 Show Posts
|
|
Pages: [1] 2 »
|
Sorry, my fault. The explorer has not been updated yet. I just randomly learned about the hard fork. |
|
|
|
|
This is crazy talk. Even someone with over 50% can't generate bogus transactions. But he can reverse them for a short period of time. A SHORT period of time. And what's the difference between a PoS attack and a PoW attack? Nothing. Neither is better or worse then the other. PoW miners have "stake" also. After all, they are paying for the hashing power aren't they? And the coins they are getting need to be worth something don't they? YACoin was always intended to be a CPU coin. Now that it has become a CPU coin we want to change it? That's nuts!
|
|
|
|
My grokonet.com servers are off line. Power and Internet service is out due to storms. Update: My power is back on but still no ISP. That wind storm was quite powerful. Update 2: http://explore.grokonet.com is back on line. Whoa. Inflation plunged! Was that the storm or did I miss an N change? Not having any luck hitting it.. and yes, you did miss an N change! Internet has been up and down. It's up again, but who knows for how long. |
|
|
|
My grokonet.com servers are off line. Power and Internet service is out due to storms. Update: My power is back on but still no ISP. That wind storm was quite powerful. Update 2: http://explore.grokonet.com is back on line. Whoa. Inflation plunged! Was that the storm or did I miss an N change? |
|
|
|
But, but but.... I was promised 100% inflation!
Actually, getting 502 bad gateway, so I didn't get to look yet...
My server ran out of memory or something and crashed. Sorry for the kill joy. I shifted things around so hopefully that won't keep happening. It's back up for now. From the charts you can see that back in November inflation was hovering around %300. Now it is hovering around %100. That's a pretty dramatic drop if you ask me. |
|
|
|
Wow I love all the discussion! So much drama! If only we had some way of ... hmmmm ... conceptualizing the inflation rate of YACoin. Wait! I know! A bunch of graphical charts! ... Introducing ... ... Drum Roll .... http://explore.grokonet.com/?inflation=1I present a collection of brand new YACoin inflation charts for all to enjoy. And may all your fears of inflation be tamed. These charts are updated once every minute. Amazing! According to my calculations, YACoin inflation is going down quite dramatically and steadily. After all, YACoin is only 1.3 years old. I also present ... http://explore.grokonet.com/?volume=1Which is a collection of volume charts also updated every minute. I think they give a kind of assurance that people are doing what YACoin was designed to do. That is exchanging currency. Not just mining it. |
|
|
|
Thanks you Groko, too! (txid 56af5a86657badb4730707dd0d51b17b2990a5c132aa8951da7551c080576cf7)
Wow thanks Beave! Your gift feels very generous now that difficulty is over 500! I was thinking of making a graphical chart for my explorer. My top 1000 chart stays up to the minute with new blocks ( http://explore.grokonet.com/?top1000=1), and it even handles a deep block chain reorganize instantly. I was thinking it it might be cool to make an up to the minute graphical chart. What kind of chart should I make? Keep in mind I don't have trade or price data. I just have the block chain and address balances. Any suggestions? |
|
|
|
Sorry, no plans for an API. |
|
|
|
Just a little over 30 minutes before NFactor of 15 becomes our new life for the next few months! Who's as excited as I am?  EDIT: Woohoo, it's here! Miners adapted and no hiccups  The difficulty is too damn high! At least the pump and dumpers will leave us for a while. |
|
|
|
 so, i'm writing the electrum-based block explorer myself now, mostly out of boredom (and also due to lack of volunteers). the code is a major mess and still WIP (eg. no search box and no address handling so far), so consider this an early alpha preview. http://yacoin.unsha.net/PS: i know that i suck hard at writing javascript That looks real nice! I throw my hat into the battle of the block explorers with my brand new super fast address balance look up with transactions: http://explore.grokonet.comHi Groko It's nice to see someone who can still write HTML One can easily add a "smidge" of Javascript to that html that would "ask" for a new block every ~30 seconds or ~minute or what ever time period you like. It could even be "done with AJAX". Where is the server getting its data from, might I ask? Is it an "ABE" YACoin talking to a yacoin daemon on the server? Ron Thanks Ron! I'm glad you appreciate my simple HTML approach. My web page design skills are a bit for wanting that is for sure. And I have never attempted to do anything using JavaScript. But I'm planning on skipping that skill-set with Ruby Rails. My yacoin forum is a Ruby Rails application. As far as my data goes, it looks like you guessed it. My data is coming from a live yacoind executable in server mode that I query using a yacoind executable in client mode doing RPC calls to the server. It's pretty simple really. But g-damn it's fast. I added address balance tracking that keeps up to the minute with every new block and responds to a block chain reorganize. I have a new top 1000 list of addresses that is also kept up to date on the fly. Check it out here: http://explore.grokonet.com/?top1000=1 |
|
|
|
so, i'm writing the electrum-based block explorer myself now, mostly out of boredom (and also due to lack of volunteers). the code is a major mess and still WIP (eg. no search box and no address handling so far), so consider this an early alpha preview. http://yacoin.unsha.net/PS: i know that i suck hard at writing javascript That looks real nice! I throw my hat into the battle of the block explorers with my brand new super fast address balance look up with transactions: http://explore.grokonet.com |
|
|
|
Looks great. What "bogus transactions " for example ? I'm not sure their exact origin. But I suspect they are mostly related to mining (PoS and/or PoW) where transactions that were supposed to be deleted due to orphanage were not deleted. I also suspect I had situations where I had two or more clients with copies of the same wallet making PoS blocks simultaneously. There was also that period of time when PoS mined blocks would get automatically orphaned because of the new trust rules automatically orphaning PoS blocks that follow right after other PoS blocks. Whenever a transaction fails to get deleted it becomes permanently inserted into the wallets cache of transactions. I discovered you can erase all cached transactions without loosing the important stuff like the wallet keys and account names. The block chain is after all the real master record of valid transactions. |
|
|
|
I haven't chimed in for a while. This single thread form of communication is ... so ... how should I put it ... primitive! Hehe. I've been busy with an exciting enhancement to the QT client I hope will get good reviews! It started with a "I wonder if would be possible to...." and then boom. I figured out how to do it. "It" is a new balance column in the transaction view that actually reports the correct balance! It's running great now, so people with their own builds and good backup habits please give it a try and let me know your feedback. Have a look at my post: http://yacointalk.grokonet.com/t/balance-column-and-rescan-revamp/90Check out the pull request: https://github.com/yacoin/yacoin/pull/49Or merge from my branch: https://github.com/grokouser/yacoin/tree/balancesAnd one more thing: YBKF8i9LR7ikHwHUqYsRjbxpFzyo4GpM6w Groko does not pump and dump. Groko asks everybody to JUST SAY NO to pumpin and dumping!!! IJD4sXLZpx3iE1mGb/XF9tdlEH7U7xXkKTKwHw3TZCTUjPW5dOKQE9aqlCIrSt234lb3vtFxiajQOqkrVcYxOLk= LOL. We need to get a spell checker in that message signer. |
|
|
|
Stop the presses! I noticed my POS mining performance fixes aren't in the testing_0.4.3 branch. I know it's a hard change to test, because you have to have a big wallet with lots of ripe POS inputs to see if it really works. So far I'm getting fantastic results performance wise. But I need a few days of non-stop POS minting to be sure I didn't break anything. See my post: http://yacointalk.grokonet.com/t/pos-mining-performance-boost/67 |
|
|
|
Hi all,
I have a question. How can I do POS that is not split into two. Meaning if I have 1,000 YAC it is not split in two 500 YAC with the Stake added to each I see these POS all the time on the blockchain were the address is kept whole and the stake is added to the one address.
I would love to be able to do this, but have no idea if it is a special wallet or what. Help!!!
