Razzia of the Blizzsters - Page 18

Warning for everyone in this thread: I WILL moderate your posts very harshly from now on if you can't have a civil discussion.

Elldar

Sweden287 Posts

April 29 2015 15:21 GMT

#341

I mean 1-3 seem to me like blatant statements rather than core conceptual theory as the author perceive it to be. If it were the case that 12 worker in the beginning and mineral changes "conctracts time"* in such a way that took away strategy from play and error were made all the time by the players, then you would see more games being very flipp-floppy where the momentum is swining like a pendulum from one side to another til the game ends/pendulum stops. Even so, a wild pendulum is funnier to watch and even play, than a pendulum that basically never moves when a mistake or build order advantage is made, llike in hots or wol.

*I would argue that it is removal of time since none of the core values like mining rate or mining time have been increased/decreased, they just cut off the early build up period, the game isn't moving faster economically you just cut of the slow period.

I applaud the Stardust accretion/last chapter it was right on.

BronzeKnee

United States5211 Posts

April 29 2015 17:06 GMT

#342

On April 29 2015 11:36 RenSC2 wrote:
Sometimes I think a whole lot of people in this thread and others think that "Blizzard ideas = Bad, community ideas = Good" no matter what their actual merits are.

You are arguing against that, but that isn't what I argued. Most of the community ideas are universally bad, most of us know that. I don't want a community made game, and I think I made that clear when I talked about my experience with a community made hero for Coming of the Horde which was nearly a disaster (my post page 16). But some community ideas are so good they are ground breaking for a RTS game.

The ability to weed the bad ideas from the good is the difficult, but it is what separates great design teams (and games) from mediocre and bad ones, regardless of where those ideas come from (ie externally from the community or internally from the design team).

When people give Blizzard a great solution to a problem, but Blizzard ignores it for months while they try their own bad solutions, then decide to implement said great solution 6+ months later, it is a sign that their design team cannot weed out bad ideas from good ones, again regardless of the source.

That is the root of the problem with SC2.

To highlight this point, I bring us back to the Warhound again (and I have many other examples):

To me, creating an idea is a process where you think of something, decide on how it fits into the game, decide what role it plays, work out the interaction between it and other parts of the game. It requires a massive amount of thought and possibly some diagramming. When I add something to COTH, I don't just throw it in there willy nilly.

And that is why Blizzard's ideas are terrible. They aren't well thought out. The Warhound never should have escaped a designers head, never should have made it to implementation. But it did, Blizzard spent a lot of time and money on that. And nothing to show for it.

Someone should have been fired. Probably multiple people, that is no way to run a business. Modern software design ensures that everything is diagrammed, written up, and well thought out before you begin actually coding.

The Warhound should have been weeded out before it was implemented, it was a terrible idea. And the fact it was not is very telling.

The only real solution here is the one I've been calling for since the release of HOTS, the design team needs to go. If it doesn't, we'll keep seeing bad ideas being put into the game.

TheDwf

France19747 Posts

April 29 2015 23:15 GMT

#343

A bit at random, I took the following high level TvZ bo7 from various periods of the game:

(1) MarineKing vs Nestea, GSL Open Season 2, November 2010;
(2) Mvp vs Nestea, Blizzcon [3 series united], October 2011;
(3) MarineKing vs DRG, MLG Spring Arena, April 2012;
(4) Mvp vs Life, Code S finals, October 2012;
(5) Bogus vs Soulkey, first HotS Code S finals, June 2013.

And I compared the average supply count of the Terran player at various points:

(Small mistake on the graphs, the Blizzcon 2011 is noted November instead of October.)

Despite the low sample size, we clearly visualize the evolution of openings and the movement towards “modern macro play”.

Note: even if “5 minutes” stands as our “point zero” here, don't think there is no hyper-development within the first 5 minutes of the game. Though dormant, the seed is fully here. It just so happens that the 2 rax of 2010 TvZ has been replaced with CC first into 1-1-1 or Reaper 3OC, which by force materialize the effect later. The consequences are best seen here:

And yet, on average, Zerg max less fast in Mvp vs Life (~18:10) than in MarineKing vs DRG (~14:15). What happened? Is time being de-contracted? Is hyper-development finally losing ground? Alas no: the phenomenon has become partially hidden. We are indeed only looking at raw supply, but we have no information about the quality of said supply. After the Queen patch, hyper-development actually became so violent that instead of maxing in 14 minutes on lings/banes/mutas—which Terran could stand—Zerg was now maxing in 15'45 on lings + broods/infests. (The drone barbecues coming from Mvp's Blue Flame openings did the rest to bend the curve.) The length of the curve increases, but its steepness too; it simply has to be converted into something else because of the protective barrier of the supply limit—whose raising would be an absolute disaster.

Past certain thresholds, investments in economy and technology become more interesting. (And actually forced!) This is why supply inflation in LotV is in fact an euphemism regarding the contraction of time between HotS and the current version of LotV. The extent of the phenomenon is best seen when comparing similar scenarii:

(1) Proxy rax attack vs 1b Roaches defence

If we compare what SuperNova owns at 5 SC2 minutes (~3'37 in LotV) with MarineKing's macro 2 rax on Steppes of War (lol), we see that SuperNova has much more despite (a) proxying further and (b) proxying harder (+1 rax). SuperNova's contain is stronger (4 bunkers to 2), yet he still starts his expand/transition (CC + dual gas) earlier. Thanks to the current LotV economy, at 5 minutes SuperNova has already 35% more resource value on the map than MarineKing. + Show Spoiler +

Values are roughly similar for Zergs. Quickly drawing dead, the poor Zergbong could not even do much with a pool/gas opening, while his spawn could afford to lose a hatch and still have much more than the Cerebrate.

