Bitcoin Forum
March 28, 2024, 02:50:32 PM *
News: Latest Bitcoin Core release: 26.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 [2] 3 4 »  All
  Print  
Author Topic: A custom designed FPGA miner for LTC?  (Read 5741 times)
iamrickrock
Sr. Member
****
Offline Offline

Activity: 476
Merit: 250


View Profile
May 25, 2013, 07:44:47 AM
 #21

This sounds interesting - watching
1711637432
Hero Member
*
Offline Offline

Posts: 1711637432

View Profile Personal Message (Offline)

Ignore
1711637432
Reply with quote  #2

1711637432
Report to moderator
1711637432
Hero Member
*
Offline Offline

Posts: 1711637432

View Profile Personal Message (Offline)

Ignore
1711637432
Reply with quote  #2

1711637432
Report to moderator
"I'm sure that in 20 years there will either be very large transaction volume or no volume." -- Satoshi
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1711637432
Hero Member
*
Offline Offline

Posts: 1711637432

View Profile Personal Message (Offline)

Ignore
1711637432
Reply with quote  #2

1711637432
Report to moderator
1711637432
Hero Member
*
Offline Offline

Posts: 1711637432

View Profile Personal Message (Offline)

Ignore
1711637432
Reply with quote  #2

1711637432
Report to moderator
soniq
Sr. Member
****
Offline Offline

Activity: 462
Merit: 250



View Profile
May 25, 2013, 07:53:25 AM
 #22

Sent some LTC your way, definitely interesting.

JohnCar
Sr. Member
****
Offline Offline

Activity: 252
Merit: 250



View Profile
May 25, 2013, 07:56:25 AM
 #23

Hi Nova following along with your thread. Are you located in the United States by chance?

TIT coming soon..............
BladeRunner
Full Member
***
Offline Offline

Activity: 205
Merit: 100

Cheif Oompa Loompa.


View Profile
May 25, 2013, 09:50:36 AM
 #24

Perhaps Nova can tell us exactly what FPGA IC costs 10,000 dollars. I have looked at the manufactures and I dont see any chip that 10K

anderl
Hero Member
*****
Offline Offline

Activity: 714
Merit: 500



View Profile
May 25, 2013, 10:07:38 AM
 #25

Perhaps Nova can tell us exactly what FPGA IC costs 10,000 dollars. I have looked at the manufactures and I dont see any chip that 10K



You haven't looked hard enough.  Some of the latest gen FPGA Stratix V development boards by Altera are running about 9k to 12k each.

I've worked with their older boards.  There is another theoretical route he can try to take which uses some of the recent innovations with SRAM.  I have my own theories and I've mocked up some code but I don't have plans to buy a board to implement them.

For Scrypt it is not worth the time and money.  The total circulation and average daily transactions of LTC does not make it a good investment.  Anyone trying to implement a FGPA for Scrypt while LTC is under $5 is a scam (not directed at OP), or has not done the math.
efx
Sr. Member
****
Offline Offline

Activity: 378
Merit: 250


View Profile
May 25, 2013, 10:09:49 AM
 #26

Yes, he did mention the extra cost due to the nature of development boards. I honestly expect any device capable of decent scrypt hashing to be rather pricey, that's one of the main points. Wink
Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 02:06:17 PM
 #27

Sent some LTC your way, definitely interesting.

Thanks what address did you send from?

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 02:17:03 PM
 #28

Perhaps Nova can tell us exactly what FPGA IC costs 10,000 dollars. I have looked at the manufactures and I dont see any chip that 10K



Yes I will tell you.  In fact that is one of the reasons I started this topic.  Anderl is right about the FPGA being an Altera Stratix.

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 02:22:18 PM
 #29

Perhaps Nova can tell us exactly what FPGA IC costs 10,000 dollars. I have looked at the manufactures and I dont see any chip that 10K



You haven't looked hard enough.  Some of the latest gen FPGA Stratix V development boards by Altera are running about 9k to 12k each.

I've worked with their older boards.  There is another theoretical route he can try to take which uses some of the recent innovations with SRAM.  I have my own theories and I've mocked up some code but I don't have plans to buy a board to implement them.

For Scrypt it is not worth the time and money.  The total circulation and average daily transactions of LTC does not make it a good investment.  Anyone trying to implement a FGPA for Scrypt while LTC is under $5 is a scam (not directed at OP), or has not done the math.

