Sparkle Alpha Test Network Live: Mining to be Honored in Real Network

[myhometalk]

under current difficulty , it is very hard to find a block .

[Frodo]

can mine on windows?

AFAIK there are no windows binaries available yet. So you have to compile for yourself.

[infovortice2013]

can mine on windows?

[AtomicBounce]

So the blocks are coming in really slow right now. What about doing another hard fork to allow the difficulty to normalize again?

To achieve that, it would be sufficient to put the logic that sets the max rate of difficulty changes in an if block:

Code: [Select]

if( block_data.block_num != 20080 ) {
    // set max rate of change
    ...
}

https://github.com/sparkle5/Sparkle/blob/f03b4e3d5ec6d2d18c28ed353de5fb5965c5e461/libraries/blockchain/chain_database.cpp#L732-L736

[turtletastic2]

I've noticed that the new commit doesn't even check the difficulty every 20 blocks for blocks < 20000 anymore. I was thinking, I might as well take us to block no. 20000 on my own and do something good with all those blocks instead of giving them to some idiot. So, either we don't honor these blocks at all, or we do it: In that case, I don't want them and I would use them for project bounties (starting services, building exchanges, etc.).

For the record: It wasn't me who exploited it in first place. But I got tired of waiting until we hit 20000 or restart.

Btw, difficulty is at 3342778 and we havn't found a single block after number 20000.

yeah seems stalled out, been over 30 mins now with no block 20001, with increasing diff to 3409633. I'm starting to think the new chain may be needed.

Thats what I wanted to show by taking us to 20000 quicker
Starting the new chain today would be nice, as most people would have the chance and time to recompile and restart it on a sunday instead of a work day.

The problem is, you've created a fork that isn't valid for all the clients still on alpha.5 so they ignore your branch.
I think there needs to be a change in alpha.7 to put back the 20th block difficulty check for blocks under 20000 (and of course keep the difficulty check on every block as in alpha.6)

Hard fork has been made and I was just using the new rules. Forking again the way you described it would not help at all, it would just give the original attacker those last 500-600 blocks I've mined. Either way, we're stuck at some irrationaly high difficulty and block 20000, it is probably wiser to just start a new chain, like the last time we got stuck with a high difficulty.

[bitder]

I've noticed that the new commit doesn't even check the difficulty every 20 blocks for blocks < 20000 anymore. I was thinking, I might as well take us to block no. 20000 on my own and do something good with all those blocks instead of giving them to some idiot. So, either we don't honor these blocks at all, or we do it: In that case, I don't want them and I would use them for project bounties (starting services, building exchanges, etc.).

For the record: It wasn't me who exploited it in first place. But I got tired of waiting until we hit 20000 or restart.

Btw, difficulty is at 3342778 and we havn't found a single block after number 20000.

yeah seems stalled out, been over 30 mins now with no block 20001, with increasing diff to 3409633. I'm starting to think the new chain may be needed.

Thats what I wanted to show by taking us to 20000 quicker
Starting the new chain today would be nice, as most people would have the chance and time to recompile and restart it on a sunday instead of a work day.

The problem is, you've created a fork that isn't valid for all the clients still on alpha.5 so they ignore your branch.
I think there needs to be a change in alpha.7 to put back the 20th block difficulty check for blocks under 20000 (and of course keep the difficulty check on every block as in alpha.6)

[turtletastic2]

I've noticed that the new commit doesn't even check the difficulty every 20 blocks for blocks < 20000 anymore. I was thinking, I might as well take us to block no. 20000 on my own and do something good with all those blocks instead of giving them to some idiot. So, either we don't honor these blocks at all, or we do it: In that case, I don't want them and I would use them for project bounties (starting services, building exchanges, etc.).

For the record: It wasn't me who exploited it in first place. But I got tired of waiting until we hit 20000 or restart.

Btw, difficulty is at 3342778 and we havn't found a single block after number 20000.

yeah seems stalled out, been over 30 mins now with no block 20001, with increasing diff to 3409633. I'm starting to think the new chain may be needed.

Thats what I wanted to show by taking us to 20000 quicker
Starting the new chain today would be nice, as most people would have the chance and time to recompile and restart it on a sunday instead of a work day.

[JWF]

I've noticed that the new commit doesn't even check the difficulty every 20 blocks for blocks < 20000 anymore. I was thinking, I might as well take us to block no. 20000 on my own and do something good with all those blocks instead of giving them to some idiot. So, either we don't honor these blocks at all, or we do it: In that case, I don't want them and I would use them for project bounties (starting services, building exchanges, etc.).

For the record: It wasn't me who exploited it in first place. But I got tired of waiting until we hit 20000 or restart.

Btw, difficulty is at 3342778 and we havn't found a single block after number 20000.

yeah seems stalled out, been over 30 mins now with no block 20001, with increasing diff to 3409633. I'm starting to think the new chain may be needed.

[turtletastic2]

I've noticed that the new commit doesn't even check the difficulty every 20 blocks for blocks < 20000 anymore. I was thinking, I might as well take us to block no. 20000 on my own and do something good with all those blocks instead of giving them to some idiot. So, either we don't honor these blocks at all, or we do it: In that case, I don't want them and I would use them for project bounties (starting services, building exchanges, etc.).

For the record: It wasn't me who exploited it in first place. But I got tired of waiting until we hit 20000 or restart.

Btw, difficulty is at 3342778 and we havn't found a single block after number 20000.

[monsterer]

The fix has been checked in for everyone compiling from source.  The mac binaries are coming soon.

Windoze plz

[AtomicBounce]

Thanks for your fast reaction. I also think that starting a new chain would help to quickly get back to a properly functioning chain.

[turtletastic2]

Yep, you're right Frodo for the fast reaction.

I'm also for starting a new chain. Since we have to wait every 20 blocks until we can proceed and diff stands at 1317968 now, it will probably take longer than Dec 14 to sort this out. And that waiting time would make the shares of this attacker even more valuable.

Might also make sense to slightly raise the difficulty amount that can be changed downwards every 20 blocks, e.g. from 2% to 5% or 10%, so that we can adjust quicker downwards than upwards.

[Frodo]

  for fast fix.

Will all blocks be honored? It makes sense as it is incentive to find exploits but then again it is probably a pretty high pay 

Regarding the high difficulty it would probably be the best to start a new chain.

[turtletastic2]

Issue might be that miners turn their miners off, because they don't earn the same amount anymore with their hash power and would wait until this issue is over... it'll probably take a lot longer than 3 days until we have regular 20s blocks. I'm turning my miners off until this is over.

[turtletastic2]

Diff was at ~100000 before it started and is now over 1mill. Hard to say what it will be when we hit 20000 blocks, but reverting back from 1mill to 100000 will take us:

wolframalpha.com/input/?i=1000000+*+0.98^x+%3D+100000

114 diff adjustments, which come every 20 blocks. Means it will take 2280 blocks.

I'd ball park the avg block time to be 5x higher over these 2280 blocks, so it will take us 100 seconds * 2280, to get there, i.e. ~
3 days. Not too bad actually, but your call on how to proceed.