Pages:
Author

Topic: First commercial ASIC miner specifications and pre-launch (Read 29365 times)

newbie
Activity: 7
Merit: 0
They should be sending out free prototype chips like other companies which develop new technologies do!

LOL, that'll be the day.  Try-before-you-buy dedicated mining hardware.
sr. member
Activity: 373
Merit: 262
They should be sending out free prototype chips like other companies which develop new technologies do!
hero member
Activity: 527
Merit: 500
Anyone got refunds in the meantime?
hero member
Activity: 840
Merit: 1000
Soo... Now i'm wondering how many people have 'invested' in this guy.

Anyone dare to speak up?
 Cool
hero member
Activity: 686
Merit: 564
Which was why I was asking him what part did he used since I expected there would be at least two trip up. The first being unless there's some really super fast breakthrough Flash chip, it would be pretty much damning evidence regardless of his answer Cheesy
Aha - cunning!
newbie
Activity: 42
Merit: 0
There's no reason why a specially-designed Bitcoin mining board would need an SRAM chip; SRAM is large and slow and a Bitcoin miner needs fast access to a small amount of data. This wasn't even a competent scammer...

Which was why I was asking him what part did he used since I expected there would be at least two trip up. The first being unless there's some really super fast breakthrough Flash chip, it would be pretty much damning evidence regardless of his answer Cheesy
hero member
Activity: 686
Merit: 564
                                                       
Use of Block RAM (BRAM) for storage of constants [8]. Reconfigurable hardware devices
such as FPGAs often have on-board memories which can be pre-loaded.
Storing the Kt constants in these memories frees up space in the device
which can then be used to implement extra logic. The free space also leads to
improved routing and, thus, a general speed-up in circuit operation.
                                                        
From "Optimisation of the SHA-2 Family of Hash Functions on FPGAs"
Robert P. McEvoy, Francis M. Crowe, Colin C. Murphy and William P. Marnane
             Department of Electrical & Electronic Engineering,
                      University College Cork, Ireland
             {robertmce, francisc, cmurphy, liam}@rennes.ucc.ie


As a person who's actually done an FPGA bitcoin mining design that uses block RAM (though not in exactly this way), I know that it's part of the FPGA itself and not a seperate chip. There's no reason why a specially-designed Bitcoin mining board would need an SRAM chip; SRAM is large and slow and a Bitcoin miner needs fast access to a small amount of data. This wasn't even a competent scammer...

Edit: Oh, and the reason it doesn't use block RAM in that particular way? Doing so is only worthwhile if you're designing a standard SHA-2 hash engine that computes the hashes of variable-length pieces of data one at a time, whereas efficient Bitcoin mining compute lots of hashes of differing short pieces of data in parallel.
member
Activity: 62
Merit: 10
Well if you try to call there, his phone is turned off cause of insuffcient funds...
i guess the HK Police Department will prosecute this, if you didnt get your refund aswell:
http://www.police.gov.hk/ppp_en/contact_us.html

Quote
Registration Service Provided By: EXOWARE
Contact: +44.3333408855

Domain Name: ASICMINER.NET

Registrant:
Shenzhen advanced electronics R&D Co. Ltd
Gerald Strobel ([email protected])
Room 823, Bd 144, Nathan Rd, Kowloon
Hong Kong
HK,00000
HK
Tel. +852606261520

Creation Date: 29-Jun-2011
Expiration Date: 29-Jun-2012

Domain servers in listed order:
ns3.exoware.net
ns4.exoware.net


Administrative Contact:
Shenzhen advanced electronics R&D Co. Ltd
Gerald Strobel ([email protected])
Room 823, Bd 144, Nathan Rd, Kowloon
Hong Kong
HK,00000
HK
Tel. +852606261520

Technical Contact:
Shenzhen advanced electronics R&D Co. Ltd
Gerald Strobel ([email protected])
Room 823, Bd 144, Nathan Rd, Kowloon
Hong Kong
HK,00000
HK
Tel. +852606261520

Billing Contact:
Shenzhen advanced electronics R&D Co. Ltd
Gerald Strobel ([email protected])
Room 823, Bd 144, Nathan Rd, Kowloon
Hong Kong
HK,00000
HK
Tel. +852606261520
sr. member
Activity: 461
Merit: 251
For anyone still in doubt, I can confirm that it is, indeed, a SCAM.
See his post on reddit:
Quote
We're currently about to close the sale of a whole first batch of devices to a private investor. We will be required not to duplicate the device for at least 3 years. Sorry guys. We will eventually refund all the people who sent the 5 btc within 48 hours.
And just last night he sent me an email:
Quote
Hello,

Thanks for taking the time to write us an email.
We will have a couple devices left to sell.

