( I was personally trying to use gekko to reinvest using micro-trades much closer to the exchange-enforced minimum 0.000001 GHS, and the granularity was causing my fees to sometimes be much much higher, sometimes 5%, 10%)
TL;DR* Granularity causes rounding errors on the fees when you're reinvesting tiny amounts with micro trades... Specifically: The "1 satoshi" granularity causes issues. ( The exchange rounds up to the next whole satoshi, and it's a common practice never waive the fee under any circumstances.) Interesting problem, I wouldn't have even thought about it if you hadn't explained it. Didn't realize it has so much effect. |
|
|
|
Fixed a bug with the "enforced" state ( old cex.io-only reinvest mode) where the code had some poorly defined behavior which could cause a stuck state with no possible exit condition. Also, spent the last 30-45 minutes auditing code in proximity to the now-fixed bug, and updated the release notes with the specific details in: https://github.com/kuzetsa/gekko/releases/tag/critical.bugfix, which is now the latest release as of: 2014 June 25th 22:00 ( local NY time)
... And https://github.com/kuzetsa/gekko/releases/latest now points to the " critical.bugfix" release. Thanks for the update. How come it says cex.io trading fee will be: 0.75%? cex.io fee is 0.2% usually. |
|
|
|
|
So is the compensation for poor people like me that only ordered a single batch 1 X1 unit from Minersource still effectively zero? I don't call a buy 2 get one free offer really compensation as it requires further outlay, it's more of a discount coupon promo then true compensation.
|
|
|
|
Are there any trustworthy group buys yet for the S3?
Why would you join a group buy when the S3 price is unknown? Last thing your would want to do is commit before knowing that! |
|
|
|
They are delaying it because of technical issues, they are test running the s3s to provide the same reliable miner s1 was with the same price that will ROI just like s1 did. Maybe just one day before diff hits 16G they would open for orders.
I am very glad they are doing this. I'd much rather wait (really, a week -that's nothing) and get quality equipment than get it now and have to ghetto engineer it to get it to hash at advertised speed. A week is probably 10% of the useful mining life of the device if you think about it, and look at the diff increases. I think they are waiting to sell all the S2 before launching the S3.
I think thatīs the reason.
That sounds plausible, I am quite sure they would have tested the S3 well by now. |
|
|
|
my deposite no show in the wallet deposite in coinex.pw I cek the blockchain it was 45 confirmation
indicated coinex.pw SCAM
Why are you trying to make deposits there? Can't you read? Coinex doesn't work. |
|
|
|
Fine with me, i will get out of the scam as soon as possible from now on im on a strict NO PREORDERS rule.
I wish more people would stop pre-ordering. It's fueling one of the worst behaved industries I have ever seen. There is a class action going against Cointerra atm. http://www.coindesk.com/cointerra-mediation-class-action-lawsuit/Most of the ASIC manufacturers have failed to deliver on their pre-order promises. This type of funding is bad for all, and should stop. Other industries rely on venture capital with strict controls not pre-orders. |
|
|
|
The jump is possibly from all the devices going in transits after next difficulty. So after next diff it will be a cool off and its better to have such a jump now.
What makes you think it will cool off after this next jump? |
|
|
|
Speaking as someone who has their products in hand, no they dont. They get effective about 1.1-1.2w/gh.
Rockminer is not a great example, they run their core VDD at 0.72v whereas the BE200 can go as low a 0.55v. What's the W/GH ratio for end system at 0.55v? Who knows, they quote only the chip specs, have to wait until someone under clocks a system properly. |
|
|
|
Speaking as someone who has their products in hand, no they dont. They get effective about 1.1-1.2w/gh.
Rockminer is not a great example, they run their core VDD at 0.72v whereas the BE200 can go as low a 0.55v. |
|
|
|
Not trying to stoke the fire, I'm genuinely curious, but are there any machines that are currently available that are using .55 W/Ghash?
The ASICminer BE200 40nm chip is 0.35J/GHash at rated voltage. Though as usual people will over clock them straight away thinking more is better. Here is a neat little video of BE200 immersion testing. http://vimeo.com/98321051 |
|
|
|
BitFury is making a push to list on the stock market and just received $20million in Angel funds or so they say, that doesn't happen without something in hand to show for the value!
