99 precision 02/13/2020 04:02 PM CST
Just wanted to let everyone know it is possible to get a max precision sigil. I harvested this about 4 days ago:

On this scroll you see an exquisite (87) congruence sigil etched in delicate detail (99).

I'm not sure if the 3rd sigil harvesting tech is active or not but I do have it. Still not easy to get an 80+ sigil. I was very lucky.
Reply
Re: 99 precision 02/14/2020 08:22 AM CST


I was seeing better than usual congruence signals too, but only congruence. Can you do the same with a decay sigil, for example?

Out of curiosity, what are your skills, which buffs, and how many hours of farming did I it take?
Reply
Re: 99 precision 02/15/2020 02:32 PM CST
Arcana: 868
Scholarship: 771
Outdoorsmanship: 664

A few days before I got the 99 sigil I was lucky enough to get a 90 integration and rarefaction. Saw from the remaining *** I had that it was possible to get higher. I want to say I spent over four hours getting the congruence sigil so not easy at all.
Reply
Re: 99 precision 02/16/2020 08:38 PM CST

TL;DR: So I've been doing some testing, and I'm convinced you've just had astronomical luck. It's so ridiculously high that I don't see repeating this unless there are secondary factors at play.




For sake of argument, let's ignore all of the ways the system can screw you over.

Potential Points Of Failure (other than the game itself):

- Initial precision is low (Range: 1-15 - low starting points GREATLY decrease your chance).

- Critical failures on an improve attempt (prematurely blow up the sigil).

- Danger gets too high due to not being presented with enough precision improvements (too many steps, indirectly affects quality)

- Low C/P ratios (pool cost / precision).

- Being left with a low number of stars in a single pool (wasted precision options, increases danger). Trivial still will work 33% of the time with 1 star, but straightforward requires 2, challenging/formidable 4, and difficult 7, minimum, with an option to blow up if they're in the upper ranges.

Additional Assumptions:

- Assume secondary sigils are not inherently more difficult than primary sigils (they seem to be).

- Assume one pool is not more heavily weighted than the others (and thus indirectly increases danger).

- Assume the sigil feats are turned on and working as intended (I'm still not convinced they are).

- Assume there aren't unseen, complicating factors such as location, time of day, or season.


Why This Still Seems Unlikely:

Now. Assuming you don't have ANY of those problems, you still only have 45 points to work with. That means, in an ideal situation, you need an average of 1.86 - 2.178 C/P (pool cost / precision improvement) to hit the 99 precision mark. I took a sampling of the last 100 improve attempts on multiple decay sigils. Here's what I'm seeing. (C/P data)

- Minimum: 0.29

- Maximum: 3.00

- Average (total): 1.62*

- Total % > 1.86**: 37%

- Total % >= 2.178**: 14%

*Note that if none of the potential points of failure existed (beyond just using the bucket), one would expect to get to find 75-85 precision sigils regularly. This just doesn't happen. Most sigils are in the 40s range, with some in the 50s, and very rarely in the 70s and 80s. I don't see how someone would expect to find multiple 99 quality sigils with these numbers with any sort of regularity. At least 50% of the time, you'd

**Note that % is misleading here, but I can't think of a better way to represent it. You need several high value, high impact improves. Multiple trivial improves with very high C/P ratios (critical success) will still blow up before they get to the desired range.

Data: https://pastebin.com/u2ZbSZUT

Suggestions For Improvement. All would be great. Any would be nice.

- Raise the floor for new sigils. Always start them at 15.

- Only blow up a sigil when danger >= 18.

- Give another source for sigils, or improving sigils.

- Reduce the range of improvement attempts to always be the minimum. No more 1-3 stars for trivial.

- Have a GM look at and confirm the sigil feats are enabled, and also confirm if some sigils will be harder to improve than others.

- Create a way where moving points from one bucket to another (ie: sanity > resolve) will always be a net positive. This lets danger be the only "gotcha" you have to watch, and it makes the game more engaging.

- Create a series of non-signature spells (or tie into existing spells) to reduce the cost/danger/RNG of certain actions. eg: having mental focus up reduces focus costs or increases focus pool to 20. Divine Guidance/new spell for sanity. Redeemer's Pride for resolve. Aura Sight could increase the starting quality. Finesse could add a few points when you scribe it. Or just a set of new spells entirely sprinkled around or AP.
Reply
Re: 99 precision 02/16/2020 08:39 PM CST


Edit: that should have been P/C, not C/P. Points of precision per star cost.
Reply
Re: 99 precision 02/17/2020 03:51 PM CST
Well here are a few tips that might help:

1. Don't bother with sigils that start at anything lower than 11 or 12 precision. Find 15 or so rooms with the sigil you want. Make a script that goes to each one and pauses when it finds a good one. Then just pause the script and improve the sigil.

2. You can still improve with max danger. Did it three times in a row once. Very unlikely but possible. Increased luck "might" help. You can also keep improving at max danger without risk until you find the desired thing you want to improve. The drawback is you will keep losing clarity.

3. Don't worry about clarity when precision sigil hunting unless it's lower than 50 or 60. I've made plenty of mastercrafted items with low clarity sigils.

4. Avoid trivial and straightforward unless you don't have enough stars to do a challenging or difficult. I'm iffy on formidable. Sometimes it will use up a lot of stars for only 4 precision but you could get 9 or so.

