[Research] Gathering - Effects of Guild Tavern and Guild Banner Buffs

[thehipone.6812]

After the big gathering update I took a slight detour from my mob drop research that was posted a little over a week ago to take a closer look at how the [Guild Gathering Boost](https://wiki.guildwars2.com/wiki/Guild_Gathering_Boost_(consumable)) and [Gathering Banner Boost](https://wiki.guildwars2.com/wiki/Guild_Gathering_Banner_Boost) influence the materials that you gather from nodes. The text on these buffs seems fairly straightforward:

*Tavern Boost: Increase chance to find rare materials from mining, logging, and harvesting by 20%.*

*Banner Boost: Increase chance to find rare materials from mining, logging, and harvesting by 15%.*

 

From this wording, and where I see these banners placed in the world, I think it is fair to say that I and many players would expect that these boosts would increase things like:

-Living World Season 3 node items like the Jade Shards that drop from mining Jade Fragments in Lake Doric, Petrified wood in Ember Bay, or Blood Rubies

-The number of gemstones that drop from mining metal nodes

-Perhaps you might expect the buffs to increase the number of the hidden bags that contain a gemstone and sometimes drop from logging trees.

-You may also expect the buffs to increase the amount of Freshwater Pearls and Maguuma Lilies that drop from Mussels/Jungle Plants, respectively.

 

Aside from the last one, where the drops are so rare and nodes spread out so much that testing is impractical, I set out to test hypotheses related to the bulleted statements and also to see if I was able to tell if the buffs stacked. To do this, I gathered a minimum of 500 strikes on nodes of each type with a) No Gathering Buffs b) Only the Guild Tavern Boost and c) With both the Guild Tavern Boost and the Guild Banner (15%) Buff. Suffice to say, the tested hypotheses were not supported by the data and the gathering buffs do not seem to operate in a manner consistent with the tooltip description. On to the testing...

 

 

# Pick Me a Winner Dude - Jade Shards

I chose Jade Shards as a model for the Living World Season 3 drops because I find them easiest and fastest to gather. There appears to be a base 33% chance to get a Jade shard from each strike of a jade fragment node. Thus, the hypothesis would be that with the 20% buff you might get 33% *1.2 = 39.6% of strikes that yield a Jade Shard. If the banner buff stacks, this number should go higher. Here are the results:

 


https://imgur.com/zO6dzkh

 

**Unexpectedly, there was no effect of the gathering buffs on the frequency of getting Jade Shards from the nodes.** I (and I think much of the playerbase) probably considers Jade Shards to be “rare materials” that would be expected to benefit from the buffs as worded in the tooltip. I do not have data yet on whether this extends to all of the other LS3 and 4 rare gatherables like Blood Rubies, Petrified Wood, or Difluorite Crystals. However, based on the following metal mining testing, it would not surprise me if there is also no effect. Again, this runs counter to my observed understanding of player expectations and behavior – there is almost always a gathering banner on the ship in Bloodstone Fen, so it seems that somebody thinks that it will increase the chances of getting a blood ruby. Observations of mapchat conversations back this up too.

 

Each strike also gives an average of 5 unbound magic. This was not affected by the gathering buffs. I also tested the hypothesis that these buffs do not increase the number of watchwork sprockets, as it was basically a freebie test while doing the other ones. As expected, the number of Watchwork Sprockets did not change with the buffs.

 

*TL;DR Guild and banner buffs did not increase the frequency of receiving Jade Shards from mining Jade Fragments.*

 

 

# More Gems for Free?!? Sure! Oh, Gemstones - _Merch_ – Platinum Mining

Other things that could possibly be considered “rare materials” for the purposes of evaluating the effectiveness of gathering buffs are the gemstones that drop from metal nodes. Ruby shard, opal crystal, carnelian nugget, sapphire orb, etc. etc. I mined platinum ore and used only regular nodes. I skipped the rich platinum veins just in case they had a higher gem chance, as indicated by the wiki. Mining was spread roughly evenly over Sparkfly Fen, Mount Maelstrom, and Timberline Falls and was performed in conjunction with the hard wood logging described in the next section. Test results were:

 


https://imgur.com/AECVaa1

 

Platinum nodes always give 1 ore per strike so nothing exciting there. The chance at watchwork was also the same, matching our initial hypothesis. As was the case with the jade shards, **there was no effect of guild tavern or banner buffs on the frequency of getting a gemstone from mining**. I don’t know about you, but this was quite unexpected for me. Even after seeing jade shards unaffected, I still expected to see an effect of the buffs on gemstones.