The idea is to have a lot of stakes competing for new POS blocks. If finding POS blocks doesn't happen often, you can just manually send the outputs to one address after they become spendable again. Yes, you can! But I want "to choose" too. I'd like to have an option in wallet. You set the manual "threshold". After that splitted amount will never be less than "threshold". In another words, output will be splitted only if amount in PoS block greater than 2x"threshold". It does make sense because if you have too many transaction in wallet for PoS you need more powerful CPU. you can find the relevant stake generation code here: https://github.com/yacoin/yacoin/blob/master/src/wallet.cpp#L1376the nStakeSplitAge variable is set to 90 days and it seems like it's pretty important for some reason... (no idea why) next at line #1475 the pos reward gets split into 2 outputs if the stake was younger than nStakeSplitAge (90 days) 0/10 would not recommend fiddling with the nStakeSplitAge variable without knowing exactly WHY that particular value was chosen in the first place (could be just a troll and/or a cool backdoor lol)I have looked at that comment with bewilderment. // The following split & combine thresholds are important to security
// Should not be adjusted if you don't understand the consequences There doesn't seem to be anything on the block validation side to enforce a convention on splitting and combining. Some have probably already figured this out and made personal modifications to the code. I'm already generating about 10 POS blocks a day and each one gives me like 0.2 YAC which seems so pointless. After they mature, I rake up a bunch using coin control and send them to myself. Unfortunately I suppose I have come the conclusion that the whole idea: POS creates security because stakeholders won't want to damage a coin they own. is a bunch rubbish. That is like saying anybody who owns some US dollars would never attack the US government. |
|
|
|
After mhps's comment I dug down into the code and came to the conclusion mhps was right. That is, a 1000 day old chunk of 100 coins will get get 1000 days worth of interest. And there is no limit to the number of days. If you don't open your client for 10 years, and YACoin is still around in 10 years, then you will get 10 years worth of interest.
Thanks for taking time to confirm. Just a comment -- the good thing about giving you 10 years' worth of interest if you open your wallet once in 10 years is that people are attracted by the easy interest and become a long term user of the coin; the bad thing about it is that the savers would tend to open wallet once in a long while hence lending little support to the network through being a POS minting node. There are some relevant discussions here in the contect of PPC https://bitcointalk.org/index.php?topic=127954.msg1360154#msg1360154That is interesting reading. I personally don't think we (YACoin) should force people to keep their clients open all the time. And I am at a complete loss how POS is supposed to be some kind of security utopia, especially when POS it nearly killed us. We neutered POS at block 420000, so It's behaving itself nicely now. What I have to say about security goes for every crypto currency. If you want security, then wait for more confirmations. Job done. If someone was buying your house with YAC, you should probably wait for at least an hour's worth of confirmations before you sign your title over. Selling someone a cup of coffee? An instant unconfirmed transaction will do. |
|
|
|
I am not familiar with YAC POS calculation. I see this line here Each chunk of eligible coins has it's interest calculated based on the chunk's age in the range between 30 and 90 days.
