Talk:Gem drop table

From RuneScape Classic Wiki
Jump to navigation Jump to search

The "ultra" rare drop table[edit source]

I've discovered that four additional monsters (King Black Dragon, Black Dragon, Black Demon, Fire Giant) appear to have access to an additional rare drop table that is much more like the OSRS version of the rare drop table. On OSRS wiki we separate these out into "Gem drop table" and "Rare drop table", but since RSC already has the idea of the "Rare drop table" associated with the subroutine that only gives gems/key halves, I've decided to call this one the ultra-rare drop table. (This is actually more consistent with the internal names Jagex used).

On OSRS, those four monsters are the only pre-RS2 monsters that have access to the URDT. They are also the only four monsters on RSC that, according to the old wiki drop tables, give rare/very rare access to an interesting assortment of items. I speculate that these are all on the same subroutine, and this is made a lot easier by the fact that this URDT has almost exactly the same items in OSRS as it seems to have in RSC.

The evidence sources here are either:

  • Wiki edits adding the same rare items to different monsters. Generally I'm looking for edits from multiple distinct users, especially on different monsters.
  • Screenshots that Luis shared with me from fire giant bots that ran from 2017-2018. Probably 100k+ kills there, but it only logs items that the creator wanted it to log. This means that certain rare items may have been dropped, but ignored.

Near-certain items[edit source]

  • 20 Silver certificate - Already present on KBD/Black Dragon/Fire Giant articles. Fairly common in the screenshots. Live on in OSRS as 100 Silver ore, 2/128 rarity. Sample mean of 2.2/128 on 1000 rolls.
  • 20 Coal certificate - Already on KBD/Black Dragon articles. Equally common on the screenshots. Does not exist on OSRS. Sample mean of 1.8/128 on 1000 rolls.
  • Dragonstone - Common in screenshots, and listed on Black Dragon/Black Demon/Fire Giant articles. Uncut dragonstone is also listed on KBD (and used to be on Black Dragon, until this edit, which makes me think Uncut/cut dragonstones are easily confused, and they are all actually regular Dragonstones. Discussions with other players suggest they have never seen an uncut drop. Dragonstone is 2/128 in OSRS, sample mean of 1.41/128 on 2200 rolls.
  • Rune 2-handed Sword - Common in screenshots, also present on all 4 monsters' pages. 3/128 on OSRS, sample mean 4.06/128 on 2200 rolls.
  • Rune battle Axe - Common in screenshots, also present on all 4 monsters' pages. 3/128 on OSRS, sample mean 2.65/128 on 2200 rolls.
  • Rune Axe - Common in screenshots, on 3 of 4 monster pages. Not in OSRS. Sample mean of 2.83/128 on 2200 rolls.
  • Dragon medium Helmet - Fairly rare in screenshots, on 3 of 4 monster pages. 1/128 in OSRS. Sample mean of 0.53/128 on 2200 rolls.
  • Rune Kite Shield - Common-ish in screenshots, on all monster pages. 1/128 in OSRS. Sample mean of 0.88/128 on 2200 rolls.
  • Rune Square Shield - Not listed in screenshots, but on all monster pages. 2/128 on OSRS.
  • Runite bar - Common in screenshots, present on all pages. 5/128 on OSRS, sample mean 2.00/128 on 2200 rolls.
  • 30 Death-Rune - Common in screenshots, 2/128 to drop in OSRS. 1.76/128 sample mean in 2200 drops.
  • 45 Nature-Rune - Not in screenshots, but added by this IP with 10000 kills to Black Dragon. Fits with OSRS quantity * 1.5, and is 3/128 there.
  • 30 Law-Rune - Not in screenshots, but added to Black Dragon by Mith. 2/128 in OSRS.
  • 3000 Coins - Only on the KBD page and added by an IP, but it's believable that people wouldn't pick up what they assume are a very small number of coins from the other mobs (and there's no other visual indication that it's a good amount). This is a 21/128 drop on OSRS, but I have no way of estimating for the available drop logs.
  • 300 Bronze Arrows - Added by Mith to Black Dragon. Not in OSRS, but there are some other arrows drops.
  • Half of a key - A bit hard to tell because it's also from GDT, but my best guess is that fire giants access the GDT about 5/128 (based on 285 rubies in a ~122k sample, and assuming 1/16 on rubies from GDT). I also think key halves are each 1/128 from GDT. This would normally put us at about 169 halves in a 278k sample, but instead we got 759. That suggests that GDT has around a 34/128 chance of giving a key half (maybe 17/128 for each?)

Probable items[edit source]

Possible-but-unlikely items[edit source]

  • 150 Steel Arrows - Added by the famous Black Demon IP in 2011, but also had some incorrect drops. This exact amount of steel arrows is dropped in OSRS with 2/128 rarity. I seriously question this editor's legitimacy because these exact quantities were on RSW's black demon page at the time of the edit.
  • ??? Rune Arrows - Added by the famous Black Demon IP as 12-150 quantity in 2011, but also had some incorrect drops. 42 are dropped in OSRS with 2/128 rarity. I seriously question this editor's legitimacy because these exact quantities were on RSW's black demon page at the time of the edit.
  • Adamant javelin - Obviously doesn't exist in RSC, but it's the only post-2003 item on the main URDT. It's possible that it was replaced by something else.

