Ethereum Core Devs Meeting 45 Agenda

[lrettig]

Ethereum Core Devs Meeting 45 Agenda

Meeting Date/Time: Friday 24 August 2018 at 14:00 UTC

Meeting Duration 1.5 hours

YouTube Live Stream Link

Livepeer Stream Link

Constantinople Progress

Agenda

1. Testing
2. Client Updates
3. Research Updates
4. Constantinople a. EIP 1014 Issues b. EIP 1218: Simpler blockhash refactoring. Looks like we are dropping this one unless someone speaks up, like, immediately. c. EIP 1283: 1283 is moving forward per discussions on the previous call and the core devs chat room.
5. Three competing EIPs to delay the difficulty bomb and reduce/maintain the block reward: a. EIP-858 - Delay bomb and reduce block reward to 1 ETH per block. b. EIP-1234 - Delay bomb and reduce block reward to 2 ETH. c. EIP-1295 - Delay bomb, keep rewards to 3 ETH, change other factors such as POW incentive structure. There is renewed interest from miners to implement ProgPoW.

Different articles/links regarding potential issuance reduction conversation:

• A Case for Ethereum Block Reward Reduction to 2 ETH in Constantinople (EIP-1234)
• Reddit discussion of the above article
• Coinvote (79 participants): What should be done regarding proof-of-work issuance until Casper is ready?
• FEM Forum: EIP-1234 vs EIP-1227: Constantinople Difficulty Bomb & Block reward
• Bitfly Twitter Post on Miner Opinion
• FEM Forum: Final Request From the GPU Mining Community
• EIP 1295 Presentation
• Ethereum’s Economic Breakpoint: An Analysis

[MoneroCrusher]

@OhGodAGirl I think it would give the community a positive impression if you could set that whole thing up in a testnet and people can see it working.

[OhGodAGirl]

@MoneroCrusher You can also make a testnet. ;)

@jean-m-cyr You are interesting. I like you. Telegram, Discord, WhatsApp?

[MoneroCrusher]

My observation and gut feeling tell me ETH devs should definitely listen to the community here, if somebody were to fork ETH and implement ProgPOW I think there would be great support for it. @OhGodAGirl tell me more tell me more about the E9

[OhGodAGirl]

I just reviewed the meeting notes very briefly at https://github.com/ethereum/pm/commit/f5805df13a76ead07b3a65dccc6760d70406d14f - as always, props to @Souptacular for doing these.

There is so much misinformation about ProgPoW, it's not even funny.

"Reduces DAG Size" - no, it doesn't affect the bloody DAG.

"Unoptimized" - actually, we've gone on record many times saying it is optimized. The only other optimization would be fully writing the damn thing in ROCm, which is on my list to commit soon. Again, we use local mem to exchange a global address, but CUDA uses __shfl to do direct. We just need a way to do a direct exchange, which ROCm provides.

"Lists of randomly selected operations" - Wut. Did you read the damn whitepaper? This has nothing to do with operations. Silicon doesn't care about your damn operations. It has everything to do with utilizing the core, as well as the memory, thus normalizing the efficiency gap when compared to putting it on a piece of dedicated silicon. The goal is to normalize the difference between a dedicated-ASIC and a GPU.

[MoneroCrusher]

Suggestion: Reduce DAG size to 1GB DAG was made with ASIC resistance in mind, if ProgPOW is implemented, this renders the "DAG ASIC resistance" argument useless. Therefore it makes sense to reset DAG size to 1GB and keep it there, this will allow smaller GPUs to mine too and further diversify the GPU landscape. Disclaimer: I own RX 550s 2GB and would love to mine Ethereum besides UBIQ & Monero.

[chfast]

@OhGodAGirl Do you have binary OpenCL kernels for ProgPoW? AMD miners depend on them.

[jean-m-cyr]

@OhGodAGirl No DAG size reduction? I must have been asleep when I read the code! It would have been a big draw for 3-4GB card owners.

