very mechanical in nature (player-mechanics) With bio it almost doesn't matter if you're doing the right or wrong thing as long as you're doing it fast. Your units are powerful but flimsy, requiring good attention and control to get the most out of them. Succeeding in gaining any level of mastery over bio will inevitably help your core mechanics as a whole. And you can get by in bio just by focusing on having efficient macro and active units.
In sharp contrast mech is much more decision-based than bio is. In tvz mech you'll have 3 techlab facts for a large percent of the game. The difference between making three thors and two thors and one tank with a production cycle could be the difference between losing horribly and winning decisively. When you complete a new CC you have to decide whether you can safely expand with it, giving you a new place to drop mules with the added requirement of more space to defend (more opportunities for your opponent to damage you), make an extra OC out of it for scanning and mule-dropping, or making a defensive PF out of it helping secure what you already have. Expanding when you can't hold that expansion is a huge waste of money, making an extra OC when you have no place to mule is a huge waste of money, and making an extra PF when you're running out of income is going to lose the game. Efficient macro isn't inherently good here if you aren't making the right thing. And often the bulk of your units aren't active, which makes getting the most out of the units which can be active far more imperative.
However, just getting to that point in the game where you have to make a decision, a choice based on your experience and the information you have specific to the game's scenario, requires good mechanics. You can't focus on your mechanics playing mech if you're losing by making poor choices, but you can't survive to learn how to make those decisions without decent mechanics. My standard tvz opening is cc first or 1rax cc into hellion/banshee. I use the hellion/banshee to scout my opponent, limit their creep spread (which effectively gives me time to react to anything), slow their economic and military growth by forcing them to use larvae replacing drones and lings instead of making new things to hurt me with, and to give me a safety net of force and distraction behind which I can secure my 3rd base allowing me to produce what I need to stay in the game. If you don't have the micro, macro, and multi-task to do all those tasks simultaneously, you're just going to lose to a single roach attack. And while you could try to develop those skills playing mech, you're better off doing so with bio imo. Being forced to multi-task like that with repetition and time will improve your multi-task, but only if you get there alive by knowing to make your bunker next to the cc instead of in part of your wall because you scouted his fast gas making a mass ling runby a possibility, and knowing to scan again at 6:30 to rule out the possibility of a roach all-in which would require you to make marauders with the rax-techlab and make the port-techlab with the port itself instead of floating to the rax techlab, etc etc.
I have no idea how viable it is to try to climb the ladder with mech. I started sc2 with the skill and experience playing bw competetively for 6 years yields and have been the equivalent of high masters since the beginning of beta. I played very nearly exclusively bio in sc2 until about last spring (except tvt, which I started meching in after mlg anaheim last year or so). Switching to mech was relatively easy for me once I came to understand the elementary decision making behind each matchup. I can't possibly say how easy or difficult that switch can be for someone at all relatively lacking in experience and skill.
Also with bio you have the huge advantage of being able to copy and mimic what progamers do. When you aren't sure how to handle a situation you can find replay or vod showing you the solution, or giving you a different build altogether to try out to avoid the scenario entirely, etc. With mech you mostly have to make up the answers for yourself since it isn't popular. Thus what games there are out there (and at whatever level) additionally include a great number of idiosyncrasies that you can't always copy for success. So if you choose mech, know that you do so at your own win% peril.
As for benchmarking economy, just compare income to necessary spending. I have games where my economy is just in shambles because I had to sack SCVs early to defend an attack, or to free up supply to adjust my composition safely to be able to stay in-game, etc, while having 10 OCs. You think, 10 orbitals? That's pretty good! But 10 OCs isn't good if I actually need 13 to support the constant production of everything I need, and thus I have no money and everything I choose to make is a potential game-ending hazard if I budgetted for the wrong thing (chose a BC when I needed a thor, etc).
Alllllllllll that being said, I think mech is wonderful style in each matchup, and that every terran should learn what they can about it to increase terran's strategic diversity as a whole across the world. If it turns out you can improve while learning mech, great! But if you choose this route, you may hit barriers forcing you to pause playing mech to improve your skill with bio.
Holy awesome answer!
I honestly have no other questions at this time... That's a first. I'm sure I can get my mech fix just supporting my bio play in each matchup without committing to the insanity that is this build.
I'm totally stealing the sensor towers bit though. Played a couple games where I had them pretty much covering any area I didn't have vision and absolutely hamstrung my opponent.
I'll never be blind again!
