Goblin Squad Member
|
| 2 people marked this as a favorite. |
Theodum was waiting for Thod to come back. The dwarf was the most skilled propspector of the whole settlement.
The surrounding areas offered some iron found in trash heaps - and defeating the goblins often offered some scrap metal - but it just wasn't enough.
The settlement was close to the plains - so it needed some mountains. Off course the mountains had been fiercly contested. In the North of them Golgotha had settled in them - in the South it was The Empyrial Order and The 7th Veil.
Theodum send Thod off to the South using a teleport spell. This was one of the old spells he had learned before entering the River Kingdoms. This magic shouldn't work any longer at all - but it seemed occasionally to work - albeit in an uncontrolled way.
Coming back after a day Theodum looked at the results of the trip of Thod. He came back with a dwarven smile all over his face - unloading his bags.
What came out of them did shimmer and glimmer in the sunshine. But something wasn't right. He had send off Thod to look for iron as this was needed for weapons and armour - but what he had brought back was mainly shimmering gold.
Foolish greedy dwarfs - you couldn't trust them. More then half his ore from the Southern mountain was Gold and another strange but useless mineral.
And then the greed must have completely overwhelmed him - the ore from the Northern mountains was three quarters gold.
Gold had no value here in these areas. Maybe down in Absalom or in the Five Kings Mountains this would represent a fortune - but here in the river kingdom it was iron which was more valuable as gold.
So all Thod had brought back for the settlement was fools gold.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
So if the hex originally has iron of 1000 and you harvest 10 units of ore, it has an iron rating of 990 and there is a 990/1000=99% chance it will get one point of iron back when the next 15 minute resource rating check comes around.
This quote got me thinking and I had to shudder. The basic idea seems good. You give it a max chance to regenrate if it is on full and the chance goes down if you harvest too much.
BUT - if I read the above correctly then there are some serious issues with the formula:
Assume the following ratings:
Iron: 1000
Gold: 100
Someone harvests 20% of each resource at the start of the day (math is easier if I assume start of the day and not random times - it doesn't matter that much when it actually is harvested)
New growth is every 15 minutes - this makes 96 growth times in a day (or 92 if you take off the 1 hour downtime).
I showed in a different thread how the math looks like. You would get a 10% growth over the whole day for iron. So you start with 1000 - it drops to 800 because of harvesting and it grows back to 880.
Now take the gold. The same harvester took out 20% of the gold. So you have the same 80% chance that gold growth back - BUT you always get 1 point of gold back. This means you actually instead of going from 800 to 800.8 in the first 15 minute you go from 80 to 80.8.
This means gold will recover after 5 3/5 hours back to 100% while iron is still at 88% at the end of the day. I'm not sure what happens after that - there are 70 15 minute intervals left. So if it isn't capped, then it goes up to 170 at the end of day 1.
So in summary - 20% of iron and gold harvested start of day. Iron ends up at 88% while gold ends up at 170%.
If you don't believe me - make a test. I did this in two hexes.
Hes 1 is the hex directly north of Keepers Pass.
Hex 2 is the hex where you can enter the mountain 2 east of Golgotha coming from Emerald Spire
I've been using a dwarf level 7 mining - miner 90.
Please keep in mind I assume a iron value of 1000 at the start and of 100 for gold for both hexes. I do not know the true values - but I can say for certain it should be a lot less for gold compared to iron.
Gathered resources in hex 1:
Antimon: 12
Gold: 26
Iron: 43
Sal Amoniac: 2
Gathered resources in hex 2:
Damn fool dwarf ... seems I handed all over and didn't write it down.
It was 3 times more gold as iron
The record find was: 1 Iron Ore, 9 !! gold ore in a single rock. From memory it was around 20 iron and 60 gold when I stopped.
This gathering by the way was inside an hour after server restart.
And I had seem a similar issue in three mountain hexes a few days ago when I also gathered always more gold as iron. There is a small chance that locust harvesting happens directly after server start by tier 1 miners - but I doubt this with the current density of players and because I see this again and again in many different hexes - some worse, some less worse.
Unfortunately this is exactly what I expect to happen with the formula as given if it is as Lee writes it. The most common resources will decline over time as on day 1 the growth (absolute) is the same for rare or common recources as you took out the same fraction and the chance of regrowth is based on fraction left.
From day 2 it gets worse as again the same amount relative is taken out - but the formula penalizes the resource that is low already.
Over time the common resource becomes uncommon and the uncommon becomes common.
Filling up at downtime seems to aleviate some of the issue - but it seems not to fix it.
Not believing me? Do some harvesting with a level 7 and monitor what you bring back.
A fix will come in the next post.
|
Goblin Squad Member
|
What do we want in regard to resource growth?
A good algorithm is simple and rectifies itself if it runs out of bounds.
1) Overharvesting will penelize you
2) There should be a cap on how much a resource can grow - it should not grow to infinity / by a fixed number per day
3) More common / important resources need to stay more common
4) Tier 1 and tier 2 (or 3) harvesters should not influence the other tier resources
A few definitions of values used:
Input values needed:
HV(r) = Hex value (resource) This is the basic value in the spreadsheet - the 1000 for iron Lee mentions - I use the (r) as you have multiple values - one for each resource.
GV = Global value A global value that can increase / decrease output across the whole map.
