No like really... you probably still shouldn't use it.
@PCHSwS
24 күн бұрын
I'm gonna use it
@T.J95
24 күн бұрын
I enjoy your videos
@Hippyflow90
24 күн бұрын
Don't tell me what to do. You're not my real dad 😂
@PotatoSofi
24 күн бұрын
Try stopping me. I'll make a 0 redstone dust storage system and you can't do anything about it.
@asphalt2554
17 күн бұрын
Im gonna use it >:3
@neohedron5126
24 күн бұрын
Should also be said, using a fully oxidized bulb is better for performance (because it sends fewer light updates) but the light updates are really not that big a deal, they’ve been heavily optimized in these versions.
@inspectortalon
25 күн бұрын
May Palla Palla forgive us for our sins...
@ryanthescion
24 күн бұрын
I would absolutely use it for the swag factor if it didn’t take such an absurd amount of quartz
@MoonyInspired
24 күн бұрын
@@ryanthescion quartzless redstoneless filter when
@ryanthescion
24 күн бұрын
@@MoonyInspired for real
@sambuma
19 күн бұрын
@@MoonyInspired Don't forget hopperless.
@MoonyInspired
19 күн бұрын
@@sambuma ofc ofc. Couldnt also forget, has to be 1 tall waterstream compatible.
@waffleawt6570
14 күн бұрын
I afked on my gold farm/bartering farm for the 5th day now for quartz 😅
@TheRealWormbo
24 күн бұрын
This filter is like modern art. It looks interesting, but you have no idea what's going on unless the artist explains it. But at least the artist did explain, instead of… yelling.
@YellowBunny
24 күн бұрын
I can't wait for Mojang to fix comparators sending out unnecessary updates next.
@MoonyInspired
24 күн бұрын
Oh hey. An upload, after many rapscallimonths.
@MoonyInspired
24 күн бұрын
That said, lets go man.
@RaPsCaLLioN1138
24 күн бұрын
haha Moony... joke's on you, it's been 2 days :P
@MoonyInspired
24 күн бұрын
@@RaPsCaLLioN1138 damn it, then youtube did not notify me! I need to go to the other video quickly so i can make that exact same joke again. Lol. Just finished watching. Its a very interesting tidbit that may help very specific usecases, maybe dustless contraptions like box loaders? But i dont really know. Good job as always. ❤️
@AgentEm34
24 күн бұрын
Should have went with rapscallEONS
@MoonyInspired
24 күн бұрын
@@AgentEm34 good one
@RyanEglitis
24 күн бұрын
Yay, not clickbait this time
@GortMinecraft
24 күн бұрын
I love that you're basically like "I made this neat new thing... you probably shouldn't use it." (Edit: and then I notice you commented almost this verbatim. XD) Sweet design. Lots of cool tricks in that!
@ribqahisabsent
22 күн бұрын
*chef's kiss More seriously, a number of acronyms were used in the video, and I think acronyms should be specified at least once in any work that uses them. Obviously, there's good reason to use acronyms within expert fields in which certain terms are mouthfuls. However, even if the acronyms being used are commonly understood within the field, using them without introduction makes it hard for laypeople and experts from other fields to understand what's being said. This is quite common within stem fields. Generally, whenever a research article uses an acronym, it's first presented in the form of " ()". This is even the case for very obvious acronyms, like ITI, CS, and SA.... Oh, sorry, I meant intertrial interval (ITI), conditioned stimulus (CS), and self-administration (SA). But you see my point: specifying the non-abbreviated terms can prevent the audience from needing to use other sources to interpret the material. Science communication is an important part of working in any scientific field because a public that does not understand the science cannot appreciate or apply it. I'd argue that's especially true in fields like technical minecraft, for which there is no proper discipline and bringing in new members depends almost entirely on how accessible public information is to laypeople. That's just one person's thoughts, though, and the video was great overall! :)
@HenryKogakumichi
24 күн бұрын
oh, an actual dustless filter, neat.
@asingleoat
24 күн бұрын
you shouldn't compare the MSPT figures directly, but rather the deltas to baseline. on that accounting the dustless one really is a good deal faster. still should probably use the standard one most of the time
@asingleoat
24 күн бұрын
more specifically, comparing the deltas will give a better picture of the contribution of the sorters and thereby a better picture of the performance characteristics of very large sorting arrays
@DatNerd
24 күн бұрын
So, the baseline/delta calculation is a good one, but I reckon the bulk of the MSPT difference between control and the experimental versions will be item lifetime and moving item entities, because items need to travel down the ice lane versus items need to fall through powdered snow. This control setup does not control for item lifetime. To weed out the differences, I'd bet the ideal test would be maybe 10^6 items per sorting array, per distribution method (water & entities vs dropper and loose items). For the control, tick warp the world with no item entities.
