Talk:Horse

I was looking at the abuse filter examining past edits and the last edit took out a part of the history section that I am not sure if it should have been removed, and if the last edition should or should not be undone, can someone check to see if it should be undone? Gameking1happy (talk) 14:57, 10 February 2021 (UTC)

Revision undone Dlljs (talk) – Undated comment added at 16:21, 10 February 2021 (UTC). Please sign comments with ~~~~

I see that the Jump Strength section says it was a cubic equation fit to the listed values, and fit especially at exact block amounts. While I'm sure this data is usable, has nobody yet found the actual algorithm used to calculate jump height? Or is it a more complex thing falling-out from ingame physics? 75.76.15.79 00:01, 17 March 2021 (UTC)

It would be incredibly easy for in-game physics for jump height to match actual physics, which is a quadratic, not a cubic formula. And there's no reason why it wouldn't be the same. So I am skeptical about this curve fitting, which is clearly someone's personal research and unrelated to official sources. I have removed it. Amatulic (talk) 21:41, 17 March 2021 (UTC)

The code for jumping in Minecraft does take air drag into account, and therefore does not resolve to a neat polynomial formula, let alone a quadratic one. Instead, the behavior emergent from the code corresponds to a formula that is exponential at heart, with a constant and a linear component thrown in for good measure. 2001:4DD6:77E8:0:D8F7:6955:AD5D:5D49 18:33, 10 May 2025 (UTC)

The foals in Bedrock Edition also had different sizes when growing. This should be mentioned at least in the trivia section. Slardna (talk) 12:15, 16 April 2021 (UTC)

No. Trivia sections should be reduced or eliminated when possible. I added a small change to the article saying that foals grow in stages. Amatulic (talk) 18:17, 17 April 2021 (UTC)

Until March 2020, the conversion factor between internal speed units (for horses, up to 0.3375) and b/s equivalents had always been listed as 43.1[7], meaning you multiply 0.3375 by 43.1[7] and receive the max. horse speed of 14.57 blocks/s.

Anyway, in March 2020, user playyip apparently did some testing and reported that this value seemed to be inaccurate, as a horse with an internal speed that should have corresponded to 8.0b/s was at 7.81b/s actually slower than a minecart.

Because of this, he adjusted the conversion factor on the Tutorials/Horses page to 42.16 and the max speed on the Horse page to 14.23b/s (42.16*0.3375). This change carried over to other pages on the wiki in the following months, such as Transportation.

The issue I have with this that this change doesn't entirely add up with other numbers on this Wiki, such as the internal player speed 0.1 everywhere being reported as 4.317b/s, because going from that number, the conversion factor would be 43.17 (10 * 4.317), not 42.16 (this entire comparision is by the way even made right on this page, it just doesn't add up any more when you check the math).

My personal guess is that user playyip simply made a mistake during testing and got the wrong result, which is plausible since - opposed to for example player speed, which has been tested many times - he is the only data point here.

This is why I would suggest to:

a) Either revert the max horse speed being reported as 14.23b/s back to 14.57b/s, and adjust the conversion factor on the Tut/Horses page to 43.17

b) Or do more testing to check if playyip was actually right and there is some fundamental inaccuracy regarding entity speeds on this wiki, though I doubt this.

I also want to emphasise that the difference between 42.16 and 43.17 is anything but minor, it translates to a quite substantial difference in horse speeds.

TilWei (talk) 02:59, 3 December 2021 (UTC)

Since noone had any objections, I made the corresponding edits on this page, Tutorials/Horses and Transportation Methods.

TilWei (talk) 01:47, 13 December 2021 (UTC)

This is easy to test. If the conversion factor is 43.17, the a horse with speed 0.1853 should be about the same speed as a minecart (8 m/s). If the conversion factor is 42.16, then a horse with speed 0.1898 should be about the same speed as a minecart.

I have tested this on 1.18 and indeed the conversion factor is 42.16. The horse with speed 0.1853 slightly lags behind the minecart and the horse with speed 0.1898 matches it exactly. You may easily make the same test for yourself.

Playyip (talk) 23:19, 10 June 2022 (UTC)

Max speed is the sum of an infinite geometric series, fiddle with the formula a bit and you get a factor of (1 / (1 - 0.6 * 0.91 * 0.98) * 0.98 * 20) =~ 42.16. Related the max jump height is the sum from 0 until the result becomes negative strength*0.98^2x-0.08*0.98^2*((1-0.98^2x)/(1-0.98^2)). I can't be bothered fixing up all relevant pages so anyone feel free.

