Talk:Fishing

Leather armor
The page says that you can catch damaged leather boots. However, I recently caught an undamaged pair of leather pants. How is this possible? The answer should be added to the page. The BlobsPaper.png 03:20, 16 March 2017 (UTC)


 * In vanilla with default tables, it's not. You can look at the loot tables yourself, found in "version.jar/assets/minecraft/loot_tables/gameplay/fishing". Ensure you are not using a custom table (located in the "data/loot_tables" folder in the world you were playing) or didn't reel in a pre-existing item entity (as fishing rods can pull item entities to you). Skylinerw (talk) 03:35, 16 March 2017 (UTC)


 * Ws this on Pocket Edition? I may or may not have had the same experience, but can't remember too well. - MinecraftPhotos4U (talk) 08:16, 16 March 2017 (UTC)


 * This was Pocket Edition., is the table any different in Pocket Edition? The BlobsPaper.png 22:23, 16 March 2017 (UTC)


 * That's really odd... possibly even a bug. Here is the vanilla junk.json loot table used in the latest version of PE; there's no mention of leather pants. In-fact, there's no mention of leather pants in any of the vanilla fishing loot tables (the vanilla fishing loot tables can be browsed here). Jocopa3 (talk) 02:56, 17 March 2017 (UTC)


 * Are you sure that table was for Pocket Edition? The BlobsPaper.png 03:00, 17 March 2017 (UTC)


 * Or possibly the loot table changed recently. - MinecraftPhotos4U (talk) 08:41, 17 March 2017 (UTC)

Ink sacs
Can ink sacs still be fished? See here: https://www.reddit.com/r/Minecraft/comments/66y01g/auto_fishing_for_20_hours_afk_found/. HalfOfAKebab (talk) 01:25, 23 April 2017 (UTC)
 * I happen to know that the fishing loot tables in Java edition haven't changed from what the wiki says, I've been keeping track – Sealbudsman talk/contr 01:59, 23 April 2017 (UTC)
 * Confirmed ink sacs can still be fished, I got some earlier in 1.12.2 during some non-afk fishing for a few hours. I do not know if it is still possible to reel in a batch of 10 ink sacs at once as a loot drop, but the item is still there with a high rarity. Jack McKalling (talk) 12:58, 11 October 2017 (UTC)

Clarity of Note 1
The note is currently "All blocks above the bobber must be completely transparent to sunlight. This includes glass and tripwire, but not leaves".

I believe this is slightly confusing. It could too easily be misread as meaning: glass and tripwire must be transparent, but leaves are not required to be transparent.

I think it would read more clearly if the second sentence was "For example glass and tripwire are completely transparent to sunlight, but leaves are not."

- 86.171.85.221 20:21, 24 August 2018 (UTC)
 * Made some changes to the note. Is it better now? --AttemptToCallNil (report bug, view backtrace) 20:26, 24 August 2018 (UTC)

Diamonds
I think I remember reeling in diamonds in old versions. Has this been removed, or is the page inaccurate? Or am I just wrong? IndigoGolum (talk) 00:08, 2 October 2018 (UTC)


 * I'm not certain, but I think it's been removed. Spiffybubble (talk) 16:35, 23 May 2019 (UTC)
 * I spent many survival in-game days fishing with a Luck of the Sea III rod (I was hoping to get a Fortune enchanted book), and never once picked up a diamond. The article doesn't say it's possible. If this was possible before, I'm pretty confident it isn't possible anymore. ~ Amatulic (talk) 18:18, 23 May 2019 (UTC)

Wrong information about fishing?
The article states that it takes between 5 and 45 seconds to catch a fish. However I found this minecraft code (version 1.12):

this.ticksCaughtDelay = MathHelper.getInt(this.rand, 100, 600);

this.ticksCaughtDelay -= this.field_191519_ax * 20 * 5;

If I do not get something wrong the first line means that it takes between 100 and 600 ticks, i.e. 5 and 30 seconds, but not 45 seconds. –Preceding unsigned comment was added by Unbekannt130 (talk • contribs) at 8:23, 13 August 2019 (UTC). Please sign your posts with

Fish mob
In bedrock edition on Xbox 1, my friends and I have found that if you put the bobber somewhat above the fish mob, the particles would appear sooner. Please tell me why. Actionjgaming (talk) 09:35, 22 December 2019 (UTC)


 * I don't this this is an intended feature, you probably just got lucky. Komorebi17 (talk) 11:26, 7 June 2020 (UTC)

Unenchanted chance to No LotS chance
I think that Unenchanted chance should be changed to no LotS chance as an enchanted fishing rod without LotS has the same catch table even if it was enchanted with other enchantments such as mending, unbreaking etc...