Sub-tables[edit source]

In OSRS, the URDT also rolls on the gem drop table and mega-rare (spears and shield half) table with 20/128 and 15/128 rates respectively. We saw 3 shield halves in about 278,000 fire giant kills. Keep in mind we think the fire giant gem table rate is 5/128 (based on rubies), and then in OSRS it's 1/128 from there to promote to mega-rare, and then 4/128 to hit to If the rates are the same as OSRS, then 278k * 5/128 * 1/128 * 4/128 = 2.65 expected shield halves just from the gem part of the table. If it's also the same on the URDT side, then it's 278k * 1/128 * 15/128 * 4/128, which would be another 7.95 shield halves. We only saw three, total, in the entire set. That makes me think URDT does not have a sub-roll on mega-rare.

As for URDT to gem, it's really impossible to tell. My hunch is that if they aren't subroutining to mega-rare, they're probably not subroutining to gem either...but who knows?

Regarding red dragons[edit source]

There is a small chance that red dragons also can roll on URDT, but I think this is unlikely.

  • There is also an Uncut dragonstone listed on Red Dragon from this edit by Stormy but I think this may be an error. Red Dragons do not roll URDT on OSRS, and the drop table on the wiki here doesn't currently mention any of the other URDT drops. Stormy's edit also included the Dragon medium helmet and Shield half.
  • The dragon med helm/shield half were removed by an IP who claimed Jagex confirmed they don't drop them, but this was reverted. * Stormy removed the dragon med helm after changing their mind.
  • Mith removed the shield half in a larger cleanup.
  • RSC2001 later said he killed red dragons for weeks and never saw an uncut dragonstone, and also never added any of the other obvious URDT drops.

Overall, I think this is fairly compelling evidence that red dragons do not roll on URDT, and likely don't drop any type of dragonstone, nor the dragon med helm. They would still access the GDT and technically give access to the shield half.

Best guesses for tables[edit source]

Gem/"Rare"[edit source]

Item Quantity Rarity
Uncut sapphire.pngUncut sapphire10.250000Common (~1/4)
Uncut emerald.pngUncut emerald10.125000Common (~1/8)
Uncut ruby.pngUncut ruby10.062500Common (~1/16)
Uncut diamond.pngUncut diamond10.015625Uncommon (~1/64)
Half of a key (loop).pngHalf of a key (loop)10.007812Rare (~1/128)
Half of a key (tooth).pngHalf of a key (tooth)10.007812Rare (~1/128)

Plus a 1/128 chance to subroutine the Mega table. The rest (65/128) is nothing.

Mega[edit source]

It is exceedingly likely that these are the only items here, and that it's accessible from the gem table. The rates are conjectural but are based on OSRS.

Item Quantity Rarity
Rune Spear.pngRune Spear10.062500Common (~1/16)
Half Dragon Square Shield (left).pngHalf Dragon Square Shield (left)10.031250Uncommon (~1/32)

Ultra[edit source]

Item Quantity Rarity
Certificate.pngSilver certificate200.015625Uncommon (~2/128)
Certificate.pngCoal certificate200.015625Uncommon (~2/128)
Dragonstone.pngDragonstone10.015625Uncommon (~2/128)
Rune 2-handed Sword.pngRune 2-handed Sword10.031250Uncommon (~4/128)
Rune battle Axe.pngRune battle Axe10.023438Uncommon (~3/128)
Rune Axe.pngRune Axe10.023438Uncommon (~3/128)
Dragon medium Helmet.pngDragon medium Helmet10.007812Rare (~1/128)
Rune Kite Shield.pngRune Kite Shield10.007812Rare (~1/128)
Rune Square Shield.pngRune Square Shield10.015625Uncommon (~2/128)
Runite bar.pngRunite bar10.015625Uncommon (~2/128)
Death-Rune.pngDeath-Rune300.015625Uncommon (~2/128)
Nature-Rune.pngNature-Rune450.023438Uncommon (~3/128)
Law-Rune.pngLaw-Rune300.015625Uncommon (~2/128)
Coins.pngCoins30000.164062Common (~21/128)
Bronze Arrows.pngBronze Arrows3000.015625Uncommon (~2/128)
Half of a key (loop).pngHalf of a key (loop)10.132812Common (~17/128)
Half of a key (tooth).pngHalf of a key (tooth)10.132812Common (~17/128)

Perhaps the one people care about most is the Dragon medium Helmet. The sample mean was closer to 1/256 (9 in ~2171 rolls), and while it's somewhat unlikely (p = 2.6%) to get that unlucky for a 1/128 drop, the fact that it's 1/128 in OSRS, and that every other drop table I've ever seen in RSC is out of 128, makes me think it's even less likely to be 1/256.

