Cryptocoin Mining?

[Attic]

Good info here. The decentralization theme is important and achieving it is a work in progress.

Will look more into Proof of Stake Coins.

There is responsibility in miners as well, to help decentralize as they are able and to not corner the market.

I'm convinced that the conundrum here is that decentralization means leaving much up to the individual, but when much is left to the individual, that inevitably a few individuals in a group will be entirely and wholly corrupt. So without centralized control, the need for some form of centralized control arises. We can look to the maleabilty bug in BTC, Mt Gox itself as indications of corruption that does not police itself well. Recently at wafflepool there were a few individuals gaming the reject rate to effectively flood the pool with shares that should have been rejected but weren't. How to solve making corruption hurt the corrupter rather than those in corruptions line of sight?

Proof of stake has an interesting take on this. The anandtech trading forums are somewhat of a decentralized place to do business and trade and it manages itself fairly well I think, but it still needs mods to enforce folks who try to game the system.

It appears that any crypto currency and/or pool can be gamed in someway and a need for centralized control arises, but centralized control centralizes power and we already have this with the government and the US dollar and the Fed.

 

[slashbinslashbash]

Good info here. The decentralization theme is important and achieving it is a work in progress.

Will look more into Proof of Stake Coins.

There is responsibility in miners as well, to help decentralize as they are able and to not corner the market.

I'm convinced that the conundrum here is that decentralization means leaving much up to the individual, but when much is left to the individual, that inevitably a few individuals in a group will be entirely and wholly corrupt. So without centralized control, the need for some form of centralized control arises. We can look to the maleabilty bug in BTC, Mt Gox itself as indications of corruption that does not police itself well. Recently at wafflepool there were a few individuals gaming the reject rate to effectively flood the pool with shares that should have been rejected but weren't. How to solve making corruption hurt the corrupter rather than those in corruptions line of sight?

Proof of stake has an interesting take on this. The anandtech trading forums are somewhat of a decentralized place to do business and trade and it manages itself fairly well I think, but it still needs mods to enforce folks who try to game the system.

It appears that any crypto currency and/or pool can be gamed in someway and a need for centralized control arises, but centralized control centralizes power and we already have this with the government and the US dollar and the Fed.

Well, the end goal is to perfect the code sufficiently that people are unable to cheat each other. I can't remember where, but I saw either a BitShares or a MasterCoin video recently where the developer talked about how in the physical world, a central authority was required in order to establish trust and provide justice, to "make things right" when somebody found a way to cheat the system or simply not live up to the terms of a contract. The new DEX systems (distributed exchanges) which are a big part of BitShares/Mastercoin/Counterparty/etc. are designed to allow "Smart Contracts" where the software/network does the enforcing. Of course these will start off very simple, but there are plans for scripting languages, etc. which allow complicated contracts to be created and enforced.

Of course there are always flaws in software, but by conducting rigorous testing, making everything open-source and responding quickly when flaws are found, the developers are doing the best they can to make this a reality.

You can modify the Bitcoin code to create an invalid transaction, or spend somebody else's BTC, but that transaction will be rejected by all non-corrupt nodes on the network immediately because they will recognize it as invalid.

So really they are trying to replace the central authority with lines of code that enforce the rules. This idea is the central theme of Bitcoin and all all Bitcoin-derived cryptocurrencies (although the "rules" are very simple and straightforward). And so far, it seems to be working ok. There have been setbacks, but things are coming along.... and the setbacks have mostly been with 3rd-party, closed-source code.

EDIT: If you think about it, MMORPG's such as WoW and EVE already have some version of this, and have for some time. Of course there is a company that runs the game, and can reset things or ban users if somebody cheats, but for the most part, the software itself is trusted to keep people on the straight and narrow.

