Bitcoin Forum
December 15, 2017, 04:20:25 PM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 2 3 4 [5] 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 ... 586 »
  Print  
Author Topic: HashFast announces specs for new ASIC: 400GH/s  (Read 875470 times)
Bitcoinorama
Hero Member
*****
Offline Offline

Activity: 532



View Profile
July 26, 2013, 11:57:22 AM
#81

Unfortunately pre-orders have been a necessary evil at this point in time. Non-recurring engineering costs are very large, especially at 28nm. The risks are compounded by the speed at which products can be created for the market that demands their use as soon as feasibly possible due to a finite window of opportunity by which they can realise a return on investment. It's one thing for you, and many other crowd sourced parties to risk your investment, as that risk is evenly distributed (within reason), but for one entity to bare the brunt of all the risk on a make or break endeavour is huge. Which is why venture capital is unlikely unless they get to use the technology for themselves, otherwise their return earned from a cut of the profit between covering NRE and manufacturing, Hashfast's own profit, and the retail price leaves their hands tied as to how effectively they can compete on price in the future. This is aside from whether such VC would be comfortable tolerating such risk, when there are far better and certain returns to be made on other endeavours when we talk of investment of several million dollars. It obviously requires less than that for 40nm+. In any case it's very unlikely Hashfast, or Uniquify will stomach such funding themselves. I don't know their personal circumstances, but such risk requires sharing, I'd imagine pre-orders are a prerequisite, then again they appear to have made some headway, and have afforded staffing already. In any case I don't believe this is a scam, Simon's clearly a smart guy and Uniquify are a legitimate ASIC design team, but everyone owes it to themselves and the companies they choose to support by accepting the role their investment makes in these circumstances, and acknowledge that they are sharing the risk in developing these products at this point in time...

Hi, I agree with what you're saying, but just playing devil's advocate for a minute, is a full pre-order (crowd-funded type) scheme really a "fair" share of risk in the BTC-ASIC world at the moment? If a company was to seek the majority (or all) of its NRE and manufacturing costs from investors in advance of having a product then the investors bear practically all of the risk. The company has the opportunity to realise a profit on their prodcut (which may not even exist) on the basis of its perceved worth at some point in the future. There is no particular incentive to make a product that is profitable for the consumer when it actually ships if the business model assumes that the majority of the profit for the manufacturer is made during the pre-order phase (depending on their refund policy of course). As we all know, it's hard to predict what will hapen in two weeks, let alone three months the way things are going at the moment. Is it worse for one VC individual/company that may be worth many millions to loose a couple of million on a high-risk high-return investment or for lots of average Joe's to loose $10000 (which may be a large chunk of their savings/credit card balance) if things don't work out as expected? I wouldn't have said that one was better than the other actually. In practice any loss is likely to affect your average Joe's life more than the VC, but that depends on the personal stakes of those behind the company etc.

Having said all of this, it seems as though some of the current wave of ASIC manufacturers are funding the chip design and initial fab themselves (e.g. bitfury, not sure about the others) and then taking pre-orders at the PCB manufacturing stage when they have demonstrated working hardware. This seems to fit in well with the notion of shared risk and "fairness". HashFast also seem to have invested in chip design and fab already (I think, based on what's been presented here) before taking orders (or pre-orders), which again would indicate a level of shared risk if they did decide to go down the pre-order route.

I guess that my point is that as a community we should be scrutinising each new offer carefully and if an offer involves a good balance of shared risk (as was your point) then this is a good thing and maybe a pre-order phase is fine/necessary. If an offer shifts all of the risk onto potential consumers then this is presumably a bad thing and should start alarm bells ringing imo. We all know of at least one example where a full pre-order (funding design, fab etc.) hasn't worked out well for consumers. Actually we have quite a lot of power to stop this happening again by not investing in schemes where the consumer bears all of the risk - regardless of how much gold is waved in front of our noses!

I appreciate what you are saying, but the more legitimate offerings want to build a company and be around for the longer term. Not get as much cash as possible, perform a lacklustre effort on production subsequent to raising funds and then disappear. The incentive is establishing a customer base and repeat business. If you upset your investors they tend not to reinvest. There's not too many that would chance Butterfly Labs a second time around after the ordeal they have suffered.