DV = Daily value - the value intended for optimal growth during 24 hours
P = Penalty A value that is needed for scaling and to have non optimal growth if a hex is depleted.
Variable value needed:
HVcurrent(r) = Hex value (resource) currently How much of a resource actually is available
Derived values:
DV15 = the growth every 15 minutes DV15 = DV / 92 (or 96 depending if you count 92 or 96 15 minute increments)
S = Scaling value A value that scales with the resource available. This will be detailed later.
Chance to grow is simple:
Chance = HV(r) * DV15 * Global value * Scaling
if chance > 1 then new resources = round(chance) otherwise it is compared to a random value and either = 1 if the random value is < chance or 0 if it is greater chance.
For nitpicks you could do at 1.5 chance a value of 1 plus a 50% chance that it might be 2.
The chance should be relative to the resource value. These are already set and this is fulfilled.
DV15 is based on how much daily growth you want. DV15 is 0.1 / 92 if you want 100 new iron (on average) per day with a base value of 1000 or it can be 0.5 if you want 500 iron to grow back per day with a base value of 1000. The value DV in principle is redundant as DV15 is what truly matters - but it is easier to think as DV and let the computer devide into DV15 as to think in DV15.
Global value: This value should be 1. It is in principle redundant. But I would build it in for flexibility. You can use it as multiplyer. If you feel not enough resources are generated - raise it to 1.5 to get global growth rate increased by 50%. You feel resources are too plentiful - lower it to 0.5 to have global growth halfed.
This leaves the scaling. The intend was that a overharvested hex should give you less resources.
Scaling = 1 - [ABS[1 - HVcurrent(r) / HV(r)] ] * penalty]
I'm sure this can be written in a better way - as it is like a double negative ...
HV(r) / HVcurrent(r) is how much resource currently is available in relative terms compared to the spreadsheet value.
ABS[1 - HVcurrent(r) / HV(r)] is how much resource is missing compared to the spreadsheet value - or how much surplus is there.
For a brand new 100% full hex the value of the inner bracket is 1 and for a 0% hex the value is zero.
But there are two issues - what do you want to do if resource has grown greater then the value in the spreadsheet - and what do you do if it is at zero.
This is where penalty comes in. If you set it to 0.7 this means that you get a total value of 0.3 at 0% and not 0. If you set it to 0.9 you will get 0.1 at 0% and not 0. This means you will still have a minimal growth. Maybe I should have named the variable minimum.
Another side effect is - and this is why I add the ABS - if resources grow above the spreadsheet value, then growth will slowly decline.
You still might want to keep the exemption handling in to grow +1 if resource current is <= 0 - just in case. But this formula hopefuly achieves what you actually want resources to do.
It should be robust as it is self correcting (slowing down growth if low on resource or very high (above spreadsheet value) or resource. And it has a simple variable you use to determine the daily growth.
And it has parameters that you can tweak to finetune it.
0.7 for penalty seems reasonable - but you might differ. And the global value allows to tweak according to growth of population in the game.
|
Goblin Squad Member
|
So if the hex originally has iron of 1000 and you harvest 10 units of ore, it has an iron rating of 990 and there is a 990/1000=99% chance it will get one point of iron back when the next 15 minute resource rating check comes around.
This quote got me thinking and I had to shudder. The basic idea seems good. You give it a max chance to regenrate if it is on full and the chance goes down if you harvest too much.
BUT - if I read the above correctly then there are some serious issues with the formula:
Assume the following ratings:
Iron: 1000
Gold: 100
Someone harvests 20% of each resource at the start of the day (math is easier if I assume start of the day and not random times - it doesn't matter that much when it actually is harvested)New growth is every 15 minutes - this makes 96 growth times in a day (or 92 if you take off the 1 hour downtime).
I showed in a different thread how the math looks like. You would get a 10% growth over the whole day for iron. So you start with 1000 - it drops to 800 because of harvesting and it grows back to 880.
Now take the gold. The same harvester took out 20% of the gold. So you have the same 80% chance that gold growth back - BUT you always get 1 point of gold back. This means you actually instead of going from 800 to 800.8 in the first 15 minute you go from 80 to 80.8.
This means gold will recover after 5 3/5 hours back to 100% while iron is still at 88% at the end of the day. I'm not sure what happens after that - there are 70 15 minute intervals left. So if it isn't capped, then it goes up to 170 at the end of day 1.
So in summary - 20% of iron and gold harvested start of day. Iron ends up at 88% while gold ends up at 170%.
If you don't believe me - make a test. I did this in two hexes.
...
Interesting. So why is the issue worse with essence gathering where you only ever get one essence per essence type per node ?
|
Goblin Squad Member
|
Interesting. So why is the issue worse with essence gathering where you only ever get one essence per essence type per node ?
This is VERY interesting. Thod is a dowser level 3 (skill 25) and he always finds 2 essence of the basic type and I don't have the issue.
I haven't done any tests on essence because I don't have a dowser level 7 or above.
Also Essence will work differently as a lot of gatherers ignore it.
Rocks - I always gather them
Piles - I nearly always gather them (80%+)
Plants - I sometimes (50%/50%) gather them
Sparks - I seldom gather them (20% max)
So sparks are the least likely to be overharvested. You don't need that much essence - so you leave it. Especially as you get the same essence in the same regions - all woods are esoteric - so if you are inside woods you very soon have LOADS of spare - and stop.
