New nightly

[Derakon]

Why does it have to end with "mind blasting"? Why not use e.g. "It tries to blast you with its mind"?

[Rizwan]

If you can think of a better phrase ending in the noun "mind blasting", please let me know. Or a better noun, even - but I suspect that's harder.

It gazes at you using mind blasting ... or
It uses its power of mind blasting ... or
It gazes at you utilizing its power of mind blasting

[Derakon]

I found a couple of pairs of boots of speed in my last game, so they're definitely still around. I can't speak to their overall rarity, but they're supposed to be rare, so it's not out of the question that you wouldn't find more than one in a game.

Actually this may be one situation where randarts are more reliable than standarts. It seems to be not at all uncommon to get randart boots with a decent speed boost on them...and when those do happen, often they also boost another stat (or, rarely, two stats). Getting +8 to speed and STR can be a bit silly...

[scud]

Count>10: 2600109
Count>15: 81634
Count>21: 1305

10, 10, 10, 8, 7, 10, 10... where's my 21, then?!

Now CL47 and only a single pair of BoS so far (and those a measly +6). Has the correction to the 'Supermarket Sweep Shoe Shop Special' made BoS less common than they were In The Pre-SSSSS Days, or have I just being unlucky so far? I'm wearing my Wormtongues which I found at CL43 – they and/or Dal-i-thalion always used to be waiting for me on the floor at 500'.

However I do have very attractive Chaos DSM of Permanence (rPois, Chaos, Disen). Which is nice.

[Magnate]

A couple of issues I have seen in the latest nightly. Playing as a hobbit mage.

Reveal monsters doesn't show invisible monsters. It says detecting invisible, but nothing shows up, even when I have a ghost of some sort right next to me

I quaff a potion of rpoison, my screen shows RPois, and the air hound just poisoned me. I know this wasn't the old behaviour, assuming it is a bug unless there was an intentional change.

Haven't seen much else different yet, but that is as it should be.

Thanks for the report - I think those are both bugs. I am confident I will be able to find and fix the rpoison one (because I caused it), but I don't know about the reveal monsters one. Will first see if I can reproduce them.

[myshkin]

I just ran my speed ring monte carlo. Of 6,000,000 speed rings generated at dlvl 91:

Mean: 9.00
Std. dev: 1.80
Count>10: 2600109
Count>15: 81634
Count>21: 1305

[Derakon]

Probably just coincidence. It's not that uncommon to find multiple rings of speed with the same pval; finding three in a row is unusual but not enough that I'd leap out and cry "bug!" just yet.

If you're really worried, you can turn on wizard mode and generate a bunch more just to check.

[scud]

Apologies if this has been discussed already and I missed the memo, but...

Playing OSX nightly of 14 April, I've progressed beyond cave spiders and have found three RoS. All three have been +10. Are RoS now coming in fewer flavours, perhaps +5/+10/+15, or is this simply coincidental?

[Derakon]

There was some discussion about DSM randarts awhile back. I doubt the devs would mind getting a patch that fixes the problem, but it's also probably not an especially high priority.

There's a one-to-one ratio of standarts to randarts, since the randarts use the standarts to decide how powerful an item should be. It's certainly not inconceivable to have a larger number of randarts, or even an unlimited number, but it would muck with game balance. At some point you just have to get in there and win the game, and that point should happen well before you start running out of artifacts to find.

[Spacebux]

Randart generation is already terribly slow (it will speed up when I rewrite it, but even then I worry). Effects have to have a power which is non-trivial but not excessive (something like between 5% and 50% of total power), and if we added a restriction not to duplicate, we would unnecessarily reject entire randarts. But I'll bear it in mind for the future - one day *all* spell-like effects will be part of the effects framework, and there will be lots more to choose from. (At least, that used to be takk's intention - not sure if it still is.)

Don't worry about the re-write of RandArts too much: I don't mind the slowness, actually, as it only happens once. A bit more even distribution of resists would be nice; or, as I wrote earlier, to give balance to RandArts players---slightly increase ego/flavor item appearances with resists. Increasing the # of potential effects sounds great.

They get [to-hit/to-dam bonuses] a lot because there are a lot of them in the standart set - the total is the same, though the randarts may be more findable. Both are likely to be reduced for 3.3

Uh-oh,.. there goes one of the alluring points of playing RandArts.



You may want to consider increasing the total # of MaxArts to something slightly larger than 137 and potentially decreasing the odds of finding an artifact that has a larger number of flags / bonuses.

BTW - any chance we could have breath affects added back into DSM artifacts on RandArts? It just seems odd the artifact DSM suits cannot be activated.....