With that in mind any VC will not necessarily be concerned about the long term customer satisfaction if it's to fulfil one run, but when you have customers funds, and they are relying on you, if you deliver, you almost certainly have established repeat custom...

Bitfury still requires pre-orders, but he has performed an outstanding job of standing on his own two feet and proving his design delivers! Every company that takes a step in the right direction is treading in virgin ground here and is creating benchmarks future competitors are expected to meet as a necessity.

Make my day! Say thanks if you found me helpful Smiley BTC Address --->
1487ThaKjezGA6SiE8fvGcxbgJJu6XWtZp
1513354825
Hero Member
*
Offline Offline

Posts: 1513354825

View Profile Personal Message (Offline)

Ignore
1513354825
Reply with quote  #2

1513354825
Report to moderator
1513354825
Hero Member
*
Offline Offline

Posts: 1513354825

View Profile Personal Message (Offline)

Ignore
1513354825
Reply with quote  #2

1513354825
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1513354825
Hero Member
*
Offline Offline

Posts: 1513354825

View Profile Personal Message (Offline)

Ignore
1513354825
Reply with quote  #2

1513354825
Report to moderator
1513354825
Hero Member
*
Offline Offline

Posts: 1513354825

View Profile Personal Message (Offline)

Ignore
1513354825
Reply with quote  #2

1513354825
Report to moderator
1513354825
Hero Member
*
Offline Offline

Posts: 1513354825

View Profile Personal Message (Offline)

Ignore
1513354825
Reply with quote  #2

1513354825
Report to moderator
klee
Legendary
*
Offline Offline

Activity: 1484



View Profile
July 26, 2013, 12:18:17 PM
#82

Great news, watching!
ziptie
Newbie
*
Offline Offline

Activity: 18



View Profile
July 26, 2013, 01:03:35 PM
#83

I just googled hashfast, and that's not what I was looking for.  Grin
RHA
Sr. Member
****
Offline Offline

Activity: 392


View Profile
July 26, 2013, 01:07:25 PM
#84

Most often confirmed skills of John Skrodenis at Linked In:

    Online Marketing
    Marketing
    Business Development
    E-commerce
    Employee Relations
    Social Media
    Social Media Marketing
    SEO
    Digital Media
    Social Networking
klee
Legendary
*
Offline Offline

Activity: 1484



View Profile
July 26, 2013, 01:07:31 PM
#85

I just googled hashfast, and that's not what I was looking for.  Grin
LMAO
greenbtc
Full Member
***
Offline Offline

Activity: 210


View Profile
July 26, 2013, 01:15:38 PM
#86

I just googled hashfast, and that's not what I was looking for.  Grin

Are you sure?  Grin
Damnsammit
Sr. Member
****
Offline Offline

Activity: 406



View Profile
July 26, 2013, 01:41:43 PM
#87

HAHA

I had to watch the video...

jlsminingcorp
Hero Member
*****
Offline Offline

Activity: 495


Hooray for non-equilibrium thermodynamics!


View Profile
July 26, 2013, 01:46:59 PM
#88

I appreciate what you are saying, but the more legitimate offerings want to build a company and be around for the longer term. Not get as much cash as possible, perform a lacklustre effort on production subsequent to raising funds and then disappear. The incentive is establishing a customer base and repeat business. If you upset your investors they tend not to reinvest. There's not too many that would chance Butterfly Labs a second time around after the ordeal they have suffered.

With that in mind any VC will not necessarily be concerned about the long term customer satisfaction if it's to fulfil one run, but when you have customers funds, and they are relying on you, if you deliver, you almost certainly have established repeat custom...

Bitfury still requires pre-orders, but he has performed an outstanding job of standing on his own two feet and proving his design delivers! Every company that takes a step in the right direction is treading in virgin ground here and is creating benchmarks future competitors are expected to meet as a necessity.

Agreed, if we see no more like the first example and more like the latter then the BTC world will be a better place. We have the power to support (or not) these ventures, I just hope we all use our heads when we choose!

Bitcoin can only be understood backwards, but it must be lived forwards.
BitFury ASIC miner hosted group buy [DONE MINING]
Ytterbium
Full Member
***
Offline Offline

Activity: 238



View Profile WWW
July 26, 2013, 01:52:33 PM
#89

I appreciate what you are saying, but the more legitimate offerings want to build a company and be around for the longer term. Not get as much cash as possible, perform a lacklustre effort on production subsequent to raising funds and then disappear. The incentive is establishing a customer base and repeat business. If you upset your investors they tend not to reinvest. There's not too many that would chance Butterfly Labs a second time around after the ordeal they have suffered.