Entry level device with 2 boards price breakdown :

Controller device $150.00
ASIC board $250.00 x 2
Shipment $50.00
Total $700

Full device with 10 boards price breakdown :

Controller device $150.00
ASIC board $250.00 x 10
Shipment $50.00
Total $2700

The case will be similar in size to a mid-tower standard PC and you will
be able to purchase 8 additional boards if you decide to go for the entry
level machine now, sorry to inform you that, in that case, there will be
one more charge of $50.00 for shipping.

In case you decide to go ahead please provide details of your order, your
full name, address and valid phone number along with the 5 btc transfer to
1FyZ3qU54jxU5tF2KHxHmnyjkAsdWuiyFr
We will let you know when the production have actually started both via
email and our blog.

Best Regards
Asicminer team
Now why would he be refunding 5 BTC back to everyone if he still has a couple to sell??  Undecided  Anyone who gave him 5 BTC - you will never see it again.  So don't hold out any hope.
hero member
Activity: 807
Merit: 500
NO REFUND TILL THIS VERY MOMEN
Do you mean "so far" (as opposed to "until this very moment") in American English, or did you actually JUST get your refund when you posted?
member
Activity: 62
Merit: 10
newbie
Activity: 14
Merit: 0
Well asicminer if you dnt plan to deliver i ask you for imediate refund

Thanks


Openrune


did you ever get that refund?
newbie
Activity: 17
Merit: 0
i've also signed. this is really interesting to mine with special hardware -> less power
sr. member
Activity: 404
Merit: 250
I signed up for your list already!
member
Activity: 70
Merit: 10
Well, ASIC miner is closing its doors http://www.reddit.com/r/Bitcoin/comments/iynpe/asicminernet_calls_it_quits_wont_sell_product_for/

they claim they just sold a batch to an investor, im gonna go ahead and call bull and say the total network hash wont increase 2.5 terahashes..

i guess we can watch block explorer and find out Smiley

LargeCoin continues to plug away quietly, and it's not a scam. For more information, http://largecoin.com.
newbie
Activity: 42
Merit: 0
Well, ASIC miner is closing its doors http://www.reddit.com/r/Bitcoin/comments/iynpe/asicminernet_calls_it_quits_wont_sell_product_for/

they claim they just sold a batch to an investor, im gonna go ahead and call bull and say the total network hash wont increase 2.5 terahashes..

i guess we can watch block explorer and find out Smiley

Alright I admit it, I'm the one who bought his entire batch. But out of love for my fellow bitcoin miners and not wishing to drive away potential bitcoin users, I had decided that I will only put one machine into operation every 3 days, which is why in turn I required that they don't duplicate the machine in the next 3 years.
member
Activity: 154
Merit: 10
will mean will just take longer to get trough 21.000.000


I will keep going Wink...


 ;Ddig baby dig
newbie
Activity: 6
Merit: 0
Well, ASIC miner is closing its doors http://www.reddit.com/r/Bitcoin/comments/iynpe/asicminernet_calls_it_quits_wont_sell_product_for/

they claim they just sold a batch to an investor, im gonna go ahead and call bull and say the total network hash wont increase 2.5 terahashes..

i guess we can watch block explorer and find out Smiley
newbie
Activity: 18
Merit: 0
It could be totally true, there was a research paper from the Chinese University of Hong Kong about using Pilchard "RAM" boards for DES encryption in ECB mode back in 2004.
http://microsys6.engr.utk.edu/~bouldin/pilchard.pdf
http://www.cse.cuhk.edu.hk/~phwl/mt/public/archives/old/ceg5010/pilchard.pdf

And they probably wanted us to believe that this is the paper they used as inspiration:
http://www.ee.usyd.edu.au/people/philip.leong/UserFiles/File/papers/sha_fpl02.pdf

Problem is that this paper is pretty old, and the hardware they mention is outdated. If other people in this forum with much more modern FPGAs couldn't come up with a feasible implementation, there is no way that this could be competitive. This paper was written before GPGPUs became common.

And I couldn't find a more modern version of this Pilchard module.
full member
Activity: 154
Merit: 100

I'm glad this thread is over... unfortunately this isn't

Pages:
Jump to: