This coin is going to around the dollar mark within the next 11 months based on my projections.  http://xkcd.com/605/ |
|
|
|
|
Well I gave them a 25% bump over straight die shrink. As for Labcoin well I don't believe their claims but am willing to be surprised if/when they show hardware operating at that efficiency. Still even if Labcoin can achieve that level of die efficiency it doesn't mean competitors will. For example if you take Labcoin's claim 130nm to 55nm is actually (130/55)^2 = 5.5x reduction in die space. Throw in 20% more for higher clock (faster switching time due to smaller process) and you are looking at closer to 314 MH/s per mm^2. When you see that Bitfury is "only" 132 is starts to show how optimistic Labcoin's claims are.
Granted I don't know they can't do it but great claims requires great proof. Labcoin's die efficiency claims are 30% to 70% better than any other announced specs when normalized for transistor size.
|
|
|
|
DeathAndTaxes sure seems to know alot about the KNCMiner... Do you have a inside hookup or do you work for them?  Nope I just read a lot (and have an eidetic memory). |
|
|
|
It has 5 USB ports. None of which are used. It has 5 ethernet ports and can do GiG ethernet. Only one would be used the rest is just wasted spec. It has an Altera FPGA on it for lots of helpful miner specific things. Completely useless and expensive. Ethernet is a good choice for interfacing with the, wait for it,,, internet! Lots of cheaper alternatives have Ethernet. Nobody would use usb for a standalone rig. Nobody has (Bitfury, Avalon, ASICMiner) The SO-DIMM: You can preconfigure/program them on existing dedicated programming stations instead of having to hook up each miner for programming. FAST!
An SD card can be loaded with a linux distro. No need to connect the miner to a programming station. USB onboard:
5ports, 4 for ASICS and one for 'users'.
We don't know the detail of the 5th port but it is possible that it could support a WiFi connection. We know none of the ASIC boards connect to the host by USB. Hint: take a look at the PCB photo. They are connected to the host using low speed, low cost serial. If simple works why go complicated. The same connectivity that Avalon, ASICMiner, and Bitfury use. The host simply needs: Small amount of computing power and memory to run a stripped down single purpose linux distro Enough storage/flash to store OS, mining software, and optionally webserver for reporting/management low speed serial interface to the ASICs (we are talking slower than a modem is fine). Ethernet connectivity Cheap. No embedded board is going to cost less than a rPi. Hate to break it to you but looking at the photos it would appear that KNC ended up going with rPi beaglebone instead of the expensive overkill SO-DIMM system on a board. |
|
|
|
|
Could just use a bitcoin address or PGP public key. This provides the site with "proof" of the request and authentication as well. Essentially if your PGP key or wallet is compromised it is your fault and the site can prove so.
|
|
|
|
So once the code is open sourced, you'll admit that Ripple is as open as Bitcoin? After all, anyone then could run their own validator if they are so inclined (though that is not too useful in some cases)...
I'm wondering where you get these statements like "Nodes only run closed source code owned by OpenCoin, Inc and operate with the express limited permission of the owner." from... please quote the section from the LICENSE file that says so as you apparently have different information than I have.
When the source code is available and it is issued under an open source license with redistribution, modification, and derivative work rights then yes I will consider it an open network. Of course I don't think that will happen. OpenCoin is already backing off from having source code available by the end of the year. OpenCoin sole when to monetize this is to sell off the coins they have given themselves. An open source version will inevitably spawn a fork/alt which has no premine there will be no open source servers. |
|
|
|
KNC Sept/Oct 500TH/s is way too low, because September are day1+2 [1,speculation] only and ~800 orders [2]. Most of the orders contain at least one Jupiter, few orders contain one Saturn which has not upgraded meanwhile [speculation]. October orders are shipped on 11 business days, so lets assume another 11*(800/2)= 4400 orders (containing Mercuries as well). According to knc-staff Jupiter will have at least 500GH/s [3]. So let's calculate with 0,5TH/s as an average per order: (800 + 4400) * 0,5TH/s = 2.600TH/sMaybe even more since there are orders with 6,8,10 Jupiters [unverified]. Cloudhashing wanted to buy from knc as well [4], don't know whether they actually pre-ordered. If so I guess it's one huge order. Those days knc is shipping ppl will think the network-graph must be broken as it is showing hashrate increase at an angle of 88° or so until October 15th when last ones are shipped [5]. At least it will not be comparable to anything we saw before during this time x_X [1] https://www.kncminer.com/news/news-32 "Finally we wish to confirm that yes we are still on track for our delivery towards the end of