Graphs if you need help https://www.desmos.com/calculator/1dpw3biygq https://www.desmos.com/calculator/9ajvzpsos9

185.218.127.224 12:45, 14 June 2022 (UTC)

There's one hole in the current reasoning: It assumes that an (empty, I presume) minecart does in fact travel at 8 m/s. Has anyone ever bothered to actually verify that?

I'm currently toying around with horse stats and devices to measure them, and am noticing that while riding my trusty test steed, at 0.3375 internal value MiniHUD clocks the two of us at 14.570 m/s peak speed, which matches the standard 43.17 conversion factor. I'm trusting that MiniHUD's display is correct, as it also clocks my walking speed as 4.317 m/s, and me riding a minecart as 8.000 m/s. Obviously I can't measure an "unpiloted" minecart this way. There are also other factors that may influence the results of measurements; for instance, MiniHUD also reveals that every(!) change in direction reduces a horse's speed by a non-trivial margin.

--2001:4DD6:77E8:0:D8F7:6955:AD5D:5D49 16:10, 10 May 2025 (UTC)

Okay, rather than having a race between a horse and a minecart, I've just run some proper objective tests:

• Minecraft Java Edition 1.21.5
• 200 blocks test track
• ample run-in of for the contestants to reach their max speed before crossing the start line
• start and finish line set up to generate pulses separated by exactly 200 blocks' worth of racing time
• start and finish pulses transmitted symmetrically to a central timekeeping station, with repeaters distributed evenly on both legs
• time between rising edges of start and finish pulse measured at intervals of 2 redstone ticks (*)
• minecart racetrack uses 199 blocks of fully powered rail framed by detector rails at the start and finish, plus a 10-block run-in of fully powered rails.
• tripwires used to detect the start and finish on the horse racetrack
• finish line is designed wide enough so that horses can be easily ridden in a straight line, with no "course corrections" whatsoever.

The timekeeping mechanism is implemented as follows: Both start and finish pulses toggle a copper bulb, which via a comparator extends or retracts a sticky piston to make or break an observer-observer clock, which in turn pulses a dropper, which deposits items into a chest. After each race, the chest is checked for deposited items, indicating the time taken to complete the track. Each deposited item, which I refer to as a "BogoTick" henceforth, should correspond to 2 redstone ticks. Since I have not fully analyzed the start and stop behavior of the timing mechanism, its results should be taken with a grain of salt, and may be off by up to +/- 1 BogoTick. A run with an occupied minecart - widely accepted to have a maximum speed of exactly 8.0 m/s, and expected to easily reach that speed on the track in question - is used for calibration.

The expectactions:

• At 8.0 m/s, the reference player-occupied minecart should clock in at 125 BogoTicks, +/- 1.
• Presuming an unoccupied minecart also has a maximum speed of 8.0 m/s, it should clock in with the same result as the reference cart. If on the other hand the discrepancy in a horse-vs-minecart race is due to unoccupied minecarts being faster than expected, the unoccupied minecart would have to be faster to explain the discrepancy, which would result in a lower BogoTicks score.
• If ridden horses use the purported speed conversion factor of 42.16 (for whatever yet-unexplained reason), an internal speed value of 0.1898 should achieve a speed of 8.0 m/s as well, and therefore also clock in with the same score as the reference. If on the other hand the conversion factor for horses is the standard of 43.17, the same internal speed value should achieve a speed of ~8.2 m/s, which should nominally clock in at only 122 BogoTicks, or 3 BogoTicks less than the reference.
• If ridden horses use the standard conversion factor of 43.17, an internal speed value of 0.1843 should achieve a speed of 8.0 m/s as well, and therefore also clock in with the same score as the reference. If on the other hand the conversion factor for horses has the purported value of 42.16, the same internal speed value should only achieve a speed of ~7.8 m/s, which should nominally clock in at 129 BogoTicks, or 4 BogoTicks more than the reference.

The results:

• The occupied reference minecart clocked in at 126 BogoTicks, 1 BogoTick more than expected. This indicates that the timing mechanism overestimates the measurement, and we should also increase our expectation for the other results by 1.
• The unoccupied minecart clocked in at 126 BogoTicks as well, consistent with it traveling at the same speed as the occupied minecart. It can therefore be ruled out as an explanation for the reported horse-vs-minecart speed discrepancy.
• With a base speed value of 0.1898, the horse clocked in at just 123 BogoTicks - 3 BogoTicks less than the reference! Note that this is the result expected for a speed conversion factor of 43.17, and completely inconsistend with the purported speed conversion factor of 42.16.
• With a base speed value of 0.1843, the horse clocked in at 126 BogoTicks - exactly like the reference! Again, this is the result expected for a speed conversion factor of 43.17, and completely inconsistend with the purported speed conversion factor of 42.16.

