Pressure Mounts for Ethereum Developers as ASIC Miners Begin Dominating Hashrates
Ethereum’s hardfork scheduled for later this month will likely have some irreversible ramifications. Ethereum mining will never be the same again — and its not because of the fallen profit margins, nor is it because of some silly dispute to create Ethereum Cash. This is a planned event supported by almost the entire Ethereum community — minus the miners. We all knew that mining Ethereum would soon come to an end; it’s in Ethereum’s roadmap to switch to a Proof of Stake consensus. And although Ethereum will still be mineable until that fateful day in which PoS comes, some decisions made for this hardfork will have dire consequences for Ethereum miners.
Ethereum’s Significance in the GPU Mining Industry
Ethereum was (and still is) the poster-child for GPU mining. It shoulders a huge symbolic image of cryptocurrency mining. By the 2017 “cryptocurrency mania,” Bitcoin mining was already in the controls of those holding large amounts of powerful ASIC miner. Ethereum, on the other hand, was very profitable to mine — and anyone could do it. From gamers with powerful GPU cards, to hobbyists with some money to experiment with, to college students “borrowing” electricity from their dorm rooms, and of course to early-adopting entrepreneurs that invested large amounts into facilities making a killing. Ethereum mining is not only symbolic to all the beneficiaries, but the entire blockchain and cryptocurrency industry: it brought the idea and concept of “mining” mainstream.
Those golden days are long gone. 2018 was a brutal year for cryptocurrency speculators, and miners are not an exception. NVIDIA was the worst performer in the S&P 500 last quarter, hundreds of mining facilities have shut down either temporarily or permanently, and tens of thousands have stopped mining at home. AMD’s RX 580, which cost almost $600 earlier last year, can now be bought for less than $150. There are huge liquidations of mining equipment which can be found on sites like Craigslist, eBay, and Reddit. The GPU shortage is over and gamers worldwide rejoice as prices have now reached normal levels.
Enough reminiscing about the past, let’s get to the present.
The Constantinople Hardfork
The Constantinople hardfork is just another step closer to Ethereum’s eventual switch to Proof of Stake when Casper is implemented. This month’s hardfork is set to occur at block number 7,080,000 (estimated to occur in less than two weeks) and will implement five EIPs (Ethereum Improvement Proposal).
- EIP 145 proposed by Alex Beregszaszi and Paweł Bylica — Simple summary: “To provide native bitwise shifting with cost on par with other arithmetic operations.” If you are not a developer, this EIP will have little to no effect for the majority, so don’t worry if that “simple” summary made no sense to you — I had a hard time understanding myself.
- EIP 1014 proposed by Vitalik Buterin — Simple summary: “Allows interactions to (actually or counterfactually in channels) be made with addresses that do not exist yet on-chain but can be relied on to only possibly eventually contain code that has been created by a particular piece of init code.” Again, if you are not a developer, you do not have to worry about this change.
- EIP 1052 proposed by Nick Johnson and Paweł Bylica — Simple summary: “Specifies a new opcode, which returns the keccak256 hash of a contract’s code.” This EIP will allow contracts to perform bytecode checks in a cheaper, more efficient way.
- EIP 1283 proposed by Wei Tang — Simple summary: “This EIP proposes net gas metering changes for SSTORE opcode, enabling new usages for contract storage, and reducing excessive gas costs where it doesn’t match how most implementation works.” Another improvement in the Ethereum network addressing high gas usage which in the macro-level is greatly beneficial, but irrelevant for the 99.9%.
- EIP 1234 proposed by Afri Schoedon — Difficulty Bomb Delay and Block Rewards Adjustment. This proposal is the focal point of this article — so let’s dive right in.
EIP 1234: Why It’s Likely to Send Ripples Throughout the GPU Mining Community
Current State of Ethereum Mining
Depending on where you live and the cost of electricity you pay, Ethereum is still profitable to mine. However, for the vast majority (i.e. people in California paying $0.13-$0.16 per kWH), it is either unprofitable or not worth the trouble to simply break even.
A mining rig with eight RX 580s today can net you around 0.13 ETH a week. That’s about $20/week at today’s price, and $185/week at Ethereum’s all time high of $1,422. A drastic difference. If someone’s mining Ethereum right now, that person either has access to very cheap electricity or has the “hodler” type mentality, taking losses now for future gains. Yet, there are still tremendous amounts of resources still being spent on mining Ethereum. It still retains the most hashing power by a large margin — after Bitcoin of course. A significant portion of that hash power comes from small-time miners and mining facilities across the globe.
All this is about to change.
Difficulty Bomb Delay
A difficulty bomb was set in place in late 2016 to slowly, then rapidly increase the mining difficulty to make the economics of mining less attractive in anticipation of the inevitable switch to Proof of Stake. This “Ice Age” has been delayed before, and is being delayed again for twelve months as the Ethereum developers are behind schedule on implementing Proof of Stake along with several other items in the roadmap that hold huge significance to the future of the network.
But wait, isn’t this delay a good thing for miners? Yes and no. It comes with a paramount caveat that has gone largely under the radar. The rewards adjustment.
Block Rewards Adjustment
Everytime a block is mined, the finding pool is rewarded a block reward in Ethereum. Block rewards adjustment is not something new, in fact, it is a key part of crypto-economics. The last reduction happened in late 2017 when the block reward was reduced from 5 ETH to 3 ETH. In about a week, this reward will be reduced again from 3 ETH to 2 ETH — roughly $1.5mil in daily rewards.