September." [2] https://www.kncminer.com/news/news-20 "The number was 890 but has dropped a little as not all bank transfers arrived in time." [3] http://forum.kncminer.com/forum/main-category/hardware/107-psu-recommendation-jupiter-the-largest-planet?p=1050#post1050 "According to our engineers an 850W, 80+ Gold certified PSU should be enough for a Jupiter running at the expected speed of 500 GH/s." [4] https://cloudhashing.com/blog/13-blog/press/65-bitcoin-mining-as-a-service-maas "He (Cloudhashing’s founder, Emmanuel Abiodun) is also in negotiation with KNC, which he says is pledging September delivery for its own ASIC products." [5] https://www.kncminer.com/news/news-33 " All current paid orders will be shipped no later than October 15th. All new sales will reflect our November price point of;" *edits: -added [speculation] and [unverified] -added an additional link -made it more readable Very good. Here is another link (distribution of 141 orders) https://bitcointalk.org/index.php?topic=249065.0. if you want to add it to your cites. It works out to 495 GH/s per order which I will round up to 500 GH/s. I don't think 11 business days in Oct means they need to ship 400 units * 11 days = 4,400 unit backlog. I think KNC is smart enough to give them some slack and then never "ran out" of Oct orders they simply closed it and pushed the delivery of new units back to Nov when the price dropped. I think this was for two reasons a) to not have their customers ask for refund and reorder (if you are late order paying full price why not cancel and maybe get a unit a day or two later at reduced price) and b) to avoid "burning" their customer loyalty. So I think assuming full queue 11 days * 400 is probably an overshoot still I do like the math and analysis. I will discount it slightly to 4,000 orders today (sept/oct) with average of 0.5 TH/s which gives us 2,000 TH/s. I have a high confidence it will be at least that much. It may be more but I will update it to that for now. |
|
|
|
When people say Bitcoin's gold to Litecoin's silver, Its more like Bitcoin's Dollar to Litecoin's Quarter, but people don't like to associate cryptos with fiat, so the analogy became a way to say that Litecoin will be Bitcoin's partner.
Well I don't find that much better. If I need to spend a quarter Bitcoin I can. The OP got it exactly right it is a simplistic saying. Silver and copper currency evolved because the purchasing power of gold was so high and the PRACTICAL LIMITATIONS OF REAL WORLD COINAGE puts a finite limit on useful coins. As an example the US gold dollar (circa 1849) it was smaller than a dime and about 1/3rd lighter. The average unskilled laborer made about a dollar a day in 1849. What the hell was he going to do with a gold coin. Can you imagine your employer giving you a coin smaller than dime and about 1/3rd as light for an entire day's salary. The workers daily expenses (food, entertainment, etc) would obviously be less than a dollar. What was the government going to do mint coins 1/10th the size of a dime containing 0.15 grams of gold? Of course not. The real world has physical limitations. It was far easier to make coins out of silver everything from the tiny 3 cent silver piece up to $5 silver coin. It is very likely a laborer never saw a gold coin .... ever. What would be point just to get it and immediately need to exchange it for silver? With Bitcoin one can spent 1/100,000th of a BTC just as easily as 100,000 BTC. The idea that it needs complimentary coinage is silly. I think some crypto-currency which compliments Bitcoin will come along but it would just be something that is a drop in replacement with 4 times as many coins. It will be some outside the box thinking. |
|
|
|
Should be good. Has 4 pairs of 2 PCIE connectors so you can use 1 from each pair without any worries of overloading the wiring. Has a single 83A 12V rail which should be more than enough. 80-Gold efficiency and more headroom then KNC recommends. Can never know for sure until you test it but I can't see any reason it wouldn't work. |
|
|
|
There has been quite a bit of discussion on which PSU to purchase at the KNC forum. The bigger 1000+ Watts PSU will not only offer better efficiency at around 50% load, but the lighter load (relative to max load) will also be easier on the PSU, extending its life and reliability. The extra wattage headroom might also be useful in case the KNC chip turns out to be highly-overclockable. On the same PSU KNC forum thread, one of the KNC staff has already l leaked Jupiter will run at an expected speed of 500 GH/s! True and maybe some people can just wait for more info but there is a point of diminishing returns. Lets say the Jupiter is 500 MH/s. >850W @ 12VDC would mean >944W @ 120VAC. That is >1.8 J/GH. Already close to double what bitfury does @ 55nm and more than double all other 28nm offerings. I certainly hope it doesn't require even MORE power than that, I am hoping it will be significantly less power even when overclocked. Then again it isn't like more capacity hurts I was just pointing out the "much lower efficiency" at high load is pretty much a misnomer at this point. |
|
|
|
What about Strangemattercoin and Neutroniumcoin? Of course, only Kryptonians can spend these...
