Topic: BitShares PTS (formerly ProtoShares) Mandatory Upgrade & Snapshot Announcement - page 57. (Read 218428 times)

Anyone on here using setgenerate 4 and found blocks? I am starting to wonder if setgenerate 4 cant find any blocks, which might be the reason I only found one block so far (and its good possibel I had setgenerate 3 when I found it).

I have three dedicated servers out at third party datacentres on good bandwidth with eight cores each, or maybe four real cores appearing as two each; each such box gets two or more times as many hashes per timespan as any of my machines here at home on my much less bandwidth; home is Halifax NS Canada, third party machines are around Germany or Denmark type of region. (Forta Trust dedicated servers.)

Not one of the third party machines has a block, I have been running them since an hour or two after mining started. Getting my home machines running took a little longer but some of them found blocks almost as soon as I got them running, one of them has found three blocks and one orphan block and two more blocks.

So either variance is skewing this or merely having high hashing power on good bandwidth isn't all it takes to get a block?

The machines at home are mostly core 2 duo type of things, all only have two cores. (Old Dell  755's for example with 8 or 4 gigs of RAM). A couple have 16 gigs of RAM. The third party machines might only have four real cores with two simulated cores/threads/whatever each. They are using setgenerate true 7 so as to leave a thread/core for what the boxes normally are used for. Two of them have 16 gigs of RAM one has 24. Could their hyperthreads or whatever it is they have two of per core be interfering with each other or something ending up making the machines unable to find any blocks at all or something?

-MarkM-

My eMu would be on harder. If do droplets cant hash till the retarget, whats gonna happen when they all come online?

also, if you are still looking to buy some pm me for I have a few.

Since difficulty will be higher it will be harder.
You will not even get stales.

So, wht do you guys think, will it get easyer (due to not geting stalles) after the retarget or harder to mine ?
Alrady now I am mining 37 hours with 17 hpm and found only one block, hope it will get easyer

Current block 3790, retarget at 4032, +- 2-3 hours, will be interesting 

How far from retarget?

How if all have the same name?

./bitcoind getmininginfo

i dont know how to specify the other instances

selling 200 PTS for 1.2 BTC =0.006 each

Have also posted this on http://bitsharestalk.org/

you can use the same command line you used to start each instace and just add getmininginfo too the end of it.

I believe the stales will only be a problem until the difficulty adjusts later today.  Blocks are supposed to be 5 minutes but the hashing power in the network has them turning every 30-60 seconds right now, thus the stales.

Can someone tell me how to check mining speed of additional instances running on linux? I have managed to open 3 on my server but dont know how to access them now

Here's how it looks.

Is there anything to be done about the stales? So far I have found 6 blocks, but 4 of them have been stale.

only reason i noticed was becuase i caught a proof-of-work in the logs with a running tail and no orphan in transactions.grep is your friend 

Yes, I have a average of 260 -320 (ping) from here to US or EU

WTS 192 ProtoShares  for ~0.8B

What do you mean with high ping? 250ms? like on 3G/4G connection?

I have a questions, dose a bad internet conection (high ping) influence the mining rate? I have 17 hpm and have only found 1 block in the las 34 hours.
If I looke into the debug log I see some "Accseptet" blocks, I guss they are stalled blocks...
So is it possibel that I am not geting any blocks becours of my internet?

- hash nonce space into birthdays space in parallel, get array of (nonce, birthday)
- sort it with bitonic
- find dublicates
Bonus: don't store birthdays at all, calculate them ad hoc during the sort
This is not an attack on the algorithm itself, it's just a way to fit existing algorithm into GPU

WTS 150 ProtoShares  for ~1BTC