@BobertPickle
24 күн бұрын
Dude this is so random, love the concept of your filter by the way. I just released a new style item filter tutorial today as well. Also using copper bulb. What are the chances
@marckiezeender
24 күн бұрын
Very simple and elegant! The reason that the dust-based filter isn't that much more laggy is that each dust only has to change by 1 signal strength, and only when the filter is turning on or off, so it doesn't actually create that many dust updates compared to a typical dust-based contraption
@inamecomeunico8935
24 күн бұрын
Hey a video follow up on the comunity post! Top
@w1thso
24 күн бұрын
I've used this comparator setup for a 3x2x1 new filter desing (with redstone) with an observer and a copper bulb, but when the items come through an hopper line or a water flow (so they comes in the hopper at a certain rate), the comparator in front of the hopper make 2 impulses of redstone, so the copper bulb turn off and on, and the filter break. How tf this turns to work with you😢
@FrunoCraft
22 күн бұрын
Unfortunately this filter will break if items come in at a specific timing: Assume the top hopper has 42 items, the bottom hopper sucks out one item and at least one new item is picked up by the top hopper within 4gt (which can happen as the top hopper is not in cooldown on account of not having pushed). Then you'll have two updates within 4 gt, the second update is not picked up by the observer and the filter ends up in a locked state despite more than 41 filter items in the top hopper. I stumbled upon the same problem when I made my copper bulb filter video a while ago (and I'd link it if yt would allow link in comments, so there you go).
@RaPsCaLLioN1138
22 күн бұрын
I don't think it's an issue. Pass all your items through the aligner at 8gt and you don't run into that problem.
@FrunoCraft
22 күн бұрын
@@RaPsCaLLioN1138 Agreed, if items are aligned at exactly 8gt intervals this should fix the issue. But it's a requirement that should be mentioned ;)
@bengoodwin2141
22 күн бұрын
This is neat. I don't like the use of the cursed budding/update stuff, but I respect the achievement.
@harvey195
9 сағат бұрын
So one block more efficient space wise and barely more efficient lag wise. Uses a ton of quartz, but that shouldn't be an issue with a barter farm.
@Inyria
8 күн бұрын
What shader pack (and possibly resource pack) are you using? 😄
@Hippyflow90
24 күн бұрын
I saw yellers video when he came out with it. Glad to see you take a stab at it
@LizQuilty
24 күн бұрын
This is great! thanks! Also my thoughts were also same as yours when i saw the other one 'wtf? this has dust though??'
@aidenjamonski2483
24 күн бұрын
At least the title got changed on the other video to not make it as clickbait-y as it was previously!
@Calenardhon314
24 күн бұрын
The marginal difference in performance seems to not be that marginal: Control: 0.57 mspt Impulse item filter: 0.66 mspt Dustless item filter: 0.63 mspt So the actual lag impact of the sorters (if the testing was good enough) is: Impulse item filter: 0.09 mspt Dustless item filter: 0.06 mspt Resulting in a ~33% decrease of lag. But yeah, probably still not worth it unless you are building some insanely lag optimised storage monster.
@SandCaribou1890
24 күн бұрын
I think he meant marginal as in 0.03ms is not noticeable at all to the player really, as though it’s 33% it’s still absolutely tiny
@DatNerd
24 күн бұрын
@@SandCaribou1890 Precisely, 0.03 ms is 30 microseconds (about the half-life of Bk 251m). Also, this control does not account for item entity lifetime and item entity movement, which experientially will be the bulk of the lag. Given such a small difference over a small sample size ~3000 items, at best the test is inconclusive, at worst the test says it doesn't matter.
@Calenardhon314
22 күн бұрын
The absolute value of 0.03 mspt difference is indeed marginal and can be ignored. Usually the whole impact of storage systems can be ignored by players, unless they built a higherr-end system. I agree, that the testing was not sufficient to determine reliable performance data. However, if the 33% lag decreade are approximately correct, then this "marginal" difference can matter to really high performance systems. There are probably very few designs and use-cases out there, that would even notice this change, but they might be there. But you are correct, the main sources of lag in item sorters are usually other things.
@drbixler
7 күн бұрын
So, unless I'm completely confused, the output is based on the redstone level of 3 from the hopper facing into the comparator. Why is it that I couldn't add more blockers to decrease the number of actual items (ex: Diamond Blocks) ? I've tried adding more blockers, but then I see the output at 4 stopping the hopper. Just wondering. Also, I've been playing around with items that stack only to 16 with no luck?
@TheRealYeller
24 күн бұрын
This is the stuff I like to see :) And I also didn't even know before this that you could updated budded droppers the same as pistons. I honestly thought it was only a piston thing xD
@jonathonathon
24 күн бұрын
4:00 Can you post or pin a link to the item aligner by the original creator?