Or is interested in it for reasons of academic curiosity and not an attempt to make a commercial product or recoup investment later Smiley
I'm interested in hearing your SRAM idea as well as your other theories.

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
Hydroponica
Full Member
***
Offline Offline

Activity: 182
Merit: 100


fml


View Profile
May 25, 2013, 02:31:36 PM
 #30

I'll invest 1 ltc.

Actual ammount, .898 LTC

Transaction sent
ID:ce81163ab1903c85b5f4b4ef79657472477eab20b7bc3bb85ca6529008fb0733

Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 02:51:44 PM
 #31

Hi Nova following along with your thread. Are you located in the United States by chance?

I'm from the USA.  Right now I'm in Ecuador, but the contract I was on ended and I'll be heading back in the next couple of weeks.
Specifically I reside in Phoenix AZ.

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 02:55:32 PM
 #32

I'll invest 1 ltc.

Actual ammount, .898 LTC

Transaction sent
ID:ce81163ab1903c85b5f4b4ef79657472477eab20b7bc3bb85ca6529008fb0733

Thanks!  I'm seeing it.

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
anderl
Hero Member
*****
Offline Offline

Activity: 714
Merit: 500



View Profile
May 25, 2013, 03:57:56 PM
 #33

Perhaps Nova can tell us exactly what FPGA IC costs 10,000 dollars. I have looked at the manufactures and I dont see any chip that 10K



You haven't looked hard enough.  Some of the latest gen FPGA Stratix V development boards by Altera are running about 9k to 12k each.

I've worked with their older boards.  There is another theoretical route he can try to take which uses some of the recent innovations with SRAM.  I have my own theories and I've mocked up some code but I don't have plans to buy a board to implement them.

For Scrypt it is not worth the time and money.  The total circulation and average daily transactions of LTC does not make it a good investment.  Anyone trying to implement a FGPA for Scrypt while LTC is under $5 is a scam (not directed at OP), or has not done the math.

Or is interested in it for reasons of academic curiosity and not an attempt to make a commercial product or recoup investment later Smiley
I'm interested in hearing your SRAM idea as well as your other theories.

When LTC gets over $5.  Right now I'm find using ASICs for BTC and GPUs for LTC and scrypts.

But look into what is coming to market right now in FPGAs.
soniq
Sr. Member
****
Offline Offline

Activity: 462
Merit: 250



View Profile
May 25, 2013, 04:36:56 PM
 #34

Sent some LTC your way, definitely interesting.

Thanks what address did you send from?

I sent you a PM

Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 06:11:32 PM
 #35

Perhaps Nova can tell us exactly what FPGA IC costs 10,000 dollars. I have looked at the manufactures and I dont see any chip that 10K



You haven't looked hard enough.  Some of the latest gen FPGA Stratix V development boards by Altera are running about 9k to 12k each.

I've worked with their older boards.  There is another theoretical route he can try to take which uses some of the recent innovations with SRAM.  I have my own theories and I've mocked up some code but I don't have plans to buy a board to implement them.

For Scrypt it is not worth the time and money.  The total circulation and average daily transactions of LTC does not make it a good investment.  Anyone trying to implement a FGPA for Scrypt while LTC is under $5 is a scam (not directed at OP), or has not done the math.

Or is interested in it for reasons of academic curiosity and not an attempt to make a commercial product or recoup investment later Smiley
I'm interested in hearing your SRAM idea as well as your other theories.

When LTC gets over $5.  Right now I'm find using ASICs for BTC and GPUs for LTC and scrypts.

But look into what is coming to market right now in FPGAs.

I can't say as I disagree with you there at all.  My only response to that would be it's pure research.
Xerox was researching GUI's back at a time when almost no one had a home computer. 
It didn't make financial sense either it was more or less a "hey cool look what we can do" sort of thing. 
That's really all this is intended to be.

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
Puycheval
Member
**
Offline Offline

Activity: 95
Merit: 10


View Profile
May 25, 2013, 06:27:35 PM
 #36

I'm new to crypto currencies and I'm looking for a FPGA implementation too.

scrypt differs mostly because it uses an entirely new list so frequently.  
I think the big problem is that you can't unroll salsa mixing because of its recursive form. Thus you can't parallelize calculations as you can do with sha256. The only thing you can do is to have multiple instance of your 'cores' run in parallel. But I don't think Stratix have enough on-die ram (52 Mbit max) to overwhelm a pool