Bitcoin Forum
May 05, 2024, 03:44:44 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: DIY ASIC's...  (Read 4370 times)
ISAWHIM (OP)
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
April 27, 2013, 07:10:33 AM
 #1

Seems that the technology for these specific chips is nothing new...
http://www.cast-inc.com/ip-cores/encryption/sha-256/

Oh, look... a real company, with actual contact information, with actual ASIC products to sell... Who would have imagined?

Why "develop" your own chip, when they already exist...

Hmmm... I smell future lawsuits here... Can anyone say multi-national copyright infringement?

So, who's game for laying down some generic schematics for a standard USB interface that uses standard serial-usb single-connect chips?

Anyone?
TalkImg was created especially for hosting images on bitcointalk.org: try it next time you want to post an image
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
ISAWHIM (OP)
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
April 27, 2013, 07:19:50 AM
 #2

Or this chip... Does more than just SHA-256, with true hardware acceleration... Not just "execution", like the ASIC's being developed now.
http://www.st.com/web/catalog/mmc/FM143/CL1814/SC1537/PF252558

Instant expandability and PCIe ready!
ISAWHIM (OP)
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
April 27, 2013, 07:28:31 AM
 #3

Wait, can't our network cards also do sha-256 encryption internally... and our sound-cards... and our video-encoders...

Why stop at just GPU's... I am sure the processors also have dedicated sha-256 commands too... why are we trying to code this stuff manually, with raw-code software? The hardware is there already. Just because OLD code from CPU "failed programs" from "unknowledgeable programmers", were using copy-pasta code that had obvious limits...

Here is another multi-range CPU dedicated to ultra high-speed encryption designed for "gigabit bandwith" encryption. (Like our network cards also do.)
https://www.power.org/press-release/amcc-powerpc-460gtx-increases-performance-and-bandwidth-for-networking-and-telecommunication-applications/
Remember remember the 5th of November
Legendary
*
Offline Offline

Activity: 1862
Merit: 1011

Reverse engineer from time to time


View Profile
April 27, 2013, 07:30:21 AM
 #4

SHA256 != Double SHA256.

BTC:1AiCRMxgf1ptVQwx6hDuKMu4f7F27QmJC2
ISAWHIM (OP)
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
April 27, 2013, 07:42:03 AM
 #5

SHA256 != Double SHA256.

Um... There is no such thing as "double Sha256", it is still sha256... and solved with sha256. You could use three, and call it triple... but that doesn't make it TripleSha256... if it was "double", it would be SHA512... Thus... stop applying "laymens terms" for post-process encryption methods that have no bearing on the actual processing/hashing component.

Tongue

I'm talking out my ass... Just throwing dust into the wind, to see if I can create a dust-storm from a hand-full of nothing.

Is that like double 128, because we do it twice, makes it 256? (Dur, hum, dum)

Someone came from the /b/ boards, right into the stereotypical pool of copy-cats... Trend much?
Remember remember the 5th of November
Legendary
*
Offline Offline

Activity: 1862
Merit: 1011

Reverse engineer from time to time


View Profile
April 27, 2013, 08:01:33 AM
 #6

SHA256 != Double SHA256.

Um... There is no such thing as "double Sha256", it is still sha256... and solved with sha256. You could use three, and call it triple... but that doesn't make it TripleSha256... if it was "double", it would be SHA512... Thus... stop applying "laymens terms" for post-process encryption methods that have no bearing on the actual processing/hashing component.

Tongue

I'm talking out my ass... Just throwing dust into the wind, to see if I can create a dust-storm from a hand-full of nothing.

Is that like double 128, because we do it twice, makes it 256? (Dur, hum, dum)

Someone came from the /b/ boards, right into the stereotypical pool of copy-cats... Trend much?
And I'll say it again Double SHA256 != DoubleSHa256. A Bitcoin block is a hash of the hash of the block header that is lower than the target difficulty.

BTC:1AiCRMxgf1ptVQwx6hDuKMu4f7F27QmJC2
fpgaminer
Hero Member
*****
Offline Offline

Activity: 560
Merit: 517



View Profile WWW
April 27, 2013, 08:06:56 AM
 #7

This has been brought up many, many times before.  While everyone appreciates new, and interesting topics, and I highly encourage fresh discussion, please search the forums for existing topics before posting.

