[ANN][BURST] Burst | Efficient HDD Mining | New 1.2.3 Fork block 92000

[bobafett]

depends on the diff. but i think about 1 block per day.

[loonb]

Mates, need help. Am I right, and 2378 burst just dissapeared from pool side?
http://cs622727.vk.me/v622727963/1776b/olzNb0Tubzc.jpg
http://cs622727.vk.me/v622727963/17774/oT3POkLCJTk.jpg

[burstcoin]

Mates, need help. Am I right, and 2378 burst just dissapeared from pool side?
http://cs622727.vk.me/v622727963/1776b/olzNb0Tubzc.jpg
http://cs622727.vk.me/v622727963/17774/oT3POkLCJTk.jpg

Looks like a block explorer error. That transaction and the one below it show as being in blocks that don't seem to exist. My best guess is that the block explorer and that pool didn't update to 1.2.1 until after the fork, the pool tried to payout and it got confirmed on the 1.2.0 fork, the block explorer picked it up, and that doesn't exist now, but the block explorer needs to reparse some stuff to be correct, since some 1.2.0 stuff is still there.

[bensam123]

I'm doing 25TB and not getting a block per day.

Could someone indoctrinate me on exactly how the mining process works? From what I understand, it looks for deadlines, it finds deadlines and if it submits it first you win the block? But if you don't process it fast enough and another block appears, you lose any chance at deadlines in the last block. If that's the case wouldn't an extremely fast processor give you a huge leg up on the competition as you'd be sending in deadlines before anyone else if more then one person has the same deadline?

Another question, if you're plotting with the GPU plotter with 'direct write' so it optimizes and plots at the same time, how much slower does it go then normal plotting? It also looks like it freezes when you start plotting, but it's still writing to the disk, no status text is updated. Does stagger size matter with 'direct write'? If it's properly defraged couldn't you effectively have a stagger size of like 1024 and it wouldn't matter?

I'm using Blagos miner and it seems like on occasion I get "Microsoft Visual C++ Runtime Debug Errors: miner.exe R6010 Abort () Has Been Called" Anyone know what this is or why it's happening? It seems to happen randomly too.

I think I fixed this by installing C++ Runtime 2013, I only had 2008 and 2012 installed.

[burstcoin]

Could someone indoctrinate me on exactly how the mining process works? From what I understand, it looks for deadlines, it finds deadlines and if it submits it first you win the block? But if you don't process it fast enough and another block appears, you lose any chance at deadlines in the last block. If that's the case wouldn't an extremely fast processor give you a huge leg up on the competition as you'd be sending in deadlines before anyone else if more then one person has the same deadline?

Every combination of accountId, nonce, generationsignature, and height will calculate out to a deadline number. The generationsignature and height will be the same for everyone each block. Every block the miner goes through all your plot files calculating the deadlines for each nonce you have plotted, keeping track of the best one(or ones if you use a pool that takes multiple). The deadline is a number of seconds that if that many seconds pass since the previous block without a new block coming out you can announce a block with that accountId and nonce. Faster speed may help you find your best deadline faster, but once it has been found you can only sit and wait to see if someone else had a lower deadline and releases a block first, or if that much time passes and you can release a block. If your setup finds your best deadline very slowly it does put you at a disadvantage, since it opens up the possibility of someone else's deadline coming up and them releasing a block before you finish finding yours, when you might have had a better one.

[bensam123]

Could someone indoctrinate me on exactly how the mining process works? From what I understand, it looks for deadlines, it finds deadlines and if it submits it first you win the block? But if you don't process it fast enough and another block appears, you lose any chance at deadlines in the last block. If that's the case wouldn't an extremely fast processor give you a huge leg up on the competition as you'd be sending in deadlines before anyone else if more then one person has the same deadline?

Every combination of accountId, nonce, generationsignature, and height will calculate out to a deadline number. The generationsignature and height will be the same for everyone each block. Every block the miner goes through all your plot files calculating the deadlines for each nonce you have plotted, keeping track of the best one(or ones if you use a pool that takes multiple). The deadline is a number of seconds that if that many seconds pass since the previous block without a new block coming out you can announce a block with that accountId and nonce. Faster speed may help you find your best deadline faster, but once it has been found you can only sit and wait to see if someone else had a lower deadline and releases a block first, or if that much time passes and you can release a block. If your setup finds your best deadline very slowly it does put you at a disadvantage, since it opens up the possibility of someone else's deadline coming up and them releasing a block before you finish finding yours, when you might have had a better one.