It should also be noted that since proto-WoL, the basal income rate was slightly increased (auto-mining at start + AI worker fix).

(2) 2b mutas into Roaches against mech

Comparing SuperNova vs bly (set 2) and GuMiho vs Solar. bly's build is less economical (3Q speed + faster Lair vs gasless 4Q); his macro is of course considerably inferior to Solar's one; he loses more in harassment, and still:

LotV timings were converted in SC2 minutes.

When Solar reaches 180 supply, on 63 drones, he has not yet completed +1 armor, he has no transition and only 3 bases. When bly reaches 180 supply (one minute later), on 59 drones, he has way more tech (Hive and 2/2 on the way, Hydra den, Burrow…) + 5 bases ready. Solar can only dream of such a transition.

Thus, on LotV, players would max faster, on higher technology, if they were “allowed” to do so. But players cannot “glide” over the development curve all the way through. Interrupting the cycles, some forces say “no”. Which are? The very game and the opponent. Look indeed at the number of times both players reached a high threshold of development in Bogus vs Soulkey:

What happened? Did we go back to the youth of WoL? Yes, in a way. Players can indeed orient the missile of hyper-development towards three kinds of entrances, according to the ETA triad (Economy; Technology; Army). To break your opponent's flow, you can invest everything in army; but since your opponent is riding the “ET” wave, you better succeed now or you'll lose the ensuing battle of curves. Enter: all-ins.

Case study (link)

Hyper-development in action.

In approximatively one minute, an extremely deadly “timing attack” is prepared and launched. The steady flow from Bogus' infrastructure kicks in too late: Bogus is wiped.

But there's more. Here, the eruption is still partially minimized, because (1) Queens don't belong to the attack and (2) Soulkey also turned all his Zerglings into Banelings. Army value thus represents slightly more accurately the productive surge:

Contraction of time at its finest.

In the span of a single minute (actually, 30 seconds for most of the herd), Soulkey's offensive army value is multiplied by 8.65. The game is riddled with surges like that (generally on a lesser scale, of course), turning the arena into a powder keg. And people wonder why SC2 is an untameable creature!?!

+ Show Spoiler +

Let us be clear here: the larva mechanic produces by force such spikes. What is problematic is not their existence, but their extent. A peaceful 2b mutas would also produce a spike… but (1) Spire takes time; (2) mutas are expensive, which limits the brutal increase in supply; (3) they can't bulldoze your base like a Roach/Baneling bust!

I took Zerg as an example, but the same principle applies for Warpgate timings (example for a 2b all-in: here and there). Terran's old SC1 tractor is more wheezy, but the locomotive ends up cranking out a lot of stuff too with MULEs and Reactors… except it obeys to the basic principle of having to walk across the map!

Last but not least, the impact on units is immense. The mould comes before the model…

SC2 is a RTS suffering from autoimmune disease. Excessive contraction of time made the “RT” part grow cancerous, and it immediately started viciously attacking the “S” part, which was forced to retract under the assault:

On April 19 2015 22:46 TheDwf wrote:
SC2 games, because of hyper-development, tend to degenerate towards (1) all-ins; (2) passive macro; (3) worker bowling.

If you have ever wondered why the frontier between timings and all-ins is sometimes so blurry in SC2 (particularly with the following characteristics), look no further: the culprit is, again, contraction of time. See Welcome to ZParcraft II for a few concrete applications in the context of the last major crisis of asymetric hyper-development.

Tons of pressure builds also died (or can't see the light) because hyper-development automatically made them become all-in… while of course they lack the strength of an all-in. Those builds thus face the ruthless following alternative: disparition or radicalization. Pressure disappears? The cursor moves towards passive macro. Pressure becomes more radical? Towards all-in. Spaghettification of strategy. + Show Spoiler +

In practice, some of those builds survive; they simply remain coinflippy/gimmicky, i. e. you just pray that your opponent is going the wrong route or blunders heavily, etc. This phenomenon is particularly visible in periods of massive imbalance, which is why players then feel that “nothing is solid”.

Hyper-development is also what makes some people say that “players have become so greedy”. Though the formulation is simplistic (as with most of the common vulgate), the idea behind it is sound: it just so happens that what was once “greed” is now standard. See for instance “One Core to Rule Them All” in Welcome to ZParcraft II.

You have surely noticed the worrying appearance of invincibility in LotV. Invincible Roach/Ravager Warpgate, invincible Disruptors. Why? Because otherwise, hyper-development defends too easily. The torrent of resources allows the defender to produce more units and defensive structures, so tech-based aggression/harassment tools have to follow the movement and become much more powerful to have an impact. And since it is still not enough in many cases, enter the ultimate “anyway”: instantaneity, invincibility.

In order to preserve “activity,” the hydra of hyper-development forces the Blizzsters to repeatedly kill the defender's advantage, replaced with the puzzling concept of the attacker's advantage.

Enter: Medivac Boost; proxy Hellbats [Hellions turning into Hellbats at Armory tech]; semi-teleporting Tanks;
Enter: various bane busts; muta regen; Ravager busts; invincible Nydus;
Enter: Warpgate; Blink; Recall; invincible Disruptor + Warp Prism picking from massive range;
Enter: worker bowling.
Etc.