Treasure Misinformation / Misunderstanding
The page says that "In order to catch items in the treasure category, the bobber must be in open water. This is technically defined as the 5×4×5 vicinity around the bobber (2 blocks away horizontally, 2 blocks above the water surface, and 1 block deep)." The "(2 blocks away horizontally, 2 blocks above the water surface, and 1 block deep)." part doesn't make sense to me as how can the bobber be 2 blocks above the water when the water is only supposed to be 1 block deep.


 * I believe there must not be any solid blocks in that region: air is a valid substitute for water. 73.26.38.97 08:53, 29 January 2021 (UTC)
 * It's 2 blocks deep, not 1 block. So the bobber has two 5×5 layers of water under it, and two 5×5 layers of air over it. Amatulic (talk) 03:50, 30 January 2021 (UTC)

Probabilities for non-open water fishing?
The page has neat tables for open water fishing, but it leaves some major questions open about non-open water fishing. I especially wonder about how, if at all, Luck of the Sea affects the Fish vs Junk probability when treasure is blocked. --CrCl3 (talk) 17:50, 12 March 2021 (UTC)
 * I know how I'd do it if I were the programmer.
 * Generate a random probability value between 0 and 1.
 * If the probability value falls in the range for fish or junk, then award the fish or junk when reeled in.
 * If the probability value falls in the "treasure" range....
 * If the bobber isn't in open water, go to step 1.
 * Otherwise award the treasure when reeled in.
 * In that way, the proportion of fish to junk remains the same with or without treasure. Amatulic (talk) 17:25, 16 March 2021 (UTC)

Lily pad
I just checked files and minecraft:waterlily is set with weight 5 in treasure table in bedrock 1.16. It's not it the junk table, nor in the jungle_junk. edit by J3 – Unsigned comment added by 212.67.87.28 (talk) at 16:42, 16 March 2021 (UTC). Sign comments with
 * Is that also the case with Java? If so, then the lily pad should be moved to treasure. Amatulic (talk) 17:30, 16 March 2021 (UTC)

Two possible additions
I believe there are two items that would do well to be added to the text somewhere.

First, it isn't possible to gain treasure fishing underground, even if one meets the 5x4x5 requirements. (My pond is 9x6x7, and in several hundred attempts, I've not yet seen any treasure.)

Second, putting in a skylight (a tunnel up to the surface) that's covered in a glass block doesn't count as allowing sun and moonlight through, for purposes of reducing the time between catches.

Can anybody confirm, deny, or speculate on either of these hypotheses?

Thank you! SirDaddicus (talk) 04:52, 26 June 2021 (UTC)


 * I tested this on a vanilla Java 1.17 server, with a fully underground 6x3x6 pool with 0 sky light and a Lure II/Luck of the Sea III fishing rod. Literally the first item I pulled up was an enchanted bow, and the fourth one was a nautilus shell.


 * So at least on Java edition it seems like sky access isn't required. --CrCl3 (talk) 01:03, 30 June 2021 (UTC)
 * Survival? I tested on Java 1.17 with a 7x3x10 pool with a 7x3x10 volume of air directly above it. I made hundreds of casts and never received any treasure. I did receive junk, however, and a reasonable approximation of the fish distribution.


 * I added a 1x1 shaft to the surface, and that didn't change anything. Except there was a very tiny reduction in the time spent on each cast when I removed the glass block at the top. I suspect this was when the sun or moon was directly overhead. I then expanded the shaft to 2x3, and still no treasure. But, again there was a significant (but not complete) reduction in casting time.


 * Finally, I expanded the shaft to 5x5, all the way to the surface. This caused treasure to start coming at the expected rates (about 10% with my LotS rod).


 * So, I seriously question your experience. Are you certain it was vanilla 1.17? SirDaddicus (talk) 01:45, 30 June 2021 (UTC)


 * Yes, it was survival, and yes, I'm sure it's vanilla 1.17. (Downloaded launcher from official website, installed game using it, downloaded the server from the official link).


