Farming Reward Calculation

#1

Farming Reward Calculation

Please note, the spreadsheet originally linked in this post and shown below is now outdated.

Is our single source of truth for farming calculations is now at simulator.grid.tf.

NEW

  • much more easy to use
  • updates done in relation to 3.0.0 upcoming farming model

Remark

  • please help us check this calculator and give us feedback
  • a web version is almost ready and will be released soon

Examples

  • the reward is about 40-50 USD per month for a machine of 1200-1400 USD without upside of
    • TFT price going up
    • people using capacity (cultivation, can be a lot)
    • income because of network utilization
  • this is without income of cultivation, bandwidth, …
5 Likes
Farming Feedback
Let's share our farming setup
#2

image
image

simulation done with TFT 0.05 -> 1, utilization, basic network utilisation

please make your own simulation, this is just an example calculation, no promise whatsoever

#3

ps we’re waiting for our web-guys to update the wiki, hope will be done asap
the numbers on the wiki are not in line with the updated TFT/USD price

see also Farming Feedback

#4

Hi!

Thank you for this post. I find it very useful.

I have a question concerning the rewards attributed to nodes. It especially concerns the parameter: “price of TFT at point of registration on blockchain” on this spreadsheet.

From DIY farming experience and what I understand, this means that when you connect a node for the first time online, the Grid will check the price of the TFT and this will be assigned to your specific node. If you hook up another node 1 month later, where the TFT price is different, the Grid will assigned this new price to “this” new node. Is this correct? And then the price attributed stays the same (constant)? Is this also correct?

So if you have 2x the same 3node set-up, but you connect them at a different time, and the first time it’s at x$ and the second time at 2x$, one node would do twice the same amount of TFT as the other, even if they are the exact same hardware set-up? If this is so, I wonder if this can lead to problems in the overall ecosystem. I mean, the two systems give the same service to the Grid but one gets more TFT?

Also, I wonder… Is this parameter built to give an incentive for farmers to be there early in the Grid? (With the starting hypothesis that the price goes up in time.) If so, is this the best way to proceed? It seems to simply create an unfair advantage to enter when the market is low and then it impacts the future of all token distributions. I really wonder if this parameter is healthy for the project.

Maybe I am missing something. I might be way off the track. Also, what if 10 farmers get in when the price is low, but offer poor services to the Grid, while 10 other farmers get in when the price is high, but offer great services to the Grid. Is this fair?

Could this lead to a difficult situation where a group of people would create a market manipulation where they “dump” the price of TFT and then set up their nodes at low prices? Then when they want to sell, they pump the price up and sell their proportionally large quantity of tokens.

I would have thought that it would be more “fair” for all farmers that the more time your node has been on the Grid, the more tokens you would get, independently of the price of TFT when a certain node has been put online. Or maybe that the TFT distribution amongst farmers is set-up to give the most TFT to all participants based on the current TFT price, and the quality of farming offered by each farmers (total time online, less % of time offline, etc.).

Building on the last parenthesis, an incentive could be to reward the farmers with the most uptime overall. (Not just the period of time since the last time the node turned offline, as this data is very local and does not represent the overall health of the node, i.e. the node could be offline 10 minutes for maintenance every 3 months but online +/-99.99% of the time for years.)

I think Threefold is amazing and I am a happy farmer! Those questions are out of curiosity and respect for this project. I hope I am being clear here.

Thank you for your time and great work!

(Edited typos)

3 Likes
#5

amazing response, sorry for not answering sooner, I missed it.
allow me to come back on it next weekend, needs further thought for sure

2 Likes
#6

This weekend we took the time to do a large update on our library.threefold.me, lots of new info around dao, planned validator nodes, gep, … Check it out please, all remarks very welcome.

created new page for dao rules in relation to farming

See https://library.threefold.me/info/threefold#/tfgrid/farming/threefold__poc_dao_rules

We realized that a dynamic TFT/USD price is not good because there are spikes and its too hard to do a reliable price detection which can be verified. Not good for a DAO. Thats why for now we have the TFT price as part of our GEP’s and start at 0.08 (why 0.08 because there is super limited liquidity (only 7000USD to bring price to USD 0.08, and 9000 to 0.1, so its prob still too low).

about the current farming rewards model:

  • the main intent was that farmers know what their farming return will be at point of connection, we got feedback from our 2.x series that this would be a good feature, but you could be right that this is suboptimal
  • as side effect it indeed gives more return if a TFT goes up in value, which I think is fine

can this be abused?