This only adds up to 86 out of 128. I think the rest is empty, but we could be missing other stuff. ʞooɔ 03:02, 5 May 2020 (UTC)

Comments on above[edit source]

Great writeup, I posted on the reddit thread about it, but I figure it's good to have it here too, I'm pretty sure that RS2 replaced the Adamantite Spear on the drop table with Adamant Javelin. Hubcapp (talk) 02:23, 6 May 2020 (UTC)

In case you're not aware, spears in RSC work the same as Javelins did in RS2, up until Jagex retconned them as ballista ammo for OSRS :-) Hubcapp (talk) 02:28, 6 May 2020 (UTC)

April 24, 2021[edit source]

Jagex's Mod Ash replied to a question regarding the rare/ultrarare/mega drop tables https://twitter.com/p2_pure/status/1385743851130802182 —Preceding unsigned comment added by 75.64.57.118 (talk • contribs) 18:13, 24 April 2021

Which allows the following comparison between guesstimated rates and real rates:

Approximation Real
Item Quantity Rarity
Certificate.pngSilver certificate200.015625Uncommon (~2/128)
Certificate.pngCoal certificate200.015625Uncommon (~2/128)
Dragonstone.pngDragonstone10.015625Uncommon (~2/128)
Rune 2-handed Sword.pngRune 2-handed Sword10.031250Uncommon (~4/128)
Rune battle Axe.pngRune battle Axe10.023438Uncommon (~3/128)
Rune Axe.pngRune Axe10.023438Uncommon (~3/128)
Dragon medium Helmet.pngDragon medium Helmet10.007812Rare (~1/128)
Rune Kite Shield.pngRune Kite Shield10.007812Rare (~1/128)
Rune Square Shield.pngRune Square Shield10.015625Uncommon (~2/128)
Runite bar.pngRunite bar10.015625Uncommon (~2/128)
Death-Rune.pngDeath-Rune300.015625Uncommon (~2/128)
Nature-Rune.pngNature-Rune450.023438Uncommon (~3/128)
Law-Rune.pngLaw-Rune300.015625Uncommon (~2/128)
Coins.pngCoins30000.164062Common (~21/128)
Bronze Arrows.pngBronze Arrows3000.015625Uncommon (~2/128)
Half of a key (loop).pngHalf of a key (loop)10.132812Common (~17/128)
Half of a key (tooth).pngHalf of a key (tooth)10.132812Common (~17/128)
File:.png[[]]1.000000Unknown
File:.png[[]]1.000000Unknown
File:.png[[]]1.000000Unknown
File:.png[[]]1.000000Unknown
File:.png[[]]1.000000Unknown
File:.png[[]]1.000000Unknown
Item Quantity Rarity
Certificate.pngSilver certificate200.015625Uncommon (~2/128)
Certificate.pngCoal certificate200.015625Uncommon (~2/128)
Dragonstone.pngDragonstone10.015625Uncommon (~2/128)
Rune 2-handed Sword.pngRune 2-handed Sword10.023438Uncommon (~3/128)
Rune battle Axe.pngRune battle Axe10.015625Uncommon (~2/128)
Rune Axe.pngRune Axe10.023438Uncommon (~3/128)
Dragon medium Helmet.pngDragon medium Helmet10.007812Rare (~1/128)
Rune Kite Shield.pngRune Kite Shield10.007812Rare (~1/128)
Rune Square Shield.pngRune Square Shield10.015625Uncommon (~2/128)
Runite bar.pngRunite bar10.015625Uncommon (~2/128)
Death-Rune.pngDeath-Rune300.015625Uncommon (~2/128)
Nature-Rune.pngNature-Rune450.015625Uncommon (~2/128)
Law-Rune.pngLaw-Rune300.023438Uncommon (~3/128)
Coins.pngCoins30000.195312Common (~25/128)
Bronze Arrows.pngBronze Arrows3000.015625Uncommon (~2/128)
Half of a key (loop).pngHalf of a key (loop)10.157471Common (~2,580/16,384)
Half of a key (tooth).pngHalf of a key (tooth)10.149658Common (~2,452/16,384)
Rune Spear.pngRune Spear10.000925Very rare (~1,940/2,097,152)
Half Dragon Square Shield (left).pngHalf Dragon Square Shield (left)10.002775Rare (~5,820/2,097,152)
Uncut sapphire.pngUncut sapphire10.039062Uncommon (~20/512)
Uncut emerald.pngUncut emerald10.019531Uncommon (~20/1,024)
Uncut ruby.pngUncut ruby10.009766Rare (~20/2,048)
Uncut diamond.pngUncut diamond10.002441Rare (~20/8,192)
Item Quantity Rarity
Rune Spear.pngRune Spear10.062500Common (~1/16)
Half Dragon Square Shield (left).pngHalf Dragon Square Shield (left)10.031250Uncommon (~1/32)
Item Quantity Rarity
Rune Spear.pngRune Spear10.007812Rare (~1/128)
Half Dragon Square Shield (left).pngHalf Dragon Square Shield (left)10.023438Uncommon (~3/128)

ZorakTalk 21:45, 24 April 2021 (UTC)

