Bitcoin Forum
November 22, 2017, 08:17:40 AM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 ... 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 [179] 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 ... 2138 »
  Print  
Author Topic: Swedish ASIC miner company kncminer.com  (Read 3011309 times)
Mota
Legendary
*
Offline Offline

Activity: 800


View Profile
July 13, 2013, 08:31:14 PM
 #3561

if what you're saying is true, and this other site is real, then hats off to bitfury, but ill stick with knc..... I'm skeptical of the post, because its not on bitfury's site even. bitfury.org doesn't mention it.

Bitfury uses a "reseller" type system, they don't sell direct.  Here is the thread where US sales were conducted:
https://bitcointalk.org/index.php?topic=251966.0

EU Sales:
https://bitcointalk.org/index.php?topic=250249.0


Like KNC, they are presales, so all buyers be warned.


EDIT: Bitfuy does post directly in those threads to validate them as official resellers.  FYI.
With all respect to bitfury for what he made, nut do you really want to compare a product which costs more than triple the price for power comsumption? I mean, if KNC went for the same price you could compare them, but try to get how long it would take KNC to break even in total cost. 15k in power cost is more than 10 years running continually in Germany... And we have one of the highest power costs worldwide....

Who wants to be a billionaire? Me!
You could help Wink 1Dvja1RFCqxdnYRgjTntwGvdCeUisU4xp
1511338660
Hero Member
*
Offline Offline

Posts: 1511338660

View Profile Personal Message (Offline)

Ignore
1511338660
Reply with quote  #2

1511338660
Report to moderator
1511338660
Hero Member
*
Offline Offline

Posts: 1511338660

View Profile Personal Message (Offline)

Ignore
1511338660
Reply with quote  #2

1511338660
Report to moderator
1511338660
Hero Member
*
Offline Offline

Posts: 1511338660

View Profile Personal Message (Offline)

Ignore
1511338660
Reply with quote  #2

1511338660
Report to moderator
Join ICO Now A blockchain platform for effective freelancing
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
titomane
Sr. Member
****
Offline Offline

Activity: 389



View Profile
July 13, 2013, 08:40:35 PM
 #3562



This is the best piece of info in this entire discussion.  Unfortunately Fabs almost never release their defect density and yields, unless you are a customer under a Non-Disclosure Agreement.  In almost two decades following CPU/IC production I have never willingly seen a fab release this info.  Very rarely you can get it leaked from a former employee, but then it is always tainted and uncertain if that person has an axe to grind with his former employer or not.  Alternatively, sometimes, you will get vague impressions from the executives during financial disclosures, but never the detailed info you want.

Now, the only thing apparently known about KNC is that the package size is about 3000 mm2.  Let's assume I am wrong, would not be the first time, and that they have a relatively small chip for the package size.  Say 15-20%.  That would be a die size of 450-600 mm2.  That would still be BELOW the lowest curve on the above diagram.  That's really scary for yields, but not impossible to obtain working product by history.  But any customer with a die size that large would have to expect very low yields, and they probably had to sign a waiver from the foundry about guaranteed yield due to "design constraints".

Here are some historical data points for 3rd party fab yield problems with some of the larger GPU dies, for bigger customers:
http://www.geek.com/games/ati-and-nvidia-have-troubling-40nm-yields-from-tsmc-815501/  which were "fixed" a year later:
http://www.dailytech.com/TSMC+Says+40nm+Problems+Resolved+Preparing+28nm+Fab+Production+/article17355.htm
http://www.eetimes.com/document.asp?doc_id=1261006  which were "fixed" about 10 months later:
http://www.pcmag.com/article2/0,2817,2411985,00.asp

Now, a TSMC/UMC is going to work with an AMD/NVidia to improve yields because they are producing 10-100s of millions of chips.  For a smaller companies, they will not go to great lengths to help them revise designs and processes to improve yields.  You get what you get.

Traditionally GPU dies have been some of the largest dies, with the high-end (low yield) parts topping out around mid-500 mm2 in size.  And the parallel structure of a GPU is what most closely resembles the SHA256 ASICs being designed and produced by companies today (Avalon, BFL, KNC, BitFury, etc.).  So from a manufacturing standpoint, we can probably infer a lot from large ASIC production issues by looking at issues that GPU manufacturers have had in the past.

The bigger the dies, the lower the yield, the more you have to charge to cover your costs on the wasted dies from a wafer.  Therefore it is always good to make your dies as small as possible.  Just 10 mm2 per die, when factored over an entire wafer, can mean tens of thousands of dollars in improve yield per wafer.