 * I have tested it both underground (with no sky access) and in my AFK fishing farm (with 1-block sky access), always getting treasure items, I also tested it in singleplayer now to make sure it isn't a server thing, and still got a nautilus shell.


 * I have never seen anyone else claim that there is a sky access requirement for treasure, indeed every treasure AFK fishing farm design I have seen has at least some light blocking blocks above the water. One example is the design here: https://www.youtube.com/watch?app=desktop&v=mnXFj0aOveE, maybe you could try building an exact copy of it in an open area to make fully sure there wasn't some problem with your test setup?


 * The casting time reduction from sky access should also be on/off rather than gradual, depending only on the column above the bobber. You asked if I'm sure my game is vanilla, but are you 100% sure yours is?


 * If we can't figure out the cause of the discrepancy, we'll probably have to try get someone else to test this.


 * --CrCl3 (talk) 03:43, 30 June 2021 (UTC)


 * SirDaddicus, I apoligize, I had reverted your edit originally because I had no idea what was meant by "shaft". It wasn't clear to me that you were referring to underground fishing. I play Bedrock Edition, and while I don't normally fish underground, when I have done so I have never caught anything but fish and junk. On the other hand, one never finds a 5x4x5 volume of water underground.


 * It makes sense to me that the 5x5 area must be exposed to the sky. It's enough simply to say that in the article: the 5x5 surface, regardless of elevation, must be exposed to the sky to catch treasure. I have added a clause to that effect to the article. Amatulic (talk) 02:16, 30 June 2021 (UTC)


 * Ideally, the question should be settled by someone with access to the source code. Which would at least tell us what was intended, though it might not rule out a non-obvious bug.  --MentalMouse42 (talk) 10:19, 30 June 2021 (UTC)


 * Amatulic, thank you! I like your change even better than my original.


 * MentalMouse, I have to agree. Clearly, there is a discrepancy here. I'm definitely running vanilla 1.17 (no datapacks, resource packs, or mods), and yet no treasure at all until I cleared a shaft to the surface. NOTE: I only did about a hundred casts between having a 2x3 shaft and expanding it to 5x5. So, it's possible that was just bad luck. Does anybody know how to check the source code? SirDaddicus (talk) 12:56, 30 June 2021 (UTC)


 * Even if the source code can't be checked, a safe assumption is to think like a programmer trying to create an efficient game mechanic. If there's a requirement for the 5x5 area around the bobber to be water, it's trivial to impose a requirement that each of those blocks must be exposed to the sky. An additional requirement that a certain percentage of those blocks be exposed to the sky is a needless complication involving a bit of extra processing; it isn't hard to do but it's unnecessary for creating an effective game mechanic. Why waste the CPU cycles on something that doesn't matter? Amatulic (talk) 13:17, 30 June 2021 (UTC)


 * Well, "trivial" if you've set up a cheap way to test exposure, but I'm pretty sure they've got that. SirDaddicus:  The people who have access to the code are those who have licensed access through Mojang, which is mostly mod developers.  A complicating factor is that before programmers can effectively use that code, it needs to be "deshrouded", IIUC by the Forge team (complicated situation), and I don't know how long it will be before they've done so for the 1.17 release (especially with point releases still coming). --MentalMouse42 (talk) 14:20, 30 June 2021 (UTC)


 * Until we get that info, let's wait to undo Amatulic's and my change. Alternately, if more people chime in, perhaps we can get a better feel for it. It IS possible that my experience is simply really bad luck, but I was well over a thousand casts with zero treasure. But, once I built the shaft, two of the first ten casts resulted in treasure. The odds against a thousand casts with no treasure is less than one in ten thousand, so while it's possible, I think there's something else going on here. SirDaddicus (talk) 14:38, 30 June 2021 (UTC)
 * ETA: There are a handful of source-divers around the wiki -- staff, admins, and regulars; I don't know which of them are currently active, and have to go to my IRL job now.  My summary line before was meant as a wave to any of those who might be following Recent Changes; when I get back tonight I may try looking up some of the folks I've worked with in the past, and messaging them. --MentalMouse42 (talk) 14:41, 30 June 2021 (UTC)
 * Thank you! SirDaddicus (talk) 15:22, 30 June 2021 (UTC)
 * So I have access to the code and I cannot confirm that skylight is needed. However not entirely shure as not very experienced. However, I cannot confirm that with testing either, as a tipp, create a cheat-world, type in "/give @p minecraft:fishing_rod{Enchantments:[{id:luck_of_the_sea,lvl:40},{id:lure,lvl:5}]}" and you have a fishing rod that gives you round about every second item a treasure. I made a datapack to see when treasure is possible (inspired by a video from Rays Works) so could someone please tell me how to upload stuff here in the talk page? @SirDaddicus: Maybe the fishing bobber wasn't always exactly in the middle? GuiTaek (talk)