Hope this helps a little. Still ALOT of luck involved.
Reply
Re: 99 precision 02/17/2020 09:12 PM CST


For what it's worth, my scripts already do 2 and 3, and have you actually tested #4 or is it an assumption? My testing show a high precision return for stars spent when using trivial and standard. Most of the danger increases come from improves. #1 isn't a bad idea. I'll incorporate it. And have you actually seen a 90 or above for secondaries? The highest I've had so far is an 85.
Reply
Re: 99 precision 02/19/2020 02:08 AM CST
The 90 integration I mentioned getting in my reply above was a secondary sigil. Easy to mistake with induction. :)

Danger increases at the same rate whether you improve, approach, or whatever. You can do it 6 times before you see the first danger star and 13 until you get to the max 18 danger stars. There is a bug with ACTION where it might not work the first two times you try it but danger is still increased.

Increasing challenging and difficult is just a strategy of mine. I figure that if you only get 13 or so to play with then using it on trivial or straightforward is a waste except in special circumstances since it only increases precision a small amount. Also if you are just increasing challenging and difficult then you just need to be lucky (getting 15 or 16 on a difficult for example) 6 or so times instead of however many it would take to get to a high number with trivial and straightforward.
Reply
Re: 99 precision 02/20/2020 08:25 AM CST


I appreciate the response. I modified my improvement script to no longer bother improving anything below 10, dropped trivial, and even played around with dropping "improve sigil action" entirely. The largest gain was time saved in not improving sigils which are dead on arrival (<10 precision). I can't tell if there's a benefit to dropping trivial improves, but it seems to be mostly lost in noise. Dropping action, even with the bug, seemed to hurt my average improve by about 5 points. I also started using speculate luck, which seems to help, but again it's unclear if that's noise due to a low volume of data.

All said, I've managed to find about 1-2 80-85 sigils after 8-10 hours of farming. Not great, honestly, but it may be on par with rare metal finds from mining. The number of total steps per sigil seems to be consistently between 9 - 13, with some outliers on both sides.

Looking at the napkin math, this is probably right.

- You only want a sigil that starts in the top 33% of options, maybe higher, honestly.

- You get 3-7 options each improve.

- Of those options, you have 60 possible options. That's 3 possible source variables, with 4 possible destination variables, and 5 possible difficulties.

- Of those 60 options, you only want to act on 12 of them. That's 20%. [ Challenging, difficult, formidable, and possibly standard improvements on precision. Assuming you can use them all thanks to buckets (you can't) ]

- Assuming you can spend 75% of your total stars, that gives you ~33 stars to work with.

- The 75th percentile on P/C (precision improvement per star cost) is 2.0, so with a 25% chance, you need to use all 45 stars to get a 99 sigil.

- The average number of steps per sigil is 12.6. The 3rd quartile is an average of 14, which is likely required to reach the right threshhold since we're including standards.

Okay, so let's sum this up (admittedly with some hand waving).

- Chance sigil starts correctly * Chance of finding a usable action * chance of action having a high enough P/C * chance of reaching a high enough number of steps =

- .33 * .2 * .25 * .25 = 0.4%. Or 1 in every 250 improve attempts.

- Removing standards, and this changes to .15 * .2 * .25 * .50 = 0.375% chance. So it makes sense to keep standards.

I feel like I'm missing something, there's one or more hidden rolls, or there's an error in my numbers somewhere, as this seems too high of a chance. It means a perfectly executed run, for 13 straight hours in a high density area, should give you a high quality sigil (on average) once every 13 hours, with techs. I'm just not seeing that. I'm seeing an 80+ find in roughly that timeframe.

TL;DR: There Needs To Be An Alternate Way To Get High Precision Sigils. This craft was designed to be a consumable craft, with the base materials being the primary limiter; however, the sigils take too long to farm (assuming everyone can) to make that an option.
Reply
Re: 99 precision 02/20/2020 09:39 AM CST
>All said, I've managed to find about 1-2 80-85 sigils after 8-10 hours of farming. Not great, honestly, but it may be on par with rare metal finds from mining.

For reference, assuming that ultra-rare are the equivalent of the mythic 99 sigil, based on 8012 rooms prospected so far I have found ultra-rare veins 3 times. That puts my chance of proc'ing a useful metal at less than 0.09%. The average of my last 5 prospect run (171 mine rooms) is 88.23 minutes. So I've spent roughly 250 hours to find 60 volumes of material.

~Hunter Hanryu
>I would like to avoid the collection of broken dreams and sorrow that is the Ranger guild.~Agalea
Reply
Re: 99 precision 02/20/2020 11:44 AM CST


The more I think about it, the more I don't think it's a great comparison. The ultra rare material is found in the same location for all crafts. With enchanting, you need a second ultra rare sigils just let you use that ultra rare material.

If we were to compare this to other crafts, it's like saying that you need to spend days with 1000+ ranks farming for an ultra rare oil to complete your forging product. If you don't have it, you're guaranteed to add penalties to your end product. It also goes dull every 6 months, so you need more of this magic oil to keep rebuilding it.

Unlike forging, the rare sigils are required to make common materials. Balsa would be LCS equivalent and require 70 potency sigils (a comparison to gold in mining). You basically need the equivalent of glaes to use a unique material like orichalcum to produce MCS quality producs. Other crafts just get to use it out of the box. You need super rare, MTX, unique materials to produce the glaes or tyrium equivalent.

I'm on the fence about whether or not I think this is good design. I mean, I actually enjoy playing with the system, but rarity of sigils seems way out of whack with the products you can create with it.
Reply
Re: 99 precision 02/20/2020 11:51 AM CST
I tend to agree that sigil precision is erring on the side of too low right now. Maybe it'll get better if the remaining harvesting techniques get turned on. It seems a little off to me that the sigil precision starts so low in the first place though. I feel like it should be possible, even common, to be able to get store-bought level without having to go through the improvement process. Maybe 15-30 would be a better starting range?

Thanks,
-Biomancer Karthor
Reply