(Notice how many of those attacks are proxy attacks to recreate windows that hyper-development + the basic time barrier that is space would otherwise shut close.)

Violence calls for violence. Contraction of time is a drug. Speed in one domain has to be echoed by more speed in another domain. Hyper-mobility is the natural child of hyper-development; nonsense, the proud third generation. + Show Spoiler +

This is also why the map equation is nigh unsolvable in SC2: maps have to endure the overwhelming burden of countless issues related to contraction of time.

Now; let us consider the past, and our bright future.

+ Show Spoiler [material] +

Again, at random:
BW: PianO vs ZerO series (april 2014) + a few NaDa's TvZ games on his stream (april 2015);
LotV: various TvZ games from MMA, Ryung and SuperNova, mostly from Legacy of the Olimoleague. I doubled the sample because the increased volatility makes the 15' SC2 minutes benchmark tricky.
Time is converted in SC2 minutes.

It is actually harder to find a regular supply inflation in LotV's games with the 5-10-15 minutes system. One would probably have to refine to 5-10-12-14. Trades indeed don't occur randomly, and hyper-development further routinizes sequences. By the old 15 SC2 minutes benchmark, (1) probably more of the LotV games are already finished and (2) the “survivors” are already diving in the downwards spiral, particularly because of the decreased stock of minerals implemented by the SC2 crew (a very astute mechanic to obtain what they want). But the phenomenon is still visible at 5 and 10 minutes, though concealed/minimized by the ET substitution discussed above.

The second game between SuperNova vs bly is a good example of the insane rate of development in LotV: from 3 bases, Lair and 78 supply at 10 minutes to 5 bases, Vipers on the way and 200 supply at 15 minutes. All of this in 3 minutes 35 of real time. See “Spaghettification” for the inevitable consequences.

It is interesting to compare the relics of the first macro games in proto-WoL to the maturity of BW. In the MKP vs Nestea series, if we consider the only macro game which welcomed a third from the Terran, i. e. the Shattered Temple game (set2), we can draw the following comparison:

Not much difference, hmm? But we would again be fooled by the facial values. Just like the evolution of openings accelerated the average rate of development, we have to compare apples with apples. The benchmarks are roughly similar, except… the BW games were 1 rax FE, while MKP opened Thor drops (= high tech) with a 9'52 natural!! Despite expanding 6 minutes later, MKP outpaced his ancestors.

Le ver est dans le fruit since the beginning. Consider for instance the first game of the very first GSL finals between FruitDealer and Rainbow, another macro game from proto-WoL. At 15 minutes, FruitDealer was already on 170 supply out of a Speedling expand into 2b mutas, with 11'30 Hive into 15'30 ultras out. As for his opponent, Rainbow opened rax fact CC into 2 fact Tanks/Thors 3 rax third and still got 130 supply of biomech (= high infrastructure costs + high tech) at 15 minutes despite only securing his natural at 8'30! Of course, in this case the values are “brute” as no fights had occurred (passive macro!), and they had considerably less than their heirs because of the uneconomical openings, but still… SC2 is simply drown in resources.

Here lies the “seed” mentioned above: regardless of the stage in which it is materialized, SC2's eco-productive system suffers from a chronic state of hyper-development.

Hence the common steep landscape of SC2 games. Think of it like a cheetah. It accelerates extremely fast from 0 to 110 km/h, but can only run on a short distance before collapsing. For the economic aspect, breaking the {2;2;1} worker triplet is sound and obvious, but we also need to bend that curve: probably a tad faster at the start, but with a slower pace afterwards. Taming the beast! LotV does not do that: the eco curve simply enters depression a few minutes earlier than in HotS—the cheetah runs even faster at the start, but runs out of steam even more brusquely.

Hyper-development is structurally incapable of producing action-packed, back-and-forth games on a constant basis; such games happen against the power of the propulsion, thanks to various counter-forces and shock absorbers—notably the defender's advantage (which, ironically, is partly fueled by the massive production infrastructure coming from said hyper-development: think for instance about the speed of remax of Z/P in lategame TvZ/P). To get such games, one has to hope that timings/all-ins damage but don't kill; that fights are “stalemates”; and that economies stabilize on a “medium” setup, ideally decreasing the army size:reproducibility ratio. Naturally, in the modern era, planets rarely line up; which is why Neanderthal games are the rule while authentic gems are happy accidents. From here also comes the dreaded phenomenon of the “single engagement into victor”.

The rocket engine of hyper-development is indeed self-destructive. The brutal contraction of time tends to generate a chronic state of instability: after the initial push, a plateau is reached from which “over-critical” action occurs (e. g. players have to go “all-in” on certain decisions, or on the engagements). + Show Spoiler +

Overall, contraction of time is responsible for drastically skewing the impact of mistakes: some things matter way too much (certain slight mistakes are punished incredibly hard: the Spark) while others are too easily forgiven (the Pardon). Here again, ironically, hyper-development has sometimes stabilizing virtues…

Then, when economies crumble in macro games, it's often too delicate to catch up: under the threat of their opponent's huge army, players cannot redevelop without risking immediate checkmate. This is also what makes games predictable and anticlimatic, on top of frequently preventing better players from coming back in the game.