Quote
Why "develop" your own chip, when they already exist...
Existing SHA-256 IP cores and implemented silicon do not have the performance characteristics necessary to compete with existing GPU and FPGA based solutions.  Bitcoin mining requires a specialized design that did not, until recently, exist.

Quote
with actual ASIC products to sell...
Though they may very well sell ASICs, the particular page that you linked to is for an IP core.  An IP core is not a physical device; it is a design that could be implemented in silicon.  Even if the IP core were suitable for Bitcoin mining (it is not), it would still require a large capital investment to produce ASICs from it.

fghj
Member
**
Offline Offline

Activity: 65
Merit: 10


View Profile
April 27, 2013, 08:29:23 AM
 #8

I think this post is relevant:
https://bitcointalk.org/index.php?topic=167571.msg1752459#msg1752459
Yes it's 5Mh/s with M
ISAWHIM (OP)
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
April 27, 2013, 12:18:51 PM
 #9

And I'll say it again Double SHA256 != DoubleSHa256. A Bitcoin block is a hash of the hash of the block header that is lower than the target difficulty.

Doing it twice, is hardly a fundamental deal-breaker...

That is like saying you can't multiply, because this uses double-multiply... C*(B*X) OMG new technology.. we can't just SHA-256 it twice... Seriously...

Which link was "not a chip"... The first link IS a chip, and a process. None of those were, "just ideas", they are used in cryptographic hardware.

ASIC's that "only do SHA-256" are new... because no-one would limit themselves to that specific of an insecure technology, when just adding a few more commands, in silicon, has no penalty.

I only linked those, of the thousands I found, which interested me the most. The point was, that they already have patents and rights, to much of the hardware which, in part, is being used on the ASIC's being developed now. Items that could already be in users hands, instead of waiting for all these potential devices to actually be "created" from scratch. (Which, it is safe to assume, may even all fail, or be insufficient to adapt to any form of "modified" bitcoin code, which will eventually be made. As it stands, these potential "cracking" ASICs coming to our hands, are going to be used for that, at some point. They have a brute-force ability above any GPU, and equal to a super-computer, for such a modest SHA-256 encryption. Which, by the way, is not hard to "fake" valid checks, when you simply have to modify the package until it results in an equivalent HASH. That is how viruses inject themselves into system-files, and avoid detection after injection. They just alter insignificant values until a valid hash is returned.)

This system will only last so long, before it has to be, "trimmed" and re-secured with more non-standard encryption. Don't forget, "crackers" don't have to do a "proof of concept", they only have to bash until they find an acceptable hash with a modified chunk of hacked data. (Essentially creating or destroying coins, which-ever attack they are going for.) And if ONE person/company, producing the "large chunk" is holding the longest chain, invalid, but the longest, that it keep re-validating... they control the banks. (Which will be those developing the ASICs now, with our dollar. You don't seriously think they are going to "give us", a majority of a market they know they can consume, do you?)

Fine, you can't, and have no actual idea how to... so why answer?

As for "archived posts", If I wanted to read old dead posts, I would have searched. Is there some rule, "one original topic per forum?" Just because someone suggested it once before, and nothing was done, why should I be damned to posting a new, up-to-date topic? Get a better search function, and I might actually use it. Then I will revive every previously buried topic known to man.

Ok, I am bored now... Back to my corner of the net.

I'm not scared, I go where the money is... No more money here, I just go. I still have not actually seen a functioning GHs rig ASIC yet. Only that one, which is buggier than an e-machine, which is still a power vampire. And now he wants $8000+ for it... lol... Defeats the purpose, really!
MeisterLone
Newbie
*
Offline Offline

Activity: 22
Merit: 0



View Profile
May 09, 2013, 12:36:29 PM
 #10

Sorry for resurrecting this thread but I couldnt help but laugh at the idiocy of the OP...

ISAWHIM, I recommend keeping your opinion to yourself for the remainder of your existence. Either that, or pick up an electronics handbook off Amazon.

Funny shit.
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!