With that in mind any VC will not necessarily be concerned about the long term customer satisfaction if it's to fulfil one run, but when you have customers funds, and they are relying on you, if you deliver, you almost certainly have established repeat custom...

Bitfury still requires pre-orders, but he has performed an outstanding job of standing on his own two feet and proving his design delivers! Every company that takes a step in the right direction is treading in virgin ground here and is creating benchmarks future competitors are expected to meet as a necessity.

I actually prefer the pre-order model.  I mean, it allows savvy investors to make their decision on who to pre-fund based on their own ability to evaluate technical proposals.  For example, I decided to go with Avalon after they'd managed to at least demo a working unit (proving they weren't total vaporware).  And so far it's paid off great.

That also means of course I'm not in the position of all these people who have been completely burned by BFL and loath the idea of pre-orders.

The way KnC presented themselves made me think that they'd be a good deal as well.

And here's the thing, remember, the less risk you take the less reward you get.  Look at the price difference between batch 1, batch 2 and batch 3 Avalons. It went from $1,200 to $1,500 to something like $7k. People buying batch 1 were taking a blind risk, people buying batch 2 knew Avalon could at least ship demo units. And people buying batch 3 knew that Avalon could deliver (at least most of their orders) in a somewhat timely manner.

The lowest risk are things like USB erupters.  You know you'll get your unit and you'll get it on time. But the ROI is basically not going to happen unless either bitcoins go way up or just ignore the electrical costs (which is easy when it's probably only a few dollars a month)

KnC's prices were ridiculously cheap compared to their competition, there's a reason for that: if they made their offer low enough that profitability would basically be guaranteed if they delivered then they would get funded for sure. If they charged high prices, people not buy in - and they'd not get funded.

So basically people want ASIC makers to shoulder all the risk, and then charge reasonable prices that allow miners to get ROI are living in fantasy land.

I'm just hoping KnC keeps rewarding it's initial customers, rather then jacking up their prices in the future.

qwk
Donator
Legendary
*
Offline Offline

Activity: 1624


Bitcoin Foundation Member


View Profile WWW
July 26, 2013, 02:15:09 PM
#90

I actually prefer the pre-order model.  I mean, it allows savvy investors to make their decision on who to pre-fund based on their own ability to evaluate technical proposals.
[...]
And here's the thing, remember, the less risk you take the less reward you get.
[...]
So basically people want ASIC makers to shoulder all the risk, and then charge reasonable prices that allow miners to get ROI are living in fantasy land.

I couldn't agree more.
I've done pre-orders before and they worked out well for me so far.
Been burnt on some, made a killing on others.

I'd love to see an opportunity to pre-order a 400 Gigahash chip from anyone as long as I believe in their capability to deliver.
And you damn well bet I'll do my very best to find out just how credible their offer is.

Yeah, well... I'm gonna go build my own blockchain, with blackjack and hookers. In fact, forget the blockchain!
AMD FTW
Sr. Member
****
Offline Offline

Activity: 319


GET IN - Smart Ticket Protocol - Live in market!


View Profile
July 26, 2013, 03:55:13 PM
#91

From reading this, it seems that this 400 GH/s unit takes 400 watts to run. This makes me think that this isn't one chip, but multiple chips on a board. Otherwise this one chip would take 400 watts which is pretty ridiculous and would need a pretty big footprint to allow that much heat into a heatsink. Don't forget most cpu's are under 130 watts and most gpu boards are under 250 watts.

GUTS | GET-Protocol ICO
smart-ticket protocol for events | live product with market traction!
BTC ANN | WEBSITE | BLOG | SANDBOX | WHITEPAPER | BOUNTY
Ytterbium
Full Member
***
Offline Offline

Activity: 238



View Profile WWW
July 26, 2013, 04:01:04 PM
#92

From reading this, it seems that this 400 GH/s unit takes 400 watts to run. This makes me think that this isn't one chip, but multiple chips on a board. Otherwise this one chip would take 400 watts which is pretty ridiculous and would need a pretty big footprint to allow that much heat into a heatsink. Don't forget most cpu's are under 130 watts and most gpu boards are under 250 watts.