As I have previously said, block rewards reduction is nothing out of the ordinary and can be beneficial economically. But this reduction, in my opinion, couldn’t happen at a worse time.
Dangerous Timing for a Rewards Reduction
This reduction will not have the same effects as the late 2017 reduction. Mining Ethereum was very profitable, and the value of Ethereum was skyrocketing. Miners were still happy, and investors were happy. However, at a time when there are so many miners on the fence whether to mine or not, a 33% reduction in rewards will absolutely positively sway them away from Ethereum mining. Miners that are breaking even will suddenly be at a sharp loss, again moving them away from Ethereum.
But wait, doesn’t that mean mining difficulties will re-adjust when miners leave the network? Yes, but there’s a new threat in town.
The Rise of Ethereum ASIC Miners
The four letters “ASIC” can make both projects as well as miners shudder in fear or disgust. Application specific integrated circuits, or ASICs, are specifically designed chips to perform one task: mine powerfully and efficiently. Once an ASIC chip is developed for an algorithm, it usually means one thing: GPU miners will soon become irrelevant for mining that coin. I like to analogize small-time GPU miners as “mom-and-pop stores,” and ASIC miners as Wal-Mart.
This has caused a constant game of cat-and-mouse where projects hardfork to a new PoW algorithm, only to have that new algorithm taken advantage of again by newer ASIC chips. On the other hand, some projects welcome ASICs, as the newfound hashing power can help secure their network even stronger.
ASIC miners for Ethash, Ethereum’s PoW algorithm, are finally here. Several companies have already began sales, and many others have announced even more powerful and more efficient Ethash miners. Ethereum, once thought as a safe-haven coin from ASIC miners, will soon see a flood of these machines enter the market.
To reiterate: as small to mid-sized GPU miners leave the network, there will soon be warehouses lined up with ASIC machines (like the ones you see in Bitcoin documentaries) mining Ethereum.
Possible Consequences & Case for ProgPOW
As a result of smaller miners leaving the network while larger “Wal-Mart” type players enter, the end result will be more centralization — something Vitalik Buterin is fiercely against. It goes against the core fundamental value of cryptocurrency. Many community members have called for a temporary implementation of ProgPOW (Programmatic Proof-of-Work) to combat the rise the Ethash ASIC miners. Ultimately, the developers did not feel it was necessary for the upcoming fork.
A few days ago during a core developers meeting, perhaps under the pressure of the massive Ethereum mining community, they had a very interesting discussion regarding this issue. “Mr Else” pointed out that the current ASICs for Ethash are only “marginally” better than a normal GPU — but the next generation of the ASICs will be about “2x better than GPUs.” They continue to call this a “slow arms race” between the GPU manufacturers and ASIC companies. In the end the decision was made that ProgPOW will be implemented in the future as long as there is nothing critical found in the largely untested algorithm. However, that could take precious months to implement, not to mention the logistical headaches and resources it would take for the network to hold another hardfork.
Truth is, no one really knows for certain what the mid to long term consequences are. But the short-term consequences are, in my opinion, quite obvious. The departure of thousands of miners from Ethereum will have ripple effects throughout the entire mining industry. When they switch to different coins to mine, the difficulty rates will drastically increase for other GPU mineable coins. To put into perspective, Monero, the next highest GPU-mineable coin in market cap, has a daily reward of approximately $120K. Ethereum’s daily rewards will be reduced by $1.5mil. Simple economics tell us that this will displace a lot of GPU miners, further consolidating the power of ASIC miners. On the other hand, this could possibly result in a surge in price as supply will be reduced. But this could also take months, as the market continues its fight with this bear market.
The implementation of ProgPOW may ultimately be a moot point if it takes too long. The 3–4 month timetable of its implementation is optimistic at best. The network would also be taking a step back, as ProgPOW’s confirmation times take 1.5 to 2 times longer than Ethash. Instead, there should be an immense pressure to implement PoS as soon as possible before the new ASIC miners begin to accumulate a disproportionate amount of Ethereum, rendering PoS more centralized than desired from the get-go. It seems that ASICs always win in the end, and the saying “the rich get richer” is certainly holding true even in the cryptocurrency space — which is ironic in every sense, but not surprising.
There’s no denying that huge ASIC mining pools can pose problems, from the centralization of power to even 51% attacks on weaker coins. However, with all this being said, I am not against ASIC mining or miners — in fact I myself own a handful. We live in a capitalistic society, and where there is profit to be made, someone will seize the opportunity. Putting the financial incentives aside, they actually provide tremendous value in securing large networks. They also spur innovation in cryptography, leading to projects like Grin which uses the alleged ASIC-resistant MimbleWimble protocol. Regardless, it’s a bit saddening to think that the most influential cryptocurrency for GPU mining may be nearing its end for most miners. Unless the core developers act fast, Ethereum will no longer be known as the “go-to” GPU mineable coin. Ethereum will be an ASIC coin. GPU mining will never be the same again. The end of Ethereum mining, as we know it, is under significant threat; the beginning of the end of Ethereum GPU mining may very well be upon us.
Feel free to follow me on Twitter or send me an email.
Pressure Mounts for Ethereum Developers as ASIC Miners Eagerly Await Next Week’s Hardfork was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.