For comparison, here is a good illustration from the 45 nm days, with Intel - who historically have much higher yields than the rest of the industry (the invest more in their process and refinement).



Again, I hope KNC brings to market a good product, and in a timely fashion.  The more competition, the better we all do in terms of up-front pricing for our miners.  I just am concerned that we haven't see any substantial data about the processor (number of SHA256 cores, clock frequency, die size, heat production, etc.).

Considering the 10+ million dollars of orders they have had, you would hope they would publish this info to keep their investors/customers happy.

You draw conclusions without knowing the chip structure. You assume will have only one die, when is can have 2,3 or 4.

Only a pedant think they have the absolute truth and the others are fanboys.
You must be the first person who is always right. Although chances are that you just believe. Roll Eyes
Remember CMOS sensors die size are biggers

dwdoc
Legendary
*
Offline Offline

Activity: 966


- - -Caveat Aleo- - -


View Profile
July 13, 2013, 08:43:54 PM
 #3563

Let's face it, until KNC provides some die shots followed in relative short fashion by a hashing video there will be a lot of teeth gnashing and second guessing regarding their design choices...
bkpduke
Full Member
***
Offline Offline

Activity: 158



View Profile WWW
July 13, 2013, 08:53:53 PM
 #3564

if what you're saying is true, and this other site is real, then hats off to bitfury, but ill stick with knc..... I'm skeptical of the post, because its not on bitfury's site even. bitfury.org doesn't mention it.

Bitfury uses a "reseller" type system, they don't sell direct.  Here is the thread where US sales were conducted:
https://bitcointalk.org/index.php?topic=251966.0

EU Sales:
https://bitcointalk.org/index.php?topic=250249.0


Like KNC, they are presales, so all buyers be warned.


EDIT: Bitfuy does post directly in those threads to validate them as official resellers.  FYI.
With all respect to bitfury for what he made, nut do you really want to compare a product which costs more than triple the price for power comsumption? I mean, if KNC went for the same price you could compare them, but try to get how long it would take KNC to break even in total cost. 15k in power cost is more than 10 years running continually in Germany... And we have one of the highest power costs worldwide....

Ah, there is a failure in your logic.

Bitfury's price CHANGES depending upon the delivery date.  It's LESS than half if you take delivery in October, vs. in August.
bkpduke
Full Member
***
Offline Offline

Activity: 158



View Profile WWW
July 13, 2013, 08:56:59 PM
 #3565

You draw conclusions without knowing the chip structure. You assume will have only one die, when is can have 2,3 or 4.

That's my point, dum@$$, KNC have given us the LEAST amount of data of any of the manufacturers.

Give me some more data, I'll happily revise my conclusions based upon the data.

I don't make decisions based upon hopes and dreams and the like.  Just upon the numbers.
titomane
Sr. Member
****
Offline Offline

Activity: 389



View Profile
July 13, 2013, 09:13:43 PM
 #3566

You draw conclusions without knowing the chip structure. You assume will have only one die, when is can have 2,3 or 4.

That's my point, dum@$$, KNC have given us the LEAST amount of data of any of the manufacturers.

Give me some more data, I'll happily revise my conclusions based upon the data.

I don't make decisions based upon hopes and dreams and the like.  Just upon the numbers.


When I show you pictures of chips erroneous calculations. Trying to show the difference between package and DIE again and again.


XBOX-WII & PS3 CPU's are BGA soldered. Their DIE size are less than 15 & 17% of package size.
XBOX
http://pictures.xbox-scene.com/xbox360/cpu/CPU.jpg

That's a 30-35% die vs. package size right there (the RAISED board is the "package" - the rectangle on it is the DIE).


I imagine without information, your calculations should be worse.

CMOS SENSORS   more than 750mm2 DIE size

DPoS
Sr. Member
****
Offline Offline

Activity: 462



View Profile
July 13, 2013, 11:06:41 PM
 #3567

You draw conclusions without knowing the chip structure. You assume will have only one die, when is can have 2,3 or 4.

That's my point, dum@$$, KNC have given us the LEAST amount of data of any of the manufacturers.

Give me some more data, I'll happily revise my conclusions based upon the data.

I don't make decisions based upon hopes and dreams and the like.  Just upon the numbers.

and no one is beating a path to convince you either...  why do you think so?

You are a time suck.... why don't you just go all in with bitfury and be done with it?