@RaPsCaLLioN1138
24 күн бұрын
Now linked in description.
@raduking
24 күн бұрын
This will work until hopefully QC is removed and the bug is fixed
@RaPsCaLLioN1138
24 күн бұрын
@@raduking please never
@raduking
23 күн бұрын
@@RaPsCaLLioN1138 was just teasing, I’d rather have QC than having to play Bedrock :)))
@mayonnaiseeee
24 күн бұрын
Awesome. Thanks for showcasing it and explaining it all so clearly. Will build at least 1 tile of this for sure.
@ragdollwilly1
24 күн бұрын
I love your videos
@timlyon1
11 күн бұрын
I,m sorry, but I just can’t get it to work. Can someone help please?
@deltamico
23 күн бұрын
Next on the list: Quartz-less item sorter (no observer and comparator) Can be built without access to nether
@Measiam1
24 күн бұрын
Thanks for the dustless idea with bulbs, I’ve used a similar configuration for auto crater setups, making them a bit smaller
@ab-vj7vc
21 күн бұрын
this does not work for me, maybe it is because im playing on a papermc server or its in the nether any idea?
@AgentEm34
24 күн бұрын
This is the goofiest shit ever 😅 good job y'all
@cytrynowykisiel3715
24 күн бұрын
Can you test standard filter with experimental redstone on? Because if its optimized more then it should be propably better than non dust version?
@RaPsCaLLioN1138
24 күн бұрын
I don't think there's a point testing it against experimental features. For context, the bundles have been an experimental feature for 4 years.
@TheUnknownFactor
24 күн бұрын
What happens if you try the dusty version in the experimental features of snapshot?
@bengineer8
24 күн бұрын
What if lights were added to stop light updates?
@EmdyMC
24 күн бұрын
It's a shame it can't be ss2 since that sends an additional update, now THAT would be a reason to use it
@DatNerd
24 күн бұрын
It can be if you manage signal strengths correctly, and send batches of ~20 items at 8gt. Even still, this circuit we made should be used only for an "it can be done"-a 1wt dustless, pistonless item filter, in the same vein as Oreo's comparator-less item filter.
@The_TermiGator
24 күн бұрын
I won't use the dustless item filter, but I want the item aligner
@CommandLeo
24 күн бұрын
It's linked in the video description
@rozbiynik
23 күн бұрын
can it be ss2?
@finnabraun
24 күн бұрын
Is the item filter only marginally faster? If you subtract the baseline, the dustless filter is 0.06 ms per tick, and the standard filter is 0.09 ms per tick. That’s a 50% improvement. The improvements would likely be more noticeable if the number of filters increased.
@DatNerd
24 күн бұрын
So, I reckon the primary cause of lag here will be item entity lifetime and item entity movement. This 50% improvement you've mathematically identified has an absolute limit-there is only so much lag is caused by dustful filters. At such a small scale, 10^3 items, the difference is 0.03 milliseconds-30 microseconds! This number is very small. I don't even have a good comparative to describe how small this number is. This difference is about the half-life of Berkelium-251m. All this is to say, the error bars will be big, and not much can be deduced from the test. If someone does want to do a large scale test of this, run at least 10^6 items. Use the entity tracker mod to check item lifetime, also, carpet's /profile health, and /profile entities has more pinpoint data.
@lucidmoses
24 күн бұрын
When I say you using 41 items still I thought. With the copper blocks can you start draining at 41 items (red stone level 3?) but only stop draining when you hit redstone level 1.
@josiahdsmith5641
24 күн бұрын
No because this relies on the filter only changing signal strength (ie: redstone level) +-1. If the hopper was at SS1 (signal strength 1) and then filled partway to SS2, and then filled all the way to SS3, then that would trigger the dropper twice, which would relock the pulling hopper and stop the filter completely. It’s kinda like the opposite problem that regular item filters have, where if you do SS1->SS3 the excess leaks into nearby slices and causes them to fully empty, but in this case it would permanently lock the current slice
@marckiezeender
24 күн бұрын
if you use oxidized copper bulbs and use sufficient torches then you can avoid light updates entirely
@lucidmoses
24 күн бұрын
@@josiahdsmith5641 This is possible. Here is a Huge version. Hook up the output of a RS-Nor latch to the bottom hopper to lock it. Pull a redcoder off the comparator from the middle hopper and and hook signal strength 1 to set and signal strength 3 to reset. Now when the hopper gets to 42 items it will start draining but when it gets to 18 items it will stop and start filling up to 41 again. I could do this but I don’t know if I could make it small enough or one block tileable for that matter.
