[X] My Assistant

There are - but they involve rather more work on the programmer's part, and why should he have more work if he can just shove the work off to us instead? (IMG:[invalid] style_emoticons/default/dry.gif)

Exactly what I was talking about. It was clear that this step involves less time and effort for programmer, but shoves all work to us. I feel like being in WoW...

This post has been edited by Kralj: Jun 3 2013, 11:52

Exactly what I was talking about. It was clear that this step involves less time and effort for programmer, but shoves all work to us. I feel like being in WoW...

It isn't really a matter of time or effort gone into programming... If it was the case, implementing and keeping track of hunger/morale leves for 27000+ monsters takes a lot more time than to leave everything as ti was.

It was a matter of nerfing those who had a lot of accounts and monster teams for the sole purpose of materials/bindings harvesting and thus amassing a lot of credits selling them at auctions for the rarest monster gifts (Binding of Slaughter, Of Focus at the time, of the Owl, Of Destruction, Of Isaac, Shade fragment, Crystallized Phazon, Kevlar piece and the High type Wood, Leather, Metals and Cloth) or forging their gear with them to ungodly proportions.

Exchanging Morale for Stamina or tying your monsters appearance at fights to the player's stamina level(and having you go to the lab to retire a monster from combats until fully rested and/or given food) requires minimal reprogramming and would serve to lessen the server loads and make multi-accounting an unrewarding scheme unless you actually spent time grinding at that account while the other accounts' monsters just don't do squat nor gather any gifts.

doomgaze: To reiterate: Not a balance change. It was just getting expensive to generate a battle. Anyway, problem with using stamina level or something like that: it is even more offensive to casuals, who currently can sell the gifts and buy food/crystals/pills without much playtime needed.

Kralj: Some things are thematic, such as morale(it is obvious that only food is needed to solve the aforementioned issue).

But tenboro is one guy with a day job whose main own project is managing the e-hentai, and HV is just a small side project of that. Frankly, I welcome chance to take load off him.
Quite an unwelcome reception he gets too, tbh, which is why he is currently boycotting HV.

doomgaze: To reiterate: Not a balance change. It was just getting expensive to generate a battle. Anyway, problem with using stamina level or something like that: it is even more offensive to casuals, who currently can sell the gifts and buy food/crystals/pills without much playtime needed.

I AM a casual player and having stamina replace morale is leagues better than what we have right now.

Your statement contradicts itself... A casual player doesn't earn enough crystals, credits, tokens, food, pills or equipment drops or have his/her monster(s) bring a lot of gifts due to lack of said resources (food, pills, crystals, tokens). It also is almost impossible for a casual player to earn as much as you state from only a few battles, arenas, IW or Grindfest unless he/she has overleveled Scavenger, Quartermaster and or Luck of the Draw (which inherently require a LOT of credits, hence fights and time spent) to be actually feasible.

If players have stamina as a kind of "showstopper" or to avoid the WoW syndrome, replacing morale with stamina will still make the hardcore players manage their monster team more often (hunting/auctioning/"shrine-ing" for happy pills as a substitute for energy drinks) and have casuals like me just send their monster team to "rest" until we can actually play or the monsters return to their "stables" automatically when low on stamina or very hungry.

That schema serves two purposes:
1. Keeps the Hardcore players' monsters active and bringing items back for as long as they have resources/time to spend on them and discourages multi-accounting.
2. Keeps monsters from casuals "on the bench" until they can actually send them into the "wild", alleviating server loads and keeping it fair across both styles since both can have items brought in... Just at different paces.

I was referring mostly to tying it to having to play - which is unavoidable with stamina.

However, you don't need to fight a single round to obtain happy pills, food or crystals. As long as your monster brings in more than 1284 credits worth[¢] of stuff per day you can still just sell the stuff monster brings[¥]. Not so with tying morale to player stamina which requires players to do rounds to have their monster appear.

¢: Current item shop bot price of happy pills/7.5 + 4*current item shop price of edibles.

¥: Albeit I'm not sure how viable this is with PL25s with their half lowgrades of mats and two days per gift, the binding chance for all monsters is equal now - and the average price[€]for one is few times higher than the cost of upkeep for two days.

€: 7776c at the moment.

But tenboro is one guy with a day job whose main own project is managing the e-hentai, and HV is just a small side project of that. Frankly, I welcome chance to take load off him.
Quite an unwelcome reception he gets too, tbh, which is why he is currently boycotting HV.