Whatsapp, preferably Signal. For email just replace the dashes in my sig with a dots and send it to gmail.

[atlanticcrypto]

The EIP discussion thread for 1295 is full of community members pushing for a PoW algo shift:

https://github.com/atlanticcrypto/Discussion/issues/1#issuecomment-416183130

Until the day that PoS comes, the mining community should and will continue to innovate and make the network as secure as possible. I believe the push to an alternative PoW algo is representative of that will, and the voices behind it are growing in number and intensity.

As @jean-m-cyr mentioned, the issuance conversation is happening regardless of the PoW algo shift right now. We believe that any reduction beyond EIP-1295 puts the security of the network at undue risk. We will be publishing our logic behind and quantification of the drivers that led us to propose EIP-1295 and our beliefs that a larger scale reduction will significantly increase the probability of a project ending attack.

Let's make that clear - if the Ethereum network is successfully attacked, the entire ecosystem will be irreparably damaged.

Our analysis will be published and linked on Medium at: https://medium.com/@brianventuro/eth-issuance-part-1-dd6e9acc00b7

[nico9111]

No reduction of issuance AND Asic resistant? Yeah sure. That’s the problem when we all try to find a fair middle ground that would work for everybody and one voice gets greedy... negotiations don’t work like that. It’s not just about miners, it’s also about devs, dapps entrepreneurs, network users, investors, etc... The community overall is voting for either block reward reduction to 2 eth or 1 eth per block. Asic resistant can be part of the negotiation but the community won’t budge, block reward will decrease before full switch to POS. With your greedy input, we, the non-miners, can’t wait for the full realignment of interest i.e hodlers and security handled by the same validating entity!!

[atlanticcrypto]

@nico9111 You clearly didn't read EIP-1295. It reduces issuance by 11.4%.

If you have a castle fortified with 3 walls, and you have an army outside your walls with siege ladders - and the person responsible for your security told you that you needed all 3 walls, would you knock one down and chance it? That is exactly what you're talking about doing. And what for? 1pct difference in coin float over 12 months? This is a probabilistic exercise - not one of compromise.

Guess what happens if the network is attacked - your devs, dapps entrepreneurs, network users, investors (including miners) all find another job.

[nico9111]

Yes we all read it. Reduction via uncles is not enough. Again it’ll be either reduction by 1 or 2 eth per block. During the call, after your lenghty background intro, you even said that you could back EIP 1234 as long as ASIC resistance would be implemented. Now you are threatening the network with your fear tactic and show greed going back to 1295. That’s not the way things work in life. You’re not alone, you can’t threat anyone when you know that POS will make you obsolete sometimes soon and that comparable chains like Bitcoin and ETC pay much less for security and function all the same. 1295 will not happen and if you don’t like it you can always mine some other chain. Other miners will take your spot!!

[atlanticcrypto]

@nico9111, I believe the developer community is comprised of some of the smartest systems designers on the planet. I fully believe they will make a decision based upon a probabilistic assessment of risk/reward, and I will respect it. Before they do, I will make every effort to quantify and defend the reasoning behind EIP-1295 for Constantinople. This is adult swim, empirical analysis is required.

[nico9111]

Some reddit link for further discussion. You will not hold us hostage Brian, not now, never!! https://www.reddit.com/r/ethereum/comments/9apnco/an_ethereum_miner_atlantic_crypto_is_threatening/

[nico9111]

@atlanticcrypto your rethoric of fear spreading and behavior don't reflect empirical analysis. It only reflects greed as you're trying to juice out everything you can from Ethereum without caring about its community and potential. This behavior doesn't belong to the negotiating table, it's dangerous. Your paper on Medium shows exactly who you are and what you're after even though you already gave us plenty of hints during the conf call!!

[atlanticcrypto]

@nico9111 With that comment and tact I will no longer respond to you.

