Recent posts by Gaminic on Kongregate

Flag Post

Topic: Realm Grinder / Drow Bloodline Bugged?

This explains so much… I’ve been trying the Offline Goblins build, but it’s incredibly under-performing, despite my near-100.000% Drow bonus. Thanks, Pseudobyte, for the excellent explanation. Guess I’m just not going to bother with those builds anymore. :/

 
Flag Post

Topic: Realm Grinder / My R17 Experience

Originally posted by Toggler:

Not sure if you were using an autoclicker or not, but I have some advice for those like me that do not use one. Use Elves when you first reincarnate so you can actually get a decent bonus for C135 when you get to research. Even if there might be something slightly more efficient, they are fast enough that the added benefit later more than makes up for any minor delay.

If you are using an autoclicker, this can be pretty much ignored as your bonus will get high enough on its own.

Elf BL is overall a great choice because Clicks are a big thing in both Merc and Research builds. Manually building up clicks can be annoying, so whenever you do a passive overnight run, those Bloodline clicks really add up. I even run the first few LF runs with Elf BL for the faster startup.

 
Flag Post

Topic: Realm Grinder / Off-line auto-cast

Originally posted by lesdrui:

Well making a game require you to NOT play it to make progress is just bad design imo .

That is the majority of [early] Research though. Most [early] builds take a few hours to build up. Set it up during breakfast, check during lunch, final upgrades/buildings during dinner, then go to bed and wake up closer to your run’s goal.

Auto-casting while offline is probably not a design choice, but a limitation of how the game works. Sure, you can perform a quick calculation to determine to uptime of a spell, but can you calculate the uptime of several spells and their overlap time? What happens when you have “non-constant” behavior, such as Magical Treasure, Brainwave or Fairy Chant? This game shows exactly why we use Simulation for complex systems: as long as each separate building block is simple, Simulation is simple, whereas mathematical analysis because incredibly complex.

It would be an option to have a sped-up simulation run at login, which basically just plays the game according to the state at logout and the time passed, at an increased speed. It may be a bit intensive for Flash, though; it doesn’t do well with long calculations.

Alternatively, Spells could have a simplified behavior while offline, effectively just turning into passive abilities.

 
Flag Post

Topic: Realm Grinder / elven heritage underpowered

Does Undead not become worth it at R16+? I haven’t unlocked all researches yet, but there seem to be some additional offline bonus things going on. Additionally, their spell is one of the best in the game. I just wish there was a Blade Dance equivalent for their spells so it would actually benefit offline play. Something like “Permanent Darkness: 50% effect while offline”.

Undead were the first race I played and I looked forward to playing Undead/Drow, only to find it was massively underwhelming.

 
Flag Post

Topic: Realm Grinder / Passive bonus to vanilla factions?

He’s at R21, I imagine he knows about Researches. Secondly, he says it goes away when taking Drow/Dwarf, but those options disappear once you upgrade your Unique building to actually get you the boost, doesn’t it?

 
Flag Post

Topic: Realm Grinder / the R15+ dilema.. any ideas?

Yeah, after a few more runs it’s safe to say my analysis was off, for the reasons mentioned in this thread (gem gain per Forge hit lowers strongly). Definitely makes me wonder why everyone is saying “LF until e72 is fine” and mocking Cialis runs, because Cialis pushed me much faster than that. :p

@Sliverin: I’ll check in the next run. I think I couldn’t get one of the Smith upgrades without the full combo. The no-excavations start is a good point; guess I can be a lot more free with my excavations during this phase.

 
Flag Post

Topic: Realm Grinder / grand balance was kill

R15 is e72. You’re not getting 2x gems on 20 minute runs. At 5e69 it was around 37min (30% gems per Forge hit except for the first, 4 total hits for roughly 100% gems), now at 30e69 it’s around 55min (20% gems per Forge hit, 6 hits for double gems).

 
Flag Post

Topic: Realm Grinder / the R15+ dilema.. any ideas?

Summary: Using only LF, you’re looking at around 7 hours for the final stretch from e69 to e72 (1Dvg to 1Tvg).

At R15, 1Dvg (1e69) it takes four Blacksmith hits to double up. The first hit doesn’t give many gems, just gives you the coins required for a few additional upgrades. The following three hits give about +30% of your current gems each.

