Right, spent quite a lot of time today (5+ hours) making changes, so here's what I've got in addition to previously mentioned:

• Added blue beacon option at the start which selects double rate + specials enabled in a single action
  
• Selection timer at the start begins at 10 seconds, after making a rate selection it will add 10 seconds, so still get the full 20 seconds as previously, just it'll end sooner if you don't select anything and go with the default (2x + specials)
  
• 10 seconds are added to the countdown timer when a night defence dies. I changed the text description to say this in 2.51, but actually it only added 5 seconds. So now the delay matches the description and is longer.
  
• Fixed boom issue during cinematic as reported by Whateverson. Could not reproduce the silo healing issue, but put some triggers in place to hopefully mitigate it.
  
• Goliath -5 hp to 125
  
• Nuke heals enemy units directly at your temple to 40% health/shields before scattering them (see previous post)
  
• Dragoons -10 shields to 40
  
• +0.05 phoenix on Stargate to 1.25 from 1.20
  
• Mothership and Thor will now kill fewer interceptors as part of their special pulse cycle; this is due to interceptors now being a lot weaker (see previous post)
  

I'm about 75% of the way through implementing what I am calling 'diffused spawning' where units spawn over 2 trigger cycles instead of just 1, so it should help make map max during spawning fairer. My testing shows it is working as intended as present, but total progress including testing yet to do is about 50-60%.

And how's that new spawn system going to work again? (During the verge of CCMU.)

At the moment, all units are created in a single trigger cycle. If you need to spawn 50 marines and 50 BC, the BC are prioritised first and 100% of them will spawn before any marines do. So if we only had space for 80 units, then we'd get 50 BCs and 30 marines. Most importantly is that 1 team must always spawn before the other team gets a chance to spawn. So say in this example the bottom team is spawning first and there are 80 slots left and they try to spawn 50 BCs and 50 marines: they'll get 80 as outlined above, and the top team wants to spawn 25 archons/high templar and 50 zealots but will get absolutely nothing because map max was already reached during the bottom team's spawn.

In contrast diffused spawning will spawn a small proportion of units in 1 trigger cycle, and the remainder of units in the 2nd cycle (note that the strict unit preference still exists: all BCs will attempt to spawn before any marines do). The maximum number of units it will spawn of any type in 1 cycle is 9, and not all units have a chance to spawn in the first cycle (weak ones like marines, reapers, marauders, zealots, zerglings, broodlings, anything that isn't very strong late-game). So in this example, on the first trigger cycle the bottom team will spawn 9 BCs, then the top team will get a chance to spawn 9 archons and 9 high templars, then the bottom team will spawn their remaining units, which means they will spawn another 41 BCs and then 12 marines before map max is met. In this case we've gone from 50 BCs / 30 marines and nothing, to 50 BCs / 12 marines and 18 archons/high templars.

In this example it may seem rather lame, but the thing to note is that it happens for all unit types: so really the top team may have gotten to spawn 9 archons/high templar, 9 dark archons/11 dark templar, 6 carriers, 4 reavers/immortals and 3 void rays and (assuming bottom units are still the same) the bottom team would only have spawned a total of 28 BCs: 9 from the first trigger cycle and 19 from the 2nd.

It's not perfect, but it's a step in the right direction. Specifically it does mean that in the late game a zerg will have a much harder time preventing you from spawning anything by building lots of cheap zerglings, broodlings and hydralisks, because all of those units will spawn strictly in the 2nd trigger cycle, meaning more powerful units will get a chance to spawn in the 1st cycle.

The existing alternation, where bottom team spawns first then top team spawns first will continue to exist, but that mechanic should be much less important because each team will get a chance to spawn a few units in the first trigger cycle. Obviously in cases where map-max is very close, eg team 1 will spawn 30 units on the first spawning cycle but there is only room for 20, will the other team fail to spawn any units. The alternation will still help with this, and obviously it also determines which team gets first chance to spawn their 'remaining' units (in my example above, bottom team got to spawn the remaining 41 BCs and 12 marines while top team didn't spawn anything else).

There is also another benefit in that because I have 2 cycles of creating units in the spawn box, I will now be able to clear out (some/most of) the units from the 1st cycle before attempting to spawn the 2nd. This should reduce the likelihood of unit unplaceable errors. This is part of the system that I haven't yet implemented.

Post has been edited 1 time(s), last time on Feb 20 2012, 9:36 pm by Lanthanide.

