Sheesh man, i tend to take this for granted sometimes, but always still realize, the amount of work you put into this stuff is mind blowing. ty

Ilmango - "Don't place 50.000 bells in your base!" Me - Ok... [sadface]

"Save on lag, become unemployed" I mean, if it helps the simulation..

The villager tests are really interesting, a thing that could also be interesting to know is if it helps to use double carpet, berry bushed or magma

I've reallllllly wanted to see a video like this for so long. There's an insane amount of speculation in the Minecraft community about what causes lag, but very few resources backing it up with evidence. Thank you!!!

2:30 The trapdoor and fencegate will be more laggy in normal worlds because their state change notifies all mobs to tell their pathfinding that a block in the world changed. The AI checks whether the block is in the range of the path and might update the path then or later. I really like that you lag-tested the hopper minecarts checking for item entities. I expect that hoppers checking for item entities cause a similar amount of lag. The amount of lag is unexpectedly high for me. It might be explained by hoppers now checking 5 times for item entities every tick (pickup volume has 5 segments) 13:50 If you want to make it even worse, place comparators reading the chest content. 15:00 Hoppers that are transferring items are on cooldown in 7 of 8 ticks. During cooldown they are as laggy as a locked hopper. 18:20 Bells cause lag because of their animation and their interaction with raid mobs (glowing effect applied after a certain time of ringing) 21:00 I think the villager comparison with/without job is unfair, because the work station search is faster when there are no workstations nearby. If there were some nearby, but the pathfinding would fail, it might be worse. Probably similar for beds and village centers. 21:20 Villagers and other mobs that are in minecarts do still pathfind, but they are teleported to the passenger position all the time The frosted ice hasn't been changed. Maybe the tick scheduler is slower with frosted ice now. In general with lag testing it might change a lot when you are in a real world. E.g. accessing blocks in a chunk section with many different blocks is slower than in a chunk section with fewer different blocks (going above certain powers of two is relevant, as more memory is required to store the block data). Accessing entities is also slower in normal worlds, because there are more non-empty entity sections. Thank you for this great lag test, I learned quite a few things from it and it really puts the lag sources in relation to each other. I suggest you also test goats and axolotls as their AI is similar to villagers/piglins.

Finally someone stands up for clearing item frame's name...

I would be intrested in Armorstands, normal ones and the ones with no gravity and so for statues

nothing like a good ol BLT to start off the day

should definitely test pistons and flying machines next episode

Is there a difference between normal items frames, glowing item frames, item frames with item, and item frames with map?

Villager : "our rights have been so much dragged in the mud, it can't be worse" Illmango : "so a villager is like 50 regular hoppers"

One test could be regarding light updates. For example a storage system/tree farm having redstone/light updates in the dark, versus a lit up contraption.

i suggest to test transfer speeds - how much lag does it cause to transfer 20 items per second over a distance of 1000 blocks. for fairness: items have to come out/go into chests ideas: - Chest Minecarts - Hopper Minecarts - Waterstreams - hopperchain -.. whatever else?

I don't play minecraft to the extent that I would ever come across any of these situations, but it's oddly satisfying watching ilmango break it down.

verry fantastic video, but i am missing two things. So two questions: 1: extinguised campfires compared to normal ones 2: What about Item frames which are holding an item

Maybe you can test tnt and withers, for witherblastchambers and compare them with tnt. Nice video by the way!

I am curious if mobs on leads cause extra lag. I ran into an issue in 1.17 where it seemed to do so in some cases but it was hard for me to prove.

I would love to see more block lag checks. Like would a lot of light emitting blocks have an effect on client and or server. Would a crying obsidian perimeter wall and ground be bad to do.

I have two ideas what can be done next: 1. move approximate lag values into a spreadsheet 2. make similar tests, but with starlight, lithium and other popular projects installed. Don't bother with testing each one alone.

Transport of redstone vertically. Examples: daylight detector, water column (soul and magma), wall block, torch tower, piston. Also should test tripwire and string to transport power (use an pistion with entity to trigger next tripwire like a repeater). Lighting updates: is it better to light up light changer sources to lessen the number of changes? or wall in the build so less blocks need to change light level?

Curious about whether having items in the item frames would have made any difference

21:05 "Alright, kids. Save on lag. Become unemployed." ~ilmango 2022

3:06 *other players wont be affected so long as their clients arent currently rendering chunks in which you have placed them (i know its kinda obvious but maybe it clears it up for someone)

this video is incredibly useful and i come back to watch it regularly, it really helps when trying to build stuff on a small server im playing on that doesn't cause major tps issues

I definitely learned quite a few things here! Super useful. I know you already tested the villagers, but I've always wondered about people doing super complex setups to "shut off their path finding" (not even sure this works anymore) and whether that actually has any benefit at scale.