The LotV economy (12 workers + 1500x4 + 900x4) is a massive regression compared with the already terrible SC2 basis. There is not the slightest ounce of “progress” here. (People who consider the LotV economy as an improvement confuse the symptom and the disease; the LotV economy fixes nothing, it simply makes the disease worse, so new symptoms arise.) I am sure people remember MULEs on gold bases and the ridiculous spikes of income on eco charts, triggering the nerf and the increasing scarcity of Gold bases on maps? Well, for LotV, the Blizzsters decided that it was Christmas for all races: MULEs on Gold are no longer the exclusive property of Korhal peasants, they are now democratized. Hurray for equality! The economic changes in LotV are simply this: a “Golden economy” with a free, permanent MULE at the start. What kind of gameplay do we get from this? Unsurprisingly, not a stable one.

The first SuperNova vs bly game is, for instance, a dazzling example of the kind of complete nonsense that LotV currently delivers: proxy 3 rax bunk contain 3OC into defensive Tank and Cloak Banshees mech vs hatch sac into 1b burrowed Roaches Lair rush into 2b turbo Roaches mass Nydus (spotted but indestructible). Yay, OK.

+ Show Spoiler +

http://i.imgur.com/bhoEL8c.jpg

So far, LotV is thus not an accident, but the accomplishment: the Blizzsters have finally the means to fully apply their original program…

On April 11 2015 06:39 TheDwf wrote:
Why did SC2 purposefully, systematically remove control from the user in various domains? Because the Original Blizzsters thought that (1) spectators come first; (2) spectators must be excited; (3) excitement = randomness.

Contracting time = less control = more “time-based” mistakes = increased randomness. (…) If the RT part of RTS is violently compressed then the S withers away too by force.

… and they have many weapons at their disposal to keep contracting time. First, since boredom “de-contracts” time, “Mr. Viewer” comes to think that contraction of time is the key to solve the problem of… contraction of time. But “turtling deathballs” are merely the most visible symptom, so the problem is not at all defending or lesser mobility. The game ended up being too “slow” (= passive) because it is… too fast! same as camping partially comes from the fact that the defender's advantage is actually… too weak! The common misunderstanding allows the SC2 crew to tell the “community” what it wants to hear, while applying more of the methods which caused all the problems to begin with. Just consider the transition between the end of WoL and HotS: wasn't it exactly the same problematics? What little dark “miracle” turned Zerg (by essence supposed to be the anti-deathball race!) into a deathballing race? Answer: this horror was the rotten fruit of that one. Same as this or this come from this, MULEs rains in mech into air lategame scenarii are self-explanatory, etc. Then, after the scapegoat units were banned for political reasons (though WoL Infestors were of course a massive joke), HotS unsurprisingly had to recreate what WoL had refused to deconstruct… Out: broods/infests. In: SHosts/Corruptors/Viper. “But with LotV you precisely can't go broods/infests!” — Who cares! Symptoms change, disease remains. Sapping the eco substratum to weaken/prevent 200/200 accumulation changes nothing to the internal logic of hyper-development. Yeah, after the Roaches/Ravagers 3b all-ins are solved, you will a-click 170 supply of GMO mammoths instead. Anything new under the sun?

Second, the true-false concession of “less extra workers at the start”. They may concede less than +6. But then what? See for instance how politicians declare that “unemployment increased slower than foreseen this month” and claim that this is a decisive victory. Well, too bad unemployement still rose. Same idea here: hyper-development would still be deadlier, but less than with the 12 workers change. “Victory!” roars the big cat. Victoire sans lendemain.

Third, the total amount of minerals in a base: a worthy debate indeed, but one that leaves untouched the central question of hyper-development. If you remove 500 minerals from each node, or x minerals from y nodes, you didn't modify at all the basal rate of development! In fact, the sinister synergy between the 12 workers change and the decreased quantity of minerals in each base is a clever coup from the SC2 crew: they make sure that people cannot escape the game of cubes. The tour de force is to combine the “Golden economy” with various nonsensical tools (incarnations of the attacker's advantage) to further dive the game into a chronic state of instability, alternating short periods of passive, turbo build-up with high-stake “action”. Brutal propulsion into spectacular crashes—exactly what “Mr. Viewer” wants! Sense and strategy will further disappear in the wake of the razzia, but one cannot make an omelette without breaking some eggs. And no sacrifice is too big to multiply Sparks and get what the Blizzsters truly want to enforce: Speedcraft.

jinjin5000

Korea (South)1368 Posts

April 30 2015 06:58 GMT

#344

Seems like sc2 is having identity crisis on what to be. Action packed fun "skill shot" micro from hero based MOBA or wc3 from game of numberous army with few high valued target but no "hero" unit.

Watching TLG today what Jakatak said resonated well with what I was brewing on.
Blizzard also seems to be confused on what mech is- Its a philosophy of units that possesses heavy zone control units (siege tanks, vulture spider mines+harass, Goliath AA range) in trade for mobility and ability to recover to certain extent if not positioned well. Blizzard seems to go instead in direction of re-skinned durable bio ball approach- adding all-purpose Cyclone into arsenal while siege tank simply isnt doing the job it is supposed to be doing in zone control.

Just because it comes out of factory doesnt mean its "mech" unit at all. Thor is just a big clunky marauder with weak hard counter AA against mutalisks. Tanks simply dont do justice vs protoss at all even with emp. Blizzard's band-aid solution of adding mobility and speed buff to mech totally goes against its ideology in favor of high intensity explosions.