[nico9111]

I say things as they are. Truth can hurt sometimes but is very much needed. Bye

[atlanticcrypto]

Part 2 - Our discussion of the Purchasing Power of ETH and the deflationary nature of ETH's Proof of Work system.

https://medium.com/@brianventuro/eth-issuance-part-2-deflation-13775c651a42

[nico9111]

All the credit goes to @econoar for this well sourced and informed paper:

https://www.reddit.com/r/ethereum/comments/9arwc2/comparing_the_issuance_eips_network_security/

[vp993]

Hi everyone,

I wrote an article on this issue. Join the discussion! https://medium.com/@whalesburg/ethereum-block-reward-suppression-part-1-90ed04475786

[MoneroCrusher]

@Snuff How can you say ASICs are only 10% of the network? I can also say it's 60% of the network. Both are opinions. Actually let me show you some arguments supporting my claim for ASICs making up ~40% of the network: According to your own article you say $1.7 billion (in GPUs) were invested from September to March. Yes, it would have cost $1.7B in GPUs if the hashrate came solely from GPUs, but it didn't.

According to Jon Peddie research ( source: https://www.jonpeddie.com/press-releases/gpu-market-declined-seasonally-in-q4-cryptocurrency-provides-smaller-offset) a total of 3 million GPUs, worth $776m were sold to miners in 2017 by both manufacturers. In Q1 2018 GPU sales went down by -10% compared to Q4 2017 (source: https://www.jonpeddie.com/press-releases/gpu-market-increased-year-to-year-by-3.4/). And according to this article (https://www.ccn.com/declining-gpu-sales-to-cryptocurrency-miners-healthy-amd-ceo/), AMD and Nvidia both confirm sales of $165m and $289m to miners, or $454m for Q1 2018. Amd reported sales of GPUs to miners in Q2 2018 to be $65m (https://www.coindesk.com/amd-sees-q2-drop-in-gpu-sales-to-crypto-miners/).

So now we can do some math: If $454m worth of GPUs were sold to miners in Q1 2018, and it was down 10% from Q4 2017, then during Q4 2017 approximately $500m worth of GPUs were sold to crypto miners or a total of $954m worth of GPUs in Q4/17-Q1/18. If we take a RX 570 as a benchmark (doing 30 Mh/s) and say it sold for $300 (that's a low price for a Q4 2017 RX 570), then the hashrate should have increased by 95 TH/s. From the 1st of October to the 1st of April 2018 (Q4-Q1) the network hashrate rose by over 160TH/s.

Now what about the rest of 2017? According to our numbers, $276m worth of GPUs should have sold during that timeframe to miners (Q1 virtually nothing, because crypto wasn't hyped back then, and then it slowly rose during Q2 and at the latest by Q3 people were crazy for GPUs and crypto).

At the start of 2017 the network hashrate for Ethereum was 5.96 TH/s (https://bitinfocharts.com/de/comparison/ethereum-hashrate.html). Therefore by the end of Q1 2018, we should have had a total network hashrate of $776m+$454m/$250 per RX 570 = 4'920'000 RX 570 or 147.6 TH/s. We had 110 TH/s more than that by end of Q1 2018.

Where did that come from? Yes, probably Bitmain and other ASIC manufacturers (as was also evident in Bitmain secretly mining Monero and owning over 80% of the hashrate). So the network hashrate of Ethereum as of this moment is more likely to be 35-45% ASICs, according to my estimations. Open for debate.

[nico9111]

Carbonvote https://www.etherchain.org/coinvote/poll/298 Almost 100k ETH participation and 78% approval for EIP-858 followed by EIP-1234

The community is voting!

[salanki]

With 100 votes it's quite a small sample. Since carbonvotes are weighted by amount of ETH someone has it doesn't necessarily represent understanding of network security and long term economic policy. It only represents "I have a lot of ETH, I don't want anyone else to get new ETH so I can sell mine for a lot of $$$".

As earlier commented, the poll doesn't include a PoW change nor does it include an option for EIP-1295. At least it should be updated to reflect the different options that people actually want.

[atlanticcrypto]

@salanki how dare you use logic. i'm going to post a reddit thread about you being a terrible human and shackling baby kittens to a one lane highway divider.

[nico9111]

Haha lol @atlanticcrypto the master engineer PHD of space and science @salanki a weighted vote is still a vote. 100k eth has been voted. It’s an economic reality that voters with more at stake have the right for a bigger vote. I’m not saying it’s the best but it also can be much argued that a non-weighted vote would be unfair. Secondly, this is a vote that the community is voting upon. Constantinople is coming up and we need to have something ready. If you wish to propose for a vote that makes more sense for you, be my guest. Nobody’s going to do it for you. If you don’t, you can’t complain. If the current proposal doesn’t include anything about asic resistance, maybe is because it might not be needed. I didn’t create this vote. In my opinion, EIP-1234 would be best and if the pill can go easier with asic resistance, so be it as long as the core dev can implement it into the update (that I can’t say...)

[MoneroCrusher]

@nico9111 so 0.1% of ETH holders should decide for everyone? In my opinion ASIC fork off + reduction from 3 ETH to 2 ETH would be a good solution for everyone, the network stays secure + is more decentralized & at same time issuance is decreased by 33%.

[nico9111]

@monerocrusher Decentralized governance is a hard thing to manage but we have tools and etherchain carbon vote is a good one. I agree it needs to better managed i.e how many are not voting because they think they’ll have to share their private keys. So yes more education is needed but as time goes by we need some result and the clock is ticking. If you don’t agree with this vote because you think it doesn’t represent the community then you’re welcome to offer solutions.

[salanki]

I wonder how people would like if we changed the way democratic voting works in parliament elections to give people with more in their bank accounts more votes.

Without verified identity it's hard to make a better anonymous voting system, that is for sure.

[nico9111]

@salanki, it’s called lobbying. At least with the blockchain only the owner of a wallet can vote so easy to verify the “being part of the community” aspect. One of the proposal I have is that for each and every vote proposals, there’s one option that always says “I don’t agree with this vote”. If this last option wins, the entirety of the vote is nullified. What can be challenging is the vote proposal step. I feel a proposal needs to be promoted and agreed by a mininum number of community members, pretty much like for presidential election where a list of candidates is proposed and only a few move on to the actual election.

[olalawal]

Again the little guy gets screwed , failure to fork off the corporate ASIC mining conglomerate is a huge mistake byt the ethdev team, no half ass vote can hide this fact. Why would a miner even bother to vote, 4 options and none carry the MAIN concern of miners.

Miners really don’t even care about issuance that much becuase you could double issuance and all that means is ASIC builders will ramp up even more and normal miners get the same tiny peice of the pie.

Forking Asics however is a Re-leveling of the playing field and would get alot of votes.

Reduce issuance with no fork will probably kill the network for gpus , If thats what the team wants have at it.

[MoneroCrusher]

Completely agree, the decentralized miner community wouldn't care if issuance was increased even to 6 ETH :-) You know why? Because they wouldn't get it, Bitmain and Innosilicon and their big partners would. Only way to avoid a community split is to respect the miners that secured the network for years and made it big. Ethereum is supposed to be decentralized after all, isn't it? Issuance reduction would put 90% of the hashrate in a such low number of participants you could count them with your fingers. I'm failing to understand how such a huge number of people don't get that concept. All they can think about is "miners are greeeeedy!!!!111, let's not give them more!!1" I think issuance reduction is out of question, and I support it as a miner. But ASICs have to be destroyed at the same time unless you want a community split. Good luck ETH devs.

[vp993]

Hi @MoneroCrusher!