Does it mean for POS generation if I have 100 YACs in the wallet for 1000 days I can only expect 100 * 90* 5% / 365 reward if a POS block is found? In PPC POS calculation the probability of finding a POS block is based on coin age capped by 90 days but the reward value itself is not capped. In the above example I would expect 100 * (1000 - 30) * 1% / 365 reward. I think you are right. I will update the post. Groko - that is a great post on yacointalk. I've been mining YAC since the launch, and I learned a few things from your overview. Thanks! And I believe mhps is right - I think YAC caps POS at 90 days. So a 100 YAC transaction at 1000 days would produce the same interest as at 90 days (100 * 90 / 365 * 0.05 YAC = 24.65 coin years * 0.05 YAC => 24 years * 0.05 = 1.20 YAC gained). Again the guide is very well written, but it maybe helpful to add an example, or two, of how the 5% interest is calculated. Things like the coin-year rounding down to the nearest whole year weren't initiative to me when I originally started. From the example above, I would have expected 1.23 YAC instead of 1.20 - it boggled my mind why all POS were 'rounded' to the lowest 0.05. I figure if we can save someone an hour of sifting through the forums, it will benefit the community. Thanks aso118. I had lots of trouble figuring out what POS was all about, especially with the confusing way the YACoin client has been reporting earnings from POS blocks. I've been working hard to fix how the client reports POS earning. It sounds like a new release of the YACoin client is coming soon. If all goes as planned, the new client will have a transaction record that matches to the last decimal point how much YACoin is actually in your wallet. After mhps's comment I dug down into the code and came to the conclusion mhps was right. That is, a 1000 day old chunk of 100 coins will get get 1000 days worth of interest. And there is no limit to the number of days. If you don't open your client for 10 years, and YACoin is still around in 10 years, then you will get 10 years worth of interest. However, the probability of finding a POS block for those 100 coins started at a low point at 30 days, and increased gradually until the coins got 90 days old. After 90 days the coin's continue to accrue interest, but the probability of finding a POS block for those coins stays at a constant maximum probability from day 90 until day 1000. If a POS block were found on day 1000, I would expect the interest to be (100)*(0.05)*(1000)/(365) or 13.698. |
|
|
|
I am not familiar with YAC POS calculation. I see this line here Each chunk of eligible coins has it's interest calculated based on the chunk's age in the range between 30 and 90 days.
Does it mean for POS generation if I have 100 YACs in the wallet for 1000 days I can only expect 100 * 90* 5% / 365 reward if a POS block is found? In PPC POS calculation the probability of finding a POS block is based on coin age capped by 90 days but the reward value itself is not capped. In the above example I would expect 100 * (1000 - 30) * 1% / 365 reward. I think you are right. I will update the post. |
|
|
|
I made a lot of progress on the stake function for Qt today. I can see why no one did it in the past, it's a real pain  but will be cool when done. Also, forum.yacoin.org is now live. Thanks to Sahtor for doing all the work to get it going again! I know your pain Joe_Bowers. I tried totaling up up all the inputs eligible for staking in CWallet::CreateCoinStake (in wallet.cpp) then printing the result out the log each time it's called. But the total just jumps around randomly. I can't figure out if there is a bug in that function, or I just don't understand how that function goes about finding inputs that should be checked for possible POS block generation. There is supposed to be some unpredictability function so people can't game their POS blocks for some kind of timed attack. Perhaps that's whats making it tricky to understand. Is it possible to run the wallet in a debugger and watch how it works? Sure there is. Follow the instructions at https://github.com/yacoin/yacoin/blob/master/doc/build-msw.txt. At step 26 use Makefile.Debug instead of Makefile.Release c:\yacoin>mingw32-make -f Makefile.Debug
c:\yacoin>cd debug
c:\yacoin\debug>copy ..\release\*.dll .
c:\yacoin\debug>gdb yacoin-qt.exe
(gdb) list CWallet::CreateCoinStake
1373 return CreateTransaction(vecSend, wtxNew, reservekey, nFeeRet, coinControl);
1374 }
1375
1376 // ppcoin: create coin stake transaction
1377 bool CWallet::CreateCoinStake(const CKeyStore& keystore, unsigned int nBits, int64 nSearchInterval, CTransaction
& txNew)
1378 {
1379 // The following split & combine thresholds are important to security
1380 // Should not be adjusted if you don't understand the consequences
1381 static unsigned int nStakeSplitAge = (60 * 60 * 24 * 90);
1382 int64 nCombineThreshold = GetProofOfWorkReward(GetLastBlockIndex(pindexBest, false)->nBits) / 3;
(gdb) b 1381
Breakpoint 1 at 0x491442: file src\wallet.cpp, line 1381.
(gdb) run This will take longer then usual because we are running in debug mode. When it finally fires up you will get: Breakpoint 1,CWallet::CreateCoinstake......
(gdb) list
......
(gdb) n
1384 CBigNum bnTargetPerCoinDay;
(gdb) print nStakeSplitAge
$1 = 7776000
(gdb)
|
|
|
|
|
