Why so many pools do not support "Merged mining"?

DrHaribo

legendary

Activity: 2730

Merit: 1034

Needs more jiggawatts

Quote from: makomk on November 03, 2011, 07:26:00 AM

Merged mining is a huge pain for pools to implement from a technical perspective.

No kidding. I just finished a big rewrite of my pool software to get merged mining going.

The worst part is that there is almost no documentation. After Luke-Jr started the page at https://en.bitcoin.it/wiki/Merged_mining_specification there is at least something. I mostly had to read source code to figure it out.

Well, we finally have merged mining up and running, and we made a block after 1 minute. Grin

DeathAndTaxes

donator

Activity: 1218

Merit: 1079

Gerald Davis

Quote from: redditorrex on November 03, 2011, 12:59:25 PM

In all fairness, I don't think the main reason MM was brought about was to increase mining profits, but too ensure the safety of the Namecoin chain by putting as much hashing power as we could into it.

I agree and it has succeeded in doing that. I was just pointing out why not ever pool is jumping on the merged-mining train. It is a lot of work still, most of the "easy money" is gone and one is risking the uptime of the pool vs a 7% and decreasing "bonus".

redditorrex

sr. member

Activity: 280

Merit: 250

Nom Nom Nom

Quote from: DeathAndTaxes on November 03, 2011, 12:13:25 PM

Also there is decreasing reward as more people "do it".

NMC is worth about 0.01 BTC. This is w/ roughly 1/8th of Bitcoin miners doing merged mining.

As more people merge mine the difficulty of NMC rises. If hypothetically 100% of Bitcoin miners merged-mined then eventually difficulty of NMC would equal BTC.

dot-bit.org/tools/nextDifficulty.php

Currently the ratio between BTC:NMC difficulty is 7.69. since 1 NMC is ~1% of BTC that mean merged mining provides a 7.69% "bonus".
However even w/ next difficult change that ratio will drop to 5.81 which at 0.01 BTC per NMC is 5.81% "bonus".
If price remains ~0.01 BTC and 50% of miners merged mined then the bonus would be ~2%.

Each additional pool gains "less" for the same amount of work. When it first started the NMC "bonus" was roughly 140%. Now it is ~7%. In a couple weeks it will be 5% and eventually could be <2%. Many pools may not see enough reason to implement it.

In all fairness, I don't think the main reason MM was brought about was to increase mining profits, but too ensure the safety of the Namecoin chain by putting as much hashing power as we could into it.

I for one love how Bitcoins design has enabled a decentralized TLD system, and think it too will be important in an ever changing government controlled internet.

DeathAndTaxes

donator

Activity: 1218

Merit: 1079

Gerald Davis

Also there is decreasing reward as more people "do it".

NMC is worth about 0.01 BTC. This is w/ roughly 1/8th of Bitcoin miners doing merged mining.

As more people merge mine the difficulty of NMC rises. If hypothetically 100% of Bitcoin miners merged-mined then eventually difficulty of NMC would equal BTC.

dot-bit.org/tools/nextDifficulty.php

Currently the ratio between BTC:NMC difficulty is 7.69. since 1 NMC is ~1% of BTC that mean merged mining provides a 7.69% "bonus".
However even w/ next difficult change that ratio will drop to 5.81 which at 0.01 BTC per NMC is 5.81% "bonus".
If price remains ~0.01 BTC and 50% of miners merged mined then the bonus would be ~2%.

Each additional pool gains "less" for the same amount of work. When it first started the NMC "bonus" was roughly 140%. Now it is ~7%. In a couple weeks it will be 5% and eventually could be <2%. Many pools may not see enough reason to implement it.

makomk

hero member

Activity: 686

Merit: 564

Merged mining is a huge pain for pools to implement from a technical perspective. Before shadders added support for it to poolserverj there wasn't actually a practical way to do it at all. Even now there are some hard-to-fix issues with merged mining.

finway

hero member

Activity: 714

Merit: 500

Just curious.

Topic: Why so many pools do not support "Merged mining"? (Read 1373 times)