|
United States7483 Posts
On July 29 2015 17:25 BlackLilium wrote:
I feel you are making many examples of how to deal with air using air.... you bring vikings, vipers, corruptors into the discussion. Of course that is viable and way to go and the game is overall mostly balanced because of those units. But it is also - in a way - symmetrical: you get air to fight air.
Wouldn't it be more interesting if air-to-air was weaker and ground-to-air stronger?
If you answer is no - and that's what I am getting from your responses so far - why did you bring the GtA and AtA into the discussion in the Start Chat #1? It becomes a bit confusing to me....
Hrm, to be more explicit, in game design for an RTS, timings matter more than flat unit designs. For most of the game, until max army supplies, in those matchups ground vs air units do just fine. While it would be more ideal if the late game super air armies could be beaten by ground based armies as well, that can't be accomplished without completely scrapping many of the unit identities Blizzard has at the moment. The viking, for example, would have to go out the window entirely.
For example, if my opponent makes mutalisks in TvZ, I don't make air units to respond, I make marines, widow mines, and a thor or two. In PvT, if my opponent is making vikings or banshees, I can just get stalkers. In ZvP, if my opponent is massing void rays or going phoenix into void ray, hydralisks do great, and then I add infestors.
It isn't until the air army starts approaching max that the ground based forces start losing to the air ones, and that has more to do with the fact that air units can clump and can all fire and ground units are more spread out, and have difficulty all attacking at once due to terrain limitations. If you opened a unit tester and put 120 supply of void rays on top of 120 supply of hydralisks and then start the fight, the hydras blow them up quite quickly.
The tempest is also a problematic unit design, although we didn't touch on it at all due to time constraints.
The only matchup that has an air vs ground problem before super late game armies exist is PvZ, with mass mutalisk being too fast and durable for the low DPS stalker, and their regen being too much for high templars to whittle them down over time.
I generally agree that mutalisks should lose their bonus regen and medivacs should have their boost removed or weakened, but you'd have to be careful about compensating elsewhere for it. The mutalisk regen was put in place for ZvT, and it's probably needed there. It just happens to break ZvP. You can take it away, but zerg would need something to compensate.
Medivac boost tends to make dropping a little too safe an option I feel, but that's an opinion and I don't have anything objective to back that up.
|
United States4883 Posts
|
Sounds really great, can't wait!
|
United States7483 Posts
Due to several issues, we have no choice but to postpone today's episode. We will be back on the air next saturday, at the usual time.
|
Now, thinking a bit slower, I realized that Warpgate does not give 1 extra round of units but around 1.75: It cuts down the reinforcement arrival time (the aspect you discussed), but also it gives the unit out before the cooldown and not after (minus the gateway-to-warpgate 10s transformation time - that gives 0.75)
By increasing warp-in time from 5 to 16 seconds (or whatever time there is), it will be tuned down from 1.75 down to 1.25.
It will be annoying but will it change the mechanic that much?
|
United States4883 Posts
On August 16 2015 23:49 BlackLilium wrote:
Now, thinking a bit slower, I realized that Warpgate does not give 1 extra round of units but around 1.75: It cuts down the reinforcement arrival time (the aspect you discussed), but also it gives the unit out before the cooldown and not after (minus the gateway-to-warpgate 10s transformation time - that gives 0.75)
By increasing warp-in time from 5 to 16 seconds (or whatever time there is), it will be tuned down from 1.75 down to 1.25.
It will be annoying but will it change the mechanic that much?
I maintain my point that instant reinforcement is the key factor here. As someone who has dealt with many 2-base all-ins in PvZ, the issue is not that the first group is too powerful, but that Protoss can warp units directly into the fights. There have been many times where I've been fighting a war of attrition and suddenly there's just more Protoss units than me because my units have to travel from my main to my third while Protoss has no travel time at all. This does not happen in TvZ because each race tends to reinforce in waves, and if the Terran is streaming units across the map, it is highly exploitable. Pretending like travel time doesn't matter significantly for warp gate timings is absolute silliness. In addition, a lot of problems were the result of scouting the pylon, witnessing the first warp-in, and not being able to really do much about it because the units would warp in before you could do anything more than shield damage. With a highly reduced offensive warp-in, the first warp-in actually has to be protected, and reinforcements aren't as powerful unless Protoss goes robo for a warp prism.
Sidenote: VoD and audio for Episode 4 are up. I think the video might still be processing on Youtube, but it should be up shortly.
|
gr8, gonna try to catch up to it tonight, probably comment too. 
(just tell me again when you're getting annoyed and I'll stop)
|
In 2-base all-in PvZ which involves what... 7? 8 warpgates? That's 8 units more on the front from the start.
No one is claiming that one extra round of units is insignificant, especially in early-to-mid game.