1: Signs ( I know that the frames aren't as bad as I thought, but there are a few admins that say you should have signs instead of frames, so I'd like the comparison) 2: Maps in frames 3: Empty vs Partial filled vs Full vs Full shulker filled storage (Does the contents of a storage block cause a difference) 4: Villagers with pathfinding ai blocked/turned off (Glass block in head, standing on double carpet, using command) 5: Pistons moving blocks/entities (Don't know if this is possible at all)

I would like to see the difference of lag between the item frames and signs. Also signs with color/glow ink.

Amazing video! Few other items I 'd be curious about: 1. Item frames with items in them 2. Signs vs Signs with with glow text 3. Allays while holding items

maybe mob heads, such as wither skulls? the regular mob heads aren’t a full block so they’d probably cause some client side lag and the wither heads might cause some server side lag by checking for soul sand underneath them?

Great video! love the effort you put into these videos!

Thanks for putting in the time to test this, its so nice to know. Ideas for more tests: Item frames with items in, armor stands with and without armor, Signs with colored text and glow ink vs just normal tekst.

The amount of lag water and light updates cause would be interesting

When comparing lag between comparators, repeaters, and observers could you show if comparators in subtract mode make any difference?

I loved the video, a lot of work went into it and it's super helpful. You tested campfires while burning, a lot of people use unlit ones for floor/ceiling decoration and I'm curious if those have an amount of lag. Killing chamber lag tests would be nice too. Some are going to be a bit mob specific, obviously campfires are more laggy than things like wither roses or powder snow, but time to kill does matter too. Water physics would be a nice test for systems that use dispensers to flush mobs away, they have a lot of water updates. Things like bubble columns for lag and whatnot. You did item frames, but seeing glow item frames would also be interesting. Same with signs vs glow signs and potentially number of characters on those. It seemed in 1.17 that glow signs were 3x worse than normal signs. Frosted ice is so unfortunate, I really wanted to frost over the entire end void in a world I have.

I like the idea of measuring lag in hoppers! I'm interested how much lag extinguished campfires create, since I use them quite a lot in my build! Keep it up as a minecraft scientist!

This gives me an idea for a lag-efficient farm: Have hoppers on the perimeter of one chunk, and surround that chunk with item-generating farms, like cactus farms, with the water streams leading over the hopper perimeter, probably in some sort of circular way

I quite like the idea of using hoppers as a unit of lag measurement. It works well for both low lag producers like furnaces by just using a fraction of a hopper, but it is significant enough that it still works well for entities like villagers as well. I would love to see a comprehensive list of how laggy different entities and blocks are both in MSPT and FPS, or at least more videos on the subject

Really interresting video (and potential series) thanks! :) I have a couple of questions: - I know composters are more efficient to place above hopper lines instead of other blocks, but what if there are villagers nearby? Even though the villagers wont be able to find a clear path to the composters, will they influence them, causing increased lag? - And now where on this topic, how does villager workstations used for decorative purposes, in builds that also houses villagers, affect the amount of lag they produce? Here I'm of course talking about workstations that are inaccessible to the villagers but still present close by to where they are. - Does mob crossing chunk borders cause lag? I'm thinking specifically about for example, animals in a pen, if the pen allows the animals to walk over chunk borders, compared to a pend that is within a chunk. - And a final question, in regards to lines of redstone, does it matter if they are in a dark enviroment where they cause a lof of lighting updates to the area compared to if they are outside in broad daylight? I'd imagine this is only clientside fps that's affected by this?

I wonder if lighting updates/moving blocks have generally improved or worsened

Thank you for the in-depth analysis!

Re: The item frame test. I thought the whole point of why item frames were considered laggy is because when they hold an item, it has many faces that need to be rendered, not that the item frame itself was laggy. I'd like to see the wall of empty item frames compared with a wall of item frames with items in them.

21:05 thank you mango for giving the needed boost for me to change my life, i can now finally become a full time reddit user

Amazing video Ilmango. Thank you. I'm extremely interested in armour stands, including those that come with datapacks (like vanilla tweaks) which allow you to lock/seal them. I'd also be interested in knowing what % of lag comes from what in your average SMP. That would give context on what to focus on.