I can speculate - but this would be how the current algorithm acts on the other side of the spectrum - when there is a lot in a hex and people harvest less as comes back new.
Edit:
Oh - one more bit. Do you actually check (tooltip) how many resources each you get? The names for all essence are too long - so you have to hoover over it to actually see if you got >1. I did this for this test and always tended to have 2 with rare occasions to have more. But I'm a level 3 (skill 25) dowser. A level 7+ should get more.
But it is painful to keep note of it as you have to move the mouse every time over it to check how much you get.
This is some great investigation work. Hopefully someone from GW can read and comment.
|
Goblin Squad Member
|
Assume the following ratings:
Iron: 1000
Gold: 100
Someone harvests 20% of each resource at the start of the day (math is easier if I assume start of the day and not random times - it doesn't matter that much when it actually is harvested)
I'd think that the obvious way to mine in your hypothetical hex is to pull out the 100 gold and whatever iron you find with it. Then the gold is depleted, reduced to 1 gold/15 minutes and you can mine iron to your heart's content.
|
Goblin Squad Member
|
I'd think that the obvious way to mine in your hypothetical hex is to pull out the 100 gold and whatever iron you find with it. Then the gold is depleted, reduced to 1 gold/15 minutes and you can mine iron to your heart's content.Assume the following ratings:
Iron: 1000
Gold: 100
Someone harvests 20% of each resource at the start of the day (math is easier if I assume start of the day and not random times - it doesn't matter that much when it actually is harvested)
But you can't just pull out gold - at least not yet. So what you end up with is the following
1 gold/15 minutes
1 iron/15 minutes
1 antimony/15 minutes
1 etc
And you only get this if you harvest 24/7.
The perfidity of the algorithm is that it is capped at +1 / 15 minutes one way or the other.
So you can either deplete it and mine 24/7 and get 96 (92) resources per day or you let it grow - say to 1500 (150% for iron) and 600 for gold (5 days zero mining) and you can pull out at leasure 96 (92) resources a day and you will have them for eternity.
Long term this makes a mockery of the tables that GW craft as the resource will go either towards 0 or towards A LOT but either won't be helpful if you need iron to craft weapons and not gold.
Medium term you see effects as elsewhere where Shedim says in the same hex where he was getting iron/silver on a 1 to 1 basis it is now 1 : 2 for silver.
Oh and there is tier 3 which we can't mine until very much later. So they will likely be the most common resource once you can mine it as they will grow 96 per day as they are on 100%+ from day 1.
Still waiting for any information that pokes a hole in this and tells me I'm wrong. But so far it seems to be able to explain exactly the issues that gatherers are seeing and that frustrates them.
A node of 9 gold 1 iron is frustrating in the current time of game.
Oh I should have added on bit - Lee said that they 'fill up' at the end of day. So what I expect is
Iron: Start 1000 -> 800 -> 870
Gold: Start 100 -> 80 -> 170
1040 of 1100 left at end of day
Fillup 60 = 55 iron and 5 gold
The fillup is what keeps iron still around and replenishes it. But the above doesn't take tier 3 into account and that might make a stop to replenish at some stage as the total number of resources will be 100%+ only the distribution will be off.
I say - I can be wrong - but that is the best way to explain what is happening on a gatherer basis and is in line with what Lee has written.
|
Goblin Squad Member
|
... pull out the 100 gold and whatever iron you find with it. Then the gold is depleted, reduced to 1 gold/15 minutes and you can mine iron to your heart's content.
Sorry if my answer above assumes too much of an understanding.
The 'whatever' would be 1000 iron. Pulling out gold is NOT increasing the iron production. Well - the fill-up might do that. But that seems a temp fix.
|
Goblin Squad Member
|
... pull out the 100 gold and whatever iron you find with it. Then the gold is depleted, reduced to 1 gold/15 minutes and you can mine iron to your heart's content.Sorry if my answer above assumes too much of an understanding.
The 'whatever' would be 1000 iron. Pulling out gold is NOT increasing the iron production. Well - the fill-up might do that. But that seems a temp fix.
That might or might not be true. I think that the mining skill check that determines how many resources are found might be done before determining what resources are found. If that is the case, then removing the gold improves the amount of iron found. One way to find out: strip a hex bare and see where the numbers take you.
If DeciusBrutus is correct, though, the numbers of what you draw out will remain in a close proportion until all the minerals are close to 0. If miners are getting 3 gold for 1 iron over an extended run, one might conclude that the numbers for the hex are more like:
Iron: 1000
Gold: 3000
|
Goblin Squad Member
|
Urman
You miss the important bit I said above
Iron / Gold might start 1000 and 100
But the way Lee described it the issue is that I am convinced it changes over time v
So you did start 1000 / 100 with lots more iron - but after a while it becomes 500 / 500 or even 200 / 600
Just plug the formula Lee describes into Excel and let it develop over a few cycles.
So yes - by now it is 3:1 gold to iron as far as I understand it - but it should not be that way - ever !!!