I agree that having a stream of units moving across a map, that can be intercepted, is a weakness. Protoss does not have that vulnerability.
We were however not focusing on tactical aspect, but just on numbers: how much more of an army you have on the front thanks to warp-in? And the answer is: 1 (or 1.75) wave of units. The benefit is constant and persistent. It does not grow over time.
On August 19 2015 01:10 SC2John wrote:
Sidenote: VoD and audio for Episode 4 are up. I think the video might still be processing on Youtube, but it should be up shortly.
Oh my... now everyone will hear my sick voice 
|
United Kingdom20318 Posts
but just on numbers: how much more of an army you have on the front thanks to warp-in? And the answer is: 1 (or 1.75) wave of units. The benefit is constant and persistent. It does not grow over time.
Warpgates themselves produce units more often than gateways, warp-in time aside.
|
On August 19 2015 05:29 Cyro wrote:
Warpgates themselves produce units more often than gateways, warp-in time aside.
That is true and we totally neglected that fact.
|
Love it, would have to add a lot more, especially on the topic of injects and mules, but very much agree with what was said.
One thing about the mentioned inconsistency of blizzard by making the game "slower" when the direction before was "faster" is that I don't necessarily think that the direction was "faster", even though it may have turned out as such.
So I think the reasoning behind the 12worker start is just that they wanted to get rid of those first 2mins of the game in which people only built 5workers and a supply depot/overlord/pylon. I think what they optimally hoped to achieve was that it turns out like when you jump into a VoD and then forward to the point where the player makes the first valueable decision of the kind of "do I build a barracks, a gas or safe money for a CC". The only reason why it is sped up a bit is because they didn't resynchronize it with the tech requirements for Protoss and Terran who still have to make pylons and depots before they can make that decision and for zerg it is a little bit of a set back to "only have 12workers when the other races also have 12workers", since zerg started with 3larva and larva spawn 2seconds faster than SCVs/probes build. So it turns out that you don't really start there, but again with building a depot/pylon/spending the larva-pool and thus everything happens of a little more money, which kind of accelerates the game a little, but that wasn't the reason why they did it.
For the faster mine-out. I think the reasoning is not to speed up the game, just to give players more angles of attack. It feels faster in a way that you want to expand faster, but it actually is also slower because you spend more money on expansions and safety measures and max out slower. Again, there is a degree to which it feels faster, but that wasn't the original point. It was about punishing sitting duck strategies more and force them into more risks when they want to get the resources to build their superarmy.
So in a way, maybe it is faster. But that isn't the main intention with those changes, but those are rather side effects.
I agree with BlackLilium about the unnecessary complexity of the warpgate change, but I also agree with Whitewing that it might be hard to find a simple solution that doesn't axe warpgate.
About the warpgate "snowball effect". I agree with Whitewing that it is "just one extra wave of units". But that extra wave of units does create a snowball effect!
So say the unit ratios at the start are 4:4 but with the frontloaded reinforcements of warp-ins it is 5:4 for protoss. Now a trade occurs that the protoss takes in slightly advantageous manner because he has more forces and say it goes to 4.5:3 (0.5losses vs 1loss). Now reinforcements arive for both sides and we go into the next trade with 5.5:4. This trade is won even more overwhelmingly because the ratios are even more in protoss favor now. And this is repeated until the defender loses.
So yes, there is a natural snowball effect to warpgate. But it's a snowball effect you will see with any type of gameplay in which you deploy more units initially and keep on reinforcing, e.g. as Whitewing mentioned the paradepush Terran-style.
The blink and forcefield obviously play a large roles here, but that's just the means in which those protoss armies are balanced. If you take that away from the units the units would have something different that would let them trade advantegously if they were in an advantegous position at the start of the battle.
The difference between the marine-baneling and the roach-stalker battle is that in the first one stuff is dying faster and it is more volatile to control I think. So what happens is that
a) there is a greater variety of outcomes in the marine-baneling combat based on micro and positioning while the PvZ roach vs stylker type of combats are a bit more numbers and timing based in my opinion.
b) it's not so much that stalkers don't die because of blink, because that would just be an issue of bringing enough units to survive the initial blink/forcefield rounds and then you are still starting to take down stalkers. It's more the fact that the whole battle is very slow because of low dps units + blink/forcefield. So what happens is that often during the skirmishing both sides' armies start growing, which is an inherent advantage for the protoss because he has more range and thus his army has an advantage if it grows evenly with the lower ranged zerg army. On the flip side, the baneling/marine/widow mine type of relations make for huge drop offs in supply, which suddenly turns the counter relations around. Suddenly the numbers get so low that the reinforcing+left over zerglings start countering the reinforcing+left over marines and the terran has to retreat.