Can I ask how you got the rates for key halves, gems, and spear/shield? Are you using Cook's estimated gem table rates from above, or is there a more updated estimate/Ash confirmation on those rates? And shouldn't the gem table when hit by the ultra rare drop table also be able to access the super mega rare table like normal, therefore giving slightly higher rates for the left half and spear? For example, to calculate the chance of hitting a loop half of a key (assuming Cook's gem table is accurate), I did (20/128)+((20/128)*(1/128)) which came out to 2,560/16,384. For the left half, also assuming Cook's gem table, I did (15/128*3/128)+(20/128*1/128*3/128), which is 5,820/2,097,152. - Xerxespersrex (talk) 02:40, 25 April 2021 (UTC)
I did indeed assume Cook's gem table, which fits the data and aligns with confirmed OSRS rates. I am not realy sure what I did there yesterday evening, you're correct! Will modify. (20/128)+((20/128)*(1/128)) isn't 2560/16384 though, but 2560/16384 + 20/16384 :) --ZorakTalk 06:31, 25 April 2021 (UTC)
Oops, you're right. Thanks for catching that. How about your gem rates? For those I was doing 20/128 times the rate in the gem table, so for a diamond I was doing (20/128)*(1/64) = 20/8192. It looks like you made a simple mistake and multiplied by 1/20 instead of 20/128. - Xerxespersrex (talk) 14:28, 25 April 2021 (UTC)
Note to self: never do math when you're not awake ZorakTalk 20:35, 25 April 2021 (UTC)

RDT/URDT/HERB access rates per monster, and sources[edit source]

Monster Combat level randomjewel ultrarare randomherb
Man 9 0 0 22/128[rates 1]
Mugger 9 0 0 13/128 (OSRS); [7.55/128 - 15.48/128] (RSC)[rates 2]
Goblin 13 0 0 2/128[rates 3]
Farmer, Head Thief 15, 34 0 0 23/128 (OSRS); [19.29/128 - 23.97/128] (RSC)[rates 4]
Barbarian, Gunthor the Brave 16, 37

1/128[rates 5]

0 0
Dwarf 18 1/128[rates 6] 0 0
Thug 18 1/128[rates 7] 0 23/128[rates 8]
Warrior 18 0 0 23/128 (OSRS); [21.54/128 - 23.11/128] (RSC)[rates 9]
Chaos Druid 19 1/128[rates 10] 0 Separate analysis
Zombie 19 ? 0 [20.27/128, 31.16/128] (RSC)[rates 11]
Zombie 24 ? 0 0
Zombie 32 ? 0 [13.44/128, 20.47/128] (RSC)[rates 12]
Rogue 21 0 0 23/128 (OSRS); [19.13/128 - 28.54/128] (RSC)[rates 13]
Dark warrior 21 0 0 Needs rewrite. 3/128[rates 14]
Skeleton, Skeleton mage 25, 31, 54 1/128[rates 15] 0 Needs rewrite. 21/128 (lvl 25)
Pirate 27, 30 1/128[rates 16] 0 0
Mountain Dwarf 28 ? 0 0
Bandit 29 2/128[rates 17] 0 Separate analysis[1]
Druid 29 0 0 26/128 (OSRS); [22.08/128, 32.78/128] (RSC)[rates 18]
Hobgoblin 32 2/128[rates 19] 0 7/128[rates 20]
Hobgoblin 48 ? 0 7/128 (OSRS); [2.39/128, 7.55/128] (RSC)[rates 21]
Giant 37 (3/128) 0 [3.62/128, 9.21/128] (RSC)[rates 22]
Tribesman 39 (updated, likely 2/128) 0 [5.35/128, 11.95/128] (RSC)[rates 23]
Black Heather, Donny the Lad, Speedy Keith 13 2/128[rates" 1] 0 15/128 (OSRS); [13.66/128, 19.23/128] (RSC)[rates 24]
Chaos Druid Warrior 44 (1/128) 0 Separate analysis
Black Knight, Jailer, Renegade knight, Lord Darquarius 46, 51, 51, 76 ? 0 3/128 (OSRS); [1.97/128, 3.93/128] (RSC)[rates 25]
Earth Warrior 52 (updated, likely 2/128) 0 14/128 (OSRS); [11.54/128, 19.83/128] (RSC)[rates 26]
White Knight 56 (updated, likely 1/128) 0 5/128 (OSRS); [2.67/128, 8.38/128] (RSC)[rates 27]
Ice Warrior 57 (updated, likely 3/128) 0 10/128 (OSRS); [6.56/128, 13.46/128] (RSC)[rates 28]
Jogre 58 (updated, likely 1/128) 0 15/128[rates 29]
Chaos Dwarf 59 3/128 or 4/128[rates 30] 0 0
Moss Giant 62 (updated, likely 4/128) 0 5/128 (OSRS), [2.75/128, 7.91/128] (RSC)[rates 31]
Shadow Warrior 64 8/128[rates 32] 0 18/128[rates 33]
Otherworldly Being 66 (3/128) 0 0[2]
Ice Giant 68 (updated, likely 4/128) 0 0
Salarin the twisted 69 (1/128) 0 39/128 (OSRS), [31.34/128, 44.55/128] (RSC)[rates 34]
Paladin 71 (2/128) 0 8/128 (OSRS), [4.69/128, 11.18/128] (RSC)[rates 35]
Lesser Demon 79 (4/128) 0 1/128 (OSRS), [0.0/128, 2.26/128] (RSC)[rates 36]
Greater Demon 87 (5/128) 0 0
Ice Queen 103 (updated, likely 3/128) 0 10/128 (OSRS); [6.56/128, 13.46/128] (RSC)[rates 28]
Blue Dragon 105 (5/128) 0 15/128 (OSRS), [8.64/128, 16.88/128] (RSC)[rates 37]
Fire Giant 109 3/128[rates 38] 1/128[rates 39] 19/128[rates 40]
Red Dragon 140 Needs rewrite. (5/128)[rates 41] 0