Antimattercoin, if it collides with a normal coin both will vanish leaving only energy. Antimattercoin with an estimated mining cost of $62 trillion per gram is obviously going to be top dog although that whole if it collides with normal coins you lose both makes me think CaliforniumCoin is where the smart money is at. Produced only in microgram "blocks" and requiring a nuclear reactor to mine it has a scarcity that is about 1 million times that of Gold. Makes GoldCoin look like dirtcoin. However I will offer all GoldCoin owners the chance to upgrade to CaliforniumCoin at a introductory exchange rate of 200,000 GLD for 1 CFC.  Wallets are going to be tough though. It takes this 50 ton container to safely transport 1 gram of Cf 252. Some might not like that it has built in demurage which is unavoidable due to the laws of physics. You lose half your coins every 2.645 years. On a related note the idea that a coin will be successful based on its name makes me concerned about the OP level of critical thinking. |
|
|
|
No. Although the OP makes it seem like that. The house is foreclosed and sold at auction. All creditors including the lien holder and mortgage company are paid any excess equity ends up going to the owner. No different than any other foreclosure.
In this mans case he owed $134 didn't pay it, the lien was sold, 6 months passed and the lien increased to ~$5,000 due to interest, penalties, and fees. That is likely predatory and some caps should be put into place but I would point out the lien holder "only" got $5K not $197,000. The amount of his mortgage and lein were more than the house sold at auction so the residual equity was zero. In other words the man was already broke before the foreclosure occurred.
I may have misread the article, but I thought it said that the mortgage was paid off, and the lien holder got to keep all the equity, not just the amount owed. If your interpretation is correct, then I don't think there is anything wrong with what happened. No it never stated the mortgage was paid off only that tax lien holders are paid AHEAD of even the mortgage. So hypothetical scenario the house is worth $197K, it sells in auction for $160K net. The lien was $5K, the mortgage was $170K. The lien holder gets their $5K (not bad if it cost them $134), the mortgage company gets $155K ($160K sale price - $5K) and take a $15K loss on the mortgage. Homeowner loses home which potentially had $27K in equity because you are never getting top dollar in an foreclosure auction. |
|
|
|
I stand utterly corrected. |
|
|
|
Add in conversion charges and bid/ask spread from local fiat to bitcoin.
Add in transaction fee.
Add in conversion charges and bid/ask spread from bitcoin to local fiat.
How 'free' is it now?
It won't be free but it doesn't have to. Bitcoin is an open network with low barriers to entry that will mean lower prices. WU couldn't sustain the sometimes 10%+ fee without high barriers to entry. The same thing applies with banking. A "wire transfer" is a copy lines of text sent over a computer network and the banks collect up to $40 to do that. Their cost is essentially $0. Why the infinite markup? Because banks are a "members only club" with nearly impossible barriers for new players and that keeps the fees high. Credit card networks are the same thing. All traditional fiat transfer networks involve massive barriers to entry which protect margins. Bitcoin is disruptive. Still the article was somewhat simplistic, it will take a while before Bitcoin is ubiquitous enough to compete with traditional money transmitters. The article says remitters collect 8% on $815B annually and it could go to zero or close to zero. Well that is kinda naive but a Bitcoin based competitor charging 4% would mean remitters save half on fees and 4% of $815B is a nice chunk of change. |
|
|
|
Alright guys, some of you have a lot more knowledge then me in the PSU area, so if you could help me out here, I'd appreciate it. I have a Saturn on order, which I plan to upgrade to Jupiter once they start offering the modules separately. The PSU that I have picked out seems like a perfect fit to me, and at a fairly good price. If you guys could look it over and let me know, that would be great. And if it is a good one, it could turn out to be a good deal for everyone. http://www.newegg.com/Product/Product.aspx?Item=N82E16817139011Should be the minimum acceptable wattage for a power supply. I'd recommend going with something around 1kw... They are much more efficient at 50 % load versus 80 - 90 % load. That hasn't been true for the last decade but it is a myth which dies hard. The minimum efficiency for an 80-Gold PSU is 90% at 50% load and 87% at 100% load. The curve is very flat. For example here is the test for the PSU linked to: http://www.plugloadsolutions.com/psu_reports/CORSAIR_CMPSU-850HX_ECOS%201464_850W_Report.pdf50% load = 90.38% efficiency 75% load = ~90.0% efficiency (estimated due to their horribly low res chart) 100% load = 87.06% efficiency So using a larger PSU would save you maybe 1% in power, 3% under the extreme example of 100% load vs 50% load. At 1% power on 600W 24/7 would save you 53 kWh per year. At $0.10 per kWh we are talking $5. Spending $50 more on a PSU makes the break even 10+ years. At once time PSU were pretty shitty with narrow "sweet spot" (40% to 60% of peak power), and efficiency tanked (<70%) outside of that. This is where the advice of buy 2x the needed power. Those days are long gone thanks to the 80 Plus program and even lower end 80Plus PSU have an almost flat efficiency "curve" for loads anywhere from 20% to 100%. It is nice companies keep making more and more efficient PSU (Platinum and Titanium are also possible) but once you get to ~90% it starts being diminishing returns. Still Google and facebook use custom PSU which only output 12V to get 95%+ efficiency. Really there is no reason for PC PSU to be as complex as they are. Maybe some future standard will improve that. |
|
|
|
yeah, so why wait 5 weeks to do a mockup when you had the materials? I disagree..... respectfully.