Loving the guests. Also 5 people feels like the right number for that long of a show.
|
United States7483 Posts
On August 19 2015 06:05 BlackLilium wrote:Show nested quote +On August 19 2015 05:29 Cyro wrote:
Warpgates themselves produce units more often than gateways, warp-in time aside. That is true and we totally neglected that fact.
It's mostly irrelevant, and isn't a product of the warp-gate mechanic inherently, and it would only matter in PvP if one player delayed warp gate for a while. You could easily just have warp-gates produce at the same rate, they just made it quicker because they wanted to make warp-gates the absolute way to produce and wanted to nerf proxy gateway rushes before warp gates.
As for your approx 1.75 number, that would be true if the warp gate attack hit the instant warp gates finished, which is generally only the case in PvP (if at all). Otherwise the initial burst production was already used and the lack of travel time for reinforcements isn't applicable.
Snowball effects occur because one player's forces are fighting more efficiently than the other player's, and is continuing to reinforce. It doesn't matter what form that reinforcing takes.
|
United Kingdom20318 Posts
You could easily just have warp-gates produce at the same rate, they just made it quicker because they wanted to make warp-gates the absolute way to produce and wanted to nerf proxy gateway rushes before warp gates.
They significantly reduced the build times of gateway units to around the warpgate speed a long time ago in WOL era i think. As a result, IIRC toss players were going nex-gateway at the start of game vs zerg/terran and being able to survive even stuff like proxy rax. I think that was talked about as one of the measures to reduce the power of warpgate timings, but it never made it to live servers because protoss was too powerful in the first 6 minutes of the game with the capability to make units that fast.
As for your approx 1.75 number, that would be true if the warp gate attack hit the instant warp gates finished, which is generally only the case in PvP (if at all).
Some of hit with the first-second wave vs zerg/terran too. I've not played at any kind of high level for a few years but i did that stuff all the time at low master in WOL and early HOTS and i saw it used at higher levels often too outside of pvp
Even stuff like parting style immortal/sentry all in against zerg would move out before or with the first warp
Snowball effects occur because one player's forces are fighting more efficiently than the other player's, and is continuing to reinforce. It doesn't matter what form that reinforcing takes.
If you look at some of the more extreme gateway all ins pvz for example where you have few units and like 8 gates, if you compare protoss reinforcement to terran:
0:00 seconds - first wave of protoss units hit. Terran production starts
0:30 seconds - second wave of protoss hits. Terran starts second wave, first wave is walking across map
1:00 - third wave of protoss hits, terran starts third wave production - protoss has 3 waves of units at the opponent while terran has only 1
Even if the warp itself took 30 seconds, bypassing the travel time would be a big advantage. I can play protoss and high mobility zerg styles with some level of proficiency but playing slower styles or terran is really hard to get into the rhythm of because reinforcing is so difficult - that feels to me like the main difference between the races. Toss has warpgates and power units, zerg can put heavy emphasis on stuff like speedlings which produce fast and run very fast, terran is stuck in the stone age when it comes to reinforcement and it takes them far longer to go from production structures finishing to having 3 production waves of stuff in their opponents face. That's really the power of warpgates, when your 6 additional gates finish and within a minute of blizzard-time you have 20-25 finished protoss units units on his half of the map.
not really balance comments, just gameplay feel stuff from what's probably diamond league player at the moment with low master experience
|
On August 19 2015 07:47 Whitewing wrote:Show nested quote +On August 19 2015 06:05 BlackLilium wrote:On August 19 2015 05:29 Cyro wrote:
Warpgates themselves produce units more often than gateways, warp-in time aside. That is true and we totally neglected that fact. It's mostly irrelevant, and isn't a product of the warp-gate mechanic inherently, and it would only matter in PvP if one player delayed warp gate for a while. You could easily just have warp-gates produce at the same rate, they just made it quicker because they wanted to make warp-gates the absolute way to produce and wanted to nerf proxy gateway rushes before warp gates. As for your approx 1.75 number, that would be true if the warp gate attack hit the instant warp gates finished, which is generally only the case in PvP (if at all). Otherwise the initial burst production was already used and the lack of travel time for reinforcements isn't applicable. Snowball effects occur because one player's forces are fighting more efficiently than the other player's, and is continuing to reinforce. It doesn't matter what form that reinforcing takes.
I don't see how or why 1.75 number is applicable only when the attack occurs the moment the warp gates finish.
If I pit two players against each other, one with warpgate and the other without them - but otherwise having the same production time - then the warpgate guy will have approximately 1.75 wave of units more on the front (assuming the front is somewhere in the middle of the map). The value is not going to increase over time, unless - of course - the battle itself goes in one's favor. It will be so because
- time of one production cycle is saved due to travel time. A newly created cycle for warpgate player is on the front, while the same newly created cycle for non-warpgate is in transit.
- time of approximately 0.75 cycle is saved due to cooldown-after rather than construction-time. A newly created cycle for a warpgate player is on the map, while the same cycle for non-warpgate is in the construction queue.
