[Research] Different Salvage Kits for Unidentified Gear and Magic Find for IDing

I posted some research data about Unidentified Gear on Reddit the last couple of days and thought I just give a quick summary here.

• First, I identified 10 stacks of Common Unidentified Gear with 311% MF, which is my permanent account bonus to it from essences of luck and Achievement Chests.

• Then I identified 10 stack of Common Unidentified Gear with 581% MF, using temporary buffs.

Result:

• Temporary Buffs to Magic Find affect gear identification

I also salvaged 1k each Unidentified blue and green rarity gear each with a basic and master salvage kit, so 4k in total.

Result:

• A higher rarity kit gives you slightly more materials

Most players will be interested in which method will be the most profitable for them to handle their Unid Gear drops but I deliberately left out information about gold values because

• I didnt check because the API was down until yesterday
• The material markets are in disequilibrium since friday and you have to make your own choice of which value you give each mat
• it depends on your personal Magic Find
• for the RNG involved my sample sizes are too small
• I havent checked with 0% magic find on my f2p account yet

But in general, you have 3 choices:

Salvage directlyThis currently seems the worst method

Sell on the TP or ID yourself?

As it depends on your MF you should most likely sell it on the tp undercutting the lowest listing or a bit above, if your account magic find is low and you are not running plenty of temporary buffs. Because someone like me with over 500% MF will most likely get 30-50% more value out of the identification process and therefore also offer more silver for them on the TP than you would get from IDing.

I know plenty of players are running their stocked up magic find/exp and other boosters atm, so if thats the case with you, you might want to consider IDing yourself to unlock some new skins.

Changes from demo weekend to PoF launch

Some changes were made how Unid Gear works from the demo to the live server, some of them you probably know already but I think its worth mentioning them again:

• The ID process has been moved from NPCs gated behind content completion (heart vendors in Crystal Oasis) to double click in your inventory
• gold costs for identification have also been removed
• Unid Gear now drops in different rarities (blue, green, yellow)
• Eyes of Kormir have been removed from the loot table of identifying gear as direct drops (unless they drop from IDing green Unids but not blue Unids)

The loot table for salvaging Unids got a complete overhaul. During demo you got:

• 1-25 unrefined common mats of all tiers (wood, cloth, leather, metal including gold and silver ore)
• all tiers had same droprate (assuming from small sample size)

Now you get:

• 1-5 common t5 and t6, way more t5 than t6 mats
• essences of luck
• eyes of kormir
• Congealed Putrescence

Happy salvaging/identifying!

Edit:If a Dev is reading this and is allowed to give out information about this, I would be interested, if these changes were made based on community feedback or if some mechanics and loot tables have been intently been altered or left out during the demo to protect the economy.

Interesting research.

Personally if selling on the TP ever loses value I'd still salvage directly for simple convenience, but it's interesting to see what the difference is.

For those interested, there's also a discussion about Wanze's research taking place on Reddit.

Identification Results Compared in Table Form

Identification Results Compared in Text Form

Advantage of adding 270% MF

• ~4% fewer T5 metal, wood, cloth, leather
• ~15% fewer T6

• 60% more rares

• <60% more exotics

Conclusions

• Magic Find makes a significant difference, but not enough that you'd easily notice (reminder: Wanze looked at 10 stacks, i.e. 2500 gear drops!)
• There are noticeably more rares & exotics, which result in somewhat fewer T5 and a lot fewer T6 mats from salvaging
• Temporary MF matters

I thought it would also be useful to compare salvage versus identification. Wanze salvaged only 4 stacks compared to 10 that were identified, so I multiplied the results by 2.5

Salvage Results

Relative Comparison of Salvage Results

• Copper Fed vs 581% MF: ~35% fewer T5 gear-related mats, ~14% fewer T6
• Copper Fed vs 311% MF: 37% fewer T5, 25% fewer T6
• Silver Fed vs Copper Fed: ~38% more T5, ~12% fewer T6

Conclusions Regarding Salvage versus Identification

Unless you're very short on time or patience, it's always worth identifying if you have any decent level of Magic Find. Wanze hasn't yet published results from a low-MF alt account, so salvaging might be ok if MF is very, very low.

And repeating from above: the market remains too volatile to say whether it's better to sell on the TP rather than identify. However, that's probably the best option for the impatient/time-crunched player if you price aggressively (i.e. closer to the lowest sell offer rather than instant-sell).

No one in particular said:That's a huge wall of text and data. Is there a tl;dr?