I have now seen multiple hexes with gold more abundant as iron. Actually I haven't seen a hex recently that is the other way round. And that feels very, very wrong.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
So you did start 1000 / 100 with lots more iron - but after a while it becomes 500 / 500 or even 200 / 600
Sorry if I'm missing something, but can you lay out - succinctly - your understanding of the process that would lead to this outcome? When I read Lee's post, I didn't see anything that I thought would result in what you're describing.
|
Goblin Squad Member
|
Is randomness a possible factor? While it could be bugged reading your description my gut says perception problem but I have not been following resource conversations closely. Are resource restocks independent of each other or is if tied together? Ex: you either get 1 gold or 1 iron every 15 mins
|
Goblin Squad Member
|
To be a little clearer, if the maximum Gold Rating in a hex is 100, how will that rating ever be greater than 100? Sure, perhaps the Gold Rating will get back to 100 more quickly than the Iron Rating gets to 100, but once the Gold Rating hits 100, it stops increasing.
I don't know how to descibe math and statistics succinctly and in a way that everyone understands. I know I failed many times before and will fail again.
I added a spreadsheet that uses Lee's formula. This does NOT include exemptiuon handling. As such it should add a max of +1 after gold reaches 100 - unless GW uses exemption handling.
The reason I doubt they use it - they didn't have it for 0 originally and only added it later when it went wrong. As 0 is the more obvious case that the formula goes wrong - why add it for > starting value. But this part is pure speculation.
The issue is - the growth is depending on the fraction of resource available - NOT on the absolute value.
Now I only did a single day. But start day 2 with 880 / 170 and do a 20% gahering again and it slowly will mean that gold becomes more common and iron less common - up to a stage where gold overtakes iron.
There is nothing at all in the algorithm that can prevent it.
hours start iron chance start gold chance
0 800.0 80.00% 80.0 80.00%
0.25 800.8 80.08% 80.8 80.80%
0.5 801.6 80.16% 81.6 81.61%
0.75 802.4 80.24% 82.4 82.42%
1 803.2 80.32% 83.2 83.25%
1.25 804.0 80.40% 84.1 84.08%
1.5 804.8 80.48% 84.9 84.92%
1.75 805.6 80.56% 85.8 85.77%
2 806.4 80.64% 86.6 86.63%
2.25 807.2 80.72% 87.5 87.49%
2.5 808.0 80.80% 88.4 88.37%
2.75 808.8 80.88% 89.3 89.25%
3 809.7 80.97% 90.1 90.15%
3.25 810.5 81.05% 91.0 91.05%
3.5 811.3 81.13% 92.0 91.96%
3.75 812.1 81.21% 92.9 92.88%
4 812.9 81.29% 93.8 93.81%
4.25 813.7 81.37% 94.7 94.74%
4.5 814.5 81.45% 95.7 95.69%
4.75 815.3 81.53% 96.6 96.65%
5 816.2 81.62% 97.6 97.62%
5.25 817.0 81.70% 98.6 98.59%
5.5 817.8 81.78% 99.6 99.58%
5.75 818.6 81.86% 100.6 100.57%
6 819.4 81.94% 101.6 101.58%
6.25 820.2 82.02% 102.6 102.59%
6.5 821.1 82.11% 103.6 103.62%
6.75 821.9 82.19% 104.7 104.66%
7 822.7 82.27% 105.7 105.70%
7.25 823.5 82.35% 106.8 106.76%
7.5 824.4 82.44% 107.8 107.83%
7.75 825.2 82.52% 108.9 108.91%
8 826.0 82.60% 110.0 110.00%
8.25 826.8 82.68% 111.1 111.10%
8.5 827.7 82.77% 112.2 112.21%
8.75 828.5 82.85% 113.3 113.33%
9 829.3 82.93% 114.5 114.46%
9.25 830.1 83.01% 115.6 115.61%
9.5 831.0 83.10% 116.8 116.76%
9.75 831.8 83.18% 117.9 117.93%
10 832.6 83.26% 119.1 119.11%
10.25 833.5 83.35% 120.3 120.30%
10.5 834.3 83.43% 121.5 121.50%
10.75 835.1 83.51% 122.7 122.72%
11 836.0 83.60% 123.9 123.95%
11.25 836.8 83.68% 125.2 125.18%
11.5 837.6 83.76% 126.4 126.44%
11.75 838.5 83.85% 127.7 127.70%
12 839.3 83.93% 129.0 128.98%
12.25 840.2 84.02% 130.3 130.27%
12.5 841.0 84.10% 131.6 131.57%
12.75 841.8 84.18% 132.9 132.89%
13 842.7 84.27% 134.2 134.22%
13.25 843.5 84.35% 135.6 135.56%
13.5 844.4 84.44% 136.9 136.91%
13.75 845.2 84.52% 138.3 138.28%
14 846.1 84.61% 139.7 139.66%
14.25 846.9 84.69% 141.1 141.06%
14.5 847.7 84.77% 142.5 142.47%
14.75 848.6 84.86% 143.9 143.90%
15 849.4 84.94% 145.3 145.34%
15.25 850.3 85.03% 146.8 146.79%
15.5 851.1 85.11% 148.3 148.26%
15.75 852.0 85.20% 149.7 149.74%
16 852.8 85.28% 151.2 151.24%
16.25 853.7 85.37% 152.7 152.75%
16.5 854.6 85.46% 154.3 154.28%
16.75 855.4 85.54% 155.8 155.82%
17 856.3 85.63% 157.4 157.38%
17.25 857.1 85.71% 159.0 158.95%
17.5 858.0 85.80% 160.5 160.54%
17.75 858.8 85.88% 162.1 162.15%
18 859.7 85.97% 163.8 163.77%
18.25 860.6 86.06% 165.4 165.41%
18.5 861.4 86.14% 167.1 167.06%
18.75 862.3 86.23% 168.7 168.73%
19 863.1 86.31% 170.4 170.42%
19.25 864.0 86.40% 172.1 172.12%
19.5 864.9 86.49% 173.8 173.84%
19.75 865.7 86.57% 175.6 175.58%
20 866.6 86.66% 177.3 177.34%
20.25 867.5 86.75% 179.1 179.11%