Why can't two tech trees remain seperate and distinct? Wouldn't make those 2 distinct playstyles be more "exciting" rather than 2 indistinguishable ball of mobility and all-roundness?

Pursuit_

United States1330 Posts

May 01 2015 09:44 GMT

#345

Wow, I finally got to play in the beta and it's more than just kinda rediculous how 'fast' the game is. My first game (so probably rediculously unrefined builds ect) and not even 4 minutes in I was dropping tanks in my opponent's natural. There's absolutely no place for passive playstyles in the current edition of LotV, my (admittedly limited number of) games feel more like clickfests than a strategy game.

edit: Everything seems to be happening so quickly there's barely room for micro. You just make the strongest mobile army you can and attack everywhere your opponent isn't while expanding everwhere. This is going to be a huge buff to units like DT's or Widow Mines that make the opponent play more cautious. Initial impressions ofc.

manwiththemachinegun

5 Posts

May 01 2015 15:41 GMT

#346

Lets compare the average number of units/buildings produced every minute in Brood war and Starcraft 2.

[image loading]

As Dwf has noted, the automation in macro and production mechanics seems to have wrecked the pacing of the game completely.

A possible solution could be to increase the overall production time proportionate to the amount of units of the same type being produced simultaneously.

An example would be a 5% increase in production time for each marine being produced at the same time.

This would disincentivize massing units; bottleneck and contain the production explosion, and actually force players to tech up organically.

NasusAndDraven

359 Posts

May 01 2015 18:18 GMT

#347

Guys why is it that always when there is a thread like this, everyone just assumes that brood war is the game developed by God himself and thus is perfect in every way. And the goal of SC2 development should be to make it be played as close as possible in every way as BW. Everything that is different in sc2 somehow is a mistake.

BronzeKnee

United States5211 Posts

May 01 2015 18:33 GMT

#348

Dwf's last post should be a thread on it's own, it is very impressive. I've been trying to put into words what he stated regarding harassment/aggressive tools for so long. They've made defensive tools stronger and stronger, but then had to make harassment/aggressive tools increasingly stronger to the point where now you really can't catch Medivac drops anymore because Medivacs are so quick. The end result is that now is that for me, a Protoss player, the cat and mouse game that was properly positioning my Observers and Stalkers around the map to pick off drops which required skill and control has been reduced to nothing. The Terran player gets caught, but boosts out greatly limiting damage taken. In the end the strategy is being bled out. It is mostly just act and react now. And we've reach the pinnacle as Dwf points out, where units are invincible.

On April 30 2015 08:15 TheDwf wrote:
Here lies the “seed” mentioned above: regardless of the stage in which it is materialized, SC2's eco-productive system suffers from a chronic state of hyper-development...To get (good) games, one has to hope that economies stabilize on a “medium” setup, ideally decreasing the army size:reproducibility ratio..

Anyway, while I do think that the quoted statement above is true, it isn't necessarily a problem given certain circumstances. At the moment with LOTV, we aren't even close, but I believe strongly we came close to those circumstances at times with WOL. In fact, I believe that Starcraft is best when it achieves that sweet spot.

The difference in supply between BW and early WOL as you showed was similar, though the BW openings were certainly far more "economical." That isn't necessarily a good thing.

In the games from the second GSL Final between Nestea and MKP, the contraction of time that reduces control certainly isn't evident. The micro war between Nestea and MKP during the final game was between Marines without stim and SCVs versus Drones, Queens and Zerglings without speed provide both players with ample control opportunities, it certainly wasn't too fast.

But what WOL lost overtime was the what kept the supply of both players in check during that series and kept the game interesting: the ability to attack and end the game at anytime. Both players had to build units throughout the game, and not just low tech units, in order to hold attacks. What kept WOL interesting from a viewers perspective, and also for a player was that if you over-stepped your bounds economically, your opponent could punish you hard.

And that isn't a bad thing. Expansions had to be earned, they weren't just the standard way to play. And although that kept supplies in check, the improved economy in WOL (compared to Broodwar) allowed access to units like Thors and Void Rays that would be more difficult to afford on a single base in BW. One base play not only was an option, but far more of the tech tree was accessible early, allowing for more strategic variation.

Problems arose when Blizzard and map makers allowed things like three CC before gas builds to exist or when abilities like Photon Overcharge were added to the game. By allowing Terran to survive early timings by only building their most basic unit or giving Protoss Photon Overcharge meant that the aggressors improved one base economy and ability to access more parts of their tech tree (which gave the game strategic depth) was rendered useless, because none of those timings could hurt an economic build. That drastically slashed strategic variation.

Additionally, without anything to hold economies in check, the improved SC2 economy exploded and things spiraled out of control into the mess we have now. We need the tools to keep economies in check back. Expanding should not be a right. And if we get those tools back, then we'll constantly be on the edge of our seat because anytime anyone tries to step out of bounds, the other guy could attack and punish them. We'll have a plethora of medium income scrappy games, which are so exciting.

So, the best way to guarantee action, is to force both sides to build combat units (not purely defensive or harassment units, but combat units) early and often, like players had to do during early WOL. So let's remove tools like Photon Overcharge would allow players to not build combat units.