KnC is anticipating at least 250 Watts per chip, and they have a heatsink that's rated for 320 watts, I think.  This company may use water cooling, there's self-contained, off the shelf water cooling equipment you can get for PCs that's pretty cheap now (a couple hundred bucks per unit).  KnC was going to do use it, before data centers told them they didn't want water in their datacenters. 

If that's not a concern for Fasthash then water cooling could be an obvious solution to the thermal issues.

Anyway, their announcement was that it would be significantly less then 400 watts. So this isn't really an issue.

Bitcoinorama
Hero Member
*****
Offline Offline

Activity: 532



View Profile
July 26, 2013, 06:18:38 PM
#93

From reading this, it seems that this 400 GH/s unit takes 400 watts to run. This makes me think that this isn't one chip, but multiple chips on a board. Otherwise this one chip would take 400 watts which is pretty ridiculous and would need a pretty big footprint to allow that much heat into a heatsink. Don't forget most cpu's are under 130 watts and most gpu boards are under 250 watts.

It's 400Gh/s per chip. Re-read the Uniquify letter. It's in there...

Make my day! Say thanks if you found me helpful Smiley BTC Address --->
1487ThaKjezGA6SiE8fvGcxbgJJu6XWtZp
jspielberg
Sr. Member
****
Offline Offline

Activity: 406


View Profile
July 26, 2013, 06:45:01 PM
#94

I am going to get on my soapbox for a sec... so prepare for a bit of suds  Grin

Hopefully they price it in BTC rather than some fiat denominated BTC.

First we are trying to promote an economy of buying things with BTC.  This does put some risk on the manufacturer, but if we want the BTC economy to successful, we all need to take some risks.  Miners are not averse to risks so we participate in pre-buys.

Machines that produce BTC should also be denominated in BTC.  People who buy miners are buying to make more BTC. If the machine doesn't make more BTC than it costs... then it is clearly a bad investment. That is, the purchaser could have just purchased BTC outright and been ahead.

Pricing in USD or EUR or whatever, muddies the water on the calculation of ROI, but ultimately for miners, the calculation is easy in BTC:
 (BTC in) > (BTC out) == bad deal, should have just bought BTC and held it
 (BTC in) < (BTC out) == good deal, buy buy buy

From a practical perspective, if folks are worried about trying to get BTC lined up before sales become available, Hashfast can communicate before hand what the pricing will be by a week or so.  Then buyers can get their ducks in a row.  Avalon Batch 3 gave less than a day which prevented some from being able to participate in the sale.

lajz99
Hero Member
*****
Offline Offline

Activity: 756


View Profile
July 26, 2013, 06:47:14 PM
#95

What a bunch of nonsense.  You don't know how to figure out what BTC was valued at when you purchased the machine?  IMO the more payment options, the better.

I am going to get on my soapbox for a sec... so prepare for a bit of suds  Grin

Hopefully they price it in BTC rather than some fiat denominated BTC.

First we are trying to promote an economy of buying things with BTC.  This does put some risk on the manufacturer, but if we want the BTC economy to successful, we all need to take some risks.  Miners are not averse to risks so we participate in pre-buys.

Machines that produce BTC should also be denominated in BTC.  People who buy miners are buying to make more BTC. If the machine doesn't make more BTC than it costs... then it is clearly a bad investment. That is, the purchaser could have just purchased BTC outright and been ahead.

Pricing in USD or EUR or whatever, muddies the water on the calculation of ROI, but ultimately for miners, the calculation is easy in BTC:
 (BTC in) > (BTC out) == bad deal, should have just bought BTC and held it
 (BTC in) < (BTC out) == good deal, buy buy buy

From a practical perspective, if folks are worried about trying to get BTC lined up before sales become available, Hashfast can communicate before hand what the pricing will be by a week or so.  Then buyers can get their ducks in a row.  Avalon Batch 3 gave less than a day which prevented some from being able to participate in the sale.