[Spacebux]

If you compare the numbers between my randart set and the standart set, you'll see that a similar number, total, have protection from confusion. The problem is twofold, really: first, most of the sources were on weapons, which absolutely have to be decent damage sources for everyone except mages and the occasional priest. Second, the items that did have confusion protection were quite rare. I counted several items in the spoiler that I would have considered using, had I found them, but the ones I did find were either unusable or late.

I don't have artifact spoilers handy for my previous games, but basically the same held true: confusion protection was incredibly rare, and when I did find it, it was typically on weapons that simply couldn't compete.

Contrast that with standart games, in which you're practically guaranteed to find one of Isildur, Rohirrim, or Caspanion fairly early on -- and Holhenneth and Thranduil, which also show up reliably and early, are excellent contenders for covering blindness resistance. In other words, the standart set is designed (intentionally or not) to give you access to blindness + confusion resistance with a minimum of fuss. And of course, all of these artifacts are, if not great, at least usable in the endgame. And eventually you'll probably find Gondor, which is great -- in fact, it's freakin' absurd -- telepathy, blindness/confusion, STR/WIS/CON/Speed, and a 250HP healing activation!

One question: is randart rarity based solely on power? I wouldn't be surprised if the rarities for standarts deviate from the formula used to generate randart rarities, if so. It'd be interesting to see which standarts are more common than they "should" be, and which less.

Indeed, the RandArts game will generate far fewer RConf items than the Standard game. The last RandArts game I played, I have 5 artifacts (if I count a certain massive iron crown) with RConf in it; and I uncovered a ton of artifacts. 95 to be exact. If there are 137 artifacts in the game (maximum), then I would have uncovered approximately 70% of the artifact items in the game. Hence, at the same rate, I only had about 2.2 more items left with RConf in them. Now, I don't think I uncovered any of the power rings, or perhaps some of the more powerful artifacts, such as Crown of Gondor's equivalent, so perhaps the rate of RConf was > 5.26% for the remaining items.

Point is, though, in the Standard Artifacts game, the player is typically able to come across at least one or two artifact items to alleviate the RConf slot: Thengel or Holhenneth, e.g., are often found in the early stages. In a RandArts set up you have no idea if/when RConf is coming. Often it will not be on head gear, or early on. Coupling that with an IronMan setting, one practically clings to any RConf item one gets. I think I nearly finished with a Suit of BronzeDSM of Speed until I found Crown of Serenity!

I think if you increased the odds or depth range of regular RConf items, that would alleviate some pressure on RConf protection. Caps of Night-&-Day and Seeing, for example, help RBlind greatly in the mid-levels, until one finds a better more permanent source. 'Shrooms of Clear Mind are nice, but they are quite rare and don't solve the problem once one is already confused. Maybe add RConf to another ego item or create a new one? Serenity crowns are too rare, too deep at the moment.

Would be nice if Mages had the spell "Mind Shield" which provided protection against mind blasting and confusion, in book A, as opposed to CLW, which never should have made it into the mage book in the first place.

[Taha]

A couple of issues I have seen in the latest nightly. Playing as a hobbit mage.

Reveal monsters doesn't show invisible monsters. It says detecting invisible, but nothing shows up, even when I have a ghost of some sort right next to me

I quaff a potion of rpoison, my screen shows RPois, and the air hound just poisoned me. I know this wasn't the old behaviour, assuming it is a bug unless there was an intentional change.

Haven't seen much else different yet, but that is as it should be.
________
LIVE SEX WEBSHOWS

[Magnate]

Further to d_m's and fizzix's replies:

* With fire immunity, when I get breathed on I get the message "The Foo breathes fire. You resist the effect!" The resistance message is new; what exactly am I resisting here that I otherwise would be vulnerable to?

Inventory damage. You now get messages about resisting side effects even if you are immune to the damage which causes them. Cold immunity, for example, prevents damage from ice bolts, but does not prevent them shattering potions. It does, however, prevent cold breaths from shattering potions. It's a complex set of permutations to communicate as concisely as possible.

* I saw the message "gazes at you with mind blasting". I'm pretty sure that phrasing is new and it could be better. What's wrong with the old mind blasting message?

It doesn't fit the verb + noun structure of the current spell messages. If you can think of a better phrase ending in the noun "mind blasting", please let me know. Or a better noun, even - but I suspect that's harder.

Please note that this is not a permanent issue, as eventually the text messages will all be in an edit file (ticket #1376). At that point the verb/noun restriction should go away.

