Fuzzy Duck

  • Content count

  • Joined

  • Last visited

  • Days Won


Fuzzy Duck last won the day on January 17 2016

Fuzzy Duck had the most liked content!

About Fuzzy Duck

  • Rank
    Active Member

Profile Information

  • Country
    Viet Nam

Recent Profile Visitors

188 profile views
  1. 10 GRC sent
  2. 10 GRC sent.
  3. 10 GRC sent.
  4. 10 GRC sent. Pay it forward.
  5. 10 GRC sent. Pay it forward.
  6. 10 GRC sent. Pay it forward.
  7. 10 GRC Sent. Pay it forward.
  8. I had to download this snapshot to get over the problem with block 1010002. Back in sync now and GRC due shortly.
  9. If GRC ever does hit $1k/GRC I hope you don't mind if I sell some and retire? ;) Personally, I have done nicely (in terms of GRC) out of the two P&D's that I have experienced, in that I have sold GRC at the pumped price and bought back in at the dumped price. Overall, there was little change of the value of my GRC holdings in £/$/Euro terms - did not actually check. BOINC, and subsequently GRC has been a hobby since 2000, when I first joined Seti@home running the command line (anyone remember that?). GRC is primarily a way of keeping score. However, when a gold plated P&D comes along, I cannot resist increasing my GRC holding as described above. Of course it is fun to trade on polo exchange, but I don't think I will ever be the next Axelrod managing my billions ;)
  10. Well I certainly did very nicely out of this P&D in GRC terms. Made a 50% profit.
  11. A quick look on Reddit confirms that this is likely a P&D.
  12. OK. What has happened to drive up the price?
  13. As a last resort, I reinstalled the wallet as a new user and sent the coins to my new wallet. Now just waiting for coins to mature.
  14. I have edited my config file to try and fix this problem without success. Beacon status as follows (note private key is OK). Magnitude (As of last superblock)" : 0.00000000, "Warning" : "Your magnitude is 0 as of the last superblock: this may keep you from staking POR blocks.", "Errors" : "Public Key Missing. ", "Help" : "Note: If your beacon is missing its public key, or is not in the chain, you may try: execute advertisebeacon.", "Configuration Status" : "FAIL" When I try and advertise a new beacon from an unlocked wallet I get the following error: ommand" : "advertisebeacon" "Result" : "FAIL", "CPID" : "", "Message" : "", "Errors" : "Error: Unable to send beacon::Wallet Locked::Please enter the wallet passphrase with walletpassphrase first.", "FAILURE" : "Note: if your wallet is locked this command will fail; to solve that unlock the wallet: 'walletpassphrase <yourpassword> <240>'."
  15. I agree that team Gridcoin could still be a target of team poaching accusations if we somehow prefer team Gridcoiners above other BOINCers. We should treat everyone equal. I agree that a GRC tag in the username would be nice free advertisement to spread word about team-removal and Gridcoin in general. I already have [Gridcoin] in my username btw. I disagree that "potential future clients" would care about boincstats. Such clients would need to know exaclty, how much computation power THEY would get from our network because in any case, economic actors have deadlines and money to earn. Boincstats only shows, how much has been done in OTHER projects. Therefore, such clients would require a testing phase or we would need to guarantee a minimum amount of computation power. Well, a client could just ask to be added to the whitelist and see, how many people jump onto their project to have an idea of available computation power, maybe this client could even use project rain to foster the miners to crunch his project. As an active way to solve this problem, we will need a kind of computation power market, maybe even in the GRC client itself. A user would offer his computation power (maybe evaluated by benchmarks) or just set a flag "If I am online and there is a commercial project needing my CPU/GPU, I will stop every but commercial projects". To be able to do that, I assume we eventually need to integrate BOINC into the GRC client and probably we also need a BOINC poject where such commercial partners would upload their files to be computed. Quez If we could combine GRC and BOINC clients, I think GRC would fly. Question is, how likely is that to happen?