Bitcoin Forum
January 16, 2018, 10:11:31 PM *
News: Electrum users must upgrade to 3.0.5 if they haven't already. More info.
 
   Home   Help Search Donate Login Register  
Pages: [1]
  Print  
Author Topic: "Efficient Bitcoin Miner System Implemented on Zynq SoC"  (Read 8953 times)
ztex
Donator
Sr. Member
*
Offline Offline

Activity: 367

ZTEX FPGA Boards


View Profile WWW
July 25, 2013, 07:01:39 AM
 #1

Here is an interesting article about an "Efficient Bitcoin Miner System Implemented on Zynq SoC": http://www.xilinx.com/publications/archives/xcell/Xcell84.pdf

1516140691
Hero Member
*
Offline Offline

Posts: 1516140691

View Profile Personal Message (Offline)

Ignore
1516140691
Reply with quote  #2

1516140691
Report to moderator
1516140691
Hero Member
*
Offline Offline

Posts: 1516140691

View Profile Personal Message (Offline)

Ignore
1516140691
Reply with quote  #2

1516140691
Report to moderator
The forum strives to allow free discussion of any ideas. All policies are built around this principle. This doesn't mean you can post garbage, though: posts should actually contain ideas, and these ideas should be argued reasonably.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
pixl8tr
Sr. Member
****
Offline Offline

Activity: 336



View Profile
July 25, 2013, 07:22:26 AM
 #2

Interesting article.  Seems that they were able to get up to 3.8 Mh/s with a single core implemented.  I didn't see any mention of how much headroom they had left on the SOC to implement additional cores. Maybe its only one core per chip.  You need a lot more cores though to make it viable in today's world.  Wonder if Scrypt could be made to work on the ZedBoard?

who | grep -i blonde | date; cd ~; unzip; touch; finger; bjobs; uptime; strip;. grab; mount; yes; umount; sleep; brun;
Donations: 18ByQvDUmaMKkQbYvUWmnPSu9BWeNxVMoc
dogie
Legendary
*
Offline Offline

Activity: 1456


dogiecoin.com


View Profile WWW
July 25, 2013, 09:31:36 AM
 #3

It would never be 1 core per chip. That's about 15 years ago in processor design time.


pixl8tr
Sr. Member
****
Offline Offline

Activity: 336



View Profile
July 25, 2013, 12:15:26 PM
 #4

It would never be 1 core per chip. That's about 15 years ago in processor design time.



Thanks for your history lesson, but I am referring to the "soft" mining core in the Programmable logic block of the Zynq-7000 discussed in the article.  Not hard-wired CPU cores. Grin

They referenced utilizing multiple mining cores connected to the embedded bitcoind node, but they did not expand on if they were able to create multiple soft cores on a single Zynq-7000. 

Just a curiosity.  As it would need about a 100 cores just to equal the hashing power of a USB block erupter. hahaha!

who | grep -i blonde | date; cd ~; unzip; touch; finger; bjobs; uptime; strip;. grab; mount; yes; umount; sleep; brun;
Donations: 18ByQvDUmaMKkQbYvUWmnPSu9BWeNxVMoc
gingernuts
Member
**
Offline Offline

Activity: 89


View Profile
July 25, 2013, 06:10:54 PM
 #5

It is an interesting article but IMHO not a very good result! The high-level synthesis tool they used made their life easier, but their hasher is nowhere near as fast as it should be...

As for the Zedboard - it is cool, once you get over the learning curve of handling both the CPU/Linux and FPGA flow, but unfortunately the board seems to be lacking a bit in the PSU department - I managed 38MH/s on mine but while there was more room for hashers, adding anymore would brown it out!
neslekkim
Jr. Member
*
Offline Offline

Activity: 35


View Profile
July 25, 2013, 07:03:31 PM
 #6

The article only mentioned speed and such for spartan 3 and spartan 6, the main point with the article, the Zync, was not mentioned at all?, would be more interresting to see what they accomplished on that platform..
Pages: [1]
  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!