• Magic Find matters for unidentified gear.
• Temporary MF buffs count, so it's worth going to Silverwastes for the Perseverance buffs then using all your temp buffs/boosters
• If you have the time, Identifying is better than salvaging.
• If you don't have the time, selling aggressively (at approximately lowest sell offer) is probably best; market remains too volatile to say for sure.

Thanks for the info. I'm generally running around with ~350% magic find lately. What this tells me is I should always open unidentified gear (which I have been doing) instead of salvaging it. I occasionally make the mistake of right-click salvage <= masterwork while I still have unidentified items in my inventory :(

Guess it's time to invest in one of the hidden item bags or so and put the unidentified items in there so they aren't salvaged. Inventory Wars!

Thanks for all the information.Just for clarification: Identifiying Unid Blue gear can yield rares and exotics as well - and is worth it?So far I've collected green and yellow Unid Gear to identify them when I stack up on mf because I was under the impression that the blues aren't worth it as long as you're not into drowning in minor sigils/runes. But apparently I'm wrong?

@biofrog.1568 said:Thanks for the info. I'm generally running around with ~350% magic find lately. What this tells me is I should always open unidentified gear (which I have been doing) instead of salvaging it. I occasionally make the mistake of right-click salvage <= masterwork while I still have unidentified items in my inventory :(

Guess it's time to invest in one of the hidden item bags or so and put the unidentified items in there so they aren't salvaged. Inventory Wars!

Yes, this is what I've been doing, except putting the unidentified thingies up in the shared inventory slots where my permanent harvesters were. I should get an hidden item bag too. Also, as IWN said a couple posts above, maybe I should let them stack a bit then head to Silverwastes for the extra magic find buff.

What do you call a low MF? 50? 150? 250?

I would need to check but I am somewhere north of 150 and always ID.

@Menadena.7482 said:What do you call a low MF? 50? 150? 250?

I would need to check but I am somewhere north of 150 and always ID.

We did some research, salvaging 1k RARE unid gear with 341% and 1089% magic find a couple of days ago and had a look at the exotic droprate (you only get rare or exotic gear from rare unids):

With 341%, we got 50 exotics and with 1089% we got 145 exotics. From this we concluded that the exotic droprate without any mf is around 1.2%.

For each 100% MF, that droprate would double.

That's very useful info; thanks for updating.

@Wanze.8410 said:With 341%, we got 50 exotics and with 1089% we got 145 exotics. From this we concluded that the exotic droprate without any mf is around 1.2%.

For each 100% MF, that droprate would double.I wonder if it's enough data points to draw that conclusion. Two samples is enough if the rate rises linearly with MF, but might not be if there's a lot more variance in rates than can be seen with a single 1000-bag test at each value, if there are 'step functions' involved, if there are nested drop tables, if there are multiple ways for some exotics to drop in the various tables, etc. From other research (much of it done by you), we know that ANet sometimes likes to use more complicated tables, to reduce the difference between the luckiest and unluckiest, which sometimes results in more variance from average than a simple coin toss.

In other words, I think it's safe to conclude that more MF has a profound impact on the drop rate of exotics and that it's worth boosting it as much as possible. I'm not sure it's safe to draw any other conclusions yet, at least not with the limited data I've seen to date.

In terms of market value:

• Lets value rares at 26 silver = .875% of ecto value + 2.5s (the loltools value per bag at L80, from salvaging)
• Rare unID gear has a market rate of ~74s (was closer to 50s before someone published data on reddit a few days ago)

Then dropped exotics need to be worth an average of nearly 5 gold each to be worth the cost of buying rare unID bags at MF 1089%. At 'only' 341%, they'd have be worth, on average nearly 14 gold. Given that some specialization weapons trade at 65-70 g, those numbers drop to about 2.5g for high MF (assuming ~4 such weapons drop) and 'only' 11g for low MF.

More simply, if you don't boost your MF to 1k, you are likely to be better off selling the unID gear on the TP and let someone who does open them up.

PS did you publish the data for that yet? I'd love to add it to my collection. Thanks

@"Illconceived Was Na.9781" said:PS did you publish the data for that yet? I'd love to add it to my collection. Thanks

I posted my data in the same reddit thread a couple of days ago that you are probably referring to.

I made the 341% mf sample. take whatever you need from there.

@Illconceived Was Na.9781 said:I wonder if it's enough data points to draw that conclusion. Two samples is enough if the rate rises linearly with MF, but might not be if there's a lot more variance in rates than can be seen with a single 1000-bag test at each value, if there are 'step functions' involved, if there are nested drop tables, if there are multiple ways for some exotics to drop in the various tables, etc. From other research (much of it done by you), we know that ANet sometimes likes to use more complicated tables, to reduce the difference between the luckiest and unluckiest, which sometimes results in more variance from average than a simple coin toss.