 

*TL;DR Guild and banner buffs did not increase the frequency of receiving gemstones from mining regular platinum nodes.*

 

 

# …alone or in pairs… - Hard Wood Logging

Wood nodes were interesting to test for two reasons – 1) they rarely drop the “hidden” bags containing a gemstone that could possibly be expected to benefit from the buffs and 2) you have a chance to “double strike” and get two logs from a single swing of the logging axe. I counted both the number of hidden bags and double strikes. Here are the test results:

 


https://imgur.com/CrbwXNI

 

The hidden hoard, despite being what I would again consider a “rare material” was not influenced by the buffs. But **finally** we have some data that suggests that the buffs seem to do something – the chance at a double strike increased with the tavern boost and increased further when the banner boost was added too. I’m not claiming that these are “true” drop rates, but the ratios scale roughly as expected based on the buff tooltip text. The tavern boost got about 22% more double strikes than with no buff. The banner appears to stack. What this does suggest is that the tooltip description on the gathering buffs may be misleading and that the buffs do not actually increase the chance of getting "rare materials", but actually increase the chance of getting more of the basic materials! (As long as the node is of a type that can give doublestrikes).

 

*TL;DR Guild and banner buffs did not increase the frequency of receiving hidden hoards from logging. The buffs did appear to increase the chance of getting a doublestrike (2 logs from 1 swing) from a tree.*

 

 

# Comments

The discrepancy may be explained by looking further at what the actual guild upgrade says:

 

https://wiki.guildwars2.com/wiki/Guild_Enhancement:_Critical_Harvesting “Add a personal enhancement at the tavern that increases the chance of getting a critical success when harvesting a node.”

 

The banner upgrades are less clear and only mention “gathering chance” : “Craft schematics to manufacture banners that offer allies a 15% boost to gathering chance and a 10% boost to swiftness for 30 minutes.”

 

If we take the doublestrike during logging as a “critical success when harvesting a node”, then the actual in-game function seems to match this description. But the in-game function does not match the tooltip descriptions of the Tavern and Banner buffs. I have not had a chance to test plant nodes, nor the other living world node types. I would expect that the buffs increase the chances to get a doublestrike from plants (i.e. 2 carrots in a single harvest), but cannot confirm. In practical terms, this means that the buff could be useless on a variety of nodes that cannot generate a doublestrike. Ironically, this includes the nodes where these buffs are arguably most used – the living world nodes.

 

*TL;DR Guild gathering tavern boost and banners do not seem to operate in the manner that I expected. They did not increase the chances of getting LW tokens (Jade Shards) nor gemstones from metal nodes. The buffs did appear to increase the chances to get a double strike from a tree. The buffs may have no effect on nodes that cannot generate a doublestrike. The error may be in the tooltip description.*

 

[Illconceived Was Na.9781]

Great research, @"thehipone.6812".

 

ANet has never been very clear about what they mean by "rare materials" from different salvaging kits nor "rare materials" from various gathering buffs. The short version is that it's used as an in-hour jargon term to refer to a specific sort of bonus drop, nothing to do with the "rarity" of the materials nor with the normal drop rate. That is: don't be misled by the wording; it's ambiguous. I don't think of it so much as an "error" in the tooltip as it is a "failure to communicate".

 

My friend & I did a less scientific study, during the final hours in which unidentified dyes would drop from plant nodes. We put on every buff we had, that might affect drop rates and went to every node farm and high-density farmable shrubberies in the game at that time. We ended up with hardly any unID dyes and (near as we could tell), no extra swings. But we did get a lot more of some mats than we expected.

 

or tl;dr: we concluded the same as you.

 

[Illconceived Was Na.9781]

For plant node testing (if you still have the energy), consider the herbs in the Mount Maelstrom run, since they drop several different sorts of mats, with different possible amounts:

* 1 or 2: Bay Leaves, Dill, Garlic, Mint, Tarragon, Thyme

* 1-4: Cloves

* 3 or 5: Vanilla

 

And independently:

* 0 or 1: Purple Pigment

* 0 or 1: Dandelion (junk)

 

It's a lot more tedious to collect the data, but I think it might expose some new details of the drop tables (or lack thereof).

[Ardenwolfe.8590]

Wow. This was incredibly helpful! Thank you for posting it. Really should be stickied.

[mtpelion.4562]

To describe the effect as "garbage" does a disservice to waste materials everywhere.

[Vagrant.7206]

I really appreciate this research. Thank you for doing it, I would've never done it myself.

[ThePlayMaker.1058]

Very impressive research, hats off to you. So exactly what is the purpose of these boosts then? Do they only affect certain nodes or what?

[Devildoc.6721]

That's a real shame, I thought that those buffs at least improved the chances of gemstones (including the valuable freshwater pearls and maguuma lilys). That's what I thought anet meant by rare materials. I didn't think of the LS3 and LS4 mats so much but I thought for sure, orbs and other yellow (IE rare quality) drops like foxfire clusters would increase in frequency. Could see how sprockets and maybe blue/green gems wouldn't be effected, as they're not "rare"

 

If anything you should try orichalcum nodes and see if it increases orb drops, along with maybe jungle plants, flax, and mussels as those also drop "rare" materials (lilies, flax blossoms, freshwater pearls)

[thehipone.6812]

> @"Devildoc.6721" said:

> That's a real shame, I thought that those buffs at least improved the chances of gemstones (including the valuable freshwater pearls and maguuma lilys). That's what I thought anet meant by rare materials. I didn't think of the LS3 and LS4 mats so much but I thought for sure, orbs and other yellow (IE rare quality) drops like foxfire clusters would increase in frequency. Could see how sprockets and maybe blue/green gems wouldn't be effected, as they're not "rare"

>

> If anything you should try orichalcum nodes and see if it increases orb drops, along with maybe jungle plants, flax, and mussels as those also drop "rare" materials (lilies, flax blossoms, freshwater pearls)

 

I think flax will be the next to test if I have time. Jungle plants and mussels have drop rates that are so low that testing is impractical due to the need to do many 1000's of gathers to see a difference. I would love to do orichalcum or ancient wood, but if I exclude rich nodes again, finding 167 nodes (three times to do each buff condition) to get the data is super painful. The platinum/wood trials got tedious near the end and those nodes are super abundant.

 

 

> @"ThePlayMaker.1058" said:

> Very impressive research, hats off to you. So exactly what is the purpose of these boosts then? Do they only affect certain nodes or what?

 

I don't want to extrapolate too much to all types of nodes, but the jade and gemstone data seem to indicate that there are nodes where the buffs have no effect. My best guess is that they provide a limited benefit to nodes that can doublestrike (wood nodes) or drop more than 1 basic material (many plant nodes) by increasing doublestrike "critical" chance. Dev confirmation would be the best, but you never know if we'll ever get that.

[ThePlayMaker.1058]

I'd love to see testing on mussels, as I've always been using the gathering buffs on them, and it had seemed to benefit the drop rate of freshwater pearls, but now I'm not even sure lol. If you'd like I could record my mussel farming sessions to add to your numbers.

[Devildoc.6721]

Yeah you really need to hit nodes that can drop yellow (rare) items. It might literally mean only the yellow drops like orbs, rather than lower tier green or blue drops.

[Inculpatus cedo.9234]

Just like with Salvaging Kits, or even Black Lion Chests, I don't think the word 'rare' is referencing Rare materials.

[Khisanth.2948]

The intesting thing with flax and jungle plants is that they have two "rare" drops. Flax fiber and the blossom. Only one is an actual rare though. Ancient wood is probably an easier test since they are similar. They have the ebony orbs and Hidden Treasure.

[Devildoc.6721]

> @"Inculpatus cedo.9234" said:

> Just like with Salvaging Kits, or even Black Lion Chests, I don't think the word 'rare' is referencing Rare materials.

 

Except that his testing already showed no increase in the exotic materials or fine/masterwork materials, so it might literally mean rare item quality. Either that or the buff does nothing. Could use an anet dev clarification

 

Always remember that developer definitions do not line up with player definitions.

[thehipone.6812]

Well, here are some initial flax results. I did get more piles of seeds with the boosts vs. the no boost condition. Fibers were almost identical between no buff and tavern, but the banner seemed to give more fibers. I guess I thought that blossoms were more on par with orbs and were more common than they actually are. Blossoms seem to be closer to freshwater pearls in terms of drop %. Yes, I did get more with the boosts, but I also got the most with the tavern boost only. I think there were too few blossom drops to really draw a conclusion. I don't think that the boosts truly double or triple your chances of getting a blossom (could be wrong, though, who knows). So, if you're bothering to gather, might as well still stack up the boosts.

 

 


 

 

[thehipone.6812]

OK, I'm back and now with more info. I'm up to 2000 flax gathers at each buff condition and the results with the higher number have shifted slightly from the original 500 swing tests. Results are here:

 


 

The number of flax per swing converged to about 1.5 for each condition. That is, you seem to have a 50/50 chance to get either 1 or 2 piles of flax seeds with each swing of the sickle. What is interesting to me is how the results evolved as testing was going on. The 2000 strike results are a bit different from what it looked like at 500 strikes, highlighting just how much gathering is needed to get firm numbers. I kept the plots below to see when things looked to be converging to long term averages. For the seeds, the curves are still fluctuating, but it is between a pretty narrow range of 1.48-1.50. Interestingly, it looks like the no buff condition was really unlucky to begin with in terms of getting 2 piles of seeds from a harvest, but this averaged out in the long term.


 

For fibers, if you assume the buffs can stack, a model of base drop rate of about 8.5% gives about 10% with the tavern buff and about 11.7% with both buffs, which isn't too far off of the data, assuming that perhaps the tavern buff only drop rate of fibers turned out to be a bit low. In the plot above, I might expect the orange line to turn up slightly, but the ordering of the different buff conditions seems to make sense.

 

Blossoms are quite rare, too rare to determine any conclusive drop rate. But the buffs did seem to increase the number, but again the tavern buff condition gave slightly more than with both bufffs (only higher by 1 blossom though). If we assume that the base drop rate is 1% with no buffs (no buff data is only 1 blossom short of this), then the data fits the stacking buff assumption OK. The tavern buff only condition produced about 4 more blossoms than would be expected. A **much** larger data set is needed to pin these number down any more. But we do now have decent estimates to figure out the value of these gathering buffs. If you are gathering, you will probably run the buffs anyways, but with the assumed drop rates above, the tavern buff is only worth about an extra 4 copper per swing on a flax node. If we assume that mussels/freshwater pearls share similar drop rates, then the buff increases to a value of around 1 silver per swing on mussel nodes.

[casualkenny.9817]

Thanks for sharing your results... I'm just wondering if you can get a double strike on lw3 wood and harvest currency nodes? I never really paid attention, though I'll try to see for myself.

 

I'm also wondering if these buffs can affect the relevant glyphs. I am guessing the leather/cloth ones would follow the sprocketer (ie 0), but what about the lucky strike glyph..? Though that's just out of curiosity - dun harvest enough to bother getting it, too hefty for me

[TheNecrosanct.4028]

> @"casualkenny.9817" said:

> Thanks for sharing your results... I'm just wondering if you can get a double strike on lw3 wood and harvest currency nodes? I never really paid attention, though I'll try to see for myself.

>

> I'm also wondering if these buffs can affect the relevant glyphs. I am guessing the leather/cloth ones would follow the sprocketer (ie 0), but what about the lucky strike glyph..? Though that's just out of curiosity - dun harvest enough to bother getting it, too hefty for me

 

You can get double strikes from LW nodes. When I go on gathering runs I use the full guild boost (20%), a banner (15%) and a gathering booster (33%) and I've had plenty of times where I could mine a LW3 again after the first 3 strikes.

