Topic: BitShares PTS (formerly ProtoShares) Mandatory Upgrade & Snapshot Announcement - page 52. (Read 218419 times)

I'm running the wallet twice.

Is this correct for my hardware:

2x Xeon 5470 which has 4 cores each.  Total 8 threads (Task Manager, Performance would show 8 threads / cores).
24GB DDR 2
I've run the wallet twice selected 'Mine 2 Processors (1.74GB Required)
Is this correct?  I'm now confused.

When you say 2 instances, do you mean you run it with (that would be two threads) or that you run 2 separate instances of the wallet ?

i hope we get a pool soon, i'd prefer to get a steady income over lottery. only found 4 blocks so far at about 30hpm

Question: why do I see in the log:
Found Hash 00049945caf793d422dccfb56212d35a3161b9b960315a806099c4df3c9e5c94
hash 00049945caf793d422dccfb56212d35a3161b9b960315a806099c4df3c9e5c94
hash2 00049945caf793d422dccfb56212d35a3161b9b960315a806099c4df3c9e5c94
ProtoSharesMiner:
proof-of-work found

but still nothing added in the QT....
(multiple-times occuring in log)

Its because you count HT as normal cores. So dont use HT or use better compiler

Interesting..

Its a Xeon 5470 (which is a bit old) 4 cores per CPU.  24GB DDR.

Its a new install (Windows .  Two instances running.  Should it really be allot more?

Luckybit forget about bitshares , and the simplr principle of interest is misunderstood.

The " lottery" effect is well known , and enters into sociological behavior,  I dont disagree with you Anonymint,  but I doubt you can catch the credit for it.

In the end , lets instead state this plainly :

There is not a massive incentives to rush for that 5k bounty because human behavior states that waiting is potentially more profitable,  if the algorithm has "sudo protection" or the perception,  there may be much larger reward to create a miner and try to monopolize the entity.

Having said that, time and information equals all, the benefits of there being an attempt/attempts will prove or disprove the degree at which the algorithm functions in the market in relation to its competitors.  

Bitshares the idea has proven Adam Smith ultimately correct.

The perception of this idea led to this algorithm.  

Having said that , this is an algorithm,  and just that, I will only state that your " lottry" principal is not fleshed out and missing some aspects.

Specifically I believe the CAS latency, yet they may not be the only factor in the random access latency seen from the CPU:

http://en.wikipedia.org/wiki/CAS_latency

Since the main momentum implementation is so memory intensive, the faster the memory the better the results.
I am testing a configuration with some overclocked memory ATM.

Xeons seem to perform kinda bad, i get 44 hpm on 32 cores (2.2 ghz each) in 32 threads (getting less with 64) thats 1.375 hpm per thread vs 2.5 hpm per thread on an ivy i3

Are you memory limited ? I was getting 2.3 hpm running a single instance on a i3, so this is pretty bad for a dual xeon.

Morning..

I have a question please..

I'm mining 3x Instances and I'm still not peeking 100% usage on 2x Xeon 3.3.  Its on 88%.

1)  Shall I run another instant and peak it?
2)  Or its better to not peak and allow it to work through?

Ta very much.

PS:

Does this look right:

because too many computer is mining it

Hi,
for 4 times when i found a block, client got stuck and block got orphaned.
I had to restart client of course it didnt sync anymore.
It happend on 2 different (64bit) PCs. One with Win7, i5 and 16 GB Ram and another with Win2008R2 XEON and 12 GB RAM.
No other stuff running on it. Also with different setgenerate settings (1, 2 and 3)

Any suggestions to avoid this strange behaviour?

Mining for more than 24 hours an still no valid block found.

lg t.

Ah, ok...now that spunds good
Lol, though it souds like this coin is going to be fully mined within a couple of week
Simmilar to BTC, hashe power increases faster then blockdifficulty

By the way, I found one more block After 49 hours now, I found 2 blocks (with 17 hpm) 

It is not entirely irrational. People spend on hobbies. And they get some coin with this hobby or curiosity. And they might get it anonymously, so it has a value higher than the FX value. Also the future value may justify the ROI. Alas it is a complex dynamic, very difficult to characterize entirely.

It is the best I can think of for PoW; I claim much better than Bitcoin. You know well from our many discussions that I don't believe PoS and variants such as your Decrits can be secure, because I believe they all suffer from the order of the block peer selection not being driven by a random entropy that PoW provides. But I am not claiming that I am omniscient on this, and I wish you the best possible results in proving me wrong. You know well that I respect your intellect.

And note that the mix of ranges of irrationality of the masses is perhaps a gradient in this case, thus it might not be too far from reasonable, thus perhaps one could still obtain virgin coins anonymously from mining by reasonably overpaying.

Also I have suggested that those who are very serious, can install micro-hydropower on a small stream and perhaps they get a reasonable ROI that way.

Please note that the profitability of the initial mining would not be adversely affected by this observation. The mad gold rush would still apply.

I've been mining around 24 hours with 30hpm. Uptill now I find that I've mined around 10 stale blocks and 0 valid block. Maybe it is because of my network connection? I have only 12 connections.

I think a low hanging optimization we can do is to stop current searching once the network has found a new block.

You are banking on the irrationality of the masses. Not necessarily a bad bet, but it is a square peg treatment again that involves wasting energy.


It is less clever than giving that value away without requiring the token of wasted energy.

I already explained upthread why I don't have an incentive to do so. There is an opportunity cost to my time. I would rather spend the time creating a better coin. I will earn much more that way, than consuming my limited time writing a miner for a coin I don't believe in.

Then use your 30+ years of coding and knowledge , build a miner , give it to NO one, mine with it yourself, gain a monopoly and achieve any objective you would want by owning a disproportionate amount of the entity.