In other words, I think it's safe to conclude that more MF has a profound impact on the drop rate of exotics and that it's worth boosting it as much as possible. I'm not sure it's safe to draw any other conclusions yet, at least not with the limited data I've seen to date.

For all i know, Anet has only put the same item in different sub-tables of a loot table to adjust volume per roll in order to have magic find not only affect quality but also quantity of your loot. So one item might be on the common loot table with a volume of 1 but the same item might be on the uncommon loot table with a volume of 5 and the rare loot table with a volume of 10.But you can only get 1 piece of gear from the ID process.

I am pretty sure that rare Unids have several different drop tables for Exotic gear with different base droprates (onto which your mf is being applied) but that doesnt really matter, if we only calculate the general exotic droprate of rare unids because our mf is applied to each loot table the same way. I think the only way to make our data inconclusive would be, if some of the sub tables had rare and exotic gear in them but I think thats highely unlikely.

In terms of market value:

Lets value rares at 26 silver = .875% of ecto value + 2.5s (the loltools value per bag at L80, from salvaging)Rare unID gear has a market rate of ~74s (was closer to 50s before someone published data on reddit a few days ago)Then dropped exotics need to be worth an average of nearly 5 gold each to be worth the cost of buying rare unID bags at MF 1089%. At 'only' 341%, they'd have be worth, > on average nearly 14 gold. Given that some specialization weapons trade at 65-70 g, those numbers drop to about 2.5g for high MF (assuming ~4 such weapons drop) and > 'only' 11g for low MF.

More simply, if you don't boost your MF to 1k, you are likely to be better off selling the unID gear on the TP and let someone who does open them up.

I just liquidated my loot table from the 341% sample, all listing prices, disregarding tp fees:

• The 50 exotics were listed for a total 219g
• elder wood: 8g
• ancient wood: 1.08g
• thick leather: 2.36g
• hard leather: 2.92g
• mithril: 1.22g
• ori: 0.4g
• silk: 0.70g
• gossamer: 0.03g
• Ectos: 221g
• eyes of kormir: 6g

So a total of 462.71g, of that are 243.71g from the rares (25.65s per rare) and 22.71g from common mats, including EoC (2.39s).

The eyes of kormir come from salvaging rare elonian weapons and we didnt record the ratio between those and other rare salvages but it doesnt really make a difference imho.

462 gold from 1000 rare unID? When I was typing my reply, those were selling at 70s, making the four stacks worth 700g (~600g after fees), i.e. better to have sold them.(As I type this reply, the price was down to 50-60s, as it had been last week, so the math is less conclusive.)

My recommendation remains: if you don't have the patience to boost your MF to optimal levels, it's financially better to sell unID gear instead of opening.

@Illconceived Was Na.9781 said:462 gold from 1000 rare unID? When I was typing my reply, those were selling at 70s, making the four stacks worth 700g (~600g after fees), i.e. better to have sold them.(As I type this reply, the price was down to 50-60s, as it had been last week, so the math is less conclusive.)

My recommendation remains: if you don't have the patience to boost your MF to optimal levels, it's financially better to sell unID gear instead of opening.

You mean it would have been better to not have have bought them in the first place. I just checked, I paid an average of 51s for them.

But I agree with your recommendation though, the vast mayority of players will be better off selling on the tp. But I could have told you that after the demo weekend. That they scratched the ID costs (2.xx silver) didnt make a difference.

I wouldnt open them with less than 700% mf and most players will only rarely have the chance of getting that high, so once they do, they shouldnt be doing anything else than IDing Unid gear.

When I was doing my research samples, it took around an hour to ID and salvage 5k Unids but keeping all the relevant drops in the process for research slowed it down quite a bit, so I would wager that you can probably manage 10-15k unids per hour, if you know your shit and have the inventory for it, if you dont bother about results.

So unless players have 20 stacks of unids lying around somewhere to make full use of that max magic find for an hour, you are only wasting your limited buffs, like celebration boosters.

I will definately use my celebration boosters from dailies to do this once a month but I have little use for it in other content. IF you still need experience buffs for other stuff, I wouldnt waste celebration boosters on idying unid gear, especially, if you arent capped on account mf and want to ID less than 10 stacks.

For the rare Unids. I would currently estimate that you break even around 700% mf with all the exotics you get, so every precursor that drops, will be your profit. But we dont really have enough data to determine the droprate for those yet.