Speaking of that, there aren't very many "Thank You" posts in the 0.76 thread for the 0 credit Ability reset and
the extra Battle Items added (IMG:[invalid] style_emoticons/default/dry.gif)

"Hey, Lama, hey, how about a little something, you know, for the effort, you know." And he says, "Oh, uh, there won't be any money, but when you die, on your deathbed, you will receive total consiousness." So I got that goin' for me, which is nice."

Bill Murray - Caddyshack

I was referring mostly to tying it to having to play - which is unavoidable with stamina.

However, you don't need to fight a single round to obtain happy pills, food or crystals. As long as your monster brings in more than 1284 credits worth[¢] of stuff per day you can still just sell the stuff monster brings[¥]. Not so with tying morale to player stamina which requires players to do rounds to have their monster appear.

¢: Current item shop bot price of happy pills/7.5 + 4*current item shop price of edibles.

¥: Albeit I'm not sure how viable this is with PL25s with their half lowgrades of mats and two days per gift, the binding chance for all monsters is equal now - and the average price[€]for one is few times higher than the cost of upkeep for two days.

€: 7776c at the moment.

Maybe an elaboration is needed...

I suggested making the monsters have their own stamina and automatically or interactively 'go into their stables' when high on hunger and low on stamina, thus being useful to both hardcore players and casuals.

Additionally, your other statement bears no logic... A 0 morale monster doesn't bring in anything at all, no matter its level, so casual players are FORCED to buy happy pills at outrageous prices, since they are figurine-like über-rare drops, which are as unaffordable on a constant basis as the thousands of crystals needed to increase the morale of a monster to at least a green bar which are even more unaffordable.

Having only one monster can be a solution... If you plan to play casual that is...

And certainly we do have to thank the GM for the 0 credit ability reset and the battle items... He only deleted our "shrined"-artifact-gained ability points without reimbursement of those artifacts and those battle items we had already unlocked/earned but were taken away god knows why...

This post has been edited by doomgaze: Jun 4 2013, 16:56

so casual players are FORCED to buy happy pills at outrageous prices

Why not buy crystals instead? They restore more per credit spent.

Why not buy crystals instead? They restore more per credit spent.

Here's an example:

My main monster:

Raidou Kuzunoha, Humanoid, PL 289.

Primary Attributes:
STR +7 92 Crystals for upgrade: 1100
DEX +5 150 Crystals for upgrade: 455
AGI +5 120 Crystals for upgrade: 455
END +7 92 Crystals for upgrade: 1100
INT +5 120 Crystals for upgrade: 455
WIS +5 105 Crystals for upgrade: 455

Elemental Mitigations:
Fire +16 20 Crystals for upgrade: 429
Cold +15 18 Crystals for upgrade: 339
Elec +14 17 Crystals for upgrade: 268
Wind +15 18 Crystals for upgrade: 339
Holy +15 18 Crystals for upgrade: 339
Dark +26 14 Crystals for upgrade: 4498

Current stock:
Crystal of Vigor: 847
Crystal of Finesse: 335
Crystal of Swiftness: 300
Crystal of Fortitude: 370
Crystal of Cunning: 78
Crystal of Knowledge: 315
Crystal of Flames: 211
Crystal of Frost: 67
Crystal of Lightning: 224
Crystal of Tempest: 131
Crystal of Devotion: 255
Crystal of Corruption: 561

Crystals' credit investment needed to restore morale to at least a green bar status: ~22150 credits spent on just my main monster. Advantages? None aside from partial availability at acquiral of crystals. Just one monster two more to upgrade which only makes things more expensive.

Total cost of 3 happy pills even at item shop bot prices: 23,115 credits and a total of three happy monsters. Advantages? None aside from the significant boost to morale but too rare a drop or to trust item shop bot to get at bid price at any given moment or players foolish enough to sell them to the shop for naught.

This situation is too unbalanced for casual play as it requires increasingly longer credit harvesting sessions just to keep up with the maintenance costs, which seem directed to make players delete their monsters.

Conclusion? Crystals are not cheaper in any way but are an increasingly more expensive way to sustain morale.

So much whining. You can have the new nickname 'NightWishMan.' Yet you continue to level up (10+ levels) since the patch was instated, which means you continue to play. You must enjoy the game somewhat, right? The new patch is meant to upgrade/maintain your monsters if you want gifts. If you don't want gifts, don't check them, they won't die. In any case, Happy Pills are dropping everywhere (that means it's free!, no credits spent!).