 * I just tested this in Bedrock Edition. Without the /give command and no code viewable, all I can do is contrive a test. In creative mode, I crafted a Luck of the Sea III + Lure III fishing rod, then I used the /fill command to create a large high ceiling 10 blocks over a deep ocean to block sky access -- large enough so that it would be impossible for required 5x4x5 volume to be outside of it. I built a platform a few blocks in from the edge of that ceiling to stand on. Then I switched to Peaceful and cast my rod toward the center of the ceiling's shadow. I caught 60 items. Four of them were treasure: a saddle, two enchanted books, and a name tag.


 * Therefore, sky access isn't required to catch treasure in Bedrock Edition. The same test would be easy to do in Java. No need to look at code. The test took me 10 minutes to prepare and catch my first treasure on my third cast. And it's a binary test: If you catch treasure, the test passes. I made 60 casts just to be sure I could catch more treasure, and I did. Amatulic (talk) 16:11, 3 August 2021 (UTC)
 * So I tested the same as you in java with LotS III and 4 treasures out of 20 casts. So I removed that wrong information. Using my datapack I found out, that with 2-3 blocks deep, you still get rarely the treasure when you wait to the end of nibbing (20 to 40 ticks, I looked that up in the code). Even with 4 block deep waters it happens sometimes that a treasure is not available. Finding out the exact range where the bobber lands is difficult. At the start of the nibble, the velocity is set to 0.4*(0.6 to 1.0) blocks per tick but I don't get from there to at maximum 2 blocks, what I observed while watching the height as a number with tellraw. Long story short, the reason he didn't get treasure is maybe that the lake he made wasn't deep enough, maybe he deepened his lake later and so he catched treasure. I'd like to add this information about the bobber moving downwards as well as the wrong information, that it is constantly checked wether the "open water condition" is true (only as long as a fish etc. is lured (when you see the moving bubble trail) and when it is nibbling (when you have to catch)). So I will do the changes as long as nobody interupts me. GuiTaek (talk) 23:16, 8 August 2021 (UTC)


 * Thanks GuiTaek for confirming in Java Edition and removing the sky exposure requirement. I had to revert one thing you added though: After the bobber submerges, if I wait a whole second, I don't catch anything. I changed it back to half a second. Amatulic (talk) 23:39, 8 August 2021 (UTC)


 * Your're welcome. Thanks for taking a look at my changes. How did you measure that you were waiting a whole second? Because I get other results: I filmed myself fishing and then went frame by frame through the video. I got following time intervals given in frames of the video between fish biting (the first frame that is under water, including) and reeling in the fishing rod (first frame with changed height of my fishing rod, excluding): 26 (catch), 42(no catch), 17 (catch), 32 (catch), 42(catch), 33 (catch), 52 (no catch), 42 (catch), 26 (catch), 47 (no catch), 34 (catch), 43 (catch), 36 (no catch). Dividing that with the fps that vlc says my video has of 30.052514 gives approximately 1.20sec to 1.43sec for the range of the time interval of where you can catch fish. GuiTaek (talk) 21:41, 12 August 2021 (UTC)
 * Are you doing this in Java Edition by any chance? I play Bedrock. I did not measure, although the time I perceive required to catch a fish is not even a whole second. I'll try it again. There may be a difference between Java and Bedrock. Amatulic (talk) 23:08, 12 August 2021 (UTC)