Glowing signs have a lot of speculation of lag so would be interesting. Thanks for all the videos illmango. I’ve relied on you a lot to enjoy Minecraft.

I love you, thanks for all the useful videos

Thank you for all of your hard work and contributions to this game kind sir. I'm fairly certain there isn't a minecraft player out there that hasn't been positively affected by the work you have done and are still doing.

Yes! This is awesome. Thanks so much for putting this together. Running a server for some friends and we don’t have the nicest pc to run it, so anything to reduce lag is a huge help. Curious, do things like bubble columns or breaking a lot of bambo/kelp contribute much?

Thanks for doing these tests, I love seeing the results. It's good to know hopper minecarts aren't as bad as people make them out to be. Also, a few suggestions for next time you test: - Compare horses, to other passive mobs. My anecdotal result is that they're laggier. Maybe also test pack animals. - Compare some simple clocks. (observer, hopper, comparator) - How laggy are torches these days?

Thank you very much for all the testing! Two questions I still have is: *1.* does the amount branches, redstone dust diverts to has any impact on lag? Say: 1. Redstione dot 2. Redstone line 3. Redstone T line 4.Redsone Kross on all 4 directions. *2.* And is a pure hopper tower or a hopper/dropper tower better? As a tower of a hopper, then a dropper, then an hopper again etc. Thank you in advance :)

I'd love to see the client-side lag tests - same thing but for the client. Great work.

Something I'd like to see for part 2 is examples of farms designed without lag in mind and the changes that can be made to reduce that lag. What's the cutoff in farm size/scale where this really starts to matter? Very much appreciate all the time and effort that went into this. I've always wondered what specific things are the largest contributors to lag and this definitely answered a lot of my questions.

Test ideas: -Does a minecart that travels through different light levels cause more lag than a minecart that has always the same light level around on its path? -Jack o lantern to suppress light updates from furnaces and Redstone torches compared to stone and glass. -Redstone lines on top of Jack o lantern vs Sea Latern vs Glass vs Slab? -How much lag does a light source on a sticky piston cause? For example, using glass vs glowstone as spawn proof block on a sticky piston to push items around in the nether. Thank you soo much for your content!

Thank you for doing these tests! I had two follow-up test ideas: 1. Supposedly redstone lighting updates can be a significant contributor to lag. Does it make a difference if redstone is placed on fully lit blocks so that no redstone lighting updates occur? 2. Do item frames with items in them (as would typically be the case) have more lag than item frames without items? If so, how much?

For villagers behaviour, could you also test, if having a bell close by to act as a village center (while villagers are contained) does make any difference? Also: do villagers that have a bed reduce the lag, due to them not looking for a bed?

Great video :) Would be cool to see how lag has evolved through the different versions of the game

Awesome job Ilmango, thanks a lot! If you would do this again, please test leaf blocks in different states (like persistent and non-persistent, attached and not attached to wood blocks etc), would be nice to see the results

You should revisit the Allay, since I would believe most people who would have lots of them is for Items sorters. I would like to see: > Holding stackables vs none-stackables > Pinging them with a note block (with or without items) > Having them in minecarts. (with or without items) If possible I would love to see: > If an allay holding a "Lag/Crash published book" would effect the world in a more serous way. > Finally, an allay holding a wheat to lure cows. (I'm just curious if the cows know that it's being held, & checks if it should follow every tick)

A video for hunting down fps lag would be a good compliment to this series.

Great video! For the next one can you do some tnt and flying machine related tests for world eaters and other tnt flying machines?

1:14 signal transmit (repeaters, comparators, observers, noteblocks, trapdoors, fencegates, redstone, rails) 4:40 minecarts with hopper 8:16 minecarts with chest 8:50 hoppers 11:34 hoppers with container on top 13:35 item transfer 15:48 different blocks 20:20 villagers 23:11 passive mobs 25:09 hostile mobs 26:26 frosted ice 27:21 itemframes

I would love if someone could put all this info together in one place, perhaps on the wiki? Also using hoppers as a reference unit is a great idea, it especially helps comparing between different types of lag (ie. differentiating client and server lag), great job mango!

Amazing array of tests, thank you for this. Also I've heard signs with glowing ink can generate lag but I have no idea if it's true or not, it'd be great if you could add that test to part 2 if you have the time for it

Looks extremely mesmerizing

The best part of your videos is the intro. It's awesome.

We need a spreadsheet or a table with all these info in hopper-lag units!

Do piglins cause more lag in the overworld than in the nether (that is, does the fact that piglins convert into pigman cause lag) Do hopper minecarts that intersect chunk boarders cause more lag than those that don't (guessing yes) Do item frames with items inside them cause more lag than regular item frames (guessing that it still won't make a difference)

Makes sense from a layman's perspective. A noteblock that can't play a note only changes state, whereas the trapdoor and the fence gate not only changes state but also there is the animation involved. I mean, that is where my bet would be.

Two lag questions: • Item frames with items (or maps) in them? • Villagers with professions next to magma blocks (blocking pathing) and with their work station (think trading hall)

I would really like to see how lag varies when changing the size of the are that contains them. Also, would love to see some graphs + a spreadsheet with the results!

This is the video we needed, I'm impatient to watch it

For the item frame test, if they were full, even more so with a complex item, it's very likely they'd lag a little more. I'd also like to see silo tests; like two stacks of barrels with hoppers vs one stack of double chests with less hoppers, to see if they cancel out the gains and losses.

God tier content as always

Would be interesting to see how diffirent light sources perform. And to see how a jackolanterns can effect a redstone torch that is being updated as well as some other light sources.

I would appreciate a test on powered rails vs redstone dust lines that are only changing by 1 signal strength for signal transfer. So a 15 block long redstone dust line that is always on signal strength 14 at its beginning and then gets powered to 15 for triggering a repeater at the end if that makes sense. I remember this used to reduce lag and hearing people say it was better than rail lines, at least in older versions. Tysm for doing all these tests!

Hey, great video thanks for the effort. I got one question: How much lag do signs cause?

The observer trapdoor test looked so trippy

I would really like to see some client-side tests as well. I know it would be different depending on hardware, but general tips about what causes insane FPS lag in e.g. shared areas like spawn would be useful. There's a ton of myths and lots of people make random rules on their servers to "reduce lag" but not much of it seems based in science.

I'd like to know what is causing less lag in sorting systems: - Hopper street with composters on top - Water stream across the filtering hoppers with ice - Rail with chest-minecart driving along the hoppers (only driving when there are items to sort in)

Iron farm would be good to test (villager scaring, etc.)

Good and informative video. I would like to see if the lightlevel is important so like if tinted glass is better than normal

The hopper test was really really interesting. I used to play on a server many many years ago that nerfed hopper item transport speed because the admins thought that they caused that much lag.

Could you compare normal villagers to villagers on carpet layers

Actually the Piglins might behave worse than hopper minecarts, because they would also have to scan for items constantly. I bet their performance depends on the amount of items around them. And technically, zombies and other mobs that can pick up/ equip stuff should too. You could even test zombies in diamond armor next to gold armor pieces etc where they have to even check if the items are better or worse than what they have.

The villagers with jobs could produce less lag with stable jobs, since there wouldnt need to be a check for reseting trades or changing jobs

Could you re-do the item frame test with items in the frames and also test glow ink signs? Otherwise, incredibly enlightening video and definitely busted some myths

Do chests still cause client-side rendering lag when they're occluded by a wall or when you're looking away from them?

I love this. Red stone stress tests, with science!

Great video. I would love a fps test.

Some ideas you can try are bubble columns, light updates via redstone lamps, and moving armor stands.

This is such good info for server owners and players alike. Could you please test with items in the item frames too? I think that would make a pretty big difference, wouldn't it?

wolfs, cats, animal pen close to cramming limit, wither mobswitch, flushing mobfarm with scaffolding spawning pads and the "new" nether portal POI searching

how laggy would villagers on double carpets be?

I assume that the sounds thing is because minecarts play a sound, but the volume is adjusted based on velocity. This means that non-moving minecarts are technically playing a sound, just at 0 volume, and because so many sounds are being played, the game doesn't play any others

Very interesting and great info! Thank you. The only *question* I had, based on the video, is how much worse are item frames with items vs. without as you showed it?

Great stuff!! What about the item frames with items it them?

Does the amount of items in a container affect lag in any way and what about signs, armorstands, cakes and trapped chests compared to normal ones?

You should test villagers on honey blocks, stacked carpets, carpet over berry bushes or single celled with their job block on magma! I've heard that putting their job block on magma when they're in single cells kills their pathing - if it's true, does it make them more lag friendly? Also please test pistons - unextended, extended, and constantly firing. Also, armourstands!

Really nice data. After the bells, I began wondering about villager POIs (bells, beds, and workstations), and whether the presence or absence of the POI changes the amount of lag that villagers cause.

You could check hopper Minecarts in scaffolding or similar to test the effect of their collision