FueledUpAndReadyToGo

Netherlands30545 Posts

May 01 2015 18:38 GMT

#349

On April 30 2015 08:15 TheDwf wrote:
(2) 2b mutas into Roaches against mech

Comparing SuperNova vs bly (set 2) and GuMiho vs Solar. bly's build is less economical (3Q speed + faster Lair vs gasless 4Q); his macro is of course considerably inferior to Solar's one; he loses more in harassment, and still:

[image loading]

LotV timings were converted in SC2 minutes.

When Solar reaches 180 supply, on 63 drones, he has not yet completed +1 armor, he has no transition and only 3 bases. When bly reaches 180 supply (one minute later), on 59 drones, he has way more tech (Hive and 2/2 on the way, Hydra den, Burrow…) + 5 bases ready. Solar can only dream of such a transition.

Thus, on LotV, players would max faster, on higher technology, if they were “allowed” to do so.

What you can see from this is that it takes an extra minute in LotV to get to the same supply. It takes longer, while you start with more workers and supply (it takes ~1m35s for Zerg to get 12 workers). So Hots reaches 180 supply 2m35s earlier than LotV. LotV is not faster. Solar could have built tons of stuff in 2m35.

Solar his income at 12.20 is 1920/576. Thus in 2m35s he could spend an additional ~4.8k minerals and ~1.4k gas. This easily pays for 2/2, a hydra den and a hive. He would finish most of these upgrades faster than bly too. But Solar has already won the game a minute later.

Grumbels

Netherlands7028 Posts

May 01 2015 20:03 GMT

#350

Some counting:
+ Show Spoiler +

brood war:
3 hatchery
3 evolution chamber
2 spawning pool
3 hydralisk den
2 spire
3 queen's nest
3 defiler mound
2 ultralisk cavern
0 greater spire
= 20 upgrades for 9 buildings and 12 units

starcraft 2
3 hatchery
3 evolution chamber
2 spawning pool
2 roach warren
1 baneling nest
2 hydralisk den
2 spire
0 greater spire
2 infestation pit
1 ultralisk cavern
=18 upgrades for 10 buildings and 17 units

Why is Blizzard removing upgrades? Did BW just have too many or is SC2 too fast-paced to really allow for multiple upgrades per unit?

Endymion

United States3701 Posts

May 01 2015 22:03 GMT

#351

On May 02 2015 05:03 Grumbels wrote:
Some counting:
+ Show Spoiler +

brood war:
3 hatchery
3 evolution chamber
2 spawning pool
3 hydralisk den
2 spire
3 queen's nest
3 defiler mound
2 ultralisk cavern
0 greater spire
= 20 upgrades for 9 buildings and 12 units

starcraft 2
3 hatchery
3 evolution chamber
2 spawning pool
2 roach warren
1 baneling nest
2 hydralisk den
2 spire
0 greater spire
2 infestation pit
1 ultralisk cavern
=18 upgrades for 10 buildings and 17 units

Why is Blizzard removing upgrades? Did BW just have too many or is SC2 too fast-paced to really allow for multiple upgrades per unit?

bw didn't have too many, i can think of situations where you would get every upgrade in ZvT i.e. if terran opened 2 port wraith into bio into lategame mech, that would hit basically every upgrade in the game for zerg.. also, there's 4 upgrades on lair in BW (burrow, ventral sacs, speed, antenna)

I think it's just because understanding how current upgrades influence how you optimally play a point of time in game is harder to wrap your head around than having a whole new unit, so sc2 focused on more units rather than upgrades to change the flow of the game. it's also potentially frustrating to new players to not have an easy visual indicator to see that your enemy has ling speed or hydra speed/range upgrade prior to engaging their forces, which can be very frustrating to new players (in bw), so they tried to get away from this with either visual indicators or just new units.

Pursuit_

United States1330 Posts

May 02 2015 01:35 GMT

#352

On April 30 2015 08:15 TheDwf wrote:
+ Show Spoiler +

LotV timings were converted in SC2 minutes.

Case study (link)

Hyper-development in action.

Contraction of time at its finest.

On April 19 2015 22:46 TheDwf wrote:
SC2 games, because of hyper-development, tend to degenerate towards (1) all-ins; (2) passive macro; (3) worker bowling.

This is also why the map equation is nigh unsolvable in SC2: maps have to endure the overwhelming burden of countless issues related to contraction of time.

Now; let us consider the past, and our bright future.

+ Show Spoiler [material] +

http://i.imgur.com/bhoEL8c.jpg

So far, LotV is thus not an accident, but the accomplishment: the Blizzsters have finally the means to fully apply their original program…

Quick question, are you comparing SC2 time to 'real' time or real time to real time?

TheDwf

France19747 Posts

May 02 2015 21:46 GMT

#353

On May 02 2015 03:33 BronzeKnee wrote:
Anyway, while I do think that the quoted statement above is true, it isn't necessarily a problem given certain circumstances. At the moment with LOTV, we aren't even close, but I believe strongly we came close to those circumstances at times with WOL. In fact, I believe that Starcraft is best when it achieves that sweet spot.

I have no problem with players rushing Thor drops or Voids on one base.

The economic development rate on a single base is not wildly off, to bend the curve you just need to cut 15-20% on max saturation and it's OK. The mould comes before the model, so units stats can be adjusted accordingly. Everything has to be tweaked when the economy changes anyway. 1 base play is 100% legit indeed, it simply went the wrong way in proto-WoL. It's a matter of finding the sweet spot, as you say. That's why bending a few curves would immediately do wonders, and then units would have fresh air again.

