Everyone knew Jupes were going to ROI, that was a given.

Haha.. I wish, I would love to do it.. $6k maybe, but not 12 lol..

For Neptune, BTC would have to be worth more than $5kUSD in April for it to be a good venture. Of course, simply holding your 10BTC now would put you ahead of the game.

I would be willing for a $5k product, but this is just too much, especially when the competition is planning to sell 100gh for less than $100 by first quarter, using 28nm technology.

Seems like KNC is leaning hard toward datacenter models, leaving other companies to deal with the individuals.


I haven't made my final projections for March yet (not till December), but I wouldn't be surprised. Cointerra's smallest first batch was over 2PH, they opened up 3 more batches since then for even more (i think 5PH each?). If ONLY Cointerra ships, we'll be close to that projection.

As it is, KNC's 2nd gen are about to kick the neighborhood difficulty square in the nuts within days..


No.. $5k for January/Feb/March batches..


Yup.. Mercury's are going for $6k right now


Hard to double the network? All they have to do is ship more units. Keep track with the preorders and shipped batches much lately? Sounds like you haven't.

The best method is to do exactly what folks on ebay are doing.. ebay/resell for 2..3...7x the original cost.

KNC should've kept a "home user" product line along with the "commercial" model. The latter will most assuredly hash more than 4TH, I bet 6TH.. but most home-users can't power that.

Doing another NRE allows them to pad the cost as well.

It's good to be the shovel retailer in a gold rush.

Here's the estimate thread I've been a part of, particularly the first round results (I was nearly to the penny on the estimate): https://bitcointalk.org/index.php?topic=272768.msg3409229#msg3409229

And my next projection: https://bitcointalk.org/index.php?topic=272768.msg3442350#msg3442350

And for anyone who can do basic math, yeah April is going to be a shitfest for difficulty. This is why I don't understand folks are thinking first month mining will be 12.. 15.. 20 btc.

..ok, and? That's what we already know

^-- This.

Doing a new NRE allows them to pad their pockets, for one ..

What on earth difficulty setting are you using? Are you ignoring the mega PH coming online within 90 days??

Issue 1: Why wouldn't they also offer 28nm technology since its' one-time cost has been paid? This would allow folks who can't afford to pay for a cheap car, to keep in the game.

Issue 2: For a $10kUSD unit which will likely arrive after Cointerra's $5k offerings, I would hope the darn thing does 6TH. Knowing KNC, I'm sure it may be capable, but powering will be an issue for most home-based setups.

Issue 3: Also, looking at even conservative difficulty estimates, a 4TH unit would produce 3-5btc in April, with a lifetime of less than 10BTC.

So...

I think we probably wouldn't be as picky except you started a topic as a guide/how-to, and then asked mods for a sticky, and yet the OP is riddled with failure.

Not knocking your efforts, just kinda saying it isn't sticky quality yet.

So is anyone seeing a preorder page yet?

It's easy to share settings, but unless you understand what's going on with them, you might end up losing with them as well because you have to adapt them to your specific situation. PM me what you've got and I'll see what I think about them.

Each exchange has its own personality, especially this past week (more than ever before). I have different groups of settings for each exchange for this reason. I lost a little on gox this week so I killed my gox bot until I had time to review my stats. I spent a couple hours today going over the data and plotted some new settings for me to try. "So far so good", no losses.

BTCE has continued to make BTC but ever so slightly, at least it's not losing. GOX requires you either go very conservative with large EMAs and short timeframes, or mid-to-long timeframes with short EMA cycles catch the roller coaster effects.

Have you gone to bitcoinwisdom and played with the EMA system there, looking at the historical data? Have you setup EMA calculation in some columns on a spreadsheet like Excel, to get a feel for how it works? If you do this, you'll learn a lot about how it works. Doing this would help you (and others who haven't) a lot. Part of what I'm suggesting is to use different settings for each exchange, if you haven't done so already. The market is too volatile for each exchange to react similarly.

Yeah, that does appear to be right. I'm not going on either side, just seeing what the data says.

BTC was in the 50-70 range in july which, if bought and held, would be around 17btc today I believe. Of course he didn't hold, he cashed out at intervals which makes it harder to determine the difference between B&H and EMA. If he provides us the dates and amounts, then it could be determined.

It seems like he's saying, he fronted $1200usd to try it out, and he cashed out profits at the intervals specified.

$1.2k front
$1k cashout
$2k cashout
$1k cashout

$4k cashed out total

$3.4k current balance = $2.2kusd profit not cashed out

$6.2k usd total profit above the $1.2k front.

Seems like simple math to me.

The bot has bought low/sold high for me enough to quintuple my BTC. I mean, I'd say it works. Not every situation, but it can work.

I see on the pool it shows a QBT/LTC exchange rate.. is this based on something or is it arbitrary? if it's based on something, I'm making quite a decent paycheck today lol

This is part of the reason I was a bit shocked with the claim, and the reason I pointed to commercial GPU's and other high-end products which favor 28 lately.

My 2 concerns with this would be the issues faced with 20nm tapeout, and firmware. The mercury's I've seen seem to have no issues whatsoever, but some Saturns and most Jupiters do, so it makes me wonder if they'll be able to stabilize the next generation before shipping.

Something I haven't seen people talk about, I think I will remind everyone about: Keep an eye on the Minimum Profit column of backtesting.

Find settings which minimize this, it will keep your BTC in your pocket. Ideal settings give you things lower than half a percent, for example.

20nm in 2014? Not only is that amazing, it's almost hard to believe. It would be beyond most current GPU OEM's specs.

Also to the gent who mentioned high VRM temps with the liquid cooling, you have to use  heatsinks on them and the memory, and then provide a 120mm or larger fan to blow on them. That's how the Accelero hybrid works, and it does a fine job.

The Prolimatech Mk26 beat out the Accelero products, including the Hybrid closed loop liquid cooler, by 1-2* in 3 independent tests. The Accelero systems were my top choice until that point.

Larger dissipation area + larger fans.

Yeah, as I mentioned above, you can get a copper shim and use most any generic cooler if you need to. I think the reason a lot of 79xx coolers have been disappearing en-masse is because of people like me who have a bunch of cards with fans dying all at once.

I could keep shipping them back to MSI, but this is just the better option overall.