Thanks for the research, it seems your version is more trustworthy than mine. When I was working on the article I took the data on a total NVidia & AMD sales, so 10% seems ok. I'm working on the second part of the article on ASICs, ProgPOW, and other possible solutions of ASIC issue. What do you think of ProgPOW?

[olalawal]

@snuff you need to go back to the drawing board, even a layman with zero research could tell you there is far far more than 10 percent of ASIC powered hash on the network

@monerocrusher your numbers are even conservative, if you consider the fact that , people buy gpus to mine neoscrypt, cn7 forks, Ethereum forks , and a multitude of other algorithms.

So even looking at the millions of dollars worth of gpus sold for crytocurrncy mining maybe only 75 to 80 percent or lesss of those are pointed to the Ethereum network which even pushes the scale of ASICS higher than 50 percent, this is an informed educated guess.

[atlanticcrypto]

Folks, I want to put the BTC vs ETH analysis in perspective. Let's compare the networks at the same age (meaning, 100 days of operation for both). If we do that, we get to see ETH issuance versus BTC issuance at the same network age.

I'm not sure the ETH network has EVER overpaid relative to BTC. It seems to be years away from doing so.

Bootstrapping network security is expensive in terms of monetary supply. The ETH dev team should be celebrated on every street corner in the digital world for how cheaply they've done it so far.

On a length of network life adjusted basis, ETH is no where near overpaying.

[chfast]

Legend:

• blue - BTC
• red - ETH

@atlanticcrypto you forgot to subtract the ETH created in the genesis block.

[atlanticcrypto]

ETH in the presale and ETH mined should be fungible. I didn't think I should subtract it?

[jean-m-cyr]

@atlanticcrypto I'm not sure using time since inception is a relevant factor unless both started at the same time. Otherwise the different start times occur in different socio-economic contexts.

[chfast]

@atlanticcrypto because you are counting percents by comparing issuance to the current supply while BTC started from 0 and ETH from 60 millions.

[atlanticcrypto]

@jean-m-cyr Absolutely. And now I get to make my point. Comparing ETH to BTC is a dumb relative metric. It is unusable. The ETH network is composed of a different type of hardware and is much more decentralized. It has a different USD street value of components. It takes longer to scale. There are so many things that are different that invalidate any comparisons between ETH and BTC in terms of "under" or "over" paying.

All I can say is, we operate a ton of mining hardware. If people believe that makes me untrustworthy, whatever. What I am yelling at the top of my lungs though is that if we cut the issuance here by more than we proposed in EIP-1295, we believe that the portion of hashrate that falls off the network is going to be large enough that it will put the security of the network at risk. That's all I can really do - I've published a ton of data to support it - now I have to trust in the developers to make the best call as they see it.

[atlanticcrypto]

@chfast I understand your point, but does issuance as a percent of mined coins matter or does issuance as a percent of all coins matter? The fungibility makes me believe it's the latter? If you mined a coin at block 2, it's not like you had 50% of the coin float, whereas in BTC you did.

[jean-m-cyr]

@atlanticcrypto Unfortunately the outcry from both miners and investors unjustly places the core developers (or whoever will ultimately decide) between a rock and a hard place...

[salanki]

Cross posting from #55 as that thread is less active, I apologize for the spam:

There seems to be concern of possible centralization risk with EIP-1295 and also concern about the network impact of EIP-858 and EIP-1234. I've submitted a EIP that proposes a reward reduction similar to EIP-1295 but without changing the reward structure. This EIP is intended as a measured decrease to provide data for making decisions on further decreases in latter hard forks.

[olalawal]

Link to 55, https://github.com/ethereum/pm/issues/55

It seems the same voices are talking to each other out there, basically thier own echo chamber.

AGAIN NO ONE CARES ABOUT THE BS EIP VOTE.

ita like a ballot with only one candidate, worthless , worse than rigged banana republic elections.

stop pointing to its results , its a farce

[lrettig]

Closing in favor of #56