 

[geokilla]

http://headlines.yahoo.co.jp/hl?a=20140321-00000038-yom-sci

http://www.reddit.com/r/Bitcoin/comments/20x0ww/mtgox_announces_they_actually_still_owned_200000/

Translation.
Bitcoin exchange Mt.Gox which collapsed in February announced on 20th that they have found they're still in possession of the 200,000 BTC out of the 850,000 BTC that was reported to be lost.
According to the lawyer, they found them on the 7th of this month by searching through a storage on the internet called a "wallet" which was being used by MtGox up until June 2011.
MtGox has reported that they have lost almost all of the 850,000 BTC owned while filing for bankruptcy protection on February 28th.
Edit : Little more clarity on the second paragraph.

 

[x3sphere]

Myriad seems interesting, 5 different algorithms each with their own independent difficulty. So it doesn't cut out ASIC miners but doesn't give them an edge, either.

May be a good buy, price has fallen a lot, but at #79 on Coinmarketcap I'd say this coin is very undervalued. It is trying something new after all and I don't see any disadvantages to the fair mining approach. Decided to pick some up will be interesting to see where it's at a month from now.

 

[Virge_]

Hmm.. calling all Scrypt-N miners! I just rebuilt a slew of USB sticks in anticipation that I might need them ready for when some of the more popular Scrypt coins fork to -N. I played around with Vertcoin a month ago but rebuilt these using 1.5.2 with vertminer built in as a toggle in the config.

I've managed to get optimal configurations to get 290's running 420-440khash and 290x's running 480-500khash, but I've noticed the cards seem to run even hotter than when mining regular Scrypt.. when a month ago I recall the cards running about MUCH cooler when running Scrypt-N. Has something changed or did I misconfigure?

 

[KlokWyze]

Anyone here interested in 2 spare PCI-X powered 1x - 16x USB risers I've got?

 

[Attic]

Hmm.. calling all Scrypt-N miners! I just rebuilt a slew of USB sticks in anticipation that I might need them ready for when some of the more popular Scrypt coins fork to -N. I played around with Vertcoin a month ago but rebuilt these using 1.5.2 with vertminer built in as a toggle in the config.

I've managed to get optimal configurations to get 290's running 420-440khash and 290x's running 480-500khash, but I've noticed the cards seem to run even hotter than when mining regular Scrypt.. when a month ago I recall the cards running about MUCH cooler when running Scrypt-N. Has something changed or did I misconfigure?

I've been 60/40 split of mining power to scrypt vs scrypt N and have been using vertminer 0.5.3 for awhile. There is an sgminer out there that will run either scrypt or scrypt n by changing one line in the conf file, "nfactor" : 10 for scrypt, 11 for n scrypt. I haven't played with this.

IME running N scrypt runs cards harder than scrypt with all else being the same. This is one of the drawbacks of scrypt N.

But, N scrypt runs better with cards not pushed as hard as when running scrypt. So lowered clocks may be worth trying vs optimal clocks when running scrypt. I have had to add more volts to my cards to get Scrypt N stable vs all else the same Scrypt configs.

What are your settings for 290's and 290x's for n scrypt? I'm tinkering now, with ~900c and 1250m and various TC.

 

[Virge_]

What are your settings for 290's and 290x's for n scrypt? I'm tinkering now, with ~900c and 1250m and various TC.

• Reference R9 290 XFX - 1010/1500@24700 nets consistent 440+khash a card across the board. These are tanks, love 'em.
• "DD" R9 290 XFX - 947/1500@24700 for one of the mem's, 1000/1250 for the other for 420-440khash. Huge PITA. See below.
• Reference R9 290x XFX - 1025/1500@24700 for 480+khash. A good 2/3rds of these cards run at R9 290 speeds. What a rip.
• Sapphire Tri-X OC: 1000/1375@32765 for 500+khash. Amazing all-arounder.. perfect for Scrypt and Scrypt-N.

