Missed signage points are causing most farmers and pools to be unlucky
on average based on their reported space.
If a farmer misses a signage point he will be unable to send his partial to the pool, so the pool won’t register the points either … points are basis of a pool’s netspace calculation, so the luck calculated luck of the pool isn’t affected by a farmer missing SP’s, and pool luck is always measured correctly (unless the pool uses very weird math) …
it’s more the other way around ,… if the POOL misses a signagepoint the pool won’t register the points either, because it will reject the partial, while the farmer in it’s turn actually could have farmed a block for the pool in the same SP … so pools missin SP’s seem luckier than they actually are … that’s why we always have 3 fullnodes up in 3 different regions so we never have to miss any signage point.
We have no strong evidence it’s not just luck. But this is a bit hard to explain as just luck. We’ve never done this well on our ETH pool and I wish we could swap
We’re like 15% ahead of the average for the past 7 days and for the past 14.
We’ve added a ton of layers to the Chia node to improve upon its reliability along with other improvements and of course our own farmer that a minority of our users use. So one of those things may be responsible.
We’re just debugging our POS library in Go now, it’ll improve our farmer over the wrapped C++ library once released in the next version. And of course it’s one more component towards Go-Chia a node that’ll be more profitable, reliable, faster at all functions, and sync faster than the current Python one.
Someone just posted this in our discord and did the math.
“ If I’m calculating correctly, we have 1/14th of the space that Space Pool has. If we look at the 7 day wins, we would have about 200 more weekly wins if we had the same space. I don’t know if it’s a valid comparison but a fun one to make.
If you multiply the 24 hour wins, we’d have 100 more per day.”
If you are missing signage points as a pool, your pool’s luck seems higher while actually it isn’t
pool-reference code has some (i.m.h.o. badly implemented) measures in place to reduce the missing of signage points… when a farmer posts a partial, and a SP is not (yet) found it actually waits 10 seconds to respond to the farmer’s and after 10 seconds re-checks if the SP came in in the mean time …
have you implemented something like that? because without it you will definitely reject more partials than other pools and then you get the figures you see in chiapool.directory website … because your pool will reject partials, while they were actually fine because your fullnode got a certain SP a few seconds later. result: you register less netspace than you actually have, so the pool seems more lucky than it actually is
Hilarious jokes aside, I’m kinda impressed with Flexpool’s farmer. It’s running on my 2GB RPi4 and farming 744 plots over 13 drives without a hitch. Once I had my NFT, got the OS flashed to SD, and set up my configs and mountpoints, I fired up the farmer and within 45 seconds I was getting partials.
I agree with you @Nelis, I am usually wary of closed source myself. But I felt like taking a risk, and I’m satisfied for the moment. Cause like so what if they were to start scraping my couple of dollars per day? Then I’d come on here and expose them, along with any other similar-minded victims, and soon it would be over. They’d have gotten a few bucks from me, sure, but like I somehow lose change forever in my sofa every damn day.
It’s not sustainable to do shady business. If they’re in it for the long haul, lets go. If not, it’ll make for a good story.
That said, I am not heavily invested in my farm nor am I focused on production. My risk model is Hakuna Matata. I would definitely not be this carelessly flippant if there were real risks involved.
But at this point the only negative I can think of is the closed-source farmer, and it’s not even that egregious being as how it’s relatively easy to reverse engineer.
But I can’t tell you how convenient it is for me to use that 2GB RPi4 that I had laying around, thinking it wasn’t powerful enough for anything I might ever need again. I got my laptop back! Plus I’ve got an opportunity to guinea pig the service, a canary in the mineshaft if you will, and hopefully I’ll be able to contribute to the community by sharing my experience, one way or the other.
Full disclosure: I’ve been pooling with Spacepool for the last 2 months.
@ thechiaplot I concur with your farmer configuration theory, it’s been my theory for a while now. When Truepool hit its lucky streak recently, I had a look at the top 50 farmers, practically all of them were 100% on partials, and being a small pool, the top 50 are a substantial proportion of the pool. Spacepool being a popular pool with a youtube driven user base is bound to be littered with noobs. My earnings on Spacepool have been 5-10% less than what they should be, given that my farm is solid, and by solid I mean if there’s a single stale partial in 7 days, a lamb is slaughtered in an offering to the Bram.
I jumped ship to Flexpool for a few hours a couple of weeks ago, but my farm is in my office, and the drive activity with the low difficulty level was driving me crazy, so I switched back to Spacepool. I do like the 80s video game graphics GUI style; whether it’s worth a 5-10% premium is another story.
The funny thing is, I still have 300 OG plots, and I just won another block on those, so I plotted another 100 solo NFT plots just for the hell of it since my pooled plots haven’t won any blocks yet.
They are just saying that, on difficulty level 1, you get a lot of partials (or whatever). I find it reassuring personally but I did see 5 proofs in one plot the other day