The warpgate vs non-warpgate scenario is present in any matchup, most often in non-PvP.
Basically I am taking your own argument you gave on the show, and adding the cooldown-after effect. Unless I misunderstood your argumentation?
|
United States7483 Posts
On August 19 2015 16:55 BlackLilium wrote:Show nested quote +On August 19 2015 07:47 Whitewing wrote:On August 19 2015 06:05 BlackLilium wrote:On August 19 2015 05:29 Cyro wrote:
Warpgates themselves produce units more often than gateways, warp-in time aside. That is true and we totally neglected that fact. It's mostly irrelevant, and isn't a product of the warp-gate mechanic inherently, and it would only matter in PvP if one player delayed warp gate for a while. You could easily just have warp-gates produce at the same rate, they just made it quicker because they wanted to make warp-gates the absolute way to produce and wanted to nerf proxy gateway rushes before warp gates. As for your approx 1.75 number, that would be true if the warp gate attack hit the instant warp gates finished, which is generally only the case in PvP (if at all). Otherwise the initial burst production was already used and the lack of travel time for reinforcements isn't applicable. Snowball effects occur because one player's forces are fighting more efficiently than the other player's, and is continuing to reinforce. It doesn't matter what form that reinforcing takes. I don't see how or why 1.75 number is applicable only when the attack occurs the moment the warp gates finish. If I pit two players against each other, one with warpgate and the other without them - but otherwise having the same production time - then the warpgate guy will have approximately 1.75 wave of units more on the front (assuming the front is somewhere in the middle of the map). The value is not going to increase over time, unless - of course - the battle itself goes in one's favor. It will be so because - time of one production cycle is saved due to travel time. A newly created cycle for warpgate player is on the front, while the same newly created cycle for non-warpgate is in transit.
- time of approximately 0.75 cycle is saved due to cooldown-after rather than construction-time. A newly created cycle for a warpgate player is on the map, while the same cycle for non-warpgate is in the construction queue.
The warpgate vs non-warpgate scenario is present in any matchup, most often in non-PvP. Basically I am taking your own argument you gave on the show, and adding the cooldown-after effect. Unless I misunderstood your argumentation?
So let's be clear: I'm referring entirely to the part of the warp gate that ignores travel time for reinforcements when I gave these numbers, the front loaded production was not included, because it's not a mechanic anyone ever complains about.
But yes, if you include the sudden transition from back loaded to front loaded production, then it's more like 1.75 waves of units out of the amount of warp gates you have when warp gate finishes.
|
On August 19 2015 23:23 Whitewing wrote:
So let's be clear: I'm referring entirely to the part of the warp gate that ignores travel time for reinforcements when I gave these numbers, the front loaded production was not included, because it's not a mechanic anyone ever complains about.
But yes, if you include the sudden transition from back loaded to front loaded production, then it's more like 1.75 waves of units out of the amount of warp gates you have when warp gate finishes.
Both aspects compound to the same effect. People are complaining about the most obvious and apparent factor, but other aspects - like the after-cooldown exists as well.
My ultimate point is that the longer warp-in on forward pylons may be less impactful as some hope it to be.
|
United States7483 Posts
On August 19 2015 23:28 BlackLilium wrote:Show nested quote +On August 19 2015 23:23 Whitewing wrote:
So let's be clear: I'm referring entirely to the part of the warp gate that ignores travel time for reinforcements when I gave these numbers, the front loaded production was not included, because it's not a mechanic anyone ever complains about.
But yes, if you include the sudden transition from back loaded to front loaded production, then it's more like 1.75 waves of units out of the amount of warp gates you have when warp gate finishes. Both aspects compound to the same effect. People are complaining about the most obvious and apparent factor, but other aspects - like the after-cooldown exists as well. My ultimate point is that the longer warp-in on forward pylons may be less impactful as some hope it to be.
The impact is mostly in the vulnerability of the units being warped in aggressively, and the extra time before the first wave is ready will make a difference at the pro level. Also, a not uncommon event is to locate a proxy pylon, start attacking it, units warp in to defend it, and the pylon lives. Now that won't happen: units will just die and the pylon will go down.
Circling around to attack the pylon will become a much more valid tactic than before.
There are some other reasons, but this won't completely kill offensive warp ins, just weaken them a bit.
|
United States7483 Posts
We had to reschedule our show for today to next Saturday, due to real life demanding otherwise on some of our crew. Sorry everyone.
|
United States7483 Posts
Our episode on Risk has ended, we'll get the VOD and audio up as soon as possible.
Next episode is in 2 weeks and will be on the subject of Tech.
|
United States4883 Posts
VoDs for Episode 5: Risk are uploaded! Enjoy and share with friends!
|
|
|
|
|
|
EPT
