Topic: Guild pool a 51% threat? - page 2. (Read 8867 times)

If one pool operator wants to go evil, he may just drop PPS fee to 0, then he will get +51% in a short time (new asic units concentrate alot of mobile hashpower in few user hands).
So, in the end, the question is how much we trust pool operators.

I have been increasingly active, especially in threads that are voicing concerns.  While it makes no economic sense to perform a 51% attack (it would kill confidence, thus tank the value of Bitcoin), it is always a concern just because the possibility it could happen is enough to cause some of that panic.

Right now, BTC Guild is a little under 40% in the last 24 hours, and had decent luck in these last 24 hours which slight boost to that percentage (assuming overall network luck was neutral).  If we see consistent upward percentage movement and reach a higher level, I will be killing the getwork side of the pool with very short notice in order to prevent 51% from happening.

I'm hopeful that the shipment of Avalons outside of China will level out the pool speeds more, and that ASICMINER's next round of speed will be added to Ozcoin or another pool, rather than stacking more onto BTC Guild.  These two factors should put the pool speeds into a more balanced state.

I do regularly chat with the Operator of BTCGuild, he is concerned at the rapid growth too, if you look in his thread he even mentions it

despite many peoples worst thoughts the vast majority of pooloperators are here to provide good quality mining experience to miners, verify Bitcoin transactions and support Bitcoin, not to destroy it...

BTC guild pool has the 40% of the total network power.

http://blockchain.info/pools

http://www.btcguild.com/index.php?page=pool_stats

The guild pool speed has increased +400% in the last 6 weeks.


I dont know how guild pool operates so i ask you:

Is there a possible threat for a 51% attack?


Lets make some calculations:


Gmax = Gn + (k x h)

X = Xn + (k x h)


Gmax :  guild pool speed for becoming  a 51%  threat.  (Gh/s)

Gn: guild pool speed now  (Gh/s)

X:  total network hash rate  if Gmax is reached (Gh/s)

Xn:  total network hash rate now  (Gh/s)

k: number of ASICs

h: ASIC hash rate (Gh/s)


Assume that the network's computation speed  (do not include guild pool speed in the networks total hash rate) does not grow so fast as the guild pool speed does.
A good assumption because the networks computation speed has grown +100% in the last 6 weeks Vs a +400% increase in guild pool speed.

http://bitcoin.sipa.be/speed-lin.png

Also the +100% increase includes the +400% guild pool increase. So the increase of the network's computation speed without including the guild pool increase is less than 100%.
Guild pool has added 16000Gh/s to the network the last 6 weeks. The rest network has added 10000Gh/s to the whole network the last 6 weeks.
So the rest network computation speed has a +40% increase Vs a +400% increase of guild pool speed.

How many ASICs  should be added in the guild pool in order to become a threat?

Gmax / X > 0.51                                            =>

[ Gn + (k x h) ] / [ Xn + (k x h) ] > 0.51             =>

.
.
.

k > [ Gn-0.51Xn ] / [ 0.51h-h ]

Gn = 20000Gh/s

Xn=50000Gh/s

h=60Gh/s
.
.
.
k>187



So 187 ASICs (60Gh/s each)  or 11000gh/s must be added to guild pool in order to become a 51% threat assuming that the rest network grows very slowly.

This means that guild pool will reach the threatening point in 3-4 weeks if the situation does not change.