Bitcoin Forum
November 09, 2024, 11:56:23 AM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: sha-256 engines  (Read 4729 times)
Un zafado cualquiera (OP)
Full Member
***
Offline Offline

Activity: 158
Merit: 100


aquí dice algo personal.


View Profile
July 12, 2011, 05:44:27 AM
 #1

I was reading a lot about FPGA miner and seeking a lot of info.

I want to ask for reasons to use ( or not ) any chip created until now to optimize or make a hardware miner. These maybe become the future of mining and there is not so much interest except for the Modular FPGA miner http://forum.bitcoin.org/index.php?topic=22426.0
gmaxwell
Moderator
Legendary
*
expert
Offline Offline

Activity: 4270
Merit: 8805



View Profile WWW
July 12, 2011, 07:31:41 AM
 #2

I was reading a lot about FPGA miner and seeking a lot of info.
I want to ask for reasons to use ( or not ) any chip created until now to optimize or make a hardware miner. These maybe become the future of mining and there is not so much interest except for the Modular FPGA miner http://forum.bitcoin.org/index.php?topic=22426.0

Because any SHA256 engine created for a purpose other than mining would be pitifully slow.

In terms of SHA256 computation,  1MH/s ~= 1Gbit/sec of SHA256 computation.   Why would anyone need chips that do much more than 100gbit/s of SHA256... and more importantly, why would they need them at price points that would make them attractive for mining?

As a result, any pre-existing sha256 engine may have good density, or good power consumption— but it won't have good performance from our perspective.
Un zafado cualquiera (OP)
Full Member
***
Offline Offline

Activity: 158
Merit: 100


aquí dice algo personal.


View Profile
July 13, 2011, 06:13:55 AM
Last edit: July 16, 2011, 04:25:24 AM by Un zafado cualquiera
 #3

On the FPGA modular miner they claim 1 Mhz is practically equal to 1MH/s. I read about a licensed Altera code with much less logic and probably more efficiency.

someone can explain me if they have the same usage? or the code is more dense than that of altera?

http://www.cast-inc.com/ip-cores/encryption/sha-256/sha-256-xilinx.htm

and this is an IP core sold by an enterprise.
Un zafado cualquiera (OP)
Full Member
***
Offline Offline

Activity: 158
Merit: 100


aquí dice algo personal.


View Profile
July 16, 2011, 04:39:25 AM
 #4

There is a posibility someone can do a compatible DIMM for the modular FPGA. But is just for showing that is not necesity to remake the well if this become cheaper.

http://embedded-computing.com/atmels-cloning-secures-data
gmaxwell
Moderator
Legendary
*
expert
Offline Offline

Activity: 4270
Merit: 8805



View Profile WWW
July 18, 2011, 05:56:30 AM
 #5

On the FPGA modular miner they claim 1 Mhz is practically equal to 1MH/s. I read about a licensed Altera code with much less logic and probably more efficiency.
someone can explain me if they have the same usage? or the code is more dense than that of altera?
http://www.cast-inc.com/ip-cores/encryption/sha-256/sha-256-xilinx.htm
and this is an IP core sold by an enterprise.

The licensed code you're looking at has less logic because it's not unrolled.  Every hash operation probably takes 128 cycles (or maybe 64) because it reuses the same logic. Unrolled bitcoin engines complete 1 hash per cycle (and have a latency of 124 or perhaps 248 cycles).

It's very slow, by our standards, as a result.

The reason they made it so slow is what I gave you above: 1MH/s of mining is approximately equal to 1Gbit/sec of general hashing. Since no one but bitcoin miners wants >100gbit/sec of SHA-256 in a chip (or even >10gbit/sec, most likely), no one but bitcoin miners would bother creating a design which is able to reach those speeds.  

You could put many such engines on a chip, but they'd end up taking more space than an unrolled engine with the same performance, because unrolling eliminates overhead and eliminates idle logic between attempts.
Un zafado cualquiera (OP)
Full Member
***
Offline Offline

Activity: 158
Merit: 100


aquí dice algo personal.


View Profile
July 19, 2011, 05:50:56 AM
 #6

Now you put one question on my mind. If code from that IP core is similar to the code of the Open Miner then all the guy are gettin in troubles because of industrial design is registered and nobody can use without permisson. I doubt both become similar but they gonna try to stop the project. Am i wrong?
TheSeven
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500


FPGA Mining LLC


View Profile WWW
July 19, 2011, 11:22:35 AM
 #7

Except for the publicly specified SHA256 algorithm, they won't have much in common. While the commercial ones will be optimized for high-speed hashing of huge amounts of data, our FPGA miner algorithms are optimized for bruteforcing bitcoin hashes, causing an inherently different design.
I don't think that commercial SHA256 implementations will interfere here. If anything, it will be more general patents on "using an FPGA for cryptographic bruteforcing" or similar crap.

My tip jar: 13kwqR7B4WcSAJCYJH1eXQcxG5vVUwKAqY
JoelKatz
Legendary
*
Offline Offline

Activity: 1596
Merit: 1012


Democracy is vulnerable to a 51% attack.


View Profile WWW
July 19, 2011, 04:29:23 PM
 #8

Now you put one question on my mind. If code from that IP core is similar to the code of the Open Miner then all the guy are gettin in troubles because of industrial design is registered and nobody can use without permisson. I doubt both become similar but they gonna try to stop the project. Am i wrong?
Unless there are patents, it's not an issue. Independent creation is a complete defense to an accusation of copyright infringement. If two composers independently write precisely the same song, note for note, they are both fully entitled to copyright for their individual creative efforts.

"I never had access to the copyrighted work" is a defense to a claim of infringement. It's a reason video game manufacturers no longer try to register a video of someone playing their game as an attempt to create a copyright on the "game experience". (Since an alleged infringer won't have ever seen that video, how could it help?)

I am an employee of Ripple. Follow me on Twitter @JoelKatz
1Joe1Katzci1rFcsr9HH7SLuHVnDy2aihZ BM-NBM3FRExVJSJJamV9ccgyWvQfratUHgN
Un zafado cualquiera (OP)
Full Member
***
Offline Offline

Activity: 158
Merit: 100


aquí dice algo personal.


View Profile
July 19, 2011, 07:00:07 PM
Last edit: July 19, 2011, 07:12:49 PM by Un zafado cualquiera
 #9

Now you put one question on my mind. If code from that IP core is similar to the code of the Open Miner then all the guy are gettin in troubles because of industrial design is registered and nobody can use without permisson. I doubt both become similar but they gonna try to stop the project. Am i wrong?
Unless there are patents, it's not an issue. Independent creation is a complete defense to an accusation of copyright infringement. If two composers independently write precisely the same song, note for note, they are both fully entitled to copyright for their individual creative efforts.

"I never had access to the copyrighted work" is a defense to a claim of infringement. It's a reason video game manufacturers no longer try to register a video of someone playing their game as an attempt to create a copyright on the "game experience". (Since an alleged infringer won't have ever seen that video, how could it help?)

If you say it is more prone to patent litigation on the next months is probably someone send a Cease and Desist letter. I was thinking who gonna test if is possible to patent all the desings put around the forum.

Edit: here is a discussion about what I was saying http://security.stackexchange.com/questions/3519/can-ecc-be-used-without-infringing-on-patents
Probably is not a problem but surely it helps  to check any dark problem
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!