... it was a low priority ... they sent it to a studio to have it done right ... they bought a lightbox as they didn't like the quality of existing photos which will be more important as they get closer to retail ... they were waiting on the cable harnesses which will be used in production from the OEM ... they had some unreported issues they wanted resolved before taking photos I don't know take your pick. Having chips in hand would be a pretty massive deal. I am pretty sure KNC isn't going to keep that a secret. Why would they? They want unsure customers on the edge to cancel orders? They are hoping Nov orders will slow down because they just have too much revenue? They want FUDers to drive sales to competitors by questioning the lack of chips? They want to take more time out of their carefree days to answer emails about "when will the chips arrived"? |
|
|
|
|
Correct me if I am wrong but k only needs to be unique, it doesn't need to be randomized each transaction. We simply use an RNG because given a large enough range (256 bit) it is likely all values will be unique. However that assumes the RNG is working properly and that may be difficult to test.
Using a RNG to seed a 256 bit nonce which is incremented on each tx "next-k" would be one way to be less reliant on low entropy/flawed RNG. The nonce could be encrypted along with private keys so any compromise of the "next-k" means your private keys are probably already stolen anyways. As an alternative XOR the rng value with an incrementing nonce to produce k.
|
|
|
|
There are hypes from many small companies that claim that they will deliver ASICs for BTC mining, so far a few delivered but many are "plan to deliver" soon.
What is the best among those already delivered? Let's not discuss "in the future", "planned to release soon", or "prototype working".
"Delivered today" isn't the same as "has delivered". BFL, Avalon, Bitfury, and ASICMiner have all delivered hardware to customers but among those only ASICMiner (which has the highest prices) has product available for immediate delivery. If you place a new order with any ASIC company today you are looking at Oct delivery at the earliest. |
|
|
|
The new.pics have the heatsinks mounted, which means the chips are under them. a 1/4 mounted alterra fpga would most likely damage the board if you tried to mount those heatsinks to them, so I'm saying, the real chips must be in, and in those pics.
And what would happen if the heatsinks were mounted on board without any chips? that could have been done weeks ago... it wasn't. Nobody wants to assemble something that will never function, and we've been waiting for this... I'd bet on it. It's an obvious teaser! I'm stoked! That is kinda a leap. KNC is still selling Nov units. Showing a populated board would boost sales as it would give new customers confidence that if they are on track for Sept/Oct then they are on track for Nov. At this point it is a race between various companies. There is only so much money which can go into hardware before ROI approaches 0% so a missed sale today may not be made tomorrow. Given that if they had the chips and the chips on a board I would have assumed they would snap a couple pics. If they had an operating board I am sure they would span a pic of it mining. Both would explode sales for Nov. Also the Alterra FPGA requires active cooling. Why would they mount it to the board if it wasn't to test the board? If they are testing the board they need to cool it and the mounting hole only allow a full size heat sink. So I am sure they intend to use the Alterra with the production heat sink as the goal of using a FPGA is to test the entire system (host, power supply, connectors, cooling, etc). It isn't that difficult of a problem. The simple but expensive option would be to just mount 4 FPGA on one prototype board which would make the most sense as it would allow the most comprehensive testing. If they are only using a single FPGA a nonconductive shim isn't that difficult. |
|
|
|
Maybe the "onboard linux" is an RPi? Early KNC indicate the host would be a embedded sodimm board so not a RPi. Then again it doesn't really matter. The host is relatively low end stuff. It needs a tiny amount of memory, a low end processor, ethernet, and some sort of serial connectivity to the ASIC boards. An embedded linux board will work fine, so will a RPi. I assume they simply have more experience with this board so they are using it. |
|
|
|
|
Show Posts