 

I can also safely say that practically all the Maguuma Lilies and Freshwater Pearls I got were when I used those 3 boosts, whereas I get practically none when I just gather randomly without the boosts, while doing other stuff on the map.

 

My experience is also that I get more extra strikes/rare drops when I use all 3. I do not have numbers or anything like that, but as soon as my banner runs out and I only have the guild boost and gathering booster, extra strikes go down noticeably.

[thehipone.6812]

> @"casualkenny.9817" said:

> Thanks for sharing your results... I'm just wondering if you can get a double strike on lw3 wood and harvest currency nodes? I never really paid attention, though I'll try to see for myself.

>

> I'm also wondering if these buffs can affect the relevant glyphs. I am guessing the leather/cloth ones would follow the sprocketer (ie 0), but what about the lucky strike glyph..? Though that's just out of curiosity - dun harvest enough to bother getting it, too hefty for me

 

Here's what I have on the LW S3/S4 nodes:

I am in the process of testing blood rubies and petrified wood, and have not looked at winterberries nor primordial orchids.

I do not believe that you can get a double strike from the LWS3 nodes in the way that I have defined double strike for the hard wood logs above - one swing of the axe that generates two separate drops of 1 log each. If you chop a few trees and watch the chat tab you'll see what I mean. You will never swing an axe once and get two petrified wood from one swing of the axe. For bonus strikes, yes, you can get bonus strikes on some nodes. Vaguely related -I do not recall ever getting a bonus strike on a rich metal node. Also strange- the coral in the water under the central part of Bitterfrost also seems to not give bonus strikes (or else I had a super bad run of luck).

 

Apparent drop rates for the LWS3/S4 items and effect of boosts:

Petrified wood - seems to be a 33% chance to get one for each swing of the axe (plus 2 empyreal per swing), but I have the least data for these because of they daily cap

*Jade shards* - I'm up to 1000 swings of the pick at each of the no buff and tavern+banner buff conditions and the chance is still 33% for both (plus 2 bloodstone dust avg. per swing)

*Difluorite* - I have 500 swings at no buff and 500 at tavern+ banner, 33% chance for both (plus 2 empyreal)

Bloodstone - still working through, but no buff is currently at 355 swings and about 10% chance of getting a ruby each swing. Full buff is around 12.5%, but the sample size is still too small to conclude

Orrian Oyster - Always get 1, so no possible effect

Fire Orchid blossom - Have not tested, I believe that they are 33% chance and give avg. of 2 empyreal

Winterberry - I need to go back and look more closely at how exactly everything drops (1 berry vs. 2 vs. 1 and 1) and how the dragonite drops

 

So there are two hypotheses that I'm currently considering, depending mainly on how the blood ruby data turns out:

1. The buffs do not have any influence on the gathering rates of the exotic-quality drops (difluorite crystal, petrified wood, etc.)

2. The buffs do influence the gathering rates, but the chance is capped at 33%. Basically meaning that only the blood rubies and possibly winterberries could benefit from buffs.

 

 

I would guess that glyphs do not benefit from any buffs, but I have not set out to test it. As an aside, Glyph of bounty is awesome, especially on plant nodes. It absolutely turbocharges the flax farm. As many as 80 piles of seeds from just hitting the Itzel, draconis mons, and bloodstone fen sets of nodes. Considering the relatively high per-swing value of a lot of plant nodes (mussel, ghost pepper, orrian truffle, snow truffle, etc.) I feel like the gathering sickle is the best place to put a glyph of bounty.

 

 

 

> @"TheNecrosanct.4028" said:

>

> I can also safely say that practically all the Maguuma Lilies and Freshwater Pearls I got were when I used those 3 boosts, whereas I get practically none when I just gather randomly without the boosts, while doing other stuff on the map.

>

> My experience is also that I get more extra strikes/rare drops when I use all 3. I do not have numbers or anything like that, but as soon as my banner runs out and I only have the guild boost and gathering booster, extra strikes go down noticeably.

 

I'd encourage you to keep numbers to really check if what you are seeing is true. When you're loaded up on boosters you will probably be gathering a lot more nodes, which will naturally produce more of the rare items simply by more rolls. Human brains and rare random occurences do not always play nice together. Toss in a bit of placebo effect from having the boosts and superstitions can be born.