Thanks,
~Lelex
|
Little update about myself : i'm going to stream a lot for the next two weeks, i found a good way to get a good stream with my low upload (http://www.teamliquid.net/forum/viewmessage.php?topic_id=272764, awesome guys). First one in ~~30 min after eating. I suggest you to follow me or just bookmark the page on facebook (link in the first post) to see when are next ones (i just post this on here, as lots of people asked me to start doing stuff for the guide :D)
stream up at http://fr.twitch.tv/Lyyna
|
please add your stream to your profile so i can add you to my TL favorites so i always see when you're online 
|
I finally just tried out this mech turtle style TvP following alot of Lynna's guidelines. I beat a 1000 pt master Protoss on Antiga in a 31 minute game. I see several ways I can improve on my play, so this seems kind of exciting knowing I can be so sloppy and still have such a strong lategame composition. I will definitely give this some more attempts. Keep up the good work Lynna.
|
On August 04 2012 00:19 Tppz! wrote:please add your stream to your profile so i can add you to my TL favorites so i always see when you're online
Done =)
and im streaming right now !
|
United Kingdom12022 Posts
I'm just going to say that I went mech TvP for the first time properly today and it ended as a tie game. He had tons of storming high templars, stalkers and archons but I ended up stuck with only my main left. he couldn't push up my ramp, the entire map was mined out and neither of us could build anything.
He tried to attack into me but failed and since the guy seemed so nice rather than go towards his base with what I had left I decided to force the draw. I must say until I walled half my tanks in my base by accident (causing me to lose the main battle) Mech was working exceedingly well for me. Thanks for the great advice/info in this thread! :D
|
Lyyna could I ask if you are able to release a replay pack of your recent mech games for the mech resource thread? You're a pretty high level player so it would be really useful :D
|
On August 08 2012 06:00 kollin wrote:
Lyyna could I ask if you are able to release a replay pack of your recent mech games for the mech resource thread? You're a pretty high level player so it would be really useful :D
Yes please.
|
...http://www.mediafire.com/?wwymibwgw3g8tq0 :D
New replays!
|
United Kingdom12022 Posts
Thanks Lyyna (I was StayDead in your stream chat before).
I'll give those replays a watch now. ^^
|
Oooooh replay pack? Might watch this when I get home.
|
Thanks Lyyna! I've updated the op now :D
|
great replays, helped me alot! thx lyyna for all the good work!
|
Trying portless midgame atm. So far so good.
I stopped getting cloak early because it was expensive and wasn't guaranteeing to make the banshees be helpful. I found the banshees increasingly less useful and just made them cuz they're part of the original lyyna build. The raven isn't inherently worth the investment of gas of port and itself. Sniping obs isn't important if I'm not relying on cloaked banshees. The pdd is only useful if they engage me with stalkers, which is relatively rare in midgame. Usually I encounter zlot/immortal/ht instead.
Not having a port lets me afford thors and upgrades initially more smoothly, then makes it easier to produce all the stuff I need as I take my 3rd (ghost acad, ghosts, 2fact, reactor and techlab, tank, siege, sensor towers, pf, +2, etc). Ebay and turret(s) provide detection perfectly well, and can be useful rest of game vs dt. Having ebay already makes it easy to pf 3rd asap.
I think portless might have the potential to be more aggressive as well.
Examples:
http://drop.sc/238598
http://drop.sc/238597
http://drop.sc/238596
|
On August 14 2012 15:30 Nightmarjoo wrote:
Trying portless midgame atm. So far so good.
I stopped getting cloak early because it was expensive and wasn't guaranteeing to make the banshees be helpful. I found the banshees increasingly less useful and just made them cuz they're part of the original lyyna build. The raven isn't inherently worth the investment of gas of port and itself. Sniping obs isn't important if I'm not relying on cloaked banshees. The pdd is only useful if they engage me with stalkers, which is relatively rare in midgame. Usually I encounter zlot/immortal/ht instead.
Not having a port lets me afford thors and upgrades initially more smoothly, then makes it easier to produce all the stuff I need as I take my 3rd (ghost acad, ghosts, 2fact, reactor and techlab, tank, siege, sensor towers, pf, +2, etc). Ebay and turret(s) provide detection perfectly well, and can be useful rest of game vs dt. Having ebay already makes it easy to pf 3rd asap.
I think portless might have the potential to be more aggressive as well.
Just a thing that springs to mind. You have to be more warry of the Protoss switching to air aka scout for it regularly. If you have to build a Starport still, when his air is already on it's way to your base, you're lost.
But nevertheless I like the portless midgame style.
|
Hmm I don't think that's too scary. If he's super fast carrier I should notice he has shit ups and no units and can just push before they're out. If they're not too fast my ghost/thor should handle whatever comes out until bcs are out.
|
|
|
|
|
|
EPT