I will probably analyze that Nestea vs MKP game in some text later, because it's an interesting situation indeed.

On May 02 2015 03:38 FueledUpAndReadyToGo wrote:
There is no time contraction

Haha, you're lovely. <3

I come back to you a bit later. I thought about our previous exchanges and now I know why you don't understand contraction of time. It's just a little confusion about the notion of development. Maybe with a visualization like this you would better understand it? + Show Spoiler +

Remember your own words:

On April 15 2015 04:16 FueledUpAndReadyToGo wrote:
A contraction of time would mean that things happen faster.

You simply isolate and consider “1” in a vacuum at the start and forget about the following notion of development (i.e. you don't simply bank minerals, you do stuff with it!). Note that “1” is contraction of time too, because here we have earlier = faster.

On May 02 2015 05:03 Grumbels wrote:
Why is Blizzard removing upgrades? Did BW just have too many or is SC2 too fast-paced to really allow for multiple upgrades per unit?

Yep. The technology tree is bound to wither. As the storm rages on, the weaker branchs fall. That's for instance why Terran lost tons of upgrades in HotS. Terran is the only SC1 race left in SC2, but SC1 is quartered in SC2. Yet obviously you have to tune the violins, otherwise acute crisis of asymetric hyper-development occur. And upgrades are infrastructure costs which slow down development, inserting plateaux on the curve (what I called “principle of cascade” in ZParcraft II). So you just remove upgrades. Thus the water falls more smoothly and more violently.

Siege Mode removed because otherwise, all 1-1-1 openings die to the insane contraction of time in Blink timings. Tons of Roaches timings were borderline broken vs mech too at the end of WoL, the margin of error was nearly inexistent (e. g. see here, answer to SHODAN). Even within TvT itself, bio attacks often killed mech with pre-Siege timings in WoL (back then, the cursor was a tad too much towards bio for various reasons).
Moebius Reactor removed against Protoss' hyper-development advantageously converted into Templar play.
Transformation Servos removed so Terran could disrupt the Zerg curve with proxy Hellbats.
Merged mech/air upgrades because mech passively suffers from hyper-development; since the midgame shrank, you fall back on lategame transitions (which actually become the focus of the strategy, instead of the “afterwards”).

Hence why LotV is already a graveyard of upgrades.

On May 02 2015 10:35 Pursuit_ wrote:
Quick question, are you comparing SC2 time to 'real' time or real time to real time?

Your question has several meanings, I'm not sure which one I should pick. Can you precise?

Pursuit_

United States1330 Posts

May 02 2015 22:19 GMT

#354

On May 03 2015 06:46 TheDwf wrote:

Show nested quote +

Your question has several meanings, I'm not sure which one I should pick. Can you precise?

For example, on the graph showing supply at various timings including BW stats, are those the supply counts at the time according to SC2's internal timer (which runs ~1.38x faster than 'real' time), or did you (or whoever collected the data) convert the time for consistency or measure it with a stop watch? i.e. when you say MKP vs NesTea looks similar to the BW Chart, are you using a consistent time scale?

TheDwf

France19747 Posts

May 02 2015 22:22 GMT

#355

On May 03 2015 07:19 Pursuit_ wrote:

Show nested quote +

Yep, all converted to SC2 minutes on the graphs.

y0su

Finland7871 Posts

May 02 2015 22:25 GMT

#356

Great stuff again, I do wish you would have created a "razzia of the blizzsters II" thread instead of getting more buried here.

chrissummers

243 Posts

May 03 2015 00:06 GMT

#357

thanks for the read. i can see you understand the game very well and especially understand to point out its flaws and problems. i absolutely agree, after playing the beta i have to say: it is even worse than i thought.

Gfire

United States1699 Posts

May 03 2015 05:51 GMT

#358

Thinking about hyper-development and such today I thought of a few things I wanted to say:

For the miltary-tech-economy triangle to work it's necessary that the military and tech can be aggressive before economy wins long game. Preferably aggression that isn't all-in. In SC2 we had all-ins and didn't like it so we tried to get rid of those and that just results in economy being the only real path.

Once players are on enough bases to be vulnerable to harassment they already have a booming economy (as much supply in workers as they'll want) and investing more in economy isn't really needed. So by the time military becomes a useful resource, economy becomes a useless thing to invest in. This totally kills the miltary/tech vs economy decision.

Players can't spread out across more harvesting locations (making military more useful for the opponent) without increasing the rate of worker production, which is one of the causes of this problem. Basically, only the econ player is vulnerable. Not only that, but they aren't really very vulnerable in SC2 until they are on 3-4 bases. By then they have as many workers as they want and investing more in econ isn't really needed. Their econ spending has payed off by the time they become vulnerable to anything but direct assaults, which would be all-in typically, at least if the third is close enough to the main and nat.

At first I was thinking this could be helped by slowing down the worker production or something like that. However, I've had another thought:

It would be nice if the shorter-term-focused player could be aggressive against any player going for a longer-term play. Expanding is the most obvious long-term play, but many others exist without spreading you out. Terran increases worker production (+mules) without expanding to a new location because of liftoff, and Zerg can pump drones instead of army without expanding. Protoss can Chrono probes which is an investment in econ rather than tech or military. Also any teching player has a long-term advantage and short-term disadvantage against a military player, right? So shouldn't they be vulnerable, too?

So it seems to me like a lower base count player should be more vulnerable to aggression. Of course, aggression designed to stop that player from running away with the game, and hopefully go into the next phase of the game with an advantage, not usually to win the game immediately.