20.5 868.3 86.83% 180.9 180.90%
20.75 869.2 86.92% 182.7 182.71%
21 870.1 87.01% 184.5 184.54%
21.25 870.9 87.09% 186.4 186.38%
21.5 871.8 87.18% 188.2 188.25%
21.75 872.7 87.27% 190.1 190.13%
22 873.6 87.36% 192.0 192.03%
22.25 874.4 87.44% 194.0 193.95%
22.5 875.3 87.53% 195.9 195.89%
22.75 876.2 87.62% 197.8 197.85%
23 877.1 87.71% 199.8 199.83%
23.25 877.9 87.79% 201.8 201.83%
23.5 878.8 87.88% 203.8 203.84%
23.75 879.7 87.97% 205.9 205.88%
24 880.6 88.06% 207.9 207.94%
|
Goblin Squad Member
|
To be a little clearer, if the maximum Gold Rating in a hex is 100, how will that rating ever be greater than 100? Sure, perhaps the Gold Rating will get back to 100 more quickly than the Iron Rating gets to 100, but once the Gold Rating hits 100, it stops increasing.
Lee never said it is the max gold rate - it is the starting gold rate.
HUGE difference. They might cap total resources to max - but that won't keep gold at max start rate if it growth faster as iron.
So Iron + gold might be 1100 max in the example I use - but the formula for growth will let gold grow faster as iron and therefore leads slowly but surely to more gold as iron.
So far I see this in every hex I have gathered in.
You can't cheat luck on a constand basis. I'm not a gambler because I know how stats work and I know I will lose - unless I rig the stats.
|
Goblin Squad Member
|
As I understand it was a check towards that starting value, and got the impression that thus it also become a maximum value, but I could have misunderstod.
This is how it should work. The issue is - the data I see is not how it does actually work.
The reason I don't think it maxes out - a low number in this case should be easy to exhaust and then even this one won't regenerate quickly.
I saw gold/iron issues several days ago and they get worse. If it caps out gold should suffer as well as 60 gold in a single hex gathered should be a lot more massive as 60 iron.
Goblinworks Executive Founder
|
But you can't get half the gold without also getting half the iron, because the drop rate is proportional to the current distribution, among eligible materials.
That would allow a t1 harvester to bring the iron down below gold, so a t2 gatherer would get more gold than iron. Running out of iron would result in the error message, expect once per 15 minutes.
A very rich hex that became fully depleted might take a very long time to rise above 1.
|
Goblin Squad Member
|
@Thod, I wasn't asking for a primer on statistics, I was asking for a succinct summary of your position, and I think I have it now.
In essence, your position is that our hypothetical 1,000 Iron Rating / 100 Gold Rating Hex, after losing 10 Iron, will more than likely end up with 101 Gold Rating after the calculations. Is that correct?
Couldn't your observations also be explained by Gold having a higher Rating than Iron even without the possibility of the current Gold Rating exceeding the original Gold Rating?
|
Goblin Squad Member
|
Couldn't your observations also be explained by Gold having a higher Rating than Iron even without the possibility of the current Gold Rating exceeding the original Gold Rating?
NO
a) it doesn't make sense to have gold higher as iron as iron is used for all basic weapons, armour and is ubiquitous in recipes. That would be very bad design to have it in every single hex I recently visited higher as iron
b) some developer himself said gold is rare and shouldn't even be in more than one mountain region. I was suprised to find it in 2.
c) some tester told me online that I was 'lucky' to get that much tier 2 compared to tier 1. This was a week ago and pales in comparison to my 'luck' that I have now
d) I scan these boards for information that refutes my hyporthesis or that strengthens it. So far I only see anectdotal eveidence that supports it - lile Schedim telling me he now finds more silver compared to iron (or was it coal) in the same hex as 2 weeks ago with the same mining skill. There are multiple more comments like that scattered around / that you hear while playing
I guess I find a few more reasons
Short
[b[NO[/b]
edit: Just checked in your formatted data and found 28 recipes including gold (bars). I stopped at 100 including steel - and I wasn't even half way through and ignored cold iron, silved iron, blended iron and I don't know what else that contains iron ore.
There just is no need for LOADS of gold compared to iron.
|
Goblin Squad Member
|
Couldn't your observations also be explained by Gold having a higher Rating than Iron even without the possibility of the current Gold Rating exceeding the original Gold Rating?NO
I'm sorry, I'm just confused why you're so certain that the values are exceeding the Original Ratings.