For the record, the "DD" model XFX are a HUGE pain in the ass. They only ever run 947/1500 after they've baked in @ 947/1250 for a good hour or so and then you can switch them to 1500 - try rebooting though and you're fucked until you step them all back down to 1250 and then bring them up to 1500 again one at a time. Given I own about 60 of these I wish I had just gone with all Sapphire Tri-X OC's from the get-go, those things are beastly and run flawlessly.

 

[wand3r3r]

Oh that sucks! My 2 best alternatives are appearing to have some negative feedback (290 gaming and 290 DD). Actually it's been mostly the 270/x gaming which I've been seeing the negativity about the fans.

 

[wand3r3r]

290x owners.

I thought I'd mention this since my Hynix card was acting funky, likely due to changing it to a 16x-16x riser. (Maybe it doesn't get enough power through the riser? Idk)

My Hynix and Elpida cards get identical hashrates ~860/ea with these clocks:
Hynix 900/1250 = 860 kHash
Elpida 900/1500 = 860 kHash

Oddly enough the TC is like this
Hynix TC is 32765
Elpida TC is 27400

Oddly the Hynix memory latency is clearly optimized at 900/1250 ratio, but the hashrate drops to 800 if I use 900/1500. If I use 1000/1500 it's 9xx but my stability issues got me tinkering to discover this.

The elpida card loves 900 core, and 1250 memory is around 800k while 1500 memory is 860k. It also chokes on >30000 TC, and 1000 core is slower than 900 core.

Just thought these little ratios could help others optimize their cards. The 900/1250 was new since I've only used the Hynix card at 1000core.

 

[Attic]

Great info, thanks guys.

 

[Mark R]

Well. It's all over for me. At $0.25 per kWh, and the end of the heating season, it's just no longer viable.

The 7950s have either gone in the trash (dead ones), or have been sold (half-dead or still working - someone paid $100 for this one).

The 280Xs have been moved into gaming duty, and the incument 7850 is being sold.

I don't know what I'm going to do with the CPU/Mobo/RAM combos. They're Core i3 Haswells + 8 GB, so they're pretty decent. One will probably be moved into NAS duty, and the others will probably be given as gifts.

 

[IEC]

Mine have been moved to the garage in open racks for cooling similar to Virge_'s setup. It works well and keeps the heat outside of the house proper.

 

[IEC]

• Reference R9 290 XFX - 1010/1500@24700 nets consistent 440+khash a card across the board. These are tanks, love 'em.
• "DD" R9 290 XFX - 947/1500@24700 for one of the mem's, 1000/1250 for the other for 420-440khash. Huge PITA. See below.
• Reference R9 290x XFX - 1025/1500@24700 for 480+khash. A good 2/3rds of these cards run at R9 290 speeds. What a rip.
• Sapphire Tri-X OC: 1000/1375@32765 for 500+khash. Amazing all-arounder.. perfect for Scrypt and Scrypt-N.

For the record, the "DD" model XFX are a HUGE pain in the ass. They only ever run 947/1500 after they've baked in @ 947/1250 for a good hour or so and then you can switch them to 1500 - try rebooting though and you're fucked until you step them all back down to 1250 and then bring them up to 1500 again one at a time. Given I own about 60 of these I wish I had just gone with all Sapphire Tri-X OC's from the get-go, those things are beastly and run flawlessly.

Are you using the stock BIOSes or Stilt BIOSes?

 

[Brekyrself]

Well, the end goal is to perfect the code sufficiently that people are unable to cheat each other. I can't remember where, but I saw either a BitShares or a MasterCoin video recently where the developer talked about how in the physical world, a central authority was required in order to establish trust and provide justice, to "make things right" when somebody found a way to cheat the system or simply not live up to the terms of a contract. The new DEX systems (distributed exchanges) which are a big part of BitShares/Mastercoin/Counterparty/etc. are designed to allow "Smart Contracts" where the software/network does the enforcing. Of course these will start off very simple, but there are plans for scripting languages, etc. which allow complicated contracts to be created and enforced.

