how to sitesearch effective ?

[Boron]

Originally posted by Cainehill:
One thing about putting Astral-3 on a pretender : my advice is to only do it on an immortal, or possibly a human rainbow.

Astral 3 on a pretender that's going to be out there in combat makes it far too easy to kill using piddling Astral 1 or Astral 2 mages. If you're immortal in friendly domain, this doesn't matter as much, and if you're a human rainbow, mostly used for early site searching, research, forging and casting from somewhere safe inside the capitol, it also doesn't matter as much (if they're coming after you in your home fortress, you have bigger problems already).

Because of this, I recommend either going all out on a combatant Pretender ( Astral 7-9, thus making it much harder to get a cheap kill against ), or none at all. After all - your pretender probably has better things to do than cast Akashic Record anyways.

If you can just get a mage with astral 1, it can be worthwhile to empower it once (for 30 astral gems) and then create a starshine skullcap (10 more astral gems), and then you have a mage who can cast your Akashic records. Considering that Akashic Record costs ... 25 astral gems a cast, 40 gems to get the ability isn't all that extreme; less than two castings worth of gems.

yeah i would do this with e.g. atlantis

but my main problem is that it seems to me in order to find enough gemincome as ermor i have to take a rainbow pretender ( most likely a ghostking ) to find sites and give him astral 3-4 that after 5-10 searches which equals turn 15-25 he can cast from time to time an accashic savely from the capitol
cause i looked at the magic sites document and there are unfortunately many many sites which produce death gems together with other gems .
i tried a death 10 rest on fire PoD and searched through dark knowlegde but found almost nothing .

so now that you don't have the option of a combatpretender in form of the vq for ermor best option is a first sitesearching than summoning rainbow ghost king would you agree ?

Are you saying that a death mage or Dark Knowledge won't find a fire-1/death-1 site? Not sure I understand.

[Saxon]

Has anyone done a cost benefit analysis on site searching with ashasatic record? At different settings of site frequency? At different turns in the game? Why did Saber Cherry get a life and leave us without such wonderful modeling skills?

My thought is that it is an expensive spell, 50 gems worth. You need to get a common currency to determine value, so I am suggesting two for one astral. It could be adjusted to take into account the frequency of astral sites, modified by clams, but that would make my head hurt.

How long does it take for this to pay for itself? Late in the game, if you find a site with 2 gems a turn, the game will be over before the 25 turns elapse and you have paid for it. At what point is it better to spend the gems on summons and items, rather than on the spell?

This would be determined by the site frequency. The higher the frequency, the faster you are likely to get your investment repaid. Has anyone looked at the code to see how the site frequency figure translates into actual sites? Also, does anyone know how the different terrain types effect the site frequency? In specific numbers? That would help determine where to start your searching.

It will also be determined by how long you think the game will Last. Map size will be important in this decision. Generally speaking, the earlier you cast the spell, the more valuable what you find will be to you.

If you can not model all of the above, which I can not, then just bear those ideas in mind as you search.

[Saxon]

Yes, a Dark Knowledge will find any site with at least one death gem associated with it. This is a nice bonus and applied to the others, like Gnome Lore and so forth.

[Boron]

Originally posted by Saxon:
Yes, a Dark Knowledge will find any site with at least one death gem associated with it. This is a nice bonus and applied to the others, like Gnome Lore and so forth.

sure ?
i read the excel chart from sunrays excellent dom site .

there is e.g. :
academy of high magics , lvl 1
type air .
produces 1 air + 1 death gem.

corpse candles marsh , lvl 1
type water
produces 1 death gem + 1 water gem

ebony tower , lvl 2
type blood
but produces 3 deathgems

and so on and so on

in a brief esteem i would say 1/3 - 1/2 of sites who produce death gems aren't death sites .

but dark knowlegde only does a search like a lvl 9 deathmage .

so all the sites like the listed examples aren't found by dark knowlegde right ?

this makes dark knowlege ihmo a pretty bad spell . at least the cost is 3 death gems und you often get luckily a d4 dusk elder which can search for deathsites as efficient as the spell .

so as a conclusion for myself if i want much magic sites i am almost forced to make a rainbow ghostking right ?