I noticed a huge bug in the game that I cannot warp in my Assimilator on the corner vespene geyser.
This has never happened in the previous versions and I can assure you the map wasn't maxed out.
I wish I can be more specific about it, but all I can say is I simply cannot gas fully.

my shit computer usually lags at the CCMU stage, meaning when there is like over 180 units it starts to lag and people accuse me of literally lagging on purpose when my computer only has 256MB RAM, not enough to handle the game at the CCMU stage, I guess this CCMU moderation of yours might reduce the lag my computer can produce, I tried scanning files an shiit but still to no avail, I'm on job hunts by around the 2nd week of March next month to get a badass computer that can handle CCMU's and over 1k units battling to the death. On that note, I'll have loads of fun without having to worry about dropping out the game from my lag an people whining to me about it lol..

This was already reported. I mentioned it a few (10s?) posts ago. I threw in a change at the last minute of 2.51 to try and correct an issue that routinely effects me, preventing me from building that gas easily (probe will walk up to it but fail to build the geyser, then I sit there gathering minerals without noticing it never built it). The change was to move all of the corner gas geysers out by 8 pixels, however I never actually tested it and it has resulted in this bug instead, where it is now much more difficult to build on that geyser at all.

However you can still do it: you simply need to click to the left a little with your probe/worker, closer to the geyser, and then it will be able to build it. In 2.52 I've moved the geysers back 4 pixels to the left and this appears to have resolved both issues.

Yep, slow computers are the main contributor to lag in SC these days, that and the 'NAT' issue that arises from ISPs and/or SC making funny assumptions about how networking should work.

However I think hydras need some tweaking(strength) to easily take on carriers without getting owned to the point where no1 has to make so many of them(like say over 100 hydra)just to combat the carriers since sairs will stand in the way of devourers that can also defend against the carriers just a little tweak for hydra wont make them that OP anyway as of why they stand no chance against reavers, and ultras becoming cheaper since protoss is most dominantly ground an has only 2 type of air units such as carriers an sairs, while both terran and zerg have more air units than the protoss do. So if zerg has strong ground against toss then it should be an even battle between them instead of reavers an chons knocking them off easily while sairs do the webbing since people say that zerg almost cant beat toss, just like terran almost cant beat zerg, regarding the P > Z > T ratio that alot of people speak about.

Conclusion: If ultras were to get a little cheaper and hydras get a little more strong for attacking carriers since sairs an chons will stand in the way of devourers, I can still fairly beat zerg when using toss. As DS is very more of a psychological game primarily which is why I luv it cause it has lots of skills to cook up against enemy's. People just have to be very highly intelligent against any seemingly OP or toughass units in battle then they're in the clear

Interceptors are significantly weakened in this next version, so if your argument is "make hydralisks stronger so they can better compete with interceptors" then this should already be accomplished without having to change hydralisks.

Hydra's real problem is that they do explosive damage which is only 50% vs small. Interceptors previously had 20 shields, 20 hp and 1 armor. This meant it would take 7 hydralisk shots to kill a single interceptor, ignoring shield regen wich would increase this to 8 or maybe even 9 hits. With 20 shields, 5 hp and 0 armor it will now only take 4 and even shield regen is unlikely to increase that to 5. To make a similar strength increase for hydras vs interceptors would require increasing hydra damage from the current 12 to 18, which is obviously hugely imbalanced (even this would be 4 shots but likely 5 with shield regen).

As for ultra's being cheaper, recently I have found that mass ultra is quite powerful vs protoss already. I'm not sure it needs to be cheaper.

As for terran vs zerg, a very common problem is that terran players are failing to properly use all of the spellcasters at their disposal: you *need* to get 8 medics with restoration and the energy upgrade. You *should* be getting 4 science vessals with irradiate and energy upgrade. I also see a lot of terran players who never build reapers, or never build marines, when really you need both. Early on vs zerg you definitely need a few marauders as well.

If you're massing Hydralisks against a P army, chances are you're doing something wrong.


I've never had a problem against Z with T. But then again I don't fall into the trap of investing too heavily in mech early game either. A lot of Ts will just throw their money into mech, and maybe even only a handful of siege tanks. Roach/Zergling will steamroll anyone who does this. You need to have the infantry to work with as a base.

I've found that reapers are okay, but not necessary. They're useful against armies that invest more into infantry, but M&M is really what's going on here. Normal damage transitions well into late game.

Post has been edited 1 time(s), last time on Feb 21 2012, 12:10 am by Sacrieur.

Yep, the reapers are more of an early-mid game option as their long range helps vs hydralisks (and they don't do terrible damage) as well as zerglings and importantly mutalisks. They also act as good HP sinks vs hydras that do explosive damage.