Thanks for the informative answer.

So processing your plots faster only helps if more then one block is waiting to hit the chain?

How would they release a block sooner? That only happens if another block appears on the chain and then has to be processed?

Also is there a timeline on PoC2? Like a few weeks, months?


I also think a GPU assisted miner would most definitely help things out here, at least till PoC2 shows up. I'm looking at having to purchase a ~$150 processor upgrade due to mining times as about half my HD space seems to be being flushed away. It seems as though a quad core is a requirement (hex or oct for AMD).

[muhrohmat]

yes theres a difenece from wallet 1.2.0 anunced in jpg thete and wallet 1.2.1 that is the correct non forked one blcokchain wallet but pools are still capting like the us.burstcoin.io soo i dont mine there for the next 2 day a small fork may have occurred too much low diff.

[pinballdude]

Ok,  So ive Finally ploted about 30tb and im starting to feel like the Pools are not paying me appropriately.  So i would like to start solo mining for a bit to compair the payouts.    How do I get my PoC miner to Read from Multiple drives.  as of now the Poc miner Only mines from the Plot folder that the PoCminer.bat is Included in.


I need it to read from about 7 different harddrives.   

Untill Now i was using Uray's Miner in which i was able to simple add the paths to each plot folder in a .cfg file  How do i do the same with POC miner?

The POC miner only looks at one location for files, so you will have to have one running for each drive, or have to use symbolic links or some other similar tech. to have the files from several drives appear in the same directory. I have one miner per drive, and it works fine, and totally spends all available cpu, reading off all drives at the same time (got a bunch of external drives connected to usb 2 and 3, as well as some internal drives) i use the -Xmx nnnnm  option to start the miner to limit/allocate its memory usage, and it seems how much ram an individual miner uses is dependent on how large the stagger size is.

here's one of my batch files for inspiration (this one is set quite high RAM wise, highest staggered file is 38K stagger) :

Code:

set WHERE=Mining INTENSO4TB#5 on P drive
set MAXRAM=1000m
title %WHERE%
:mine
"%PROGRAMFILES%\Java\jre7\bin\java.exe" -Ddrive="%WHERE%" -Xmx%MAXRAM% -cp pocminer.jar;lib/*;lib/akka/*;lib/jetty/* pocminer.POCMiner mine "http://127.0.0.1:8125"%*
goto mine

[pinballdude]

So processing your plots faster only helps if more then one block is waiting to hit the chain?

(this is for solo mining)

suppose a deadline exist at the end of the last file the miner will mine might. that deadline could be a 20s deadline. You want that deadline found before 20s has gone by, so your wallet can announce it at 20s exactly, otherwise some dude with a 21s deadline might win the block even though you had the better deadline, because you had not found it yet.  If you find a 1s deadline after 30s, it will only win the block if noone found a 29s or below deadline, so you need both to be lucky to have the lowest deadline of all, and also to have read it faster than the next lowest announced.

So the faster you can read all your plots the better - it will get you more blocks, as you will be able to post more deadlines in time for them to be "used"

[mmmaybe]

I would like the PR team to please get their terminology straight about the "smart contracts" as some of the articles I've read are just plain wrong.

For a start please use the term AT - Automated Transaction(s) when referring to the "smart contract" technology (i.e. what flavour of "smart contracts" does Burst use) and a specific use case (or "smart contract") such as the Lottery AT.

Thus "smart contracts" would be the most general term - so you'd say that "our platform has implemented AT for its smart contracts" (vs. say Ethereum).

An "Atomic Cross-Chain Transfer" is just a use case (it is not a technology like AT is) so don't confuse that (treat ACCT just as you do the Lottery).

It would also be appreciated if "created by CIYAM Developers" could be included in the PR material about AT as we really do need to try and get other coins to also adopt AT (so that we can actually start using ACCT ATs) and we have no PR team ourselves.

Thanks, we will be more careful with the terminology! As for credits for AT we always link to your site in the sources, but "created by CIYAM Developers" is of course possible to include in the material write (or get to check before publishing)

When in comes to journalists though we learned the hard way that some doesn't even contact us before publication, despite we end each press release with contact info and "Looking forward to hear from you!". Can't do much there, but hopefully it will get better if we are stricter with the concepts used.

