Private Pool(s) using custom block generation with 2^n transactions

[xdigital]

This has been discussed on reddit
http://www.reddit.com/r/Bitcoin/comments/20y0nq/why_do_all_the_blocks_hashed_by_unknown_miners/

I can't find it post here, I think this forum has more technical crowds than reddit

If you look at the history of blocks found last weeks, there are a lot of blocks found by unknown pool(s)/solo miner
has 2^n transactions in them (64, 128, 256, 512,....)

This is interesting! I don't think any other reasons to implement a custom block generation than efficiency.
If so, how they have done it?

[DeathAndTaxes]

No other reason.   2^n transactions is the most efficient merkle tree (balanced tree).  By most efficient I mean the ratio between # of transactions and # of hashing operations is the lowest.  Now how much of a real world difference does it make to orphans I can't say but I would guess it would be negligible.

[xdigital]

Now if someone can estimate how much power this guy has

All of the blocks found by this guy (https://blockchain.info/blocks/5.9.18.66) have 2^n transactions in them.

I assume he's a solo miner, since all new generated coins go to this address
https://blockchain.info/address/1A73ExsM2doRwTLp82rv5U36QHbBFmHD1X

Since 2014-03-24 10:13:14, he has found

Code:

292323 (Main Chain)	2014-03-24 23:56:11
292322 (Main Chain)	2014-03-24 23:42:10
292317 (Main Chain)	2014-03-24 23:24:02
292311 (Main Chain)	2014-03-24 22:37:02
292310 (Main Chain)	2014-03-24 22:35:40
292303 (Main Chain)	2014-03-24 21:21:45
292301 (Main Chain)	2014-03-24 20:15:35
292298 (Main Chain)	2014-03-24 19:53:42
292296 (Main Chain)	2014-03-24 19:35:56
292293 (Main Chain)	2014-03-24 19:13:29
292291 (Main Chain)	2014-03-24 19:03:41
292290 (Main Chain)	2014-03-24 18:59:25
292277 (Main Chain)	2014-03-24 17:25:47
292201 (Main Chain)	2014-03-24 10:13:14

In the last 24 hours
This miner has found 14 blocks in total of 188 blocks = 7.44% of the network

Total Hashrate about 46791 Th/s (estimated by blockchain)
So his power should be around 3484 Th/s

But Based on Eligius
http://eligius.st/~wizkid057/newstats/userstats.php/1A73ExsM2doRwTLp82rv5U36QHbBFmHD1X
His real peak power on Mar 18 is 1.63 10^8 = 163 Th/s

It's a big jump! unless my math is all wrong, or he did not use all of his power on Eligius

added:
I extended the math to 4 days
This miner has found 15 blocks in total of 600 blocks = 2.5% of the network = 1169.77 Th/s

[DeathAndTaxes]

Per the link ...

12 hours   3,476.30 Gh/s

Which is pretty close to your estimate.  The hashrate average seems to have declined since then so either the miner has moved to a new pool, split up his hashing power, or is experiencing problems.

[xdigital]

Another block #292334 found with 512 transactions. and Coins goes to 1Nbq2XZaRsKknf5fcT2wTXvBS31PaUWSeX

This address and above 1A73ExsM2doRwTLp82rv5U36QHbBFmHD1X has the same owner.
Since they both used in this transaction to move coins to 1zxSr679G9rvhwF6hz4Vuj7PqBnH7LsUM

[smolen]

This has been discussed on reddit
http://www.reddit.com/r/Bitcoin/comments/20y0nq/why_do_all_the_blocks_hashed_by_unknown_miners/

I can't find it post here, I think this forum has more technical crowds than reddit

If you look at the history of blocks found last weeks, there are a lot of blocks found by unknown pool(s)/solo miner
has 2^n transactions in them (64, 128, 256, 512,....)

This is interesting! I don't think any other reasons to implement a custom block generation than efficiency.
If so, how they have done it?

https://blockchain.info/address/1BGbGFBhsXYq6kTyjSC9AHRe1dhe76tD6i
This address has been used by a certain unknown miner to collate his earnings from mining blocks. For every block this person finds, it sends a satoshi to this address.

So how can Blockchain.info associate this address with double spends if the address hasn't spent anything and only newly generated coins are attributed to it?

Thanks in advance for your help

p.s. Interesting note about this "Unknown" miner: all the blocks mined sent to this address have a certain amount of transactions... 1,2,4,8,16,32,64,128,256,512....

Let me put my tin hat on and guess... Imagine that they have some miners with unlimited hashrate hosted on unreliable cheap VPS. They want to keep constant hashrate (or percentage of the total Bitcoin hashpower) and to avoid huge hashrate downspikes were some miners lost internet connection. They want to avoid detection. They want to avoid direct IP connections between miners. So they have to somehow mark each mined block, monitor their total hashrate on each independent node and adjust the power of each working miner to keep them undetected.

[fubly]

I think it´s KNC: 1A73ExsM2doRwTLp82rv5U36QHbBFmHD1X
KNCMINER

because I got today one replacement controller board from them back and this bitcoin address was inside the settings
for the eligius pool.

[philipma1957]

I think it´s KNC: 1A73ExsM2doRwTLp82rv5U36QHbBFmHD1X
KNCMINER

because I got today one replacement controller board from them back and this bitcoin address was inside the settings
for the eligius pool.

speaking of knc or any really big mining op.  security has to be a huge concern for them.

lets say the mystery hasher here has 5% of the network. sabotage has to be a concern..  by sabotage I mean physical destruction of the gear.

If 3 or 4 of the biggest ops were physically wrecked it would really help  the next 3 or 4 top miners.

Knc would be quite a target along with any physical location with 1000 plus th's of hash.  right now we are at 95,000- 99,000 th  .   I am surprised a direct physical destruction of a big mining site has not occurred as of yet.