Indeed that could be the problem. I only play java and have only for java the code. I have bedrock I can run this test also for Bedrock. I will tell you the results. GuiTaek (talk) 23:44, 12 August 2021 (UTC)
 * OK, I ran some tests with Bedrock. I observed:
 * When the bobber dips underwater, it stays underwater for a random time ranging from about 0.3 sec to 2.3 sec (timed on a stopwatch). You can just let the bobber sit there and watch it as each new wiggly trail approaches.
 * If you reel in the rod while the bobber is on its way back up, it's too late.
 * The average time the bobber stays down is less than 2 sec, seems to be about 1.2 sec.
 * It seems you have less than half the time the bobber is underwater to reel it in. It's hard to tell when it's too late, because the underwater time of the bobber is random.
 * On average, that comes out to about a half second to reel in a catch, which matches my experience. Sometimes less, sometimes more. Amatulic (talk) 02:49, 13 August 2021 (UTC)
 * (Moved from below; I had added it to the wrong section.) I'm OK with that. It must have been an aberration. (The ONLY change I made to go from no treasure to treasure was digging out the shaft of air 5x5 blocks, all the way to the surface.) SirDaddicus (talk) 22:03, 10 August 2021 (UTC)
 * Oh thanks a lot Amatulic, I really appreciate it. I didn't feel like doing this tedious work again. So basically maybe we write it down as you found out with a verify to hint somebody to look that up in the code?
 * @SirDaddicus: I'm actually happy that you come here again. I think the problem was, that your little lake wasn't deep enough. Because the fishing rod falls down when you don't react fast enough. Maybe because you had the speed bonus of open sky you reacted faster (at least that is what I found out for me: when it needs a lot of time, I react much slower) and therefore your bobber didn't went 2 blocks down (what it does very often). I'm not very happy with the explanation as you just have to sometimes be in time with reaction but I don't find a better explanation. You said you tried thousands of times, didn't you? Because as probability, this gives 5 x 10^(-23) (with 5% for treasure) so it's not just an anomaly, it has to have a reason but I can't find any, I'm sorry. Again, did you always cast in the middle of your lake? GuiTaek (talk) 23:12, 15 August 2021 (UTC)
 * So @Amatulic, I saw that you stopped the time it is underwater and not the time you click. So I had to redo the experiment and the experiment result was very clear. The catches were ALL between 0.37sec and 0.47sec and the not-catches were all between 0.67 sec and 0.73 catches and I only made 8 casts. It shows so clear, that it's always exactly half a second that I think we don't even need somebody with code access for bedrock. I will now change it so it is correct now. GuiTaek (talk) 23:44, 15 August 2021 (UTC)
 * That experiment confirms mine. There will be errors due to my reaction times, but it seemed pretty clear that the time needed to catch anything is half of the time that the bobber is submerged, and the submerged time is random between 0.3 and 2.3 sec. The bobber spends about half its submerged time descending and half the time ascending to the surface. I observed I could catch something only while the bobber was descending, and never when it was ascending. Therefore you have to yank back the rod while the bobber is descending. That gives you about half a second to react on average. Amatulic (talk) 19:06, 16 August 2021 (UTC)
 * very good GuiTaek (talk) 20:26, 16 August 2021 (UTC)

Those tables...
Those tables are a complete mess; there is no need for two tables that are completely identical aside from lily pads being added. Also, the random differences between editions result in tons of spam to the weights which is just impossible to read. I have no proposal for a solution but yeah Tryashtar-gpuser (talk) 04:59, 12 July 2021 (UTC)
 * I wouldn't call them a "mess", but they could be consolidated and cleaned up; I agree with that.


 * Regarding the duplication, all that needs to be done is add bamboo and cocoa beans to the main table, but add a note specifying that it only applies to Jungle biomes. Then, the second table could be removed.


 * Regarding the weightings, I think they should just be percentages. A note could be added showing the general nature of "parts per 120" or "parts per 110", if someone thinks that kind of precision is necessary (I don't). SirDaddicus (talk) 15:42, 12 July 2021 (UTC)


 * I spoke too soon. The weightings would have to be modified as well. I would recommend using common format, like W% (BR), X% (BRJ), Y% (J), and Z% (JJ), and then explain the abbreviations in the notes below.


 * Alternately, separate columns could be used for the four weightings possibilities. This would allow consolidation of the headers, too. (There's no reason to have both "Chance in Category" and "Weight" titles at the top of that column. The extra cell could be used to distinguish between Java/Bedrock and Jungle/All other biomes.) SirDaddicus (talk) 15:51, 12 July 2021 (UTC)


 * I'm pretty sure that the weightings are the actual raw values from the code, so they should definitely stay in. Percentage chances are easier to understand, but if they're computed values they don't get to replace the weights. --MentalMouse42 (talk) 11:15, 13 July 2021 (UTC)