You've demonstrated that the Gold Rating will dominate the Iron Rating when it comes to recovery. If the Hex started out with 5,000 Iron Rating and 300 Gold Rating, what makes you so certain you're not seeing the effects of Current Ratings of 100 Iron and 300 Gold?
|
Goblin Squad Member
|
It is a possibility that it is at 300. The main reason I don't think so it that I expected enough level 7+ miners close to Golgotha or Keepers Path and if I gathered 50 in half and hour while trying to hunt down iron and not gold then a single gatherer from these places should be able to strip harvest gold to a level it doesn't easily come back from.
Get it down to 20% and even gold won't easily recover. But no - I can not be certain. But that bit is minor - the important bit is that even if it caps out at max then finding more gold as iron is very problematic
I'm not harvesting in these region on a regular basis - I take a few while passing - or in this case after a teleport I just did a systematic test - mainly to see the ratio gold/iron.
Another reason is that Lee (?) said that gold was only in the mountain range in the south - I was never sure if he meant the one around Keepers Pass or Phaeros - surely he didn't mean both.
I also expected low values for gold. It is tier 2 and Gold seems very useless
Using Nihimons formatted data I found for
Ingredient 1 Number or recipes total needed
Gold bar 17 98 (23 of these for the ornate chain belt)
Iron
Blended Iron 11 35
Cold Iron 8 29
Dwarven Steel 32 371
Silvered Iron 25 73
Steel 30 879
I haven't checked ingredients 2, 3 and 4. Did this semi manually by sorting and counting / using the sum function.
|
Goblinworks Executive Founder
|
Thod: I just saw you provide fairly reasonable reasons why the maximum rating of gold probably isn't generally higher than the maximum rating of iron.
However, your current observations strongly support a current rating of gold higher than the current rating of iron, but would be equally well explained regardless of maximum ratings, particularly since one of the high-prior hypotheses is that the ratings are getting reset every morning.
... I remember a whiteboard indication that escalations would take from the hex they were in. Is that perhaps interacting here? What is the escalation type and strength in the hexes where you are getting this data?
|
Goblin Squad Member
|
This effect will suddenly become a good thing once we get to T2 weapons and need lodestone not iron.NO
Dwarven Steel needs 2 times as much iron as lodestone.
Finding more lodestone as iron won't be good.
ah ...
well will be a good thing for me with my several thousand iron stockpile though :D
|
Goblin Squad Member
|
Decius
Yes - it is possible that gold just stays at max - because it recovers more easily and not many people actually mine it.
The reset won't apply for that reason because it seems not to work for iron. That is why I think the reset fills out the whole hex to the total resource but isn't taking anything away.
The 3:1 gold to iron test was done inside 1 hour past reset.
What data has Phaeros, Brighthaven, Golgotha?
There must be level 7+ miners with observations. I try not to mine in their areas. They must have some observations even if they don't write down all the numbers.
And the hexes around their settlements are likely the most interesting. I find some hexes with sensible tier 1/2 ratios - guess where? To the west of EL > 2 hexes from any settlement. These are 'undisturbed' hexes.
If it is just and 2 or 3 other level 7+ miners then off course it is likely just at max for gold.
I hope I'm not coming over as zealot here. I noticed in at least one or two instances I wasn't reading every word as I should and misinterpreted some concerns.
Anyhow - late here - should be gone a while ago.
|
Goblin Squad Member
|
... the important bit is that even if it caps out at max then finding more gold as iron is very problematic.
It seems to me this is only ever likely to occur under specific, Alpha-like conditions. Namely, that there was enough sustained activity to significantly reduce a Hex's Ratings, and then enough non-Activity to allow the lower-rated Resource to dominate the Recovery.
The only reason I keep coming back to this is because I don't think whatever you're trying to report is clear enough for the devs to take action on, and I'm trying to understand it myself so I can perhaps restate it in a way that is easier to understand.
It sounds to me like your main concern is that the recovery rate for resources with low original ratings will dominate the recovery rate for resources with high original ratings, when there is not sufficient activity to continue depleting them. Is that correct?
|
Goblin Squad Member
|
Nihimon
The way I understand the formula given by Lee
a) recovery is absolute depending on fraction of resource left
b) this leads to common resources to recover more slowly to the max / start value as uncommon
c) this leads long term to uncommon resources to become more common as common ones - gold/iron is just one example.
Go out with a level 7 gatherer and note what you find and ask yourself does this make sense
If it goes above max or not and how the reset each day work is secondary.
The whole issue is in a - is recovery absolute or relative
Is it as Lee says:
80% = an 80% chance to increase by +1 then there is a problem
or do they apply a scaling factor to the +1
I don't think the discussion makes much more sense. I should have filed this to a dev after the first postings - actually I wanted but then got drawn back here again to post more and more which is likely less consturctive as we need someone who can see the code and can confirm which parts are a problem - and a problem there is - that is what I'm absolutely convinced.
|
Goblin Squad Member
|
Decius
What data has Phaeros, Brighthaven, Golgotha?
There must be level 7+ miners with observations. I try not to mine in their areas. They must have some observations even if they don't write down all the numbers.
My miner is level 7 and has the achieves/abilities/XP to get to level 8 (possibly 9 by now) if I could be bothered leveling his mining however I moved him back to Thornkeep from Keepers Pass because Keepers has sorta been placed on hold until EE .