[Arralen]

Originally posted by Boron:
i read the excel chart from sunrays excellent dom site .
there is e.g. :

academy of high magics , lvl 1
type air .
produces 1 air + 1 death gem.

you'll need a level-1 air mage to find this,

corpse candles marsh , lvl 1
type water
produces 1 death gem + 1 water gem

.. a level-1 water mage here..

ebony tower , lvl 2
type blood
but produces 3 deathgems

and a level-2 blood mage to find this site.

Gems produced have nothing to do with what kind of mage you need to find the site, ATOWR .

in a brief esteem i would say 1/3 - 1/2 of sites who produce death gems aren't death sites . But dark knowlegde only does a search like a lvl 9 deathmage . So all the sites like the listed examples aren't found by dark knowlegde right ?

right.
It's just the same with the other path-specific site searching spells.

this makes dark knowlege ihmo a pretty bad spell . at least the cost is 3 death gems und you often get luckily a d4 dusk elder which can search for deathsites as efficient as the spell .

I wouldn't say it's a bad spell per se. As nearly everything in DOM, it depends. If you have a 3gems/turn income and level-1 death mage (e.g. you're playing TienChi S&A), you'll prefer casting the spell to moving the stratmove-1 commander around. With death-4 dusk elders, things look quite different.

so as a conclusion for myself if i want much magic sites i am almost forced to make a rainbow ghostking right ?

Not necessarily a Ghost King, but a rainbowish pretender for shure.

posted by Saxon:
Has anyone looked at the code to see how the site frequency figure translates into actual sites? Also, does anyone know how the different terrain types effect the site frequency? In specific numbers? That would help determine where to start your searching.

.. you may want to search this forum thoroughly, the following is just off the top of my head:

For each province and each magic path there's a (general %tage + terrain %tage) chance that there is a magic site.

general %tage is set in the game options.

terrain %tage depends on terrain (guess!!), the highest number is used when mulitple terrains apply.
numbers are something like this:
high magic (code 1024): 30%
mountain, swamp, : 20%
woods 10%
grass -10%

or something similar (wouldn't be on the numbers)

[ July 01, 2004, 10:30: Message edited by: Arralen ]

[Massena]

From Sunray's magic site database, if all sites are equally probable then:
Level 1 mage finds 40-50% of gems in a province
Level 2 mage finds 70-80%
Level 3 mage finds 90-95%
Level 4 mage/Spell finds 100%

So...a Rainbow mage (level 2 in 7 types of magic) will find 5 times (75%x7) more gems than a specific spell - they won't necessarily be all as useful though!

[Vynd]

The problem with taking a Rainbow Pretender as Ermor is that your Pretender will be the only mage you have who can make use of the gems that you find, since Ermor's mages are among the most limited that I'm aware of. So most of these other gems will probably just get converted to Death gems. That's useful, but at a 4x1 cost (2x1 for Astral gems) is it really worth building your Pretender around? Probably only over the very long run.

My advice is build a Pretender that helps you more over the short-term, and find your magic sites by taking over someone else's territory.

[Taqwus]

If Ermor can get to Conj 6, Spectres (D-1, two independent random wizardry picks) will help dispose of other gem types; and if one has a surplus of non-death gems before that empowerment or alchemy is an option as well. Plus, there are non-Death sites that can offer substantial benefits; for instance, there are non-Death sites that offer free fortifications (very nice since Dead Ermor may be hurting for gold), and sites that offer forge bonuses will be very helpful for boosting research via mass-produced Skull Mentors.

[drool]

illuminated ones and members of second and third tiers find magic sites quite easily when walking trhough provinces. no matter if its your own, neutral, or enemy. i`m not sure if all spies are able to do this. black forest ulms are, thats for sure.