Xian01
Legendary
*
Offline Offline

Activity: 1316


󮀓1刎


View Profile
July 26, 2013, 06:54:18 PM
#96

Is Bob Smith really his name ?! The only thing that would be a bigger red flag is if he signed his name John Doe...
DPoS
Sr. Member
****
Offline Offline

Activity: 462



View Profile
July 26, 2013, 07:19:32 PM
#97

I appreciate what you are saying, but the more legitimate offerings want to build a company and be around for the longer term. Not get as much cash as possible, perform a lacklustre effort on production subsequent to raising funds and then disappear. The incentive is establishing a customer base and repeat business. If you upset your investors they tend not to reinvest. There's not too many that would chance Butterfly Labs a second time around after the ordeal they have suffered.

With that in mind any VC will not necessarily be concerned about the long term customer satisfaction if it's to fulfil one run, but when you have customers funds, and they are relying on you, if you deliver, you almost certainly have established repeat custom...

Bitfury still requires pre-orders, but he has performed an outstanding job of standing on his own two feet and proving his design delivers! Every company that takes a step in the right direction is treading in virgin ground here and is creating benchmarks future competitors are expected to meet as a necessity.

I actually prefer the pre-order model.  I mean, it allows savvy investors to make their decision on who to pre-fund based on their own ability to evaluate technical proposals.  For example, I decided to go with Avalon after they'd managed to at least demo a working unit (proving they weren't total vaporware).  And so far it's paid off great.

That also means of course I'm not in the position of all these people who have been completely burned by BFL and loath the idea of pre-orders.

The way KnC presented themselves made me think that they'd be a good deal as well.

And here's the thing, remember, the less risk you take the less reward you get.  Look at the price difference between batch 1, batch 2 and batch 3 Avalons. It went from $1,200 to $1,500 to something like $7k. People buying batch 1 were taking a blind risk, people buying batch 2 knew Avalon could at least ship demo units. And people buying batch 3 knew that Avalon could deliver (at least most of their orders) in a somewhat timely manner.

The lowest risk are things like USB erupters.  You know you'll get your unit and you'll get it on time. But the ROI is basically not going to happen unless either bitcoins go way up or just ignore the electrical costs (which is easy when it's probably only a few dollars a month)

KnC's prices were ridiculously cheap compared to their competition, there's a reason for that: if they made their offer low enough that profitability would basically be guaranteed if they delivered then they would get funded for sure. If they charged high prices, people not buy in - and they'd not get funded.

So basically people want ASIC makers to shoulder all the risk, and then charge reasonable prices that allow miners to get ROI are living in fantasy land.

I'm just hoping KnC keeps rewarding it's initial customers, rather then jacking up their prices in the future.


you should copy and paste that automatically in every thread once a week since 90% of the rants ignore all of it

~~BTC~~GAMBIT~~BTC~~Play Boardgames for Bitcoins!!~~BTC~~GAMBIT~~BTC~~ Something I say help? Donate BTC! 1KN1K1xStzsgfYxdArSX4PEjFfcLEuYhid
jspielberg
Sr. Member
****
Offline Offline

Activity: 406


View Profile
July 26, 2013, 07:24:11 PM
#98

What a bunch of nonsense.  You don't know how to figure out what BTC was valued at when you purchased the machine?  IMO the more payment options, the better.

Hmm... I was hoping for something a bit more meaty for a rebuttal. 

I agree that multiple payment options could/should be available... but the PRICE should be in BTC.
The machine makes BTC... if more comes out than you put in... you win!
If less comes out... you should have bought BTC directly.

Seems like a straight forward/simple concept...  take 5 minutes to think on it.

nightengale
Hero Member
*****
Offline Offline

Activity: 560


View Profile
July 26, 2013, 07:48:09 PM
#99

IMHO, this won't be much of an announcent until they release pricing information. If they have a 400 Gh/s chip and end up charging ASICminer prices for it, it obviously ends up being cost prohibitive. But if they're able to significantly beat KnC's $/SHA, we may have something here. I am encouraged by what they say about helping to secure the network. I'm definitely watching...
Damnsammit
Sr. Member
****
Offline Offline

Activity: 406



View Profile
July 26, 2013, 08:27:58 PM

The machine makes BTC... if more comes out than you put in... you win!
If less comes out... you should have bought BTC directly.

Seems like a straight forward/simple concept...  take 5 minutes to think on it.

I don't know how people fail to understand this...

If someone was selling a machine that only had one function and that function was to print out $100 dollar bills, would you buy it for $500?

what if I told you it could only print out four bills before it self-destructed?  Would you still buy it for $500?
Pages: « 1 2 3 4 [5] 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 ... 586 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!