It only seems to be a 10 minute run from TK down to KP/Brighthaven if you run down the side of the mountains and avoid the escalations so if I get a chance will level him to 8 and head back down to the southern mountains and see what I can find out.
|
Goblin Squad Member
|
Another thing to keep in mind is that Gold is Tier 2, so it's really not appropriate to compare it to Iron. It's not like your chance to get Iron is going to decrease your chance to get Gold.
His point appears to be that ultimately his chance of getting gold is decreasing his chance to get iron.
|
Goblin Squad Member
|
Another thing to keep in mind is that Gold is Tier 2, so it's really not appropriate to compare it to Iron. It's not like your chance to get Iron is going to decrease your chance to get Gold.His point appears to be that ultimately his chance of getting gold is decreasing his chance to get iron.
Surely that would only occur if their was a cap on total lootable items or the items appearing in the node are not independent.
otherwise you would end up with a stupid amount of gold but the same amount of iron as normal.
|
Goblin Squad Member
|
Can you point me to this Lee posts, please?
If really the resources recover at most 1 point every 15 minutes (even if we don't take into account the chance of recovery), the end result is that after a hex has been depleted it will, at most, recover 96 points of each kind of resource.
96 point of gold isn't bad. 96 points of iron is a very small quantity.
So unless the capping out is substantial it will make little difference, and even if it is substantial it will benefit mostly people that play just after DT, while people playing at the end of the day will find nodes with very few materials.
If is a "good" thing for me (and other European players) as I can shift my schedule a bit and play an hour or two just after DT.
If it work that way you will get an influx of East Europe iron diggers selling what they gather for real money.
|
Goblin Squad Member
|
@Nihimon
You really seem not to get it. Finding more gold as iron shouldn't happen ever - apart maybe some freak dice rolling.
But I have seen it on >1 occasion, in two diferent mountain ranges and 6 different hexes and 100% of mountain hexes that I have mined in lately.
Why should it never happen?
a) If it is by design (original tables) then the design is wrong, because gold compared to iron is useless and should be less common
b) If it is because of tier 1 vs tier 2 distribution then it is equaly wrong as tier 2 needs to be more rare as tier 1 - especially as tier 2 refined goods use tier 2 AND tier 1 ingredients and dwarven steel needs twice as much iron compared to lodestone
c) If it is because tier 1 is exhausted and isn't regenerating and it is temporarily then it is euqally wrong as this means I'm just ahead of the curve. Iron gathering now is tough but gold is fine. But once tier 2 gathering goes into full swing and the relationships readjust then tier 1 gathering still is tough but tier 2 will become nearly impossible
The post 1 by Theodum describes this issue in a humourous way why it is just wrong and shouldn't happen at all. Maybe I shouldn't have written that as a RPG post - but I felt it lightens the mood instead of pointing the finger towards another proble,.
Post 2 by Thod outlays information gathered and the formula Lee has posted and what actually happens if you use the formula assuming it is used as posted. It exactly explains why it happens, why it gets worse and that it will shift over time towards an outcome which is detrimental for long time gaming.
Post 3 by Thod suggests an alternative formula which would fix the issue
There is no way that I can proof it isn't by design - apart of saying that GW tends to be quite good an all the posts about resources indicate they spend a lot of time thinking how it should work.
I can also not proof that Tier 1 vs tier 2 is not by design. But it would be an equally bad design choice and is contrary to some blogs I'm sure to have read that stated tier 2 should be less common as tier 1 and tier 3 exeedingly rare. You are master blog knowledge - try finding it and related info about design of resources to be constructive.
This leaves the exhausted options as the most likely culprit - but then don't let me say a few month into the game - I told you so. If exhaustion of hexes move the ratios of raw materials to find then all the design put in the original tables is meaningless. Worse - you might even be able to manipulate a hex to produce what you want - assuming you have full control who gatheres what and how much.
In summary:
To me it is absolutely clear that resource regeneration isn' working as intended or described by communications from GW. Gold vs iron is only one symptom of it - albeit one that can be measured - there are more I mentioned and that you find them across chatter on the boards if you look for them.
I think I identified the root cause in the algorithm - and I suggested a fix.
It is possible I'm wrong on some details as I don't have access to the code - but that is irrelevant as the symptoms clearly indicate it isn't working as intended and that it is getting worse over time.
|
Goblin Squad Member
|
Can you point me to this Lee posts, please?
If really the resources recover at most 1 point every 15 minutes (even if we don't take into account the chance of recovery), the end result is that after a hex has been depleted it will, at most, recover 96 points of each kind of resource.
96 point of gold isn't bad. 96 points of iron is a very small quantity.
So unless the capping out is substantial it will make little difference, and even if it is substantial it will benefit mostly people that play just after DT, while people playing at the end of the day will find nodes with very few materials.
If is a "good" thing for me (and other European players) as I can shift my schedule a bit and play an hour or two just after DT.
If it work that way you will get an influx of East Europe iron diggers selling what they gather for real money.
Someone understands how to profit from it and sees the bigger picture.