Regard,

[CIYAM]

@mmmaybe - much appreciated - we are in this together and hopefully AT and Burst will be a successful combination (both myself and vbcs are very pleased with how AT has been integrated with Burst).

Looking to making more history with an actual "atomic cross-chain transfer" hopefully happening within the next month!

[mmmaybe]

Burstdev, is it ok if we re-organize the OP a little bit? Make it easier to edit for you and some new resources. If ok, I'll PM you something in like 12h.

Would be nice to mention LuckyAT or the lottery in the subject line but I see that it's already long...

[catbref]

Beta pool at pool.burstcoining.com:8124 is back up. Looks like the wallet was upset due to that v1.2.0 fork so I cleaned out unconfirmed transactions and rescanned the blockchain.

[mmmaybe]

@mmmaybe - much appreciated - we are in this together and hopefully AT and Burst will be a successful combination (both myself and vbcs are very pleased with how AT has been integrated with Burst).

Looking to making more history with an actual "atomic cross-chain transfer" hopefully happening within the next month!

That would be amazing with a ACCT so soon! vbcs and I was talking about trying crowdfunding, and I might have a case. Are you on irc? I'll contact you by email or PM otherwise.

[CIYAM]

Are you on irc? I'll contact you by email or PM otherwise.

I use Skype mostly so you can PM me to get my Skype id.

[catbref]

I'm seeing NXT log entries like this:

Code:

2015-01-27 14:41:42 INFO: tx with id 0 found
2015-01-27 14:41:42 INFO: Transaction to 4697159292272720752 amount 0

Plus I submitted a transaction to my wallet, called getUnconfirmedTransactionIds then getTransaction on that transaction only to be told:

Code:

{\"errorDescription\":\"Unknown transaction\",\"errorCode\":5}

Possible race condition in the wallet software?

[mtwelve]

I've known about burst for  while now, seriously considering a "bursting" rig I guess. Anyway to calculate best bang for buck? Obviously storage capacity is king so like 4tb blacks but what gpu for plotting? Does it matter? Thanks in advanced

The plotter is for Plotting your hard drives,   This is what creates the file you will mine from.  The plotting software you use Has no effect on your mining.   For a while GPU plotter was the fastest way to plot drives  but atm i think Wplotter is King,  Atleast its was for me.

Simple answer to your question,  No.  The plotter you use has no effect on your mining.  Simple the speed of your plotting which can take weeks depending on how much HD space you have to Fill!

So is the way I described above with 4tb hard drives the best way to go?

[vbcs]

I'm seeing NXT log entries like this:

Code:

2015-01-27 14:41:42 INFO: tx with id 0 found
2015-01-27 14:41:42 INFO: Transaction to 4697159292272720752 amount 0

Plus I submitted a transaction to my wallet, called getUnconfirmedTransactionIds then getTransaction on that transaction only to be told:

Code:

{\"errorDescription\":\"Unknown transaction\",\"errorCode\":5}

Possible race condition in the wallet software?

These are AT txs, 4697159292272720752 is account number. When an AT program seeks for a tx after a timestamp and there is none you see that debug message. These outputs will be removed in next releases.

[bensam123]

Another question, if you're plotting with the GPU plotter with 'direct write' so it optimizes and plots at the same time, how much slower does it go then normal plotting? It also looks like it freezes when you start plotting, but it's still writing to the disk, no status text is updated. Does stagger size matter with 'direct write'? If it's properly defraged couldn't you effectively have a stagger size of like 1024 and it wouldn't matter?

Bump on this... It takes so long to plot, a few questions help a lot.

[Merick]

Another question, if you're plotting with the GPU plotter with 'direct write' so it optimizes and plots at the same time, how much slower does it go then normal plotting? It also looks like it freezes when you start plotting, but it's still writing to the disk, no status text is updated. Does stagger size matter with 'direct write'? If it's properly defraged couldn't you effectively have a stagger size of like 1024 and it wouldn't matter?

Bump on this... It takes so long to plot, a few questions help a lot.

https://burstforum.com/index.php?threads/gpu-plot-generator.45/
This thread should answer your questions.

Specific details in the middle of that thread
https://burstforum.com/index.php?threads/gpu-plot-generator.45/page-12#post-3252

Hopefully these conversation can move to the forum and off of this thread for better organization.

Good-Luck