 

[Khisanth.2948]

How are you recording your data? A notepad(both digital and physical) got a bit annoying.

[thehipone.6812]

> @"Khisanth.2948" said:

> How are you recording your data? A notepad(both digital and physical) got a bit annoying.

 

I keep the overall data in excel to analyze. For recording the data while gathering - As long as you prepare your inventory beforehand and pay attention to not do things that could influence results (i.e. don't accept/open chests, salvage, open bags, deposit all, don't kill elementals while testing crystal drops from ore, etc.), it really isn't much work to keep track of. If I'm collecting a living world node I just count the number of nodes that I hit and multiply by 3 to get the number of strikes. I have a solo guild, so to keep a running tab while gathering I just open up that chat tab and every time I start gathering a node I type the running total number of nodes into that chat tab while the gathering animation is playing. At the end you just read off the final number of nodes and count up what you have in inventory and punch it into excel. You can always go back in the chat tab to check the count, if needed. Things like the 50 node/day cap on jade shards also act as a check to make sure the node counts are right. For basic ore (I ignore all rich veins while testing), you don't even need to count because you only get 1 ore per strike, so the number of ore that you have at the end equals the number of strikes. Count the ore, count up the number of crystals and you're good. For logs I keep a running tab in chat of how many times I get a double strike. I can always double check the number of logs in inventory to confirm. If I have counted 4 double strikes, then the number of logs in my bags better equal a multiple of 3 plus the extra 4 logs. I never run glyph of bounty on my axe or pick to make it really easy. Otherwise you have to count swings to account for bonus strikes. For flax, I just count the number of strikes as I go through each farm and type running totals into personal guild chat before moving to the next map. Winterberries look like they will require a bit of more complicated recordkeeping though.

[Khisanth.2948]

Ah ... I am mainly interested mussels so that has a similar issue to winterberries.

[Tanner Blackfeather.6509]

> @"thehipone.6812" said:

> For basic ore (I ignore all rich veins while testing), you don't even need to count because you only get 1 ore per strike, so the number of ore that you have at the end equals the number of strikes.

 

I haven't keep track enough to be absolutely certain this happens with non-rich nodes too, but for Iron ore I fairly often get 2 per swing. Not a double strike, but an actual loot result of "2 Iron Ore" rolling by on the screen.

Rich Iron vein usually nets me 13-14 Iron Ore with 10 actual strikes, while Rich Platinum always give exactly 10.

[thehipone.6812]

> @"Tanner Blackfeather.6509" said:

> > @"thehipone.6812" said:

> > For basic ore (I ignore all rich veins while testing), you don't even need to count because you only get 1 ore per strike, so the number of ore that you have at the end equals the number of strikes.

>

> I haven't keep track enough to be absolutely certain this happens with non-rich nodes too, but for Iron ore I fairly often get 2 per swing. Not a double strike, but an actual loot result of "2 Iron Ore" rolling by on the screen.

> Rich Iron vein usually nets me 13-14 Iron Ore with 10 actual strikes, while Rich Platinum always give exactly 10.

 

You are correct. I have not tested iron, only platinum, mithril and ori. Iron is an exception that you would need to count the number of strikes or nodes.

 

> @"Khisanth.2948" said:

> Ah ... I am mainly interested mussels so that has a similar issue to winterberries.

 

Mussels should be straightforward if you just keep a running total of the number of strikes and then count up the number of mussel, mother-of-pearl, and freshwater pearls.

 

Winterberries are annoying because the second drop slot is either dragonite, or 1 winterberry or 2 winterberries. If you only care about average number of winterberries, then you can just count the swings and be done with it. If you want to see, say, if the buffs increase the chance to get any additional berry vs. dragonite or if they change the frequency of 1 vs. 2 berries in the second drop slot then you need to track more info. If you get say, 10 more berries than the total number of swings you can't figure out if you got 10 single berries in the second slot, 5 double berries, or any result in between.

 

[Neural.1824]

This is not surprising at all really. Thank you for doing the work and providing the results to support it.