I was wondering a) how long it would take to reach e72 starting from e69 (=x1000 gems) and b) what the optimal number of Forge hits per abdication are. Assumptions/stats (R15, e69):
- I have around 6400 max mana and 30mana/s (2 Ruby upgrades, the rest Reincarnation).
- A “full combo” (LS + CtA + GG + SS, FC after 10sec and 6x TC ) costs me 7000 mana (6400 max + 20sec regen during combo).
- I’m assuming a flat 30% of current gems per combo except the first (setup combo, 0%). It’s not always constant (depending on upgrades) and wavers off the higher you get, but I’m pretending it’s constant.
- The gem gain per abdication is (#combos-1)x30% (e.g. 4 combos = 90% = 1.9 multiplier)
- Assuming average luck (11 LS casts = 1 Forge hit).
- Assuming x1000 multiplier required (e69 to e72), log(1000; multi) as number of abdications required, rounded upwards unless the last decimal <2 (so 26.2- = 26, 26.3+ = 27) because rounding with exponentials is a bit unfair.
- 1 minute setup time (excavate, select upgrades, buying buildings & upgrades).

A single “round” (10 LS casts + 1 Combo) costs 16000 mana or ~9min of regen.

Results:
- 2 Forge hits: ~19min per round, 1.3 gem multi, 27 abdications = ~8.5 hours.
- 3 Forge hits: ~28min per round, 1.6 gem multi, 15 abdications = ~7hours.
- 4 Forge hits: ~36.5min per round, 1.9 gem multi, 11 abdications = ~6.7hours.
- 5 Forge hits: ~45.5min per round, 2.2 gem multi, 9 abdications = ~6.8hours.
- 6 Forge hits: ~54.5min per round, 2.5 gem multi, 8 abdications = ~7.2hours.

So the optimal seems to be around 4 or 5 hits, depending on your estimates. Time per round is pretty solid, gem gain is a bit iffy. Early rounds may be a bit better, but later rounds do seem to get worse. Longer runs may be even more beneficial because you get a ton of Faction Coins from LS (even if it “misses”), which increases some bonuses significantly (Goblin(1,1), Dwarf(1,1), Drow heritage).

Overall you’re looking at around 7 hours for the final stretch from e69 to e72.

Additional tips:
- If you haven’t yet, a Max Building run is definitely worth it. My Overgrowth is around 3400%, compared to a ~3000% Gold Synth.
- Excavations boost some upgrades; if you know you’re not going to do any other runs, you can spend some extra excavations during a run to build up bonuses. You only need 1 successful excavation per run to get started.
- Use the weather service to spot bad sprees (20+ no Blacksmiths), then build a neutral faction mana regen build to churn out Lightning Strikes until blacksmiths are closer.
- Don’t abdicate when Blacksmith is in the next 3 hits. You need a few strikes to build up and a Forge hit would be a waste.

 
Flag Post

Topic: Realm Grinder / grand balance was kill

What other option is there for the end of R15? IGOT wouldn’t get you there all the way either and takes longer to setup.

I started this run at 5e66 (out of 1e72 required to Reincarnate) and after around 18h idle setup I could reach 300e66 per Pyramid hit combo. That’s a pretty easy road to well into e69 allowing me to ease up on the IGOT duration.

The only real downside of this build is the Angelline; I ran into the problem that my Flash player crashed and my Angelline got reset. Luckily the buildup is entirely passive and I’m now at 600e69 per Pyramid hit.

(Edit: use description removed; apparently I’m doing exactly what the Wiki suggests.)

 
Flag Post

Topic: Realm Grinder / How "random" is LS really ?

Originally posted by leheibrog:

Not completely,

Thats wrong, algorithm works differently. Based on random seed its determined which building should be hit with each strike. If that building is not build algorithm generate another pseudorandom number to hit building until it finds one that is build. You can see that in weather service when you look at save, build additional building and look at save again some strikes become ones to hit that building but rest of strikes is unchanged.

So only way to cheat with rws are quick runs where you buy trade lords and upgrade to generate fc with tc, cast tc for 1000 gems, cast ls, abdicate, repeat until you are next to smithy strike.

