Make Plotting Hard: Proposing K35 minimum Plot

General Chia Discussion
1

In my opinion, Chia is having an existential crisis among all the other troubles.

Chia is the Green blockchain with brand new amazing technology ==> No it is not, it is yet another Proof of Work chain and it is becoming more and more GPU based solution.
Chia has the highest Nakamoto Coefficient ==> Well not anymore.

We need to find a path forward to bring stability and differentiate Chia from all the other 100+ block chains. We have been replotting every other 3 months at least as more and more GPU based solutions coming. They will keep on coming because compute is getting cheaper. There is already talk about having ASICs; probably not going to happen due to the financials as token value is still on downward slope.

CNI tried to dissuade this path by announcing plot filtering reduction. Yet the trend didn’t stop.

I am proposing that we need make plotting really hard. I haven’t studied the Chia blockchain algorithm but IIUC, as k number go up, memory and compute requirements are going up exponentially. So I suggest we take 2 or 3 notches above to take advantage of that feature.

Why make plotting hard?

  • With plot filtering halving, it will add tremendous pressure on compute based solutions. Space usage for each k increase is linear while compute and memory usage increase is exponential.
  • It adds a huge barrier of entry, rewarding existing farmers. This will bring major stability to the chain and incentivize farmers to move.
  • With adding highest amount of CapEx or upfront costs, the chain will have the leanest OpEx as few will be interested in joining to the chain. Remember the early days of the chain when plotting was extremely hard and expensive; that prevented an exponential growth of the chain. This was a good thing as it kept the chain truly green to operate.

I may be technically wrong here; happy to learn and be corrected but the direction wise, I suggest we take a path which makes barrier of entry high.

Edit:

Lots of farmers think making plotting hard is a bad thing and they also think we should not replot.
- First and foremost, everyone will need to replot again and probably multiple times more until things settle. Bram already made that clear, CNI is pursuing for hard fork.
- Making plotting hard is good for farmers. Farming has no upside, if chia coin prices goes up; netspace will grow as farm size will go up. If plotting is hard and costly, this will add extra friction. It is in the best interest of farmers for plotting to be hard.
- CNI was very happy and supportive for easier plotting because it helped with the netspace growth. They are not invested in making plotting easy. That’s why they spent a lot of community money how to create plots efficiently and easily.

Let’s use the hard fork discussion to ensure plotting is hard. CNI is definitely not invested in making plotting hard. GH, NoSSD; they leverage our farms for their influence but they are also not vested into farmers advantage. All these entities care about low entry because they want network to grow hence less advantage for the farmers. CNI wants network to grow so they can announce how many Exabytes are backing their network as most people see that as the security of the network.

Let me also explain this in a much simple example. What does farmers have:

  • Hard Drives, GPUs and other hardware equipment
  • Plots
    Equipment constantly depreciates. Unfortunately Plots too. In the early days, plots used to cost a few dollars to purchase, now they worth a fraction of it as it is very easy to create plots. Anything that is easy to generate worth less.

If plotting becomes hard, this will help farmers to maintain the value of their plots or appreciate just like real life plots.

2 Likes
2

It adds a huge barrier of entry, rewarding existing farmers. This will bring major stability to the chain and incentivize farmers to move.

You mean punish existing farmers who just finished replotting to compressed k32 plots.

2 Likes
3

Everyone needs to replot if they want to stay competitive. I am not sure if you have noticed but most of us have replotted 4 times already. OG → Pool → C1-C10 compression → C10-C20 compression and now we need to go through C30+ compression to stay relevant.

You will see Netspace keeps on increasing as people are moving to higher “compression” levels. I am not sure how much better the new compression levels are but it is significantly better.

1 Like
4

Netspace is currently dropping, probably due to the halving, those replotting to higher c levals are likely slowing this decline. I really can’t see netspace increasing at the current price.

The filter change in June will also likely cause a drop.

3 Likes
5

Sure, I noticed that drop. My guess is that with GH releasing its own advanced compression algorithm; we will see increase again.

Note that, NoSSD is losing netspace (I read it so) as farmers moving to more advanced compression algorithms provided by Dr Plotter.

The essence of the post is though; this is not what Chia was supposed to be. I remember in the early days how Bram was boasting that plots cannot be compressed further. Look where we are now; this was not the vision of Chia. If we follow this path, Chia will be yet another blockchain among 100’s of others.

3 Likes
6

You will make it almost impossible for MOST current miners to continue, given the HUGE ram requirement to plot such large plots at anything resembling a decent speed.
I’m not talking GRAM on a GPU, I’m talking RAM in your SYSTEM.

Note that both DrPlotter and GH C30+ compression have been out for long enough now that significant netspace would be available from both - from those that can afford it.
Yet the netspace is still dropping, probably due to the rewards halving making the current environment VERY close to the “lowest profitability of all time” even for folks that can run high compression efficiently.

It’s all going to be moot anyway - Bram has already announced a roughly 18 month timeframe to move Chia to a new plot format that is supposed to be VERY GPU “compression” resistant.

2 Likes
7

All farmers and future farmers who may be willing to invest in chia have to be very careful. This is our 5th re-plot, a 18tb HDD takes at least 24hs on pretty much any compression scale(imagine having 200 HDD). If you have invested let say $5k to get 360tb, ROI will be on neverland. Plus, now, you must add a good GPU, blabla. For those who have like me, 2PB+ I must upgrade network, add more GPU’s, more Power, stressing all system to keep things running and be at least competitive. Staying on C1 / C8 / C18 make no sense since, you pay same exact fee and lose reward. This is a freaking nightmare, just thinking about replotting again in June / July or whatever happens in the next 3 month or so.