2/128 (OSRS), [0.38/128, 3.06/128] (RSC)[rates 42]

Black Demon 156, 175 (1/128) (5/128) 23/128 (OSRS), [19.83/128, 30.29/128] (RSC)[rates 43]
Black Dragon 200 (2/128) 2/128 - 3/128(3/128)[rates 44] 0
King Black Dragon 245 2/128[rates 45] 8/128[rates 45] 0
  1. ↑ Based on a combined sample (1 and 2) of 16517 kills in which 2819 herb drops occurred, the drop rate for herbs is 21.85/128, 95% CI [21.09/128, 22.62/128]. This means that it is very likely that the drop rate for herbs is 22/128, instead of the drop rate of 23/128 in OSRS.
  2. ↑ Based on a combined sample (1 and 2) of 339 kills in which 30 herb drops occurred, the drop rate for herbs is 11.33/128, 95% CI [7.55/128, 15.48/128]. This aligns with the OSRS drop rate for herbs of 13/128, but more data is necessary to make any substantiated claims.
  3. ↑ Based on a combined sample (1 and 2) sample of 3746 kills in which 52 herb drops occurred, the drop rate for herbs is 1.78/128, 95% CI [1.3/128, 2.26/128]. This means that it is very likely that the drop rate for herbs is 2/128, just like OSRS.
  4. ↑ Based on a combined sample (1 and 2) sample of 1586 kills in which 268 herb drops occurred, the drop rate for herbs is 21.63/128, 95% CI [19.29/128, 23.97/128]. This aligns with the OSRS drop rate for herbs of 23/128, but more data is necessary to make any substantiated claims.
  5. ↑ Based on a combined sample (1 and 2) sample of 19913 kills, we can calculate the following:
    Gem Drops 95% CI Rate in gem table CI (access)
    Sapphire 28 [18, 39] 32/128 [72, 156]
    Emerald 22 [13, 32] 16/128 [104, 256]
    Ruby 8 [3, 15] 8/128 [48, 240]
    Diamond 8 [3, 15] 2/128 [192, 960]

    During the collection of this dataset, players seem to have been extremely lucky in obtaining diamonds. Disregarding the diamonds, the interval of [104, 156] (max(min CI), min(max CI)) translates into a drop rate between [0.66/128, 1.00/128]. As a result, the drop rate for gems must probably be 1/128. This aligns with the OSRS drop rate for gems of 1/128.