@josiahdsmith5641
24 күн бұрын
@@lucidmoses I was saying it’s not possible with the design shown in this video. You would have to use some detection method other than a togglestate though, something with more granular control that can recognize the difference between SS1, SS2, and SS3, which means dust, which means it can’t be 1wt or the signal will leak into other tiles. You can make it SS1SS3 and dustless, but then you have to sacrifice 1wt. You can make it 1wt, but then you have to sacrifice SS1SS3. You can make it SS1SS3 and 1wABt, but then you have to sacrifice dustless. There’s no way to have all three (SS1SS3, 1wt, and dustless) all at the same time. It’s a fundamental issue due to the nature of dust, the nature of tileability, and the loss of precision by using non-dust components
@lucidmoses
24 күн бұрын
@@josiahdsmith5641 I'm suddenly less worried about dustless every since they put out that videos about FINALLY addressing the redstone block update performance. Course there are some nuts out there complaining that they will not be able to make doors as small so they shouldn't do it. Sheesh.
@cswertwertwert
24 күн бұрын
Nice. But are light updates causing any additional lag since? Just a thought. And have you tried classic with the experimental snapshot redstone changes?
@RaPsCaLLioN1138
24 күн бұрын
Replace the concrete with light source blocks (froglight or something) - again, probably marginal gain, if any at all. Mentioned this in another reply also... I don't think there's a point testing it against experimental features. For context, the bundles have been an experimental feature for 4 years.
@NonJohns
24 күн бұрын
tell me about it i lost my mind when i saw they changed the recipe from RABBIT LEATHER RABBIT ffffing LEAHRER
@TheRealWormbo
24 күн бұрын
Use oxidized copper bulbs (light level 4, IIRC) to make the light updates even less relevant than they already are since 1.20.
@CommandLeo
24 күн бұрын
Since 1.14 the light engine runs on a separate thread, so light updates cause no server lag at all.
@cswertwertwert
24 күн бұрын
@@RaPsCaLLioN1138 You say marginal gain, but there is already a marginal difference from the baseline. Wasn't sure if that would eliminate that difference altogether.
@EightBitRG
24 күн бұрын
Would this still have the problem of item overflow if you were to use eggs in the filter item slots in order to minimize how many items have to sit in the filter? Being dustless it might no longer have that problem
@RaPsCaLLioN1138
24 күн бұрын
Yes and no. You could use 16-stackable items as fillers, but once it overflows, it'll just set itself to the next highest number of items / ss change. i.e. If you set it as 18-1-1-1-1, and it overflowed, it would just set itself to 41-1-1-1-1 and stay there forever.
@josiahdsmith5641
24 күн бұрын
@@RaPsCaLLioN1138 Wouldn’t the slice lock itself up tho? Since it goes from SS2->SS3 (unlocks), then SS3->SS4 (relocks) and then nothing can trigger it to unlock again? Since it’s still using a togglestate-based system
@dagisinmines3412
24 күн бұрын
@@josiahdsmith5641I think so too when hopper is filling up slowly and not from a big stack
@HyperlightSniffer
24 күн бұрын
Exactly right. I've been playing around with a toggle system and broke it right away just by feeding it manually. You have to set up a system of gates to bunch your items up first, then release them in clumps.
@XepptizZ
24 күн бұрын
@@HyperlightSnifferI have hooked up copperbulb based filters in my goldfarm that uses icepaths and indeed processes bunches. It hasn't had any issues in many hours of afk. But yeah, feed 1 item at a time in succession easily breaks it.
@T.J95
24 күн бұрын
I like it very much and think that there is another good side, since there is no redstonedust, there is a possibility that it does not happen like with the standard itamfilter, when it reaches the max 64 itams in one hopper, the other one leaks
@dagisinmines3412
24 күн бұрын
Yep I don't think this leaks. But wouldn't the extra update from comparator still disable the filter?
@marckiezeender
24 күн бұрын
It wouldn't leak, but it would break, because the filter would turn off again at signal strength 4.
@Dpostel
24 күн бұрын
It can never reach signal strength 4. That’s the whole basis of the design for the standard filter with overflow protection. 64 + 4 unique items are not enough to output a signal of 4. You need a total of 69 items to reach signal strength 4. So one more then your hopper can have.
@dagisinmines3412
24 күн бұрын
@@Dpostel Yes that's the idea of overflow protection. But you would agree that this filter design needs overflow protection? Thus if it had the '1 sorted item and rest fillers configuration' to reduce the need to have multiple of some expensive filtered item. Then it would brake?
@marckiezeender
23 күн бұрын
@@dagisinmines3412 yes, it would break. Because if the comparator goes from 3 to 4, the filter locks again, which will permanently lock it.
Пікірлер: 99