  • if people would be able to influence the price this would not be ok
  • maybe we need to do a suggestion to the DAO once we are life with mainnet that the TFT price is not a point price but a price of a certain period e.g. last 2 months and future 2 months, so it re-balances better
  • or maybe we need the DAO to set the price every 3 months for the TFT as used in farming
  • or maybe the TFT/USD price needs to be re-adjused every X months, but this would require grants to be higher

Suggest to start with new GEP’s which are our change requests about this topic in Q1 2022

note in TFGrid 3, our aim is to operate as a DAO (Decentralized Autonomous Organization) see https://library.threefold.me/info/threefold#/tfgrid/tfdao for more info about the DAO.

I love the idea of using performance metrics like uptime, performance,network latency, … lets all discuss in GEP’s

3 Likes
Grid 3 Migration News
#7

Hi,

I was wondering if it would be a good thing to add a parameter: Internet cost per month on the simulator https://simulator.grid.tf/. Then we could add this as a cost. The simulator as of now does include the cost of electricity, but Internet costs are not to be neglected for a proper financial prediction.

There might be a reason why it isn’t there though.

We could even add a parameter/calculation that would take into account the minimum Intrenet bandwidth needed for a given farm [3node(s)]. This could really help farmers out there that want to have more than a single Titan 3node (or the DIY equivalent) and wonder the bandwidth needed.

What are the factors that come into play?

  1. The farming vs cultivation modes
    a. farming mode is simply available resources on the grid
    b. cultivation mode would need more bandwidth as users use the network to access the nodes
  2. The compute units
  3. The storage units
    a. HDD
    b. SSD

Then, we would need to settle (approximations) the bandwidth needed per TB of HDD, per TB of SSD and per GB of RAM. Here is a sketch of an equation:

min_bandwidth = bandwidth_per_TB-HDD * Qty_TB-HDD + bandwidth_per_TB-SSD * Qty_TB-SSD +
bandwidth_per_GB-RAM * Qty_GB-RAM

This would be in cultivation mode, as, in theory, one does not need much bandwidth in farming mode, as it simply is needed to “show” the grid what we have as resources, without needing to share actively on the network.

Then people could check with their Internet supplier the cost per month for the min_bandwidth and simply enter in the Simulator the cost ($) per month for the parameter Internet Cost.

This has many advantages. One farmer could check if he has enough Internet with his/her supplier, buy the 3nodes (DIY or Titans) then farm with a cheap Internet deal. Then when cultivation kicks in, the farmer could simply call his/her supplier and ask to upgrade the Internet bandwidth deal based on the current cultivation level. It would also be a great data to have when it comes to plan ahead upgrades of one’s farm.

It would also be more accurate when doing predictions for investment.

What do you guys think? Please tell me if my reasoning is wrong. I’ll gladly edit the post.

Any comments?

Later!

2 Likes
Storage / Bandwidth Ratio
#8

agree with this, lets do this anyone?
goes together with other remark we need to make sure that farmers cannot abuse the system by hosting lots of storage capacity which cannot effectively be used because internet line is too small.

2 Likes
#9

Hi Kristof!

I think you bring a very crucial point between storage/compute capacity availability VS the actual (or reserved) cultivation.

I think what some farmers would do is this:

They check their Internet bandwidth potential, and then scale their farm accordingly.

But when there is no cultivation, there is no “need” for having the maximum bandwidth capacity as there is no actual transfer of data. My point is, I don’t think farmers in general would want to abuse the system. I think the situation would be: they want to put more money into Internet bandwidth once there is an actual need, in order to reduce costs.

So my reasoning is this: once we have a clear calculation to know the bandwidth needed per compute/storage capacity, farmers could adjust their Internet bandwidth based on reserved and actual cultivation.

Does this make sense?

For example, you start with 25/25 mbps for general farming, but you know you can go to 1/1 gbps if needed.

When the farmer goes on the explorer and sees that there is an increase in cultivation reservation, they can calculate how much more bandwidth would be optimal (or the simulator shows it), and then they could ask their Internet company provider to increase the upload/download ratio accordingly before the cultivation actually increases (when the reservation starts cultivating).

There could even be a smart contract on the Grid that sends alerts to farmers when their Internet line becomes too small for the reserved cultivation. Obviously, knowing in advance that someone will reserve CU and SU on your farm could help the farmers to get the bandwidth increased before the users need it.

I hope I am being clear here. As farmers do not get direct money from farming while in farming mode and not cultivation mode, setting up a 1/1 gbps Internet bandwidth from the start might not be wise in terms of short terms economics.

Maybe I miss something or this method would not be optimal. If so, I’d be happy to know where I go wrong.

Thank you for reading!

2 Likes
#10

I also think there should be a better balance in rewards with capacity and actual use. Now the rewards are mainly driven by the on capacity. Only there is not taken into account if the capacity could actually be used for the network. So there could be a situation where there are many rewards given to 3nodes which have a high capacity but cannot contribute this capacity to the network, because there bandwidth is way to small. Besides the rewards, this can also give a distorted picture of the capacity of the network that can actually be used.