  6. ↑ Using data from a log of 1218 kills in Cook's replay drops dataset, the gem table drops are: 1 uncut sapphire (estimated 4 hits on the gem table), 1 uncut emerald (8 hits) and 1 uncut ruby (16 hits). As such, the droprate must be 1/128.
  7. ↑ Using data from a log of 21473 kills in Cook's replay drops dataset, the gem table drops are: 45 uncut sapphires (estimated 180 hits on the gem table), 18 uncut emeralds (144 hits), 12 uncut rubies (192 hits) and 3 diamonds (192 hits). As such, the droprate must be 1/128.
  8. ↑ Based on a combined sample (1 and 2) sample of 21773 kills in which 3929 herb drops occurred, the drop rate for herbs is 23.1/128, 95% CI [22.45/128, 23.78/128]. This means that it is very likely that the drop rate for herbs is 23/128, instead of the drop rate of 24/128 in OSRS.
  9. ↑ Based on a combined sample (1 and 2) sample of 16520 kills in which 2880 herb drops occurred, the drop rate for herbs is 22.31/128, 95% CI [21.54/128, 23.11/128]. This aligns with the OSRS drop rate for herbs of 23/128, but more data is necessary to make any substantiated claims.
  10. ↑ A log of 11569 kills combined at User:Cook_Me_Plox/replaydrops and bishopds log shows: 18 uncut sapphires (estimated 72 hits if on droprate), 4 uncut emerald (32 hits), 4 uncut ruby (64 hits). All seems kind of unlucky, but must be 1/128
  11. ↑ Based on a combined sample (1 and 2) sample of 341 kills in which 68 herb drops occurred, the drop rate for herbs is 25.52/128, 95% CI [20.27/128, 31.16/128].
  12. ↑ Based on a combined sample (1 and 2) sample of 619 kills in which 82 herb drops occurred, the drop rate for herbs is 16.96/128, 95% CI [13.44/128, 20.47/128].
  13. ↑ Based on a combined sample (1 and 2) sample of 435 kills in which 81 herb drops occurred, the drop rate for herbs is 23.83/128, 95% CI [19.13/128, 28.54/128]. This aligns with the OSRS drop rate for herbs of 23/128, but more data is necessary to make any substantiated claims.
  14. ↑ A log of 86431 hits xp (about 5576 kills) without picking up guams shows: 24 Marrentill (about 125 hits), 18 Tarromin (about 125 hits), 7 Harrlander (about 62.5 hits), 11 Ranarr (about 125 hits), 13 Irit (203 hits), 8 avantoe (167 hits), 4 Kwuarm (100 hits), 4 Cadantine (125 hits), 4 Dwarf Weed (167 hits). Not much data, but it will have to do... Average of about 133 hits on like 5500 kills would be 2.41%. Means that herb table access rate is 3/128 (2.34%).
  15. ↑ A log of 9269 kills at User:Cook_Me_Plox/replaydrops shows: 17 uncut sapphires (estimated 68 hits if on drop rate), 10 uncut emeralds (80 hits), 4 uncut rubies (64 hits) and 2 diamonds (128 hits). This means that the hit rate for RDT for skeletons must be 1/128.
  16. ↑ Using data from a log of 4825 kills in Cook's replay drops dataset, the gem table drops are: 10 uncut sapphires (estimated 40 hits on the gem table), 2 uncut emeralds (16 hits), 3 uncut rubies (48 hits). As such, the droprate must be 1/128.
  17. ↑ A log of 1906 bandit kills at User:Cook_Me_Plox/replaydrops shows: 7 uncut sapphires (estimated 28 hits if on drop rate), 3 uncut emeralds (24 hits), 3 uncut rubies (48 hits). Seems to be really lucky on the rubies. However, this means that the hit rate for RDT for bandits must be 2/128.
  18. ↑ Based on a combined sample (1 and 2) sample of 371 kills in which 80 herb drops occurred, the drop rate for herbs is 27.6/128, 95% CI [22.08/128, 32.78/128]. This aligns with the OSRS drop rate for herbs of 26/128, but more data is necessary to make any substantiated claims.
  19. ↑ A log of 34018 kills at User:Cook_Me_Plox/replaydrops shows: 122 uncut sapphires (estimated 488 hits if on droprate), 65 uncut emeralds (520 hits), 31 uncut rubies (496 hits), 8 uncut diamonds (512 hits). This gives a ~1.47% hit rate. This means that the hit rate for RDT for hobgoblins is 2/128 (1.56%).
  20. ↑ Based on a combined sample (1 and 2) sample of 34384 kills in which 1863 herb drops occurred, the drop rate for herbs is 6.94/128, 95% CI [6.63/128, 7.23/128]. This aligns with the OSRS drop rate for herbs of 7/128.
  21. ↑ Based on a combined sample (1 and 2) sample of 322 kills in which 12 herb drops occurred, the drop rate for herbs is 4.77/128, 95% CI [2.39/128, 7.55/128].
  22. ↑ Based on a combined sample (1 and 2) sample of 389 kills in which 19 herb drops occurred, the drop rate for herbs is 6.25/128, 95% CI [3.62/128, 9.21/128].
  23. ↑ Based on a combined sample (1 and 2) sample of 407 kills in which 27 herb drops occurred, the drop rate for herbs is 8.49/128, 95% CI [5.35/128, 11.95/128].
  24. ↑ Based on a combined sample (1 and 2) sample of 965 kills in which 124 herb drops occurred, the drop rate for herbs is 16.45/128, 95% CI [13.66/128, 19.23/128]. This aligns with the OSRS drop rate for herbs of 15/128, but more data is necessary to make any substantiated claims.
  25. ↑ Based on a combined sample (1 and 2) sample of 1367 kills in which 31 herb drops occurred, the drop rate for herbs is 2.9/128, 95% CI [1.97/128, 3.93/128]. This aligns with the OSRS drop rate for herbs of 3/128, but more data is necessary to make any substantiated claims.
  26. ↑ Based on a combined sample (1 and 2) sample of 355 kills in which 43 herb drops occurred, the drop rate for herbs is 15.5/128, 95% CI [11.54/128, 19.83/128]. This aligns with the OSRS drop rate for herbs of 14/128, but more data is necessary to make any substantiated claims.
  27. ↑ Based on a combined sample (1 and 2) sample of 336 kills in which 14 herb drops occurred, the drop rate for herbs is 5.33/128, 95% CI [2.67/128, 8.38/128]. This aligns with the OSRS drop rate for herbs of 5/128, but more data is necessary to make any substantiated claims.
  28. ↑ 28.0 28.1 Based on a combined sample (1 and 2) sample of 390 kills in which 30 herb drops occurred, the drop rate for herbs is 9.85/128, 95% CI [6.56/128, 13.46/128]. This aligns with the OSRS drop rate for herbs of 10/128.
  29. ↑ Based on a combined sample (1 and 2) sample of 437 kills in which 48 herb drops occurred, the drop rate for herbs is 14.06/128, 95% CI [10.25/128, 17.87/128]. This does not align with the OSRS drop rate for herbs of 6/129. Analysis of the OSRS droptable shows that at some point part of the chance of getting a herb was changed into the chance of getting a nature rune. Refitting the known RSC values (and essentially rebuilding the droptable) results in a 15/128 drop rate.
  30. ↑ A dataset in which 190 muddy keys (~3474 kills if on droprate) and 167 mithril bars (~3562 kills) were farmed shows: 25 uncut sapphires (estimated 100 hits if on droprate), 15 uncut emeralds (120 hits), 6 uncut rubies (96 hits) and 1 uncut diamond (64 hits). Assuming that there are about 100 droptable-hits in about 3500 kills, this gives a ~2.86% hit rate. This means that the hit rate for RDT for chaos dwarves is either 3/128 (2.34%) or 4/128 (3.125%).
  31. ↑ Based on a combined sample (1 and 2) sample of 372 kills in which 15 herb drops occurred, the drop rate for herbs is 5.16/128, 95% CI [2.75/128, 7.91/128]. This aligns with the OSRS drop rate for herbs of 5/128.
  32. ↑ Currently no drop data for shadow warriors is available. However, analysis of the OSRS droptable structure indicates that the droptable has not been changed since its introduction. As such, we can assume that the gemtable dropped with a 8/128 drop rate.
  33. ↑ Currently no drop data for shadow warriors is available. However, analysis of the OSRS droptable structure indicates that the droptable has not been changed since its introduction. As such, we can assume that herbs dropped with a 18/128 drop rate.
  34. ↑ Based on a combined sample (1 and 2) sample of 339 kills in which 101 herb drops occurred, the drop rate for herbs is 38.14/128, 95% CI [31.34/128, 44.55/128]. This aligns with the OSRS drop rate for herbs of 39/128.
  35. ↑ Based on a combined sample (1 and 2) sample of 355 kills in which 22 herb drops occurred, the drop rate for herbs is 7.93/128, 95% CI [4.69/128, 11.18/128]. This aligns with the OSRS drop rate for herbs of 8/128.
  36. ↑ Based on a combined sample (1 and 2) sample of 453 kills in which 4 herb drops occurred, the drop rate for herbs is 1.13/128, 95% CI [0.0/128, 2.26/128]. This aligns with the OSRS drop rate for herbs of 1/128.
  37. ↑ Based on a combined sample (1 and 2) sample of 326 kills in which 32 herb drops occurred, the drop rate for herbs is 12.56/128, 95% CI [8.64/128, 16.88/128]. This aligns with the OSRS drop rate for herbs of 15/128.
  38. ↑ The same log of approximately 176991 kills with approximately 1382 URDT accesses in which sapphires and emeralds werent logged shows: 72 diamonds and 272 rubies. ~1/400 URDT accesses gives a diamond, and 1/100 URDT accesses give a ruby. As such, the RDT produced about 69 diamonds (4416 hits) and 258 rubies (4128 hits). This gives a ~2.41% hit rate. This means that the hit rate for RDT for fire giants is 3/128 (2.34%).
  39. ↑ A log of 2108.78 hours of kills (approximately 176991 kills) in which sapphires and emeralds werent logged shows: 16x silver cert (estimated 1024 hits if on droprate), 17x coal cert (1088 hits), 12x dragonstone (768 hits), 22x rune 2h (939 hits), 27x rune battle axe (1728 hits), 20x rune axe (853 hits), 16x runite bar (1024 hits), 30x bronze arrows (1920 hits), 11x death rune (704 hits), 7x rune kite (896 hits), 8x dmed helm (1024 hits). As such, in about 177000 kills the URDT was probably accessed at least like 1100+ish times, but not much much more. Given that we know that all these access rates are a x/128, x must be 1, which would translate to about 1382 accesses to the URDT during this sampling.
  40. ↑ Based on a private sample of 2108.78 hours of fire giants (approximately 176991 kills) in which the high-level herbs were logged, we can calculate the following:
    Herb Drops 95% CI Rate in herb table CI (access)
    Ranarr 2314 [2226, 2408] 11/128 [25903, 28020]
    Irit 1614 [1534, 1697] 8/128 [24544, 27152]
    Avantoe 1228 [1157, 1292] 6/128 [24683, 27562]
    Kwuarm 963 [904, 1027] 5/128 [23142, 26291]
    Cadantine 783 [726, 835] 4/128 [23232, 26720]
    Dwarf Weed 578 [532, 623] 3/128 [22699, 26581]

