Is it better to...

[Celroc Amaul]

... try to mine coins using a lower GHz/second speed but mine for a longer time?

Or

... try to mine using a higher GHz/second speed for a short burst?

In short, I set up my own P2Pool node and am considering buying some mining contracts to have people mine for me there.  I'm not entirely sure if I should go for the one-day contracts at a higher GHz or perhaps the one-week ones at a lower GHz?  I'm not expecting a ROI on this, but am just more curious as to which will be more likely to find shares and get coins.

Thanks!

Celroc

[byt411]

In the long term, it would be the same.

1TH/s for 24 hours = 24TH/s for 1 hour. No difference.

[DrG]

In the long term, it would be the same.

1TH/s for 24 hours = 24TH/s for 1 hour. No difference.

There is only no difference if there is not difficulty change in the coin being mined.  Consider than if somebody could mine with 1 Petahash for a day right now versus mining with ~ 3 TH for the next year, the 1PH today would make tons more.

[philipma1957]

In the long term, it would be the same.

1TH/s for 24 hours = 24TH/s for 1 hour. No difference.

There is only no difference if there is not difficulty change in the coin being mined.  Consider than if somebody could mine with 1 Petahash for a day right now versus mining with ~ 3 TH for the next year, the 1PH today would make tons more.

yes this is true.  if he keeps 10 th for tues the 22nd of july vs 10th  at 1 th a day for the next ten days  he is better off.  since there will be a diff change in 4 or 5 days.

and the diff change is not a negative one.

Another factor is uptime on your server.   the chance of a server crashing for 1 day in a row is higher then the chance for a server crashing for 10 days in a row.

So it is less likely his node has 100% down time if he mines 10  days.
 it is more likely his server has 100% down time if he mines for a day.

that factor would flatten out over time.  but on a 1 shot deal the extra risk does exist  with the shorter time.

[jjc326]

In the long term, it would be the same.

1TH/s for 24 hours = 24TH/s for 1 hour. No difference.

There is only no difference if there is not difficulty change in the coin being mined.  Consider than if somebody could mine with 1 Petahash for a day right now versus mining with ~ 3 TH for the next year, the 1PH today would make tons more.

yes this is true.  if he keeps 10 th for tues the 22nd of july vs 10th  at 1 th a day for the next ten days  he is better off.  since there will be a diff change in 4 or 5 days.

and the diff change is not a negative one.

Another factor is uptime on your server.   the chance of a server crashing for 1 day in a row is higher then the chance for a server crashing for 10 days in a row.

So it is less likely his node has 100% down time if he mines 10  days.
 it is more likely his server has 100% down time if he mines for a day.

that factor would flatten out over time.  but on a 1 shot deal the extra risk does exist  with the shorter time.

Right, OP should check out variance.  it used to drive me crazy when I mined only overnight on certain pools.  I swear they would find more blocks during the day...it should even out over time but some nights I went all night and they found no blocks so I had just wasted electricity and time.  I think the slower over a longer period makes more sense, as long as you're talking a short time frame, because you could lose out due to increase in difficulty.

[DrG]

In the long term, it would be the same.

1TH/s for 24 hours = 24TH/s for 1 hour. No difference.

There is only no difference if there is not difficulty change in the coin being mined.  Consider than if somebody could mine with 1 Petahash for a day right now versus mining with ~ 3 TH for the next year, the 1PH today would make tons more.

yes this is true.  if he keeps 10 th for tues the 22nd of july vs 10th  at 1 th a day for the next ten days  he is better off.  since there will be a diff change in 4 or 5 days.

and the diff change is not a negative one.

Another factor is uptime on your server.   the chance of a server crashing for 1 day in a row is higher then the chance for a server crashing for 10 days in a row.

So it is less likely his node has 100% down time if he mines 10  days.
 it is more likely his server has 100% down time if he mines for a day.

that factor would flatten out over time.  but on a 1 shot deal the extra risk does exist  with the shorter time.

Right, OP should check out variance.  it used to drive me crazy when I mined only overnight on certain pools.  I swear they would find more blocks during the day...it should even out over time but some nights I went all night and they found no blocks so I had just wasted electricity and time.  I think the slower over a longer period makes more sense, as long as you're talking a short time frame, because you could lose out due to increase in difficulty.

You could always use the load balancing features of CGMiner and submit to multiple pools and reduce you variance risk that way.  Either way the option to spread out mining depends on how much the future difficulty increases will be - seems to be about 10% this round, effectively almost a 10% drop earnings with neutral luck.

[Celroc Amaul]

Thank you all for your replies.  I think I understand a bit better now