I agree with idrnd to show in the calculator the ratio between bandwidth and capacity of CU and Su. And it would be really nice if the calculator would show what the bandwidth would be if the CU and SU from the calculator are fully used. If the needed bandwidth is visible in the calculator, there could be looked if the rewards are in line with the capacity which could be really used. Since from that moment people are able to see what the required bandwidth is for their setup. At this moment this is unknown, so you can’t blame them.

2 Likes
#11

There is one thing which promotes “proper” balancing between hardware and network resources available to the farm / 3nodes:. the farmed tokens are staked up to the point where a farmer is getting capacity consumption over 30%. See here. So I would expect that nodes that have a poor connection are less sought after than 3nodes that are well connected. And when tokens are staked and redeemable/usable for the farmer I would hope he figures out a way to make his farm be used more.

For this mechanism to work - we do have to report on the connectivity number/quality of the farm for capacity consumers to be able to make their decision.

1 Like
#12

Hi, just made my first farm.

I have a question though - few that is. Why in the Farming Reward Simulator total amount of TFT is multiplied by 60? (=F43*60) is it because of staking for the period of 5years? Can that factor vary or is fixed @x60 for farmers? …can someone just buy and stake or this is farmers exclusive?

I must say i did not went in to details and might have missed something.

Thumbs UP for the project though. I hope it will make some difference and be a success.

Thak you.

1 Like
#13

It seems that one use case for threefold nodes with typical residential network connections is for computationally heavy workloads that don’t need to transmit data overly quickly. Scientific research, video rendering, etc. Threefold makes me want to figure out how to get my municipality to get a fiber optic cooperative started

1 Like
#14

Hello and welcome :wave:

First off, any spreadsheet calculators are now outdated, and I’ll update the original post to reflect this. The only up to date simulator at this point is the online one at simulator.grid.tf.

In either case, five year reward totals are shown because rewards are fixed for five years under the minting v3 model. Nodes that continue farming after five years will have their rate reset according to current TFT price at that time.

Staking of rewards is for two years or until the node reaches 30% utilization for three months.

Congrats on getting your farm online, thanks for supporting the project, and let us know if you have anymore questions :slight_smile:

1 Like
#17

Hi Scott,

Thanks for the reply. However I think the answer to my question was in the question. …if that makes any sense :blush: x 60 is actually 5years x12 months =60

So even in the simulator in advanced section “Total Farming Reward In TFT” value is actually per month value. So for example if that field for the DiY option calculates a value of 2.500 the end quantity of the tokens after 5years(or 60months that is) will be 2.500 x60 =150.000TFT - …if al parameters stay the same in that period I mean all other things considered.

“Total Farming Reward per month In TFT” would have been nice to avoid confusion. :slight_smile:

correct me if I’m wrong here.

Hurray for the community and the Threefold project !
I hope It will have a significant impact in future. Will stat my second farm 320gb RAM, 40vCores, 6TB NVME. Happy farming to all… :smiley:

2 Likes
#18

I have question about the farming rewards, I think it fits on this topic.

If I have a node, already farming, and I upgrade (e.g.) the RAM, will the monthly rewards be upgraded accordingly or will it stay the same for the five years?

Thanks!

#19

Hi @jvsc0. Good question and we have not yes an answer to it. We have received similar questions specifically for SSD and HDD expansaion (see here).

So for now there is no facility that will take into account hardware upgrades and increase (or decrease) the monthly rewards received. The things to consider to get to an answer to you question are:

  • does adding (or taking) hardware to an existing 3node reset the 60 month period?
  • does adding (or taking) hardware to an existing 3node reset the start token price?

For now we have the simple reasoning that all hardware added is there for 60 months and when you want to add hardware, add a complete new 3node. This might change going forward, as you might know we are establishing a DAO and this is a very good topic for a DAO to vote on.

#20

In my opinion (I work in healthcare, so it’s worth what it is :joy: ) it would make sense that if hardware is added (or removed) from a node, there should exist some kind of mechanism to verify this (every month?) and readjust the rewards accordingly, but keeping the original token price. It would prevent someone from trying to time token price (when adding or removing hardware) and reward correspondingly to the hardware that is effectively contributing to the grid.

From what I understand, at this moment, if I add or remove hardware, the grid creates a new node, as if the previous one was taken offline, right?

Thanks.

#21

If I read this right this is big news for a lot of us. We have been operating assuming our hardware upgrades are automatically factored into our rewards. If this is not the case, myself and many others need to wipe our upgraded nodes and re-register to the network. Let us know soon, its the beginning of the month and a good time to do this if we need to.

3 Likes
#22

I agree that this is a big news for farmers in general. It may not apply to everyone but it can have negative effects.

I had read that adding hardware (upgrading the 3node) was viewed as a good thing from the Grid’s perspective. In that case, it should be welcomed.