Just-Dice.com : now with added CLAMs : Play or Invest

[infested999]

21:38:33 (1) <dooglus> site will be back soon...

[dooglus]

21:38:33 (1) <dooglus> site will be back soon...

Thanks guys.  I should have posted here.

It's talking a little longer than expected to migrate the db to the new server, but it shouldn't be long now.

Sorry for the inconvenience.

[g83]

was pretty laggy today

[bucktotal]

more ram in computer, more gas in generator, more coffee in dooglus

[infested999]

If you are going to buy any hosting I know a good host.

[dooglus]

was pretty laggy today

We're back.

I think what you're seeing may be the forced bet delays I put in place to slow down dust bets:

< 0.00000001 0.50s
< 0.00000010 0.40s
< 0.00000100 0.30s
< 0.00001000 0.20s
< 0.00010000 0.15s
< 0.00100000 0.10s
< 0.01000000 0.05s
  else       0.00s

ie. if you bet 0, your bet takes half a second.  If you bet single-digit satoshis, it takes 0.4 seconds, etc.

If you bet 0.01 BTC ($1) or more, there's no delay, and if you bet 0.0001 BTC ($0.01) or more, the delay is only 0.15 seconds, so you can place more than 6 $0.01 bets per second.  That seems fast enough to me.  The huge number of tiny bets is becoming a problem (since I made a kinda dumb decision originally to store the bets in the fast datastore, and keep the datastore in RAM, we're running out of RAM!)

I will implement a solution that writes old bets to disk, and then I won't have to worry about running out of RAM, but for the time being I've had to limit the rate at which the 'dust bets' are made.

On the plus side, this should help alleviate the "lag" people have reported where the server stops responding for a few seconds at a time, since it will be less loaded now.

[infested999]

As always, doog handles the situation perfectly!

[dooglus]

As always, doog handles the situation perfectly!

Well, a little advance notice might have been nice, and a post to the forum thread before taking the site down.  But thanks anyway.  I'll be sure to do better next time.

[Dabs]

I think what you're seeing may be the forced bet delays I put in place to slow down dust bets:

Oh. I thought it was just my interwebs slowing down. Well, you will always have more dust bets than anything else, simply because you will have more people have only just dust.

[MPOE-PR]

was pretty laggy today

We're back.

I think what you're seeing may be the forced bet delays I put in place to slow down dust bets:

< 0.00000001 0.50s
< 0.00000010 0.40s
< 0.00000100 0.30s
< 0.00001000 0.20s
< 0.00010000 0.15s
< 0.00100000 0.10s
< 0.01000000 0.05s
  else       0.00s

ie. if you bet 0, your bet takes half a second.  If you bet single-digit satoshis, it takes 0.4 seconds, etc.

If you bet 0.01 BTC ($1) or more, there's no delay, and if you bet 0.0001 BTC ($0.01) or more, the delay is only 0.15 seconds, so you can place more than 6 $0.01 bets per second.  That seems fast enough to me.  The huge number of tiny bets is becoming a problem (since I made a kinda dumb decision originally to store the bets in the fast datastore, and keep the datastore in RAM, we're running out of RAM!)

I will implement a solution that writes old bets to disk, and then I won't have to worry about running out of RAM, but for the time being I've had to limit the rate at which the 'dust bets' are made.

On the plus side, this should help alleviate the "lag" people have reported where the server stops responding for a few seconds at a time, since it will be less loaded now.

Why not just implement martingaling as an option, and run the entire thing on the server as indicated by the player rather than getting called on each bet?

You could do one million bets in less than a second and just report the result.

[Bugpowder]

was pretty laggy today

We're back.

I think what you're seeing may be the forced bet delays I put in place to slow down dust bets:

< 0.00000001 0.50s
< 0.00000010 0.40s
< 0.00000100 0.30s
< 0.00001000 0.20s
< 0.00010000 0.15s
< 0.00100000 0.10s
< 0.01000000 0.05s
  else       0.00s

ie. if you bet 0, your bet takes half a second.  If you bet single-digit satoshis, it takes 0.4 seconds, etc.

If you bet 0.01 BTC ($1) or more, there's no delay, and if you bet 0.0001 BTC ($0.01) or more, the delay is only 0.15 seconds, so you can place more than 6 $0.01 bets per second.  That seems fast enough to me.  The huge number of tiny bets is becoming a problem (since I made a kinda dumb decision originally to store the bets in the fast datastore, and keep the datastore in RAM, we're running out of RAM!)

I will implement a solution that writes old bets to disk, and then I won't have to worry about running out of RAM, but for the time being I've had to limit the rate at which the 'dust bets' are made.

On the plus side, this should help alleviate the "lag" people have reported where the server stops responding for a few seconds at a time, since it will be less loaded now.

Why not just implement martingaling as an option, and run the entire thing on the server as indicated by the player rather than getting called on each bet?

You could do one million bets in less than a second and just report the result.

SPOILER ALERT :
Total bets : 1,000,000
Total BTC wagered : .01483050
BTC remaining : 0

[dhenson]

Why not just implement martingaling as an option, and run the entire thing on the server as indicated by the player rather than getting called on each bet?

You could do one million bets in less than a second and just report the result.

That's an amazing idea.  Present the gambler with a set of parameters and let them tweak away.

[dooglus]

Why not just implement martingaling as an option, and run the entire thing on the server as indicated by the player rather than getting called on each bet?

You could do one million bets in less than a second and just report the result.

Dabs suggested this to me a while ago, in quite some detail.  It's a good idea.

[dooglus]

That's an amazing idea.  Present the gambler with a set of parameters and let them tweak away.

I think that's the entire premise of DiceOnCrack.com.  They martingale so you don't have to.  Only it's on the blockchain like SatoshiDice.

[dooglus]

Over 1500 BTC bet with three clicks of the mouse:

[Rippyzippers]

Over 1500 BTC bet with three clicks of the mouse:

More money than some people make in 4 months, in 3 clicks.

[Bugpowder]

Over 1500 BTC bet with three clicks of the mouse:

More money than some people make in 4 months, in 3 clicks.

More money than the average american makes in 3 years.

[Zaih]

Over 1500 BTC bet with three clicks of the mouse:

More money than some people make in 4 months, in 3 clicks.

More money than the average american makes in 3 years.

I think he was referring to the profit

[PrintMule]

If only.. that bet would fail.

[Moebius327]

If only.. that bet would fail.

soon