It is pretty much.


listunspent RPC is an incredibly slow thing. It takes maybe 20x more time than the Coin Control to deliver a result. I think a custom RPC call could do better. Load all eligible inputs into a std::map using hash and weight, sort it in a descending order and display with some additional data such as value, date or number of confirmations.

700 coin days in a single input should stake soon. However if this weight is shared between many inputs, there is more to wait. As it has been mentioned before, staking is a stochastic process. In other words, it's a matter of luck and patience. I see a few "unlucky" inputs in my wallet which haven't staked since late May. Those "lucky" have staked 3 times in this time frame.

Rebasing won't help. Most of memory consumption comes from the block index. The chain is over 4 million blocks long and keeps growing fast. Deal with it.

I have committed the automatic reindexing to GitHub: Automatic Reindexing Implemented

Previously, if your block index got corrupted somehow due to a hardware error, OS crash or whatever, you could see a pop-up saying "Error loading block index, see debug.log for details". Not very informative and probably not helpful much. You were left with a broken block index and had either to resync from the genesis block (a matter of a day or two), download a bootstrap.dat and let it ride (a matter of half a day) or get a block chain snapshot somewhere. This trouble is no more. Now, the wallet detects a broken block index and starts reindexing automatically. It isn't a quick process as our block chain is almost 2 million blocks, but 20% faster than bootstrapping. You also waste no traffic as the wallet scans your existing block chain files to rebuild the block index together with the UTXO data base. You can force reindexing any time with the -reindex option.

If anyone can compile it yourself, try this option before it gets into a production release. It works well for me, but more testing is better.

No, the Coin Control works only through the Qt GUI. There was no demand for any command line interface.


If a tribar in the lower right corner is yellow rather than gray, staking works. Mouse over to see the details.

It's definitely possible. There are 192 scratchpad regs for Tahiti. It's also important to keep the ISA size small enough to fit the GCN code caches. I have no RX 480 and cannot optimise for it right now. If you care to share your work, I'm sure some donations and thanks will go your way.

Zpool is an autoswitching pool. NSGminer doesn't like this. It works best with a single chain.

This is an old guide. Coins are eligible for staking after 1 day now, but it doesn't make much difference. There is weight displayed in the Coin Control window. I see my inputs are staking successfully with their weight of 800 on average currently. An input of 20 ORB will not exceed the weight of 300 (20 coins * 15 days). Do your math.

It wasn't optimised for anything. The hard fork was delayed already. The initial OpenCL miner produced something like 150KH/s for an R9 280X. No CUDA miner, it appeared a few months later. It was private OpenCL kernels and fast CUDA miners which hurt NeoScrypt OpenCL mining in 2015.

Slack has been abandoned a year ago when a group of community members tried to take over the development in a somewhat aggressive way. Slack was a rather uncensored battlefield. When the argument was over, the current development team left it.

A status update. I'm trying to add automatic reindexing to the client, so if a block index fails for some reason, the client rebuilds it automatically. No need to download the whole block chain again. Anyway, we are going to have a new release in about 2 weeks, so everyone can update in time before the block #2M.

Got two used R9 280X for cheap. Both with the Elpida/Micron memory (EDW2032BBBG). No surprise they only do about 455KH/s at their default clocks, 1000MHz core / 1500MHz memory. That's because of shitty timings set in the BIOS by greedy and clueless OEMs. Those Radeons with the Hynix AFR memory (H5GQ2H24AFR) produce about 480KH/s at the same clock speeds thanks to AMD providing better tuned memory straps. So I hacked their BIOSes to set correct memory timings. Hash rate improved to 495KH/s at the same clocks or 510KH/s with a core overclock to 1050MHz. +9% performance on memory timings alone, definitely not bad.

Seems so. If there are people out there mining it, don't you think they do it for pure fun?

A light wallet is of no help to the network. Some people run full nodes because they want to support the network. They spend disk space, CPU time and network traffic yet it makes them feel better.

You're welcome to start any threads you like. A good laugh never hurts. I don't care what you do with that coin. The hard fork looks suspicious at the very least, but I don't really care to investigate any further. This thread is about Hobonickels.

4% to 5% interest every week and a multilevel referral programme are clear signs of a fraud.

Not very kind of you to come here to promote your coin after a rather questionable hard fork.

Exclusivecoin / kingscrown / richiela "reboot" scam <BITTREX ENABLED SCAM>

-----

Guys, if you really want to see any HBN improvements, your best bet is to motivate Tranz for the future development. No one will come out of nowhere to work for all of you holding bags and waiting for a pump. Be generous.

Nothing wrong actually on my side. We celebrate Orbitcoin's 3rd birthday and I have forgotten to order a domain name renewal in the process.

Mostly. There are a few smaller exchanges, too.

http://coinmarketcap.com/currencies/feathercoin/#markets

It would probably. Those light weight wallets are dependent on server infrastructure to get block headers and other necessary data from. It means a few more VPS to spin up. In addition, someone has to take care of this wallet's code base and follow the general development. I'm not sure if there is enough demand right now to justify costs and time required for this task.

On the other hand, if our current wallet is ported to Android, it's pretty much a one time effort. No additional servers required and no separate code base to maintain. More or less.