The crafting & gathering thread.

[Niiro]

Confirmation bias, string of bad luck, etc.

If you don't want to risk the chance then use Steady later in the synth and suffer a wasted turn when activating Strides so it's there for Advanced.

I can use Steady II and Hasty Hand (80%) and fail none throughout a synth then turn around and fail 3 in a row in the next.

RNG, operative word: random.

[Kawfee]

Confirmation bias, string of bad luck, etc.

If you don't want to risk the chance then use Steady later in the synth and suffer a wasted turn when activating Strides so it's there for Advanced.

I can use Steady II and Hasty Hand (80%) and fail none throughout a synth then turn around and fail 3 in a row in the next.

RNG, operative word: random.

That's why I'm asking, fearing confirmation bias or whatever the psych term is for that failure is more painful, so more memorable.

Maybe I'll start a spreadsheet tracker to see what it parses out at over time, it does seem like its much higher than 10%. Perhaps the tooltip % is not a hard number, but is modified by synth level? (I would imagine synth results, e.g. quality gained or progress made are affected by level, but not success/failure given tooltip language).

[Niiro]

Success rates are not effected by synth level. They are what they are.

If quality gain and success rate had an inverse correlation my Great Strides + 800 control Byregots would never succeed.

[ksamim]

Do we know the exact probability by which a craft will change condition to Good or Excellent? I'm trying to build a simulator with the intent of optimally solving a craft using machine learning, but I need those transition probabilities.

[Niiro]

That would be some very tedious data collection, it can't be parsed because crafting condition doesn't display in the log so you'd have to do it manually.

[ksamim]

That would be some very tedious data collection, it can't be parsed because crafting condition doesn't display in the log so you'd have to do it manually.

I presume that there is no built-in parser, as it would probably be against the ToS for hooking into the game? That is going to be problematic, then.

[Twilight Sky]

Must have done 2,000 synths and got exactly 1 excellent.

[ksamim]

Must have done 2,000 synths and got exactly 1 excellent.

You probably live in the boonies of the probability function. You need to pray to the RNG gods more, obviously.

[Niiro]

You've probably gotten plenty and simply not noticed them.

There's no skill that lights up to Excellent like TotT and when going thru the motions it can be very easily overlooked.

[Raldo]

I'd have to second what Niiro said. I've actually had several synths that threw Excellent up twice.

For those unaware (since I don't see it in the OP):
• Poor: Quality x0.5
• Good: Quality x2
• Excellent: Quality x4

[Seravi Edalborez]

Had 3 Excellents once. Felt so good. Or just being lucky enough that a random Great Strides is up during Excellent.

[Raldo]

Had 3 Excellents once. Felt so good. Or just being lucky enough that a random Great Strides is up during Excellent.

Ooh, speaking of this... Does anyone know if Great Strides and Synth Condition are additive or multiplicative? Any time I think about it, the synth maxes out at 100%. It probably won't change the way I do things, so I'm merely curious.

[Kawfee]

Great Strides stacked with Excellent gives x6(?) (multiplicative). Big winna! Similarly GS & Good gives x3. GS & Normal gives x2. GS & Poor, you're bad.

edit: fixed to reflect #'s from below post. I'm not sure the exact percentages but I know its multiplicative.

[OnlyAura]

I'd have to second what Niiro said. I've actually had several synths that threw Excellent up twice.

For those unaware (since I don't see it in the OP):
• Poor: Quality x0.5
• Good: Quality x2
• Excellent: Quality x4

This is incorrect. Good is 1.5x, not 2x.

I have graphs of the Bonus Exp% formula and the HQ% formula, but I can't post it as you need 10 posts to post a url. Maybe later I guess.

TL;DR Bonus Exp looks like an inverse logarithm. HQ% looks like about half of the period of a sine wave.

[OnlyAura]

I presume that there is no built-in parser, as it would probably be against the ToS for hooking into the game? That is going to be problematic, then.

You could always hook the game and just not release the hook. I thought about doing something similar, but I'm not sure it's going to be worth it (and it might be prohibitively difficult, as I expect the condition string is stored in Flash UI code, and you can't rely on the address being constant every time). Cost / Reward ratio is significantly higher if you just use experimental numbers for the transition probabilities. I feel like it's about 25% for Good, and about 1% for Excellent. But I also think it depends on the level difference.

