Thread by @NicoDeva_

@filecoin has transitioned from calibration net to mainnet at block 148,888 after a 200M ICO and 3 years of dev.

Within 24h of launch a majority of miners are already on strike and talk about a failed wedding and forks.

Why? (1/n)

Filecoin is very complex and requires high-end hardware, including min 128GB mem.

Example of 1 miner, that is 3 servers with 2 computing and 1 storing, from @6block2
Price is a very high retail $40k (includes hosting and maintenance ), regular prices min. $20k, 350TB (2/n)

Sealing speed is the bottleneck, you can provide a lot of space but with 2 computing servers you can still only seal about 1TB/day for your 336TB of storage, and it can take from 210 to 350 days to use and mine with the actual full space. (3/n)

Meanwhile, you need to stake $FIL as collateral for the space provided, this is a guarantee and is slashed if space commitment not honoured (the goal being to insure that data isn’t lost). But that’s a chicken and egg pb, as nobody has $FIL to start mining and get $FIL. (4/n)

The calibration net solved that by being a final test net, and included a “space race”, aimed at rewarding early miners who invested in expensive hardware for testing, and basically took all the risks, with 1.5M $FIL distributed but vested 540 days (5/n)

Calibration net also allowed miners to start sealing witout having to provide $FIL

Sounds good so far, let's see some numbers. (6/n)

One miner, sealing 1TB/day since 8/27 calibration start, who has sealed about 50TB at mainnet launch on 10/15 (about 1/7 of total capacity), needs 5 FIL/TB =5FIL/day as collateral to keep increasing until reaching full capacity. (7/n)

Current mining returns hovers around 0.25-0.27 FIL/day (filfox.info/en/ranks/power)

This mining reward is vested for 180d (as per the SAFT agreement) (8/n)

So our 50TB miner is on ramp up to mining capacity but also on ramp to full returns. While on ramp, as of 50/180 days, he gets 50/180 of the reward or 0.075/TB/d, 3.75 $FIL for 50TB.

Wait is that below the 5$FIL/day to keep sealing? (9/n)

Right, miners need to *buy* $FIL to keep ramping up to full capacity, or just stay at 1/7 capacity. (10/n)

Here in China people are disgusted, 恶心, this is the word repeated. A napkin calculation shows you early on that your mining system that requires $20k hardware also forces you to *buy* more coins. In a country where ponzinomics is an art, the 2017 poster boy just ble it (11/n)

Take the largest miner, with 40 PiB of storage, who is around the 1/7 sealed ratio, that means they have 280 PiB of storage space ready, at a total price of $16M. filfox.info/en/ranks/power (12/n)

Total network capacity is currently around 600 PiB, for 4200PiB installed, or $240M worth of hardware.

filfox.info/en
(13/n)

Stop, read it again. $240M worth of hardware, ready at mainnet launch. Not including all the one waiting to be installed. Not some market valuation, real money changed hands, real servers. This is a conservative estimate.
(14/n)

Miners invested early on in that much hardware, but can’t keep ramping up capacity because they are forced to buy $FIL. At $250B fully diluted valuation. (15/n)

Calculations may be off somehow but sure thing is many miners have stopped adding more capacity and drives are idly waiting, as you can see the daily power growth of 0 for many top miners (16/n)

Meanwhile, while all the coins are supposedly vested for periods ranging from 6months minimum to 6y for the team tokens, 1.5M $FIL where sent to Huobi and OkEx, issue ironically raised by Justin Sun. (17/n)

Miners provided huge support to the Filecoin team during the dev years, and now are rekt.

The much anticipated Filecoin launch has turned to a shit show and disappointment. Mood is gloomy around here (18/n)

The liftoff 1-week Zoom event that Filecoin has prepared may look a bit grim. A talk like “how I won the space race” may turn to “how I came to invest $16M in useless servers”. (19/n)

The FIL004 Filecoin improvement proposal aims at changing the vesting for miners from 100% to 75%. You’d get (50/180*75% + 25%)*0.25-0.27 ~= 0.12 FIL/T/d.
github.com/filecoin-proje…
The proposal was accepted (20/n)

After the update your 50TB-filled miner will get you 6$FIL/day, out of which you can either lock 5 and keep ramping capacity and splurge with your 1 FIL/day from your $20k hardware, or just sell everything at the current dumb fully diluted market cap. (21/n)

Anyway, looks like miners got quite rekt. This is doing the news in China, but nobody’s talking about it in the West. Rekt connoisseurs @bneiluj @RektHQ, but also others like @CoinDesk @decryptmedia @MessariCrypto @TheBlock__ could investigate a bit. (22/end)