Conclusion: As can be demonstrated under controlled conditions, the speed of horses is in fact entirely consistent with the standard conversion factor of 43.17.

This still leaves the question open why a speed difference is reported when racing horses against minecarts. However, I am quite convinced by my results, and the setup leaves little room for interpretation, so my money is on flaws in the test set-up for the horse-vs-minecart races. Most notably, I suspect small "course corrections" during the race - which cause a non-trivial blow to riding speed, possibly more so than the experimentors expected.

Therefore, any reference to the reported value of 42.16 should be stricken from the Wiki; where appropriate, it might be replaced with a note emphasizing the importance of letting a horse run perfectly straight (even more so than aligning it with the axis the racetrack is built along) when measuring its speed.

--2001:4DD6:77E8:0:D8F7:6955:AD5D:5D49 18:15, 10 May 2025 (UTC)

Just noticed a minor blunder in the above tests: Instead of the value of 0.1843, the correct value should have been 0.1853. I did in fact run the original test with 0.1843 as stated; for the sake of completeness, I've now re-run that particular test under the exact same conditions with 0,1853 instead, and can report that this minor difference had no impact whatsoever on the result. 2001:4DD6:77E8:0:D8F7:6955:AD5D:5D49 19:03, 10 May 2025 (UTC)

I've read that the mechanics changed in 1.19.4, that the new breed is a variation of the average of the two parents rather than the average of the two parents plus a random horse. But did anybody traced what is the actual algorithm?

Isaacto31415 (talk) 12:27, 27 May 2023 (UTC)

user pink_cow_moo created a psuedocode formula for this on reddit

base_value = (|x - y| + (MAX - MIN)*0.3)((rand(0,1) + rand(0,1) + rand(0,1))/3 - 0.5) + (x + y)/2

if base_value > MAX:

return 2*MAX - base_value

if base_value < MIN:

return 2*MIN - base_value

else:

return base_value

Scutoel (talk) 17:34, 25 August 2024 (UTC)

Please avoid replying to topics over a year old. -~- Nerdyguy2000   talk   edits 18:26, 25 August 2024 (UTC)

I was just looking for this info. It's perfectly fine to reply to things years after the fact in case your comment is relevant, which in this case it very much is. | violine1101 ^(talk) 18:05, 25 October 2024 (UTC)

Um ok... But it should generally be avoided, no? -~- Nerdyguy2000   Talk   Edits  23:35, 25 October 2024 (UTC)

This is important information that's currently missing from the article. Yes, MCW:TALK says that in general you should avoid posting on old topics, but it's not like you should never do it. "generally" implies that there can be exceptions, especially in cases where you're providing important relevant information, such as here. I'd prefer if we wouldn't reprimand users for something they haven't done wrong and instead be grateful that they chose to provide the missing info. | violine1101 ^(talk) 23:45, 25 October 2024 (UTC)

Okay. I'll keep that in mind. -~- Nerdyguy2000   Talk   Edits  23:46, 25 October 2024 (UTC)

Wait, the talk guidelines don't say that anymore... -~- Nerdyguy2000   Talk   Edits  23:51, 25 October 2024 (UTC)

I just bred 2 horses that can jump 5.5 blocks, I'm guessing this must have changed with 1.19.4. This article should be updated with the new breeding mechanics, since there's no source out there that describes the 1.19.4 horse breeding update in detail... – Unsigned comment added by Manchega (talk • contribs) at 01:38, 28 June 2023 (UTC). Sign comments with ~~~~

I have tested on Java 1.20.4 with a horse of jump strength set to 1.0 (the supposed max value) and I am only getting 5.29996630062955 blocks of height. As far as my testing has gone, it is impossible to get a jump strength higher than 1.0 by breeding, though I've noticed the article doesn't specify the "random variation" that is added to the average of the parents' stats. I assume that however this variation works, the total jump strength can never exceed 1.0. – ZacNVR (talk) 10:00, 4 January 2024 (UTC)

I also have a horse that jumps 5,5 blocks Ugglis2 (talk) 15:08, 1 February 2025 (UTC)

It is on bedrock Ugglis2 (talk) 15:09, 1 February 2025 (UTC)

I'm on bedrock, and I summoned horses until I had a max jump height(i think) horse. It can jump up 5 blocks + an end portal frame, which is 5.8125 blocks. It can't jump up 5 blocks + soul sand which is 5.875 blocks.