Then if you want to optimally solve each craft, just make a weighted decision tree (Keep in mind a good will always be followed by normal, and an excellent will be followed by a poor if it succeeds or fails, and a normal if it is skipped.)

[OnlyAura]

Also, here's another useful tidbit. In terms of pure Quality / CP, Basic Touch is *always* better than Standard Touch, even with Goods / Excellents. If using only Basic Touch would leave you with a CP surplus at the end of your durability then by all means Standard Touch is better, but only in this case is Standard Touch better. Here's some tables that show how much quality Basic and Standard touch will return.

              Normal     Good     Excellent
Basic           N        1.5N        4N
Standard      1.25N     1.875N       6N

Here N is just whatever the +Quality formula is going to return based on your current Craftsmanship value and other variables.

Taking into consideration the Success Rates, averaged over an infinitely long period of time, we would expect each ability to return the following amount of Quality per attempt:

              Normal     Good     Excellent
Basic          .7N       1.05N      2.8N
Standard        N        1.5N       4.8N

Finally, if we divide each one by the amount of CP required for the ability, we get the following expected Quality / CP

              Normal     Good     Excellent
Basic         .039N     .058N      .156N
Standard      .031N     .047N      .15N

So as we can see, Basic Touch always returns more Quality / CP than Standard Touch.


One of the things I'm working on is a table comparing every ability to every other ability using similar methods. Unfortunately my highest craft is 20, so most of my skills aren't even unlocked yet, but I hope to have more results of this kind in the future.

[Seravi Edalborez]

The lower tier touches have worse success rates though (70/80/90 for Basic/Standard/Advanced). When I don't have Steady Hand up I don't want to take a 70% chance at blowing CP.

Steady Hand'd Hasty Touches are also more efficient than spamming Basic Touch in any case, a big part of why Hasty Touch is essential.

[OnlyAura]

The lower tier touches have worse success rates though (70/80/90 for Basic/Standard/Advanced). When I don't have Steady Hand up I don't want to take a 70% chance at blowing CP.

Steady Hand'd Hasty Touches are also more efficient than spamming Basic Touch in any case, a big part of why Hasty Touch is essential.

I know it feels it breaks all the time when you're using the lower tier touches, but I'm only comparing Quality / CP. In terms of Quality / CP, and without stacking anything else like Steady Hand, Basic Touch is still better because you can do it again.

Basic Touch is 18 CP and Standard Touch is 32 CP (I believe). Even if you breka the first Basic Touch, you can still do another one and only have used up barely more CP than if you had used Standard Touch. So the fact that you still can use the ability a second time makes up for the lower success rate.

In any case, this isn't 100% of the story, as durability comes into play (if you only have 20 durability left, for example, then you should obviously use Standard Touch as this is the last action before you need to complete the synth).


In any case, like I said I'm only comparing Quality / CP here, and in terms of Quality / CP, Basic is always better than Standard (I don't have Advanced yet, so I don't know what the stats are and how it compares).

Sure you can use Steady Hand, but that also uses up CP. Would that CP be more efficiently used on something else? It's not immediately obvious. Needs more research. the easiest place to start though is just by comparing abilities against each other directly without considering stacking.

[Jaybar]

Why you should never take crafting out of context of other abilities:

Steady Hand 22CP / 5 Turns
Hasty Touch 0 CP 50% [70%]

SH + HT = .7N / 4.4CP (per turn) = .159N

Hasty Touch Superior (Steady Hand II even more superior)

Best used on 70/80 duras anyway. Doesn't take into account TotT procs where you waste a turn of SH, but then it essentially becomes a 2CP Steady Hand anyway.

[OnlyAura]

Why you should never take crafting out of context of other abilities:

Steady Hand 22CP / 5 Turns
Hasty Touch 0 CP 50% [70%]

SH + HT = .7N / 4.4CP (per turn) = .159N

FWIW, SH + HT >= .159N, not =. Because some of the 5 synths may be Goods or Excellents.

In any case, I agree that you shouldn't take crafting out of context other abilities, but it's also true that you can't put all abilities into context with each other unless you first understand their properties in isolation. You can't just jump to the end without going through the beginning. I suppose you can in isolated scenarios (such as this one), but that's not really very interesting. I think it's better to start with basic comparisons (which by the way are still useful for new players who perhaps don't have every ability yet) and build up to more complicated relationships.