A lot more details from Lee including the formula
The node 9:1 gold vs iron was near Golgotha and that quote was the reason I checked both mountain ranges and why I assume resources can go above the max cap as the cap for gold around Golgotha should be very low but I could easily take out 60 gold in 25 minutes in a single hex.
edit: This was to rule out Lee just mixed up both mountain ranges and what I saw was by 'design'
|
Goblin Squad Member
|
Another thing to keep in mind is that Gold is Tier 2, so it's really not appropriate to compare it to Iron. It's not like your chance to get Iron is going to decrease your chance to get Gold.His point appears to be that ultimately his chance of getting gold is decreasing his chance to get iron.
Again, Gold is Tier 2. As you open up Tier 2 gathering, it can feel like you're no longer getting as much Tier 1 Resources because those had been artificially inflated. One of the devs has talked about this somewhere, but I don't have the time to search for it right now.
|
Goblin Squad Member
|
@Nihimon
You really seem not to get it.
Just trying to pin it down. When I first saw the quote that prompted my reply, I thought you were confused and trying to understand something, so I tried to understand what you were talking about to see if I could help. Then it seemed like you were trying to report a bug, so I tried to understand it more clearly in case I could express it clearly enough that the devs could understand and act on it. Now it sounds like you're just expressing your opinion that Gold should always be rarer than Iron. If that's all it is, I kinda feel like I've been wasting my time. And apparently, you feel like I've been wasting yours, too.
|
Goblin Squad Member
|
@Nihimon
It is inbetween a bug and a design feature that has (in my view) uninteded consequences
Have a look at the comment from Diego Rossi
Do you understand why this 'feature' makes it an advantage for European players and why he predicts East European iron miners for money in the future if this will persist (no - this won't happen as it will change before this happens).
There are many issues - most of them bad for the game - related to a slow change of the ratio of resources away from the spreadsheet mentioned by Lee.
a) Different times of mining result in more common resource gathering
b) More and more tier 1 minor get messages - not good enough to harvest here
c) Resources not useful in higher numbers become common
d) Important resources like iron, coal, pine and yew become more and more difficult to harvest
e) The equilibrium between tier 1 / 2 gets out of sink
I don't know the spreadsheet Lee has. I use gold / iron as proxy for something that just shouldn't make sense.
I see other imbalances slowly developing - but they are not as pronounced yet and if I can't convince people here that a gold:iron imbalance is a long term issue then I will fail to convince them of issues with other imbalances that are only slowly developing and might have less of an impact.
Copper/Onyx close to TK seemed to develop more and more towards the latter - I haven't been there lately to generate numbers but I notices a worrying trend a while ago.
I noticed a hex in the north/East mountains that gave 80% gems (tier 1) also already a while ago
Silver/iron(coal?) as mentioned by Schedim
How can you help?
Report / note down 'weird' gathering results - best with numbers and hexes. Weird = results you feel seem 'wrong' or 'odd'. Don't just shrug them of as being 'as intended' or 'random' or 'lucky'.
I don't want that we have to wait for weird and odd to become the new norm.
I have send Lee a message but wasn't aware you have a holiday today - at least Paizo seems to have one.
|
Goblinworks Lead Game Designer
|
| 3 people marked this as a favorite. |
Hey guys
First off some info:
1. Starting value and max value are the same, so if a hex has Gold at 200 it starts at 200 and never goes above it.
2. I have target numbers of resources for the whole game that vary according to how the project demand is for that resource. For example, the projected demand for iron is such that the total iron value for all hexes is 343,152, while the project demand for gold is such that the total gold value for all hexes is 33,383. This is because the projected demand for iron far, far outstrips the demand for gold. The only things that come close to the demand for iron is coal and wool.
3. Mountains Ranges and what they should have:
Splinter Peaks (Near Callambea and Freevale): Iron, Silver, Lodestone
Northern Thorncrags (Near Golgotha): Copper in the northern end, Gold at the southern end, Iron
Southern Thorncrags (Phaeros and Brighthaven): Iron, Gold, Londestone
Southern Echo Peaks (Forgehelm): Iron, SIlver, Lodestone
I agree with Thod there is a problem with the long term harvesting of resources in that if you have two resources with differing values the lower ranked one will come to dominate more than it should. The current fix I'm working with as it is simplest (at this stage on the game programmer time is incredibly rare so I have to do the simplest implementation if I expect it to get done any time soon) fix I've got so far is changing the rate of increase to 1% of original value (which mirrors some of Thod's math, using 1% instead of DV15 in his original equations I believe which was close to 1% to begin with). Long termWe may be able to do something more complicated, and this may not make it into EE so we may do a fix shortly after EE starts and do a resource reboot, but we'll see how things shake out.
|
Goblin Squad Member
|
Lee
Thanks for the post and sorry to summon you here. I wasn't aware it isn't a normal working day in the US.
Your fixes are likely fine for a while and I will monitor if I see issues. Seems most people haven't noticed any issues / are unaware of them - so it can't be that bad yet.
But I felt it needed some work before it truly became an issue and production of new resources would get out of sync with demand.
Okay - can't resist - another desync issue - but at least a desync that seems to be squashed.
|
Goblin Squad Member
|
For Tier 1 ink: the Green dye material can be found as salvage from goblins (bags of itchy stuff), the Yellow found in the swamps or even in plant nodes near Thornkeep. Even along "patrolled" roads. There's also a plant item that can be used as either Green or Yellow, but I've found only a little bit of that.