The wiki's listed max jump of 5.25 blocks is wrong. Thetoiletslayer (talk) 18:23, 23 February 2025 (UTC)

Try checking the source code of Bedrock Edition. I'm not sure how, but I'm certain it's possible as other users have mentioned doing it and I've seen YouTubers do it too. I'm not certain the jump strength attribute even exists in Bedrock Edition, as attributes work differently there. Since the /data command doesn't exist in Bedrock Edition, there isn't a good way to test without checking the code. --MinecraftExp123^{(talk|contribs)} 01:17, 24 February 2025 (UTC)

I did a bunch of research into it and updated the values for both versions

Mooncatcher (talk) 08:46, 5 March 2025 (UTC)

Thanks! --MinecraftExp123^{(talk|contribs)} 08:58, 5 March 2025 (UTC)

The way the player's ability to charge a horse's jump is described as "jumping charging" sounds a bit off to me-- I would think it would be "jump charging". So, a) is this just me, and it doesn't really sound odd, and b) is there any other reason for it to be called that, like that's what it's called in the code, as opposed to some wiki editor from a while ago needed something to call it? If neither of those are true, perhaps it should be changed. -- Neverlix | they/them | (talk) | 02:49, 22 January 2024 (UTC)

Yeah, that sounds off to me as well. I'm going to make an edit to improve it but that entire little section could be rewritten. -BrianGLHF (talk) 03:56, 22 January 2024 (UTC)

Horses in LCE have exclusive sounds effects not found in other editions of the game.

See https://youtu.be/eJj3zgI4v9A?si=62eRF6NdAwNitZ4r&t=52 for an example.

The article doesn't list them nor mention their existence. Even with LCE being discontinued I think that info about these sounds should be available. 95.220.90.150 23:34, 2 November 2024 (UTC)

Unclear what exactly this means, does "it" refer to the resulting value or the statistic's bound:

"If the resulting value is

• greater than the allowed maximum,
• or smaller than the allowed minimum,

the difference towards that allowed extreme is respectively subtracted/added from/to it."

I raise this because after several attempts at breeding I have not had any horse that reaches the absolute maximum speed even though that should be quite likely. Sightnado ( talk / contribs ) 22:57, 30 August 2025 (UTC)

I'm assuming "it" refers to the resulting value, as I don't think the stat's bound would change, though I could be wrong. --MinecraftExp123^{(talk|contribs)} 01:58, 31 August 2025 (UTC)

@Sightnado Actually, according to the pseudocode on the page "it" refers to the resulting value. I'll clarify it on the page. --MinecraftExp123^{(talk|contribs)} 02:08, 31 August 2025 (UTC)

There isn't a render of a horse wearing copper horse armor in the gallery. CrockCraftMC (talk) 19:05, 6 October 2025 (UTC)

Please provide said video proof. BDJP ^(t|c) 19:52, 16 October 2025 (UTC)

How did you measure the speed? I've checked a horse with the maximum possible value for the movement speed attribute (which is 0.3375 as correctly reported in the page), and it moves at 14.57 blocks per second according to the Mini HUD mod. The blocks per second value listed in the article seems incorrect in either case.--Capopanzo (talk | contribs) 20:03, 16 October 2025 (UTC)

Whoever normally makes these renders really be letting the horse page go 👁 Image
BrandonEpicGamer (TALK) 21:46, 19 February 2026 (UTC)

This post sounds like you're accusing somebody of not doing their job. The wiki is made up of unpaid volunteers who often have other things going on in their lives. If you would like additional renders made, you could reach out to someone who often makes them, learn to make them yourself (check out Help:Isometric renders for an overview), or try a more polite talk page request. Rampage455 (talk) 05:48, 20 February 2026 (UTC)

You mean in the gallery? Those aren't vandalism. 👁 Image
NmF (talk) 00:34, 6 March 2026 (UTC)

Variant renders are both in Horse § Appearance and Horse § Renders. - Zamburger (talk) 16:59, 6 April 2026 (UTC)

Someone added the desired text, marking as resolved. Catscratcher07 (talk) 14:58, 17 April 2026 (UTC)

I've bred about ~560 horses according to the in-game statistics in a single-player world (for stats), and I've gotten twins and this wiki doesn't say anything about that existing or what the chances are. This has happened exactly 3 times in my horse breeding campaign! (twins as in breeding two horses and getting two foals at the same time (it's normally one)) Glimmer (talk) 10:26, 9 May 2026 (UTC)

Do you have evidence? Are you playing with mods? It is very unlikely that such a mechanic would go unnoticed long enough to not be on this page. Catscratcher07 (talk) 13:01, 9 May 2026 (UTC)

Retrieved from "https://minecraft.wiki/w/Talk:Horse?oldid=3567626"