Bitcoin Forum

Details here:

https://medium.com/@antsankov/the-what-why-and-how-of-astor-testnet-e7366ba2a730

In short, new algo would be SHA3 aka Keccak256

oh wow
this implementation will really push cpu mining back like in the glory days in the past
gonna get my threadripper ready for this

Is there a way to implement this algo on GPUs?

The usual thing in etc project. Now the one who is the first to re-arrange in the first week will get the best profit.
Many miners did not hear about this news at this stage, and the sites where mining profit is calculated have not yet rebuilt their algorithms.

It's a testnet, there is no date of fork or height announced yet. So you can held your horses ;)

Great news, mining with CPU will be profitable again, no Asics for a period of time, I'll jump in ETC for sure  ;D
ETC is a big coin and other devs will be looking to see what is going to happen with this change of algo, let's see how can we mine

Algo for FPGA. This will hurt gpu mining even more by moving gpu miners off that coin onto something else.

But it takes time to make bitstreams to FPGA or to create Asics, they are doing this to prevent 51% atacks, they will need the community to achieve this, not only big miners

Throughput comparison graphic in the article shows a xilinx FPGA running various algos including keccak256. It'll be a matter of month if not less before the network is controlled by FPGA. Given that its a large enough coin, bitstreams will be developed while its on the testnet. ASICs will follow within a year. This change does nothing for gpu miners. Just another fpga/asic mined shitcoin.

I was wondering why the price of ethereum classic was rising, must be related to the prevention of asics. CPU miners will be more abundant but their power will be less, interesting to see what happens

As rdluffy said, the aim is to mitigate 51% attacks, which completely makes sense. ETC is totally at risk on current algo given it's value and nethash vs ETH, and progpow or not that's not gonna change. Moving to a proven algo does fix it.

Sure, asics might follow much later, but there's a deal of difference in needing to maliciously engineer asics to conduct an attack vs popping onto Nicehash for a minute.

I think by the time of the transition from testnet to the mainnet, various GPU miners will appear.
Perhaps even at the moment they are being developed.

that's already developed from a long time ago. CG and cuda miner already supported keccak and this has already confirmed. Some people already anticipated the release of keccak or especially this day when there was crypto that will use a new algo caused by the security breach.

If this is old algo, then it is not clear why the article article says that the only way to mine is use Python CPU miner.

Those articles are not relevant for me but just try to visit its official announcement will bring you to the algo that supported by cuda miner and keccak already supported by cuda from 2014. https://cryptomining-blog.com/tag/cgminer-3-7-3/
cg has already supported keccak for maxcoin mining too.

thanks so much for sharing! I feel exited. I just wonder when? anyway, it seems to me i'll have to reconsider my equipment. i hope to get some good profit from it. Bitstream will be developed in a due course, i guess? Really, news are great but it rises lots of questions.
whether it's me who'd dull here or it's something that has been hidden on purpose.

What I think ETC developers are trying to do is to position themselves in a position to capture Ethereum miners when Ethereum switch over to POS, but the team need more than this to protect this project long term they need to get Dpps running on their blockchain, this is very important

Oh my god, New algorithm SHA 3 is FPGA algo.
This mean GPU Miners era has been ended?

ETC have 10 TH/s network hashrate this mean :

10 TH/s = 1,048,576 MH/s

each RX570 have 30 MH/s, so this time ETC have about 34,952 RX 570 miners moved to another GPU mineable coins.
:(

I might be wrong, but can't you dual mine Ethash and keccack on Claymore?
Mine Eth and etc at the same time.

You can still mine etc and eth at the same time because claymore is also supporting keccak right now. I guess that already mentioned in the claymore's main thread about what's coin that already supported for dual mining. Remember the main point is CG and CUDA have supported keccak algo / SHA3.

From what I understand the algo will be CPU friendly but also GPU, FGPA and very easy to fabricate an ASIC for in the near future.

The algo won't be memory hard like DaggerHashimoto.

The threat of ASICs will only exist if ETC ends up actually being worth spending millions for to design an ASIC, since its not cheap. So I guess they don't worry about this now, maybe in the near future.

Should be an interesting developement none-the-less.

There is already an asic for sha3 (keccak), it is the silicon fusion X2 with 75 gh / s.

As far as i remember they faced a %51 attack on their blockchain so just changing algo would save them from future attacks?

I think they should choose a method like dpos which is nearly impossible to face that kind of an attack.

It's much better for you to read the whole article and that is very well explained by the author about the viability of SHA3 to be used in the POW consensus. if you are suggesting you about DPOS and have you seen the latest news about EOS and it's the possibility of centralization? I thought that is worth to read that news.

Well, it'll mean that the asics and gpu farms on NiceHash can't just be pointed at ETC to attack it, as they can currently on the current algo.

Sure, SHA3 asics (if/when they exist) might be used to do so, but it's a much smaller pool (to use the fish analogy), which means that if you did own a viable asic you'd be financially better off mining ETC rather than attacking it - thus protecting the network. Ie: currently ETC is a small fish in a big pool, by switching algo it would be a big fish in a small pool - thus likely to be dominant.

I wonder how or if the change in mining algo will affect it's price. ETC has a very strong group of devs, but IDK what the change would mean for the every day miner since IDK of any precedent coins.

If the purpose is to avoid 51% attack, I don't think changing the mining algorithm would work a lot since 51% is came from the hashing power of the miner.

The change of algo will not give any impact to the price of ETC because it will not create a change on technical or fundamental of the platform. Just like when you are replacing the algo with another algo but you still used the same mechanism without any change on it. But this can bring trust by investors to put their money on ETC and this will be a difficult task to attract them back again.

It is interesting to see how many miners will actually switch ethereum to ethereum classic. They share the same name but thats about it. Good time to mine early though for etc fans

The ETH miners will not switch to ETC, the asics are on the move and the big mining farm will invest in asics. gpu miners will go to ETH.

You don't know of any coins that have changed their algo? Pure ignorant gibberish! XMR, MBC, MAX, Vertcoin - there's a huge list, due to new asics - it's one of the biggest underlying forces in crypto today. Jesus.

they are unlikely to start investing their money in the project, because there is not even a question in the technical part, but in the reputational

Well, it's all in the perception. Is it better to take proactive action against 51% attacks (especially when you have been the victim already), or continue as before?

Personally I think taking action looks better, and choosing a highly mature and proven algo is a solid choice, but certainly that is up for debate...

Ultimately very much down to investor confidence, as you say.

I think it would have been more judicious to choose an anti-asic algorithm such as X25X that SHA3 which may be mined by asic.

These three articles lay out the rationale behind Ethereum Classic using SHA3:

(1) Introduces ECIP-1049 and the proposal itself.

https://medium.com/@antsankov/ecip-1049-why-ethereum-classic-should-adopt-keccak256-for-its-proof-of-work-algorithm-e45aee32d8a9

(2) Why SHA3 is better algo than Ethash.

https://medium.com/@antsankov/pt-2-ecip-1049-why-ethereum-classic-should-adopt-keccak256-for-its-proof-of-work-algorithm-24052ea6eed1

(3) Mining you Internet Connection. Why SHA3 mining on ETC will be different than SHA256 on Bitcoin.

https://medium.com/@antsankov/pt-3-ecip-1049-mining-your-internet-connection-why-ethereum-classic-should-adopt-sha3-71ac74a94868