That’s provably wrong. If that were the case, building a previously unbuilt building would simply insert that building into the list (because “failed” seeds are skipped). This is most definitely not the case, because I use the delay-build trick all the time and it can have a drastic effect on the forecast (eg. building Deep Mines may put a Smith hit early on).

The “interpretation of RNG result” tactic could be correct (eg. if the RNG interpreter maps [1,100] to buildings by simply splitting it up in ranges, such as [1,10] for Farms if only 10 buildings types are built, then building the 11th building may change that range to [1,9], changing any 10’s previously mapped to Farms into Inn hits).

Either way, you can cheat the RWS by skipping buildings. For LF, I skip Deep Mines, Alch Lab, Monestary and Labyrinth, skill LS and check the RWS. If there is no Forge hit within X hits, I build Deep Mines and recheck. Then I build Alch Lab and etc.

I have noticed that Deep Mines seems to more strongly affect the forecast for Forges, so the range-mapping proposed by themthe may be correct. Mapping ranges by integer division or modulo operations is pretty common and in this scenario the division method is much fairer/more stable.

 
Flag Post

Topic: Realm Grinder / grand balance was kill

300k clicks is about 5 hours with Elf Bloodline & upgraded Spell on autocast. If you can unlock research at e69, that seems pretty pointless.

 
Flag Post

Topic: Realm Grinder / grand balance was kill

LF barely gets me x2 at e69 (R16) and that’s with constant babysitting. Moonshine is worthless until R18.

Maybe I’m missing some trophies, but I doubt there’s any major boost I could still be getting at my stage.

 
Flag Post

Topic: Realm Grinder / How "random" is LS really ?

Wait, what?

Originally posted by NiborRis:

Considering that your seed doesn’t reset when you abdicate,

How sure are you of that? People do the multi-tab trick, which according to you shouldn’t work. I sometimes just abdicate before getting a single Forge hit just because it’ll take 20+ casts. I’ll check next time, but I’m under the impression it does change.

Either way, you can “break” sprees through abdications by skipping buildings. Even if the seed is the same, the result is different. 100 cast results are not very relevant.

(If the seed is the same, that would actually make it easier to exploit the Weather Service; may be interesting to look into that.)

 
Flag Post

Topic: Realm Grinder / How "random" is LS really ?

True, but you’re generally not doing 100 cast LF runs, and if you are, those 22 miss sprees won’t be your main concern.

 
Flag Post

Topic: Realm Grinder / grand balance was kill

It’s not horrible, is it? Using it for e66 to e69 before going to IGOT for e69 to e72 for R16. So far the progress is pretty steady and a 32h run will easily get me to e69. The Wiki notes a 32h setup time for IGOT and that you need e69 for IGOT to get you to R16. None of the other Wiki builds work well. Drowcenaries is horrible, Priapism is easily dominated by Lightning Forge and falls off quicker, Dwarf Star was never great to begin with and who knows how well it does since the GB change.

 
Flag Post

Topic: Realm Grinder / How "random" is LS really ?

Assuming the actual odds of hitting a specific building are 1 in 11 (it should be, but all implementations have some bias), we can calculate the odds of a bad spree (eg. no Forge hits in 20 casts) easily. The odds of “failing” (don’t hit forge) are 10/11. The (a priori) odds of failing twice in a row are 10/11 * 10/11. This generalizes to P(f fails in a row) = P(f)^f. E.g. OP expects to see 1 Forge hit for every 11 casts; however, the odds of 0 Forge hits in 11 casts are (10/11)^11 = 35%. For 0 hits in 22 casts, that’s still over 12%. As why_amihere stated, to be 99% sure of hitting at least one Forge strike, you need to do more than 48 tries.

 
Flag Post

Topic: Realm Grinder / The (more) Reliable Lightning Forge build

Originally posted by digitrev:

Well, good news. A new weather service is up and running.

Oooh brilliant! Thanks for the heads up!

 
Flag Post

Topic: Realm Grinder / grand balance was kill

It bothers me that Lightning Forge is basically the only viable build during R13-15 now, aside from the final IGOT/Cialis run. I’d rather have seen some change to break Lightning Forge to shake up the entire process of R3-15, but Mercs don’t receive any buffs from Challenges until R25 so no other builds have become useful to replace LF.