~~BTC~~GAMBIT~~BTC~~Play Boardgames for Bitcoins!!~~BTC~~GAMBIT~~BTC~~ Something I say help? Donate BTC! 1KN1K1xStzsgfYxdArSX4PEjFfcLEuYhid
Sugarman
Newbie
*
Offline Offline

Activity: 5


View Profile
July 13, 2013, 11:17:20 PM
 #3568

Correct me if i am wrong,

but wouldn't not having to pay the VAT on 1 single jupiter cover half of the 6 month term (depending on your local VAT)?
kano
Legendary
*
Offline Offline

Activity: 2282


Linux since 1997 RedHat 4


View Profile
July 14, 2013, 01:47:49 AM
 #3569

Just a little bit of info regarding KnC
They contact ckolivas and myself at the end of May regarding cgminer and related information.

They then stated they'd be sending (in July) mock up devices (with an internal RPi) for us to work with on cgminer, then in September the real (faster) devices to tune cgminer to.

I have, however, heard nothing since, but no doubt expect to hear something soon.

I'll keep everyone posted if anything happens Smiley
I've been waiting 5 days for a reply (to an email I sent Marcus and Sam) from KnC about their status ... still no response ...

Pool: https://kano.is Here on Bitcointalk: Forum BTC: 1KanoPb8cKYqNrswjaA8cRDk4FAS9eDMLU
FreeNode IRC: irc.freenode.net channel #kano.is Majority developer of the ckpool code
Help keep Bitcoin secure by mining on pools with full block verification on all blocks - and NO empty blocks!
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 02:25:18 AM
 #3570

Anyone considering their hosting option? Seems like a large amount of money to front coupled with the fact that you need to accept a 6 month contract...
For me it depends on the difficulty when my unit becomes available, the USD price of BTC at that time, as well as whether they will accept payment in BTC. If difficulty is way high and the profit margin is slim, I can run it less expensively at home, but would prefer it run there b/c of the guaranteed uptime. If they do come up with optimizations, I'm assuming they will make them available to everyone and not just hosted units...
I assume they'll be using something like Bitpay to let you pay them, naturally. What kind of asic seller wouldn't accept bitcoin on top of other options.

Democracy is the original 51% attack.
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 02:30:27 AM
 #3571

one question
if u choose hosting u don't pay shipping, electricity and VAT 
(u will pay VAT when u decide to import hw in ur country, right?)
Almost assuredly it includes electricity, and you might not even need to pay VAT if they don't ship it out? Dunno.

I priced out hosting at a major California data center, they wanted $330 a month (electricity included), and that was with me admining it, and twice that for the first month's setup.

As for KNC, they told me in an email that we would be able to switch from delivery to hosting, and that the cost we've paid upfront for delivery will be applied to the hosting cost initially. I assume this means the $130+ we paid will be used to buy the power supply. Then we'd be billed at the end of the month for the hosting cost with a choice of options, and likely a good $50 or so extra for the cost of the power supply.

Looking like an attractive option to me. Imagine if something broke down in your machine and you had to send it back? Weeks lost. Imagine too if they are overclocking it for you in their datacenter, and they're the pros on the machine!

Democracy is the original 51% attack.
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 02:39:12 AM
 #3572

We need to ask support for a hosting plan for the Saturns.

says right in news release that KNC will only host Jupiters

They take feedback just like they have with hardware im sure they can figure a solution if enough people ask for it. Doesnt seem to hard the same space is going to be taken up with less power.

Cant hurt to ask, but that is their published response right now

I know i asked them anyway, i plan on upgrading my saturn to a jupiter anyway within the first month, dont see the logic behind not hosting saturns or mercuries if people want to pay for it.
For one thing, they take up the same amount of space as a Jupiter but produce half the output, and datacenters are charging you by the cabinet. So...

Democracy is the original 51% attack.
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 02:41:33 AM
 #3573

Just as I thought... 2 gh/s per chip, about 50x slower, you had me there
:| you don't really know what a 'wafer' is in chip fabbing, do you? It's the large pizza-sized single-crystal of substrate (silicon) all the chips are lithographed onto before being cut into individual chips.

Democracy is the original 51% attack.
pajak666
Hero Member
*****
Offline Offline

Activity: 497



View Profile
July 14, 2013, 02:44:51 AM
 #3574

As for KNC, they told me in an email that we would be able to switch from delivery to hosting, and that the cost we've paid upfront for delivery will be applied to the hosting cost initially. I assume this means the $130+ we paid will be used to buy the power supply. Then we'd be billed at the end of the month for the hosting cost with a choice of options, and likely a good $50 or so extra for the cost of the power supply.