    The interval of [25903, 26291] (max(min CI), min(max CI)) translates into a drop rate between [18.73/128, 19.01/128]. As a result, the drop rate for herbs must probably be 19/128. This aligns with the OSRS drop rate for herbs of 19/128.

  41. ↑ in 330 kills at bishopds log there were NO RDT drops. In osrs its 5/128, which is probably higher than it was in RSC.
  42. ↑ Based on a combined sample (1 and 2) sample of 335 kills in which 4 herb drops occurred, the drop rate for herbs is 1.53/128, 95% CI [0.38/128, 3.06/128]. This aligns with the OSRS drop rate for herbs of 2/128.
  43. ↑ Based on a combined sample (1 and 2) sample of 355 kills in which 69 herb drops occurred, the drop rate for herbs is 24.88/128, 95% CI [19.83/128, 30.29/128]. This aligns with the OSRS drop rate for herbs of 23/128.
  44. ↑ A dataset of 15984 kills shows several URDT drops: 4x coal certificates (~256 hits on the table), 3x silver certificates (~192 hits), 1 dragonstone (~64 hits), 9 rune 2hs (~384 hits), 5 rune sqs (~320 hits), 3 rune kites (~384 hits), 8 rune axes (~341 hits), 5 rune baxe (~320 hits), 2 d med (~256 hits). As such, it is likely that the URDT has been hit somewhere around 256-384 times. This would translate in 2/128 - 3/128.
  45. ↑ 45.0 45.1 https://twitter.com/Oriental_Ramen/status/1386741254025846786

