Topic: HashFast announces specs for new ASIC: 400GH/s - page 575. (Read 880461 times)

Here you go, read all about it (was this posted already? sorry if it was -- their index page is a dead "coming soon")
>>>
 http://hashfast.com/news

Never mind, they fixedized it.  Here's a partial screencap ...

My advice is avoid any large scale mining purchases that have the slightest sniff of a centralized mining farm. It's bad for the BTC network. Just look at what ASICminer have done to spoil mining for all. Keep that hash rate spread around as much as possible.

The going rate on new ASIC releases should be less than $20 per GH/s or it wont be a happy ROI.

I was talking about whether it or not someone who could afford $500k and wanted to invest all of it in bitcoin mining would want send that money to gridfinity, tomorrow, or whether you'd be better off ordering 62 KnC Jupiters, or even developing their own chips instead. even if we were to assume that HashFast and KnC were both have chip designs that work which I think is likely - although gridfinity's ability to execute is still questionable (IMO, which doesn't matter that much because I don't have $500k anyway).

Anyway, you seem pretty stupid, so clearly arguing with you would be a waste of time.  You're certainly wrong on the facts, at least. I certainly don't care if you pre-order from KnC or not.  If you want to run around saying KnC is a scam you can do it in the KnC thread.

Even if HashFast is not a scam (let's assume it for a minute just for the sake of the argument), if this is the best entry strategy that the management team thought of - it sure places big question mark on their abilities. Wouldn't it be totally different if you had *something* to show to the interested public before you ask for that kind of money?

Milestone                    KnC Miner      HashFast
-----------                 -----------    ----------
Announcement             Yes               Yes
Details of Chip              No                No
Working Chip                No                No
Video of unit hashing     No                No
Delivery of units           No                 No
Percentage of network   0%                0%


My post above said "about equal".  I stand by that appraisal. 

Well, let's hope they were - because if Gridfinity's business model is sound and there are lots of people looking to spend $500k on bitcoin mining hardware then we may be completely screwed as far as future difficulty  

I also think Gridfinity would have put more thought into their announcement and built a better website - the just asked for money, and didn't even mention they were selling hosting as opposed to raw chips, complete units, or whatever.

Notice was short to bitcointalk, but I doubt that they were hasty decisions by Hashfast or GridInfinity.

Uh, what you said was this:


You said the odds were equal. But just because two things are not 100% certain does not make them equivalent. Usain Bolt and Rosanne Barr can both claim they'll win an Olympic medal in 2016.  The fact that neither is 100% certain doesn't mean that the two claims have equal weight.

Before HashFast clarified the situation, the Gridfinity announcement made it seem like they didn't know what they were doing.

Are you kidding me?! Scamming aside, when you come on to any form of advertisement, asking for HUNDREDS OF THOUSANDS of dollars, in less than 48 hours, you damn sure better have done your homework. If I had that much money to throw around, I would definitely look at this as an indicator, these aren't $150 Jalapeno's these guys are trying to sell, these are the equivalent of purchasing a McMansion... FastHash, you guys are losing major street cred by supporting these guys on such short notice...

I never said KnC is a scam.  My claim was that KnC and HashFast share several potential problems delivering a bitcoin hashing machine to your doorstep.  Neither one has shown

"evidence that they'll actually be able to deliver what they promise when they promise."

Okay, now it makes sense.  The new guy wants to do big hosted farms, according to Skrodenis, and he didn't make that clear this morning.

It's apparently a side deal in HashFast's plan, and they are still going to go ahead with retail themselves in some shape or form, including the potential for hosted buy-ins.

Nothing more to see now until the price and specs announcement.

Basically a venture capitalist is going to have to look at this the same way a customer decides whether to pre-order a miner from a company. What is their pedigree? What is the product? Is the timetable realistic? How will future difficulty affect ROI? etc.

*yawn*.  The "KnC is a scam" people are the ones who seem naive, frankly.

I'm not trying to say that KnC is "real" and that HashFast is a "scam", I am talking about the evidence that they'll actually be able to deliver what they promise when they promise.

Of course, so far Gridfinity isn't promising anything at all, other then maybe to return your voicemail and to take your money in $1/2 million chunks over the next 36 hours.

And the question is, why would anyone with that kind of money to throw around, and willing to take a risk with it, give it to them instead of buying 62 KnC Jupiters? Or developing their own chip, for that matter?

They haven't provided a reason.


That's good.

...And so the decentralized network concept became centralized...

Hi Everyone,

First, we want to assuage all of the speculation. HashFast did in fact do a limited distribution deal with Daniel Bateman's company Gridfinity.  
While Gridfinity has chosen to focus on large hosted solutions, this will by no means be HashFast's only distribution channel.

We will soon be opening public sales on our website, (a few details are still being worked out for this launch) which will include a smaller mining machine, designed for small 'family' mines, and a more dense machine designed for more aggressive miners. We are also in talks with several potential partners for hosted solutions at much smaller entry points, as has been requested by many members of this forum.

We are sorry if the timing of the deal with GridFinity caused confusion. That was not our intention. HashFast is committed to working with small miners and maintaining network diversity.

