GPU Mining Is Here To Stay

Ethereum has had quite the month. Between a price bull run, $7 dollar transaction fees, Defi heating up, and the launch of the Medalla testnet for proof of stake, there has been no shortage of news and conversation. A recurrent theme in the conversation among miners most often takes the form of the following question:

“Will GPU mining be profitable given Ethereum’s planned transition to proof of stake?”

The short answer is yes, we believe it will and this article will outline some of the reasons why our CEO, Mark D’Aria recently made in his talk at Mining Disrupt Live.

Creation & Rise of Ethereum on Proof-of-Work

Over the past 5 years, we’ve seen the revolutionary creation of Ethereum, to its rise as second only to Bitcoin in terms of its capability to stand the test of time as a decentralized cryptocurrency. For its first five years, Ethereum’s blockchain has been secured by proof of work (PoW aka mining). And although plans for proof of stake were in the original roadmap, the core developer team has experienced many setbacks and even Vitalik himself has acknowledged that a transition away from the proven PoW is more difficult than anticipated. That said, it’s likely that it will be implemented, although likely with a few delays, as we’ve just seen with the Medalla Testnet.

So, from a miner’s perspective, the past 5 years has been fun, but time to find a new hobby, right?

As with any decentralized system, things are just as likely as not to proceed according to plan. And the factors outlined below will help explain why GPU mining is likely to remain profitable at least for the next two-to-three years.

Ethereum must remain decentralized to succeed

Centralization would spell the end of Ethereum as it would not function as a cryptocurrency and would then be faced with becoming a bad fiat currency or a slow token platform, losing out to AWS or another competitor. So it must remain decentralized to retain any value.

Decentralization means no one faction has enough control to completely eliminated any stakeholder.

For Ethereum, decentralization has meant that every decision has been fraught with debate and delay due to the fact that no one, even the founders, can completely dictate the path that the project will ultimately take. This is a good thing, as it has brought with it a kind of predictability that could not have been won through human intervention. With predictability comes investment and a foundation for the growth of the ecosystem.

Time increases decentralization

As more time passes, more stakeholders see the value in Ethereum and invest their resources in the project. As the base of stakeholders broaden, so does their variety. Over the past 5 years, Ethereum has a wide variety of interests due to its multifaceted capabilities. There are miners, investors, developers, cryptocurrency mavens, and a wide variety of visionaries, as well as the hackers, hucksters, scammers, and those outside the law that see Ethereum’s potential for benefit along one axis or another. All of these factions have a vested interest in Ethereum’s continued existence and there is just as diverse of a set of opinions on how to best accomplish this goal.

Decentralization favors the status quo

Just as with the separation of powers in a democracy decreases the unilateral power of any single branch of government, separation of opinions in the stakeholder groups tends to delay & discourage sweeping change.

A cursory review of past change time period proposed for implementing a change to Ethereum vs the actual time to implementation reveals a track record of months & years of delays. For example, the transition to proof of stake was planned for the end of 2018. As of writing in Q3 2020, the final testnet, Medalla, has just been delayed for 3 months and will likely be delayed further.

Ethereum’s foundation on Proof of Work has brought incredible value to entities outside of the crypto space under the current conditions and market players will act to keep conditions favorable.

  • GPU manufacturers Nvidia and AMD achieved record profits during 2017 and have launched specialized product lines for miners, primarily mining Ethereum, as the coin with the largest market cap.
  • Institutional investors will favor stability for their investments
  • Developers that are building on Ethereum’s robust blockchain depend on the uninterrupted function of the blockchain for their projects.

Proof of Stake & Proof of Work Hybrid

Ethereum’s transition to proof of stake will include a period of running both proof-of-stake and proof-of-work alongside each other in a hybrid system. The rationale for this system is that it will allow Proof of Stake to prove itself and establish a track record with the integrity of the blockchain guaranteed by the protection of miners executing proof of work.

In practice, it is likely that this hybrid system will continue in perpetuity. Due to the reasons mentioned previously:

  • Risk-aversion of institutional investors. Exchanging a low cost, proven system for a slightly lower cost unproven system.
  • Entrenched mining market interest.
  • Low cost, inflation approximately 5%, lower than the price volatility, of proof of work to the ETH user given the value of a stable /secure blockchain.
  • Delays due to decentralization/stakeholder consensus. One need only refer to the ProgPow debates to see how contentious debate delays change.

Given the factors above, it is likely to expect at least another 2-3 more years of Ethereum mining profitability.

Why GPU over ASIC?

This is a topic into itself. Our general take on the topic is available in GPU vs ASIC Profitability for those interested.

The short version is that Ethereum’s blockchain is fairly ASIC resistant. Many miners do quite well with GPU on Ethereum. If there were an ASIC that could have knocked all of the GPU mining Ethereum, it would have been available already.

Additionally, given the wavering uncertainty of Ethereum’s continued support of Proof of Work, most miners would be hesitant to invest in a machine which may be useless for any purpose in a three-month time frame. GPU can always be set to mine another coin or sold for residual value.