SC's core design, in SC1 and SC2, seems to have a hard time supporting harassment or some type of aggression that isn't all-in against a lower base count. You defend at a single location, and if they can bust that they've killed your whole force so they just take the game, probably. You must have air units to harass usually, and even they don't seem that effective against two bases in SC2. BW, if nothing else, has the benefit of big army attacks being less effective since units don't work as well in larger groups... It's hard to actually identify too much that's better than SC2 in BW at this stuff, though. So I'm certainly not just wanting SC2 to be more like BW here...

Actually, I'll bring up the comparison to Age of Empires II. There you usually have a quite wide open space around your town center (basically a CC) and must expand your harvesting locations outward very early to get resources, as there are only a few trees and such near the town center. However, you take these locations with cheap, quick-to-build drop-off camps that don't train workers. You'll need to do this for each resource separately generally and there are 4, so it's like the equivalent of 1-basing is still spread between several locations. Even a tech/military player will be spread out. To stop rushes you can pop your workers into your town center and while inside they can shoot at nearby enemies. Therefore, it's very hard to kill a player outright but to deny harvesting or maybe kill a worker is totally possible. At least, this is my theory based on the mechanics. I don't really know how it is in practice in competitive AoE2. Maybe I should check that out some more.

manwiththemachinegun

5 Posts

May 03 2015 07:27 GMT

#359

On May 03 2015 14:51 Gfire wrote:
Thinking about hyper-development and such today I thought of a few things I wanted to say:

For the miltary-tech-economy triangle to work it's necessary that the military and tech can be aggressive before economy wins long game. Preferably aggression that isn't all-in. In SC2 we had all-ins and didn't like it so we tried to get rid of those and that just results in economy being the only real path.

Once players are on enough bases to be vulnerable to harassment they already have a booming economy (as much supply in workers as they'll want) and investing more in economy isn't really needed. So by the time military becomes a useful resource, economy becomes a useless thing to invest in. This totally kills the miltary/tech vs economy decision.

Players can't spread out across more harvesting locations (making military more useful for the opponent) without increasing the rate of worker production, which is one of the causes of this problem. Basically, only the econ player is vulnerable. Not only that, but they aren't really very vulnerable in SC2 until they are on 3-4 bases. By then they have as many workers as they want and investing more in econ isn't really needed. Their econ spending has payed off by the time they become vulnerable to anything but direct assaults, which would be all-in typically, at least if the third is close enough to the main and nat.

At first I was thinking this could be helped by slowing down the worker production or something like that. However, I've had another thought:

It would be nice if the shorter-term-focused player could be aggressive against any player going for a longer-term play. Expanding is the most obvious long-term play, but many others exist without spreading you out. Terran increases worker production (+mules) without expanding to a new location because of liftoff, and Zerg can pump drones instead of army without expanding. Protoss can Chrono probes which is an investment in econ rather than tech or military. Also any teching player has a long-term advantage and short-term disadvantage against a military player, right? So shouldn't they be vulnerable, too?

So it seems to me like a lower base count player should be more vulnerable to aggression. Of course, aggression designed to stop that player from running away with the game, and hopefully go into the next phase of the game with an advantage, not usually to win the game immediately.

SC's core design, in SC1 and SC2, seems to have a hard time supporting harassment or some type of aggression that isn't all-in against a lower base count. You defend at a single location, and if they can bust that they've killed your whole force so they just take the game, probably. You must have air units to harass usually, and even they don't seem that effective against two bases in SC2. BW, if nothing else, has the benefit of big army attacks being less effective since units don't work as well in larger groups... It's hard to actually identify too much that's better than SC2 in BW at this stuff, though. So I'm certainly not just wanting SC2 to be more like BW here...

Actually, I'll bring up the comparison to Age of Empires II. There you usually have a quite wide open space around your town center (basically a CC) and must expand your harvesting locations outward very early to get resources, as there are only a few trees and such near the town center. However, you take these locations with cheap, quick-to-build drop-off camps that don't train workers. You'll need to do this for each resource separately generally and there are 4, so it's like the equivalent of 1-basing is still spread between several locations. Even a tech/military player will be spread out. To stop rushes you can pop your workers into your town center and while inside they can shoot at nearby enemies. Therefore, it's very hard to kill a player outright but to deny harvesting or maybe kill a worker is totally possible. At least, this is my theory based on the mechanics. I don't really know how it is in practice in competitive AoE2. Maybe I should check that out some more.

In my opinion, the key problem is the exponential growth.

Opponent interaction cannot smooth out the graph; it cannot make the growth linear.

So, it would end up having a binary impact of either succeeding or failing to prevent the critical mass required for economic explosion.

Even generally speaking, am skeptical of the approach of preventing faulty game states through adversarial player agency.

This reminds me of a discussion I had before the widow mine revert. Someone strongly asserted that the counter to mass mutalisks as terran is to stop the zerg from even reaching that point, which struck me as a perverse way to look at it. Understand where you are coming from, but not sure where I stand on this kind of thinking.

00higgo

Australia119 Posts

May 03 2015 09:56 GMT

#360

Its all just word salad to me, too hard to read but i think i agree with the article for the most part.

Prev 1 16 17 18 19 20 27 Next All

Please or register to reply.

Razzia of the Blizzsters - Page 18

Completed

Ongoing

Upcoming