 
Flag Post

Topic: Realm Grinder / [Minor Feature Requests]

1) I’d love the ability to set the contingency/planned/whatever casting point to a specific number rather than fiddling with the arrow. It gets a bit difficult to be precise when you have 3-4k mana, and since the spell scheduler only checks once per second, you can lose a full tick of mana easily.

2) Average/expected number of Faction Coins for multi-excavations. It’s an easy formula for a computer, but difficult to do in your head.

3) Auto-buy function. Just like auto-casting will cast a spell whenever mana is available / the contingency threshold is reached, the auto-build function would just buy the building whenever sufficient coins are available. This gives a bit more power to idle builds, especially at lower R & gem levels where a few extra buildings still has some effect. A similar feature for Faction Exchanges would be nice too.

4) Maybe against the design concept of the game considering formulas are “hidden”, but it would be nice to have a “required until next” stat on bonuses that have discrete steps, such as assistants. E.g. the Undead Heritage could show (17) (4h to next), Dwarf Heritage could show the clicks, buildings, spells and excavations to reach the next assistant, etc. There’s a few skills that have similar workings that could benefit from this.

 
Flag Post

Topic: Realm Grinder / fastest way to get efficient autocasting? (highest mana build)

How on earth did you get to R24 without efficient autocasting. :/

I’m at R14 and already running long-term runs that depend on autocasting…

 
Flag Post

Topic: Realm Grinder / New alignment/factions idea

Dragons seem pretty weak, Basilisk really powerful. +1 base per second is really weak. The T1 Fairy skills add minor values, but those skills are useless by themselves and mostly taken for the reduced building cost. The Demons have a +200.000 base for Hell Portal & the other building. Dragons would need 200.000 * 2/11 ~= 10hours run to match it. A few Dragon skills would be incredible for Merc builds combined with Titan & Demon traits, but by themselves they don’t really seem strong. Might be underestimating the way bonuses stack (the Dragon (*, 3) traits are hard to judge).

I like the concept of new factions (always interesting) but they won’t last long. I’m kind of disappointed overall with the fixed Vanilla → Ancients → Prestige → Merc line. I know Research shakes things up (haven’t reached it yet, so can’t comment much), but adding a “okay, now use these base factions again” phase doesn’t really add much to the game. Either these new factions’ skills are unavailable to Mercs and you’ve just added a new phase, or you unlock their skills for Mercs and it just becomes a quick unlock before going back to Mercs.

I think expanding the game at this point is more about new mechanics, not just new “parallel” content. Guessing Merc-Research will be next.

 
Flag Post

Topic: Realm Grinder / Goblin 1st Challenge

In my experience, Drow/Dwarf doesn’t matter for T1 and T2 challenges (haven’t reached T3 yet). As long as you have the right faction and their spell upgrade, and are in R2+, you can do the T1. Add in R8+ and their Bloodline and you can do T2. I haven’t had any problems with having to deactivate challenges or anything, I got Angel2 as a Dwarf and probably most other T2s as a Dwarf/Drow too.

Goblin1 is no different from Undead1: just wait it out. I had it in one short night (woke up randomly after 4hours, noticed I had it). According to the Wiki the proc is uniformly distributed in [1,1200], so chances of >1000 are 1/6 every 10 minutes. If you’ve spent 16h online and didn’t get it, you’re definitely missing a requirement (Spell upgrade/R2+).

 
Flag Post

Topic: Realm Grinder / [Solved] Bug? Lost Gift of the Heroes.

Restarting browser did it. Thanks, Shello!

 
Flag Post

Topic: Realm Grinder / [Solved] Bug? Lost Gift of the Heroes.

I purchased Gift of the Heroes (the first tier, 50Kreds premium upgrade) at R0. Now, after unlocking Mercenaries in R3 I noticed I no longer have the upgrade. I don’t have the “boosted start” after abdication, it’s no longer showing in my Upgrades tab and it’s up for purchase again in the shop.

Is there any place I can report this to?

 
Flag Post

Topic: Crusaders of the Lost Idols / Suggestions

Can we get the fully detailed number of “Crusader levels bought this reset”? It will never reach ridiculous numbers and the last digit is still pretty useful to know how far off the next Idol you are.