You can also donate, accept bounties, complete taggings, etc, to get Hath/credits, instead of complaining about a free (It's Free!) game. This game is more interesting than some $50USD Wii games, and it's completely FREE!

Also really fun for those people who bought artifacts to farm ability points. (IMG:[invalid] style_emoticons/default/mad.gif)

I had 36 artifacts saved, now they're all useless.

I had 36 artifacts saved, now they're all useless.

You have maxed out bonus stats?

So much whining. You can have the new nickname 'NightWishMan.' Yet you continue to level up (10+ levels) since the patch was instated, which means you continue to play. You must enjoy the game somewhat, right? The new patch is meant to upgrade/maintain your monsters if you want gifts. If you don't want gifts, don't check them, they won't die. In any case, Happy Pills are dropping everywhere (that means it's free!, no credits spent!).

You can also donate, accept bounties, complete taggings, etc, to get Hath/credits, instead of complaining about a free (It's Free!) game. This game is more interesting than some $50USD Wii games, and it's completely FREE!

I "continue to level" mainly due to the outrageous amounts of XP at each Dawn Of a New Day, some hourlies fought before noticing they were at hard difficulty and the IWs were completed whilst oblivious of the disappearing of the "easy" difficulty... I continue to play to see if the patch changes into something not so unfair or if it will be time to abandon it completely. EDIT: At the "hourlie" fought at 9:49 pm, my time, 43896 XP got on normal difficulty... GEEZ!

My complaint is about a feature that was once fairly available to all and then got costly all of sudden. And by costly it refers to "time costs"... I'm a casual player and it shows by my current level vs time of joining... Unlike others, I can't dedicate countless hours to grinding for XP, Credits, HATH, Crystals, Food or Happy pills... My use of HV is as a relaxation tool and to literally waste time when in between processes at my line of work when there's nothing else to do but wait.

I am a player that has been around quite some time and this is the second time Dawn of a New Day XP saw such a ridiculous increase... Sweetybell... Honestly... 8 or some days ago I got 49k some XP @ DOAND and it has almost tripled since... Today almost 110k XP from DOAND alone. Some would welcome such a chance to easy level up, for me it is most inconvenient since Monsters get increasingly more powerful and my gear just can't keep up... Now my monsters can't keep up either...

And for the record, it is not a matter of credits (Credit balance is 1,200,000~some) or earning credits (Still got some 80 older artifacts and 10 new ones that could be sold for quite some credits and almost 600 pieces of Fine and Superior gear pieces)... It is a matter solely of time and fairness across playing styles: Hardcore and casual.

This is a pasttime for me so for it to turn into something only dedicated players can enjoy is certainly a reason to complain.

Why not consider what the other players like me's situation is like instead of using the hardcore player's as a basis?

This post has been edited by doomgaze: Jun 5 2013, 04:50

I am a player that has been around quite some time and this is the second time Dawn of a New Day XP saw such a ridiculous increase... Sweetybell... Honestly... 8 or some days ago I got 49k some XP @ DOAND and it has almost tripled since... Today almost 110k XP from DOAND alone. Some would welcome such a chance to easy level up, for me it is most inconvenient since Monsters get increasingly more powerful and my gear just can't keep up... Now my monsters can't keep up either...

My favorite part of the whole circumstance is that you're probably at active status just because of your complaints about the patch, if you weren't before (the 20-posts in a 30-day window dealy), which raises your XP gains for DOAND.

In other words, thanks to your posts, your unhappiness with the patch gives you more XP.

I can't help but laugh at that.

From the title alone, I shouldn't be posting in this thread. It's working fine for me, but I was aiming towards light before (which got boosted) and I'm low enough level that sub-optimal strategies work fine for me. The only thing I dislike is that before the patch, I'd choose spell selection based on what monsters were weak to what, and now, I choose spell selection based on what I can hit, because I level too quickly to get to know player monsters.

I'm eagerly anticipating the monster type icons. They'll help with tactical maging (as opposed to strategic, like it was before the patch).

That is intentional. There are about 44000 pets in the game. This creates significant computational expense every time a battle is generated that was unforeseen when Monster Lab was added in the game. Furthermore, many of them belong to inactive or banned accounts - about week after the patch, only 27 thousand were fed.

If monsters are being stored in such a way that a lookup on 40k data entries creates any sort of meaningful overhead, then there is something very, very wrong with the algorithms/database being used.

I don't know the actual programs behind the code, but you're forgetting:
a) there's math done on those 40k monsters.
b) packet size limitations.
c) it is not done once, but by ~1k times for each user who plays HV and does daily arenas. Several times that on users who do grindfest/item world.
d) the core system was originally intended for around 25 monsters.

This post has been edited by Lement: Jun 5 2013, 09:32

If monsters are being stored in such a way that a lookup on 40k data entries creates any sort of meaningful overhead, then there is something very, very wrong with the algorithms/database being used.

60k, actually. So tell me, oh great and wise one, how would you dynamically pick n weighted random entries from a table with 60k rows while satisfying a binary condition (must be active), variable range constraint (powerlevel) and an exclusion constraint (not encountered recently) with near-zero overhead? Keep in mind that it's already split out as a separate lookup table with only active monsters, and indexed for range scans on the powerlevel.

It wasn't an major issue yet (~0.01 second query execution), but considering that this was done dozens of times per second, and these servers are supposed to be doing other shit as well, it was fast becoming one due to the ballooning monster count.

Besides, it was not the driving reason for the morale/hunger system, but it was a factor.

I dont forget and i appreciate what tenboro did and does for us
But as mentioned earlier i spend my time here for fun and relaxing
That is gone
i dont want to get that ambitious but if Tenboro wants to lead this
into this direction - its his right

60k, actually. So tell me, oh great and wise one, how would you dynamically pick n weighted random entries from a table with 60k rows while satisfying a binary condition (must be active), variable range constraint (powerlevel) and an exclusion constraint (not encountered recently) with near-zero overhead? Keep in mind that it's already split out as a separate lookup table with only active monsters, and indexed for range scans on the powerlevel.

It wasn't an major issue yet (~0.01 second query execution), but considering that this was done dozens of times per second, and these servers are supposed to be doing other shit as well, it was fast becoming one due to the ballooning monster count.

Besides, it was not the driving reason for the morale/hunger system, but it was a factor.

Hey man, don't get so defensive. I'm no more wise than you are, However, if operations on tens of thousands of data points are causing speed problems, that is usually a red flag that somewhere in the code you are probably doing something very inefficiently.

It happens to everyone, so don't feel like it is a personal attack for pointing it out. As for solutions, I couldn't tell you without look at the code. Generating the round compositions in batches might work better than doing them each round. Looking for ways to store the keys of active mobs in global memory, hashed on PL, might also be good, if you aren't already. There are all sorts of tricks you can do to reduce computational overhead.

Then again, this is a side project, so it may require more time than you are willing to invest to optimize. (IMG:[invalid] style_emoticons/default/smile.gif)

Hey man, don't get so defensive. I'm no more wise than you are, However, if operations on tens of thousands of data points are causing speed problems, that is usually a red flag that somewhere in the code you are probably doing something very inefficiently.

Not necessarily. Picking random elements from a set is an age-old problem in relational databases, and while variants of ORDER BY RAND() are fairly expensive, they work perfectly fine for the few-tens-of-thousands rows it was targeted for. As of the usual work-arounds..

Just picking one PL to do a direct index lookup instead of an index scan would require running the query once for every monster you wanted to pick, which would be slower.

Building a table with sequential IDs and rolling on that doesn't work because of the range and exclusion requirements.

So we're stuck with that.

It happens to everyone, so don't feel like it is a personal attack for pointing it out. As for solutions, I couldn't tell you without look at the code. Generating the round compositions in batches might work better than doing them each round. Looking for ways to store the keys of active mobs in global memory, hashed on PL, might also be good, if you aren't already. There are all sorts of tricks you can do to reduce computational overhead.

Precisely - which is why you don't state with such certainty that "there is something very, very wrong with the algorithms/database being used" when you don't actually know how it works.

Generating monsters for several rounds at the same time might delay having to find a real solution, but it doesn't solve the underlying problem, and would require carrying those monsters around in the battlestate, adding overhead there. If HV used dedicated server software it might have been a good idea, but since we're running it as PHP processes with no internal persistence between turns, you have to take stuff like that into account.

(And no, I'm not writing a custom server infrastructure for it.)

And the actual initialization data for the monsters is in fact already stored in memcache. But anything that requires range scans on what is strictly a key-value store is very inelegant, and additionally requires overhead (though mostly development-wise) for code to maintain the values in question. It would also increase intra-node communication on the cluster instead of using the handy-dandy local replicated database copy, so all in all, I don't think you end up saving much in the end.