They probably need the money to fab the next generation. Their 55nm is pretty old now. |
|
|
|
They delayed the launch.
yes as they are mining at the lower diff rate. this is an edge that only huge asic builders can do. very simple. the exact day diff jumps put out 20 percent more gear then the entire network has.. mine that gear at the current diff for 10 to 11 days then release it to us. so they mine 20 to 30ph at 13.7 we get it and the network is 16.2 . figure out how much that is? it is 1100btc a day for 10 to 11 days. or 11000 to 12000 btc. this happens as I type . it is AM and Bitmaintech combined. what is that in usd 6.6 million dollars. The rich get richer. Stupid conspiracy theory from people that haven't even spent 5min to work out that it's far more profitable to keep shipping the boxes out the door than to mine with them. Once upon a time when you could make 0.3BTC per day off 32watts with a BFL Jalapeno it was worth manufacturers running a mining farm but not anymore. An S2 makes what 0.037 per day, but selling it makes 3.131 BTC, when you can sell all you can make, go do the math. |
|
|
|
just my opinion... if they put S3 below BTC1 then what'll happen with the S2 buyers? well, i think bitmain will adjust the price above BTC1 about ROI, if u want ROI in the short time (1month? 2month?), well, u'll richer than bitmain  if i'm Bitmain, the hell i care about ur ROI, its about risk, afford what u can losebig profit always followed by big risk if in the end u cant ROI, re u going to blame Bitmain too? nah they're not ur friends who'll make u rich bro... the damn they care S2 is a lot of hardware, 10 blades and a PSU I would expect it to cost way more than a couple of S3s. 5 x S1 was also a lot cheaper than an S2. |
|
|
|
I'm also a bit afraid of the massive 20+% diff jump plus government auctioning 30,000 BTC. For some reason it doesn't feel like the right moment to get in.
Agreed 100%. Gonna be a crazy time for mining the next few months! 1 TH get's you .037 per day today. After the next difficulty adjustment, I cannot imagine what that will fall to! Best of luck to all!  30,000 BTC isn't much, eg okcoin traded over 57,000 in the past 24hours, also the auction is hardly going to be a dump. The return on mining will reduce until it's negative for all those who overclock or have expensive electricity. To make an S3 pay for itself, it would need to be priced around 0.75BTC or less. That's going by the mining calc on https://tradeblock.com/mining/ |
|
|
|
While I hope the S3 pricing is low, I also hope that I have PMa's discipline if the initial price is too high. The fundamental problem for the home miner is that some fraction of the buying population is throwing money at equipment. I think the price of the S3 in China will be lower because miners there will be more hard nosed about the way they spend money. This is in contrast to purchasers who have bet on the come by funding huge preorders and also haven't quite gotten their arms around what's involved in breaking even.
The fact the existing, new and coming on line mining companies are thinking they can sell $1.5 - $2.0 per Gh/s all day is further evidence that some western purchasers of mining equipment have already demonstrated so far that they have more money than sense.
If BITMAIN comes in with a low price, this will indicate that he is thinking long term.
The pricing will be in BTC not $ per GH/s. so when you do your ROI calculation it's best to only think in terms of BTC. |
|
|
|
I think that miners has to focus like a laser on what's the most important-highest ROI and drive manufacturers to provide one through the power of our purse (don't get on on sucky deals). If it is a hobby (and I respect this very much), then every new hardware is exciting as well.
The focus is certainly not on the highest ROI, typically there will be a claim about some marvelous power efficiency, then the bottom line will be the miner still draws several hundred watts. Overclocking is FOTM not underclocking, and that's whats driving up the difficulty spoiling it for all. The first thing I want to know about the S3 is can you get a good underclock happening like you could with the S1 and the pencil mod? 0.75v VDD core strikes me as a bit high. ASICminer are talking 0.55v with their 40nm offering. |
|
|
|
Bitmain are you going to reward your loyal customers with some good coupons without expiration dates for the S3, and S3 updates from S1 ?
The way I read it was an updated heat sink for the S3 which could be fitted to the S1. Is it something more than that? |
|
|
|
The S3 doesn't look as good value as the S1 was, specifically the S1 had 32 ASICs per blade and the S3 only has 16 so you are getting half the ASIC hardware that the S1 had. That's why it's only rated at 504GH/s when it would have been up around 800GH/s with better power figures if it were 64 chips like the S1. The S3 will be clocking the chips harder I bet to make up for only having 32 of them.
Does this leave an opening for the S4?  For sure, the main thing about the S2 was that it used a lot more ASIC chips than the S1, but at a more sensible lower clock speed for power efficiency. The S3 uses more power than the S1, I was kind of thinking they might have gone down the path of a S1 speed unit, but less than 100watts. There is no point adding more net hash, it just makes the diff go up, but there is a lot of point in replacing your 350watt S1 with a similar speed unit at less than 100watts. Considering the ASICminer 40nm chip does 0.35J/GHash at the rated voltage, it's should have been easy to beat with 28nm. |
|
|
|
|
The S3 doesn't look as good value as the S1 was, specifically the S1 had 32 ASICs per blade and the S3 only has 16 so you are getting half the ASIC hardware that the S1 had. That's why it's only rated at 504GH/s when it would have been up around 800GH/s with better power figures if it were 64 chips like the S1. The S3 will be clocking the chips harder I bet to make up for only having 32 of them.
|
|
|
|
|
Show Posts