Looking like an attractive option to me. Imagine if something broke down in your machine and you had to send it back? Weeks lost. Imagine too if they are overclocking it for you in their datacenter, and they're the pros on the machine!

Yeah and imagine all the decentralization gone. Imagine a scenario that one day bitcoin becomes major problem for anyone with bad intentions, much money and power in his hand, they are very desperate to bring bitcoin down. What would be smart move? destroy or somehow make some mess at KNC hosting datacentre and at some other major network hasher like ASICminer, what will happen then? due to lack of hashing power and high difficulty it will take ages for any transaction to confirm. Taaa Daaa here goes down your super hosting option.

        ▄██████████████    █████████████▄
    ▄██▀                                                       ▀██▄
   ██▀                                                              ▀██
 ██                                                                     ██
██                                                                       ██
██                                                                       ██
██                                                                       ██
██                                                                       ██
 ██▄                                                                   ██
  ▀██▄                                                            ▄██▀
     ▀██▄▄                                                    ▄██▀
          ▀█████████████   █████████████▀▀
                                         ██
Cash Poker Pro
   ♠    ♦    ♣    ♥ 
              ▒▓██ ░░░ ▓█▓▒             
           ▓██████     ██████▓         
        ░▓███▓▓███ ░░░ ▓███▓████▒       
       ▓████▓▓████████████▓██████▓     
      ░░░▓▓█████▓▓█▒▒▒██▓█████▓▒░░░     
          ░███▓▒▒▓██████▒░▒███▒         
    ▓▒   ░██▓▓█████████████▓▒██▒   ░▓   
    ███▓▒██▓░▓████      ▓███░▒██▒▒███▒ 
   ▓███████▓██████  ██▓▓█████▓████████ 
   ███▓███▓▒██████      █████▒▓██▓▓▓██ 
   ▓██▓███▓░██████████  █████▒▓██▓▓▓██ 
   ▓███████▓▓█████      █████▓███████▓ 
    ███▒▒██▓░▓██████████████░▓██▒▒███░ 
    ▒▒   ░██▓▓█████████████▓▓██▒   ░▒   
      ░   ░███▓░▒▓█▓▓▓██▒░▓███▒   ░     
       ░▒▓██████▓██▒▒▒██▓██████▓▒░     
       ▒████▓▓▓███████████▓▓▓████▓     
         ▓██▓▓▓███  ░░ ▓██▓▓███▓░       
           ▒██████     ██████▓         
              ░▒▓█ ░░░ ▓█▓░             
Anonymity Fair Play Cryptocurrency New Platforms
А platform that will change the world of poker!
               ▄▀▀▀▀▀▄▄▄▄
           ▄▄▄▄█  ▄█     ▀▀▀▀▄▄▄▄
▄▄▀▀▀▀▀▀▀▀▀   █  ▀ █            ▀▀▀▀▄
█  █▄        ▄▀                    ▄▀
▀▄ █ ▀       █         ▄           █
 █          █        ▄██▄         ▄▀
 █         ▄▀     ▄██████▄       ▄▀
 ▀▄        █    ▄█████████▄      █
  █       █      ██████████▄    ▄▀
  █      ▄▀        ▀████████   ▄▀
  ▀▄     █        ▄████▀▀▀▀    █
   █    █          ▀▀▀█       ▄▀
   █   ▄▀                    ▄▀
   ▀▄  █                     █
    █  ▀▄▄              █ ▄ ▄▀
    █     ▀▀▀▀▄▄▄       █▀ ▄▀
    ▀▄▄▄▄▄▄▄▄▀▀▀▀▀▀▀▀▀▀▄▄▄▄█
  ♠    ♦    ♣      JOIN US!
ICO Starts: Oct 26th 2017 
JohnyBigs
Sr. Member
****
Offline Offline

Activity: 364


View Profile
July 14, 2013, 02:45:00 AM
 #3575

We need to ask support for a hosting plan for the Saturns.

says right in news release that KNC will only host Jupiters

They take feedback just like they have with hardware im sure they can figure a solution if enough people ask for it. Doesnt seem to hard the same space is going to be taken up with less power.

Cant hurt to ask, but that is their published response right now

I know i asked them anyway, i plan on upgrading my saturn to a jupiter anyway within the first month, dont see the logic behind not hosting saturns or mercuries if people want to pay for it.
For one thing, they take up the same amount of space as a Jupiter but produce half the output, and datacenters are charging you by the cabinet. So...