Of course there are always flaws in software, but by conducting rigorous testing, making everything open-source and responding quickly when flaws are found, the developers are doing the best they can to make this a reality.

You can modify the Bitcoin code to create an invalid transaction, or spend somebody else's BTC, but that transaction will be rejected by all non-corrupt nodes on the network immediately because they will recognize it as invalid.

So really they are trying to replace the central authority with lines of code that enforce the rules. This idea is the central theme of Bitcoin and all all Bitcoin-derived cryptocurrencies (although the "rules" are very simple and straightforward). And so far, it seems to be working ok. There have been setbacks, but things are coming along.... and the setbacks have mostly been with 3rd-party, closed-source code.

EDIT: If you think about it, MMORPG's such as WoW and EVE already have some version of this, and have for some time. Of course there is a company that runs the game, and can reset things or ban users if somebody cheats, but for the most part, the software itself is trusted to keep people on the straight and narrow.

Bitshares along with their Keyhotee product can really move us into a true decentralized system.

I3's product list is truly amazing and can't wait for these to be launched. BitShares-PTS are something everyone should hold onto for crypto diversification.

 

[njdevilsfan87]

Anyone who wants to throw some some hashes to help test out a new kind of P2P node I am working on, I appreciate it : http://www.cryptoalts.com/index.php
(just make sure you have some failovers set in case I need to take it down temporarily for updates)

Once Mintpal releases their trading API, I'll be able to fully automate the payments, which will make future deployed nodes as set and forget (like regular P2P nodes).

 

[geokilla]

Does anyone know how much BTC ~650KH/s gets you on Scyrptguild? They don't have the estimate calculator so I have no clue. Right now, I get anywhere from 0.004 to 0.006 BTC mining and selling Tenfivecoin. I think.

 

[Erenhardt]

Damn vert coin goes vertical again. Would love to get some help with the increasing difficulty at Sunny's pool: vert.poolminewith.us

 

[Zargon]

I could probably swing my last card that's mining MMC to that later but its just a 5870

 

[IEC]

Stick a fork in it. Scrypt mining is effectively dead for the majority of people at current rates.

You should be questioning your ROI and whether your few $/mH per day is worth the time, effort, and heat.

 

[AyashiKaibutsu]

Electricity is included in my rent so my calculation is pretty easy ^.^

 

[chimaxi83]

Stick a fork in it. Scrypt mining is effectively dead for the majority of people at current rates.

You should be questioning your ROI and whether your few $/mH per day is worth the time, effort, and heat.

Yeah, I don't really bother mining anymore. I just buy BTC if I need to on Coinbase and buy coins directly. I just wish I had more time to play around making tens of satoshi

 

[Virge_]

I'd be incredibly surprised if DOGE didn't fork to Scrypt-N or Scrypt Jane. It's the only way they're going to prevent failing into oblivion vs, Litecoin.

 

[taltamir]

Does anyone know how much BTC ~650KH/s gets you on Scyrptguild? They don't have the estimate calculator so I have no clue. Right now, I get anywhere from 0.004 to 0.006 BTC mining and selling Tenfivecoin. I think.

my 650 kh/s doing 24/7 mining on them got me 0.01 BTC on the following timestamps
2014-03-21 22:30
2014-03-18 10:20
2014-03-15 08:20

 

[geokilla]

Stick a fork in it. Scrypt mining is effectively dead for the majority of people at current rates.

You should be questioning your ROI and whether your few $/mH per day is worth the time, effort, and heat.

Why is it dead? Heat isn't a problem for me since my basement is just as cold year round. In fact, it's colder in the summer than the winter..

my 650 kh/s doing 24/7 mining on them got me 0.01 BTC on the following timestamps
2014-03-21 22:30
2014-03-18 10:20
2014-03-15 08:20

So you mean it took 3 days to make 0.01BTC? I just checked my history and there aren't many payouts because I hold DOGE.. But took at least 5 days for me to payout 0.01 BTC.