Watch this space for more information, coming soon.

John Skrodenis
HashFast.com

If you go back to the original announcement thread, I asked right up front if at t=7 (today, I think) that pre-orders would start.  The answer was yes.

Here's what might be happening.  In the intervening days, it became clear that a retail pre-order was going to be a very tough run.  They shifted gears, looking at the bitfury model, where you have Dave here in USA running the order book and taking the pre-order risk. 

Whether bitfury or metabank needed money up front from Dave or not, evidently these guys do.  And they probably need it by the end of the month to meet some invoice, deadline, or agreement.  Their new guy doesn't have the money, but he solicits for a big wholesale buyer - someone whose role would be isomorphic to Dave's role in the bitfury topology, including having credibility and trust inside the bitcoin community. 

They're evidently willing to give this "angel" a smokin' deal on the chips, or, rather, they probably are going to have to. 

That said, in one or two day's time on-site, someone who brought along his tech talent could probably get satisfaction from the design teams, including Uniquify's, check the fab scheduling, look at the books, make a decision as to how hard it would be to build out the enablements, and give it to the lawyer. 

It's not being pulled off with a lot of aplomb, but it still seems as plausible as many other deals in this crazy bitcoin universe.       

Does this prove anything?


"real working units"?  In my opinion the odds of looking at your KnC unit hashing away in your spare bedroom are about equal to looking at your HashFast unit hashing away in your spare bedroom.

They are both theoretical at this point.  Claiming one is real and one is not just seems naive.

Also, I think it's kind of funny that you're offering to send just one buyer, the one who orders the most, out to San Jose to meet the team.  As if the kind of person who has money to throw around in $500k increments is going to worry about travel expenses, lol. Bitcoinorama traveled to Stolkholm on his own dime to check out KnC, and he only ordered (N?) Jupiters (IDK how many he ordered, it's not listed here)

Plus, let's keep it real - San Jose isn't exactly a top vacation hotspot

My impression is that if you're selling tech items at these kind of prices, you still have to make a big investment w.r.t sales.  Instead of buying ads, you hire sales guys to fly out to meet clients, offer to fly them out to see you - etc. Check out this video on McLaran factory.  They're selling $250k cars, my guess is that one of the reasons they made the factory so carefully designed is that they expected potential customers see it as part of the sales experience.  It wouldn't be surprising at all if someone who could afford that much would want to check out a factory tour. And you guys are charging twice as much!

The other problem is that as you invest more and more capacity into mining, you actually start to see diminishing returns, as you contribute more and more to the difficulty increases yourself.

And honestly I kind of doubt you'll be able to find many buyers at $500k, KnC hasn't even sold out of their October units yet. I think a lot of people are nervous about where the difficulty is going to be in October or November when KnC and a bunch of other units ship.

If I had $500k to invest in bitcoin mining, I'd be spending the money on my own chip designs, which would allow me to add more capacity for the cost of the wafers and packaging, ala ASICMiner and some of the other companies with that model (BTCGarden, for example). Or, I might consider designing an optimized Litecoin miner chip instead (which would basically be a CPU with some litecoin specific instructions and lots of really fast RAM).

___

There's also the question of, well, if I have $500k, why not just order 62 KnC jupiters? Why should someone buy your product instead?  If someone were to order from them today, they'd get real working units in October, where as with you they would get... what, exactly? Raw chips? Will they need their own PCBs? Will they need to design their own PCBs? What about software?  KnC supplied sample data to CGMiner developers to code against, so CGMiner should support KnC devices when they're released. And their own mining software works with the FPGA prototypes.  Do you guys have FPGA prototypes to test with?

And you guys aren't even giving a hashrate or chip count number for this money, which makes planning impossible.


I wonder what's actually going on with these guys. It seems like they funded the design, through tape out themselves, yet now all of a sudden they're working with this new company that just started and all of sudden needs $500k in the next 36 hours for chips, but isn't even telling you how many chips you'll get for that much money, let alone what exactly supposed to even do with them once you get them.

I wonder if there's someone involved here with more money then sense:  Maybe HashFast had the funds to fund a high-end ASIC design, but doesn't have enough money left over for wafers or to design whole systems? That seems strange given they would have spent (based on my understanding) millions on the design.

Or maybe Gridfinity came in, got in contact with HashFast and bought up a bunch of chips and now doesn't know what to do with them, hoping to flip them for a quick profit without doing any work (not even bothering to setup a nice website or hire someone to answer the phone!)

It's also possible they simply licensed the design, and need $500k per order because they're planning on ordering the chips from the fab as the orders come in.

Again, the funds are supposedly going to be put in escrow, which would mean that they supposedly have the money they need.

Either way, doesn't inspire a lot of confidence.

Have to raise my hand, admittedly I'd totally missed that!

The answer's in the footnotes:

"*For those interested in orders below USD $500k, send an SMS (text) message to (415) 480-4921 with the approximate USD amount that you would like to contribute to a group buy. We will get back to you if a credible group buy arrangement materializes."