--ZorakTalk at various moments in 2021


  • King Black Dragon URDT We only have 448 kills for, (maybe more if desperately picking off kills from YouTube videos) so we were never going to be able to say with any confidence the RDT or URDT rates, or even the normal rates on the table. Ash was very kind and provided these rates[3] for us, to preserve the memory of the game. Hubcapp (talk) 18:33, 26 April 2021 (UTC)


Needs work:[edit source]


TODO cook's replay drops: bandits herbs. TODO http://www.bishopd.com/mainpages/games/rsc/drops/rsc_drops.html#begin : [rengade knigth + black knight + lord daquarius + cooks' jailer data.]


Chaos Druid HERB[edit source]

Okay, this needs a proper write-up. The issue here is that chaos druids can drop double drops, with the second drop being a herb. Luckily, we have bishopds log which gives info about double drops in 5534 kills. Out of the 5534 kills, (5900-5534=) 366 kills saw a drop of an additional herb. Unlike assumed earlier, only in 212 kills this additional herb was dropped to a first drop being a herb. What makes this complicated is that some "1 herb" drops can actually be a roll on the second herb table as well, as 1665/5534 kills were a nothing roll, suggesting that this is actually a very common drop. So... let's see what we can make of this. *sighs*.

Type of drop # Drops Additional herb roll?
Nothing 1665 -
Nothing (z) Herb
High-tier herb 791 (q) - the high tier part of z -
High-tier herb 0 Herb
Guam/Marr/Tarr (y) -
Guam/Marr/Tarr 212 Herb
Air/Mind/Body/Earth rune 378 -
Air/Mind/Body/Earth rune 0 Herb
Other item 1406 (x) -
Other item 154 Herb

(x = 356 [hightier runes] + 624 [coins] + 500 [vial of water] + 56 [bronze long] + 48 [snape grass] + 43 [unholy symbol mould] - 154) (q = 232+179+134+123+80+77+56-[double rolls accessed by low-tier herbs] 29-14-12-9-5-10-11)

IT IS NOT KNOWN WHETHER GEM TABLE GIVES ACCESS YET (12 drops no access)


We can assume that the percentage of drops with a herb in the other item category will be roughly the same for the nothing and the other item category. Which leads to: (1406+154)/1406*1665-1665 = 182 nothing, but 2nd herb drop drops.



With these 182 drops we can calculate how many drops actually were a 1st high-tier herb roll as well, assuming Template:HerbDropTable is about correct, 51 out of every 125 rolls are high tier. So about 74 additional herb rolls to the nothing drop should be high tier. means that high-tier herb original role in 791 - 74 = 717.

Type of drop # Drops Additional herb roll?
Nothing 1665 -
Nothing 182 Herb
High-tier herb 717 -
High-tier herb 0 Herb
Guam/Marr/Tarr (y) -
Guam/Marr/Tarr 212 Herb
Air/Mind/Body/Earth rune 378 -
Air/Mind/Body/Earth rune 0 Herb
Other item 1406 (x) -
Other item 154 Herb

Under this assumption the number of Guam/Marr/Tarr roles is 5534-(1665+182+717+212+378+1406+154) = 820. We can validate whether this no of roles is correct because we know that the ratio high_tier/low_tier is about 51/74 ~= 0.69. With the above assumptions we do indeed end up at, 717/(212+820) ~= 0.69.

However, we must test whether a nothing drop CANNOT roll a second herb, so the high tier herb roll becomes 791 again, and nothing-herb 0.

Type of drop # Drops Additional herb roll?
Nothing 1665 -
Nothing 0 Herb
High-tier herb 791 -
High-tier herb 0 Herb
Guam/Marr/Tarr (y) -
Guam/Marr/Tarr 212 Herb
Air/Mind/Body/Earth rune 378 -
Air/Mind/Body/Earth rune 0 Herb
Other item 1406 (x) -
Other item 154 Herb

In this case, (y) becomes 928 (5534-(1665+791+212+378+1406+154)). Ratio test again: 791/(928+212)=0.69.

As such, it is impossible to tell whether a nothing roll can result in an additional herb roll, also.. it's unclear whether the assumption that the ratio for nothing and for other items would be the same.--ZorakTalk 09:44, 26 April 2021 (UTC) ---


  1. ↑ Double herb drops, http://www.bishopd.com/mainpages/games/rsc/drops/bandit29.js
  2. ↑ Unlike OSRS, in which Otherworldy Beings drop herbs at a rate of 10/128, they do not drop herbs in RuneScape Classic. A combined sample (1 and 2) of 353 drops did not contain any herb drop.
  3. ↑ https://archive.is/0RHZf


Cite error: <ref> tags exist for a group named "rates"", but no corresponding <references group="rates""/> tag was found

Retrieved from ‘https://classic.runescape.wiki/w/Talk:Gem_drop_table?oldid=134414’