Chia has just become a very well-orchestrated joke, period. I was not looking for a magic trick ROI in 2 months, I can deal with 3 / 4yrs in support of a project, anything that’s give you 10%+ APR is a good return. But, for this, I must be looking every houry, every time to see if plotter is working, transferring, deleting, plots, OMG this is nuts, worst project ever, is like having a full job and get $15 a day. Are we all idiots?

Just as an example, I have invested 3k on a NEOXA NODE, plug and play, and forget about it, I’m getting more return from NODES than if I invest $5k on 360Tb plus all this hustle.

When we invest in Crypto, we must accept that is 99% chance to lose the money, fine, I can deal with that, but not the time, this is a waste of time most of the time. So, just be careful guys, many farmers, including me, have put lot of money, time! That is priceless on a project that have just become a joke.

And please! Stop asking Bram about Chia future because, he only have one word in his mind: “Merengue”

3 Likes
8

Hi!
And I stayed on C0…
Even though I have completely free electricity and can afford to buy additional HW.
You will say that it is unreasonable - but when I see this arms race, I think that if I go further into this trend, I will have similar feelings as my colleague @Videce. The next step would probably be to close down my farm.

Sticking to the original rules - I am still interested in the project. In a sense, I am “moderately happy” that we may return to the starting point and the “raw capacity of plots”, although of course I appreciate the work of my colleagues in constant changes and new plots (it also creates a certain “noise” and interest in the project).
I’m not malicious or jealous!

As I mentioned in an earlier post - it doesn’t matter whether I earn pennies (if the project fails) from 0.6PiB or 4x pennies from 2.4PIBe,
or 1M dollars or 4M.
This 1M will be a sufficient rate of return for me on this particular investment…

Have a good day to everyone!

1 Like
9

Dude did you watch the 3years anniversary “keynote” and what Bram was talking about? if just going ups to 35k would solve the issue chia inc would probably go that way. can we first have some faith in Bram and then acknowledge that no human could forsee all the ways people would try to make chia PoW. I am confident Bram will figure out and restore/ensure the spirit of chia. having skipped repotting so far. after already having replotted OG plots to Pool standard I am obviously a little biased and not affected BUT I think every chia fan can appreciate that the last word in the PoWfication of chia has not been spoken. So lets all relax. If Bram makes chia PoW resistant that would be a huge deal! not saying Chia is full PoW even at the point we are now but if he is confident he can prevent this than us mortals should have some fait.

2 Likes
10

Well, Max stated on his first plotter that he will go for GPU, Max worked with Chia about GPU’s then they moved with Blade bit, so, this did not catch them by surprise. They have enough time to do something about this. I’m ok with the innovation until, it becomes a re-plotting full time job, having 24hs heavy loads on network, GPU’s, HDD, SSD and so on, and as we can see, this is just the begining.

2 Likes
11

Is not about losing money, is about time consuming and that we must be reinvesting all the time to keep going. CPU / GPU / Smarnodes / ASIC’s you name it, you may only need to update software or change protocol to mine another coin, and they all share the same problem, good days and bad days, chia, has become a nightmare for farmers.

12

Bram suggested to switch to a new format and I support that. I am also asking for additional incentives for farmers or he will have hard time to convince the network to follow his suggestion. If GH (Madmax) , NoSSD and Dr Plotter does not agree with him, network will not shift.

4 Likes
13

Not sure these players are that powerful. Reality is once the format comes they are the ones that will have to move. Chia at this point is Bram if you like it or not. Will change once day but hopefully not soon.

14

Give the guy some slack. One man and you expect him to be in front of every development. Get real. If he solves this it would be huge. And it is in all our interest it happens. Pretty sure Bram was always thinking about the plot compression and just missing some pieces. On this level things don’t get resolves by just putting some more work into them - sometimes it needs a perspective switch etc. seeing things in another combination etc. He told us everything in his last talk.

15

This discussion has nothing to do with Bram’s capabilities. My proposal is about how to make Chia greener and appealing to farmers who are early adopters.

16

How is it technically possible to have existing farmers’ plots valid but have a barrier to entry for new farmers? (I could just grab the old plot code from my local copy of git and plot the easy old style plots)

17

Not sure what the question is but let me try to answer it.

  • “Old style” and “all the existing” plots are going to be deprecated based on CNI’s proposal. This is where we are now; when and how it will happen is unknown. All of us will have to go through the pain of replotting at least once more.
  • The new plots will be much different and old tools cannot be used as they are .None of our existing plots can be moved forward.
  • Since we will be all replotting again after the 4th or 5th time; I propose new format rewarding to the existing the current set of farmers as we are the early adopters.

I am suggesting K35 or something similar:

  • There is no technology to plot k35 fast because it is computationally very expensive and GPUs cannot do it efficiently due to high memory requirements.
  • Everyone will be forced to flash based plotting. This will slow down and it will require significant plotting hardware such as Optane memory, PCI-E NVMe etc. Unlike GPUs, these dedicated hardware has a shallower market, expensive and makes it infeasible for new farmers to join without significant investment.
  • Huge upfront costs, which I would call as capex, will significantly reduce the Netspace size which in turn reduce the power consumption or opex of the network as few will put the necessary investment to increase the netspace size. This is basic economics.
18

I think I understand…you mean it will benefit early adopters (not existing farmers)

1 Like
19

How about, when you create a plot you have to register it on the blockchain. After 1000000 k32 are registered then only k33 can be and so on so it gets harder and harder to increase netspace

20

That sound pretty awesome for farmers, I doubt CNI will go through with it . yes that would be a much “stronger” version of my proposal.

Unlike early days of Chia, there is now very little incentive for farmers took another action of going through pain. There is no early adopter benefit as well since hike of coin prices passed, there are lots of xch out there over 3 years of farming. CNI has sold and will be selling their gigantic pre-mine which will cause more depreciation of the token.