Having a good group of reapers can be the difference between your spawn holding off or slowly pushing the enemy (and therefore being able to grab your 2nd or 3rd gas) or the enemy pushing back towards your silo and them being able to gas. Obviously once they get a gas advantage it can be impossible to recover.

@Hallucination a while ago: It occurs to me that you could use Mojo for Hallucination, and simply set his damage to 0 and give him less armor than the normal Scout, but not actually make him a Hallucination. That would make it appear to be a Hallucination, while still allowing Give Units to work. Though the death animation would be wrong.

Also, as Hallucination is a Templar ability, perhaps the Hallucination Mojos could spawn from the Templar Archives rather than the Stargate? It makes sense, imo, and, also imo, is more interesting strategically. Though perhaps the Archive's cost is too high for that to work well... I don't remember.

Main problem with adding more units to the existing buildings is trying to find a name that will fit in the available space.

Yah, but it's associated with the Hallucination upgrade; you don't really have to put it in the building if you mention that Hallucination makes Hallucinated Scouts from the Archives elsewhere.

That would be breaking the existing design paradigm where every building shows what unit(s) it creates.

I wish Starcraft was more flexible when it came to textual feedback to the player, but it isn't and we have to live with that.

If you're playing Zerg and your opponent is going for mass carriers, you simply have to spawn a critical number of hydras. Carriers are just empty soda can without even using devourers against mass hydras. It seems like the only way to win against Protoss as a Zerg is to build as much air as possible with a strong late-hydra force that will do the damage. If the protoss smartly spawns 8~10 reavers at once, you should make even more units and maybe a couple of ultras so they can get hit instead of the hydras.

The reason I would produce ultras against P is because of their high HP, not the DPS. Mass (huge mass) Hydras and air (along with plague and ensnare) is really the only way to win against protoss in army to army battle from my experience.

Latest changes:

• Zerg larva will be removed from ally bases if present (eg, if you built hatchery/lair/hive in your ally's base)
  
• Maximum of 6 zerg larva in your own base; newly spawned ones will be removed (left-most removal)
  
• Fixed an exploitable spawn bug for one of the units (built 8 buildings, received 10 units instead of 8)
  
• Carrier -$30 to $850 again; justified due to dramatically weakened interceptors
  
• Tweaked the way probes are spawned; it is no longer possible for a single protoss player in a 3 player game to build 12 pylons in each base and therefore spawn 18 probes per player. They will now spawn a total of 18 per complete spawning cycle (same mechanic as Tempest Carriers).
  

Finished implementing the spawn diffusion for Zerg, so now all races are done. Not sure if anyone particularly cares about the nitty gritty, but here's the limits in terms of base buildings that can spawn in the first cycle. This is also the order that buildings are checked for spawning:

All hero units spawn before all other units, in the first cycle.





Post has been edited 3 time(s), last time on Feb 21 2012, 8:19 pm by Lanthanide.

So we gonna be seeing DS 2.52 today?

No. Hopefully by the weekend.

Since I'm looking at all this spawning stuff again, I'm wondering whether I should implement a stricter per-player spawn limit. At the moment if you spawn (that is, successfully create in the spawning boxes) more than 511 units, the excess are simply removed and they won't be moved to the field for you and also won't show up in your 'units spawned' counter.

I could put in a limit here of like 400. I think that normally only those sorts of numbers would be reached in 1v1 games, especially no specials, or in cases where a zerg player is simply being a jerk and building lots of zerglings to intentionally create map max for their opponent. Limiting the spawn size to 400 would help for map-max in 2 ways: if top player has just spawned 500 units, 100 of them will be removed from the bottom player gets their chance to spawn, which will let them use up those 100 unit slots, and obviously if we only move a max of 400 units per player to the field every spawning cycle then it'll take longer before map max is reached in general.

Post has been edited 2 time(s), last time on Feb 21 2012, 8:15 pm by Lanthanide.

I know you've rejected the idea before but maybe it's time to place build limits on zerglings?? Or at the least reduce them to spawning 5 - 5.5 per pool....

They currently spawn 5.33 per pool. There is actually an upper limit on the number of buildings of any type that will be spawned from, anyway. This was 63 (1+2+4+8+16+32) but with the inclusion of the diffused spawning system, some buildings (notably the ones that take 9) will now be handled up to 72.

400 / 5.3333 = 75 spawning pools worth, so you couldn't reach this number through pure spawning pools anyway because they stop spawning at 63.