Maybe some saturn owners still wouldnt mind paying the cost
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 02:48:42 AM
 #3576

BS. You are totally lying now. Welcome to the list of trolls.(bdpuke)
Bitfury's 110gh rack is HUGE, and consumes 10 kilowatts!!!!
http://www.bitfury.org/bitfury110.html


Obviously you haven't kept up much.

That's a rack of FPGAs, not ASICs.  It's over 2 years old.

This is what they are actively selling:
https://megabigpower.com/shop/index.php?route=product/product&path=59&product_id=50
Doesn't it seem ridiculously overpriced with KNC's offering on the table?

Democracy is the original 51% attack.
nmat
Hero Member
*****
Offline Offline

Activity: 602


View Profile
July 14, 2013, 02:55:38 AM
 #3577

Maybe some saturn owners still wouldnt mind paying the cost

They probably don't have enough space or enough people to manage lots of devices. They could hire a big team and rent a big space, but they probably want to focus on production instead.

EDIT: Actually, there's a poll right now to vote for this feature in their frontpage, so they might allow it if people ask for it.
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 02:57:35 AM
 #3578

As for KNC, they told me in an email that we would be able to switch from delivery to hosting, and that the cost we've paid upfront for delivery will be applied to the hosting cost initially. I assume this means the $130+ we paid will be used to buy the power supply. Then we'd be billed at the end of the month for the hosting cost with a choice of options, and likely a good $50 or so extra for the cost of the power supply.

Looking like an attractive option to me. Imagine if something broke down in your machine and you had to send it back? Weeks lost. Imagine too if they are overclocking it for you in their datacenter, and they're the pros on the machine!

Yeah and imagine all the decentralization gone. Imagine a scenario that one day bitcoin becomes major problem for anyone with bad intentions, much money and power in his hand, they are very desperate to bring bitcoin down. What would be smart move? destroy or somehow make some mess at KNC hosting datacentre and at some other major network hasher like ASICminer, what will happen then? due to lack of hashing power and high difficulty it will take ages for any transaction to confirm. Taaa Daaa here goes down your super hosting option.
Meh, the difficulty is just over ~24 million. It can change by a factor of 4 every 2 weeks. So it could go down to 6 million. Not really that bad at all. It's a risk, to be sure, but so is hosting at home or in a local data-center. It's all a risk. If returns were guaranteed this would be a loan or a bond, not an investment, and the return would be correspondingly low.

Democracy is the original 51% attack.
Anenome5
Sr. Member
****
Offline Offline

Activity: 252



View Profile
July 14, 2013, 03:06:50 AM
 #3579

More exciting to me is that they offer 6 month lease. I think my local datacenter wanted a 36 month lease! That's clearly not bitcoin-mining friendly.

But that also implies something else. If KNC gets the Jupiter out the door on time they will have proved they possess a development pipeline that no one can match and they instantly become the market leader ASIC provider.

I expect that by the time the Jupiters actually ship, assuming they ship on time, Knc will have begun work on their gen-2 device in the terrahash range, to be announced shortly after their first Jupiters ship, perhaps a month after, with a release date in another 4-6 months.

That will likely crush the Avalons and BFL rigs of the world, possibly even put pressure on AsicMiner, since they aren't the most efficient miner out there cost-wise, no doubt. Bitfury's design would probably withstand a gen-2 assault somewhat well, being the market leader efficiency-wise currently, but I'd expect a gen-2 to beat the Bitfury chip handily on that metric too.

If Knc can produce a gen-2 device in another 4 months after shipping, all of us will be able to afford to upgrade, most likely. And then if they produce a gen-3 device in another 4 months, they will really have blown the competition out of the water.

One of the reasons I think Knc chose a large package size is that it's a lot easier, cheaper, quicker, to solder four large chips on a board and ship the device than do deal with multiple daughter-boards and PCBs, like an Avalon has. That leads to production and shipping delays.

In fact, it's probably more accurate to think of KNC's first device as a gen-2 device already, though it's their first product. Knc and Bitfury I'd put in the gen-2 category already, with Avalon and BFL being gen-1 (even if Bfl has hardly shipped).

Hard to say what the parameters of a gen2/3 device will look like, what it's specs will be. Personally I'd love to see blade-style devices designed for datacenter installment, ala AMiner. Knc may yet do that with these devices. We shall see.

Democracy is the original 51% attack.
ASIC-K
Sr. Member
****
Offline Offline

Activity: 280


Hell?


View Profile
July 14, 2013, 03:42:42 AM
 #3580

we should just lock this once useful thread, and start a new one.
Pages: « 1 ... 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 [179] 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 ... 2138 »
  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!