* I got two randarts that activate for polymorph. It'd be nice if randarts didn't get duplicate activations. The only duplications in the standarts are between the *thancs and the Paur* gauntlets, if I recall correctly.

There aren't enough different non-negative non-inhibited effects to do this. Randart generation is already terribly slow (it will speed up when I rewrite it, but even then I worry). Effects have to have a power which is non-trivial but not excessive (something like between 5% and 50% of total power), and if we added a restriction not to duplicate, we would unnecessarily reject entire randarts. But I'll bear it in mind for the future - one day *all* spell-like effects will be part of the effects framework, and there will be lots more to choose from. (At least, that used to be takk's intention - not sure if it still is.)

* Randarts seem to get off-weapon combat boosts a lot. Though I recall Magnate saying this is going to change.

Indeed. They get them a lot because there are a lot of them in the standart set - the total is the same, though the randarts may be more findable. Both are likely to be reduced for 3.3 ...

* I detected a Quylthulg, after which it blinked. I saw it at its previous location, then saw it update to the new location (which was still not normally visible to me) after it blinked since my next turn hadn't come yet. Is this intended?

No, but it's part of a very large can of worms we've been discussing recently. We basically need to split the storage of the dungeon into a "real" array (i.e. what is actually in the square) and a "seen" array. Although simple in concept this requires a whole lot of tidying up of the dungeon code that clearly hasn't been done for years. So pls don't hold your breath.

* When squelching ammo, the top setting is for all non-artifact [arrows|bolts|etc.]. Given the total lack of artifact ammo as it stands, this seems a bit odd. Then again, maybe there's plans to introduce artifact ammo in the future?

Definitely. But again, don't hold your breath.

* From my randarts, 9 weapons, 2 launchers, 2 pieces of armor (body and shield), and 2 rings provided protection from confusion. Compare that to the standart set, which has protection from confusion on 1 amulet (Ingwe), 4 pieces of body armor (Bladeturner, Rohirrim, Isildur, and Caspanion), 3 hats (including Gondor), 1 pair of boots (Dal-I-Thalion), and 1 weapon (Totila).

This is the fourth randart game in which protection from confusion has just not been available on anything useful -- though, this time right at the endgame I got an aggravating pike that had it. I generally consider getting blindness + confusion resistance to be an important turning point for my characters, since it allows them to ditch non-0% failure escapes. This turning point happens routinely for standart characters somewhere before 3000', if I recall correctly. Blindness and confusion are important protections, and while I don't expect every randart game to provide them, going 4 in a row without them tells me that something is wrong with protection distributions.

Not really. 4000 in a row, yes. 400 in a row, maybe. But at the moment the randart generator doesn't test for per-slot availability of any flag, neither does it test for the overall rarity of a flag (only the number of occurrences in the set). Given these two limitations, 4 or even 40 consecutive games with hard-to-find pConf is not really surprising.

I am nearly ready to rewrite the randart generator, and it will behave fairly differently. It will certainly do per-slot flag tracking, but overall rarity tracking is potentially a lot of overhead for little gain. I'll try it, and see what it costs - but it seems to me that the slot distribution is the main issue here.

Quite separately I have implemented per-slot power ratings, and it occurs to me that most resists should be rated less on most weapons. (At the moment I've only reduced the ratings on weapons for stuff like ESP, hold life, FA, SI etc.) But ironically this will actually *increase* their appearances under the current randart generator! Ho hum. One thing at a time.

Also, I just want to say that the devs are doing a great job. Keep it up, guys!

Thank you. We're back up to five or six active devs, out of a max of eight in my memory, so things are looking good.

[Derakon]

The problem is specifically it not showing up on otherwise-usable items, yeah. I don't have artifact spoilers for my previous games so I can't look at what they got, but in this game the only other artifacts I found besides my aggravating pike that had pconf were weak weapons, and Zilla's game (which I posted about in the AAR forum awhile back) the only artifact with pconf at all that I found was a pathetic axe of some kind.

[fizzix]

I'm seeing more like a 75% chance though, which is what bothers me. Again, four randart games, only one of which had pconf on an (i.e. single) otherwise-usable item, and even then only in the endgame. This is not tracking even close to how standart games normally go. I agree that randart games should be able to have different "it is hard to cover this hole" issues than standart games, but they shouldn't always be the same issues.

that could be problematic. I don't think we yet have a good randarts stats generation so it's hard to tell exactly how prevalent this is. I haven't noticed a glaring lack of rconf in my randart games, but there might have been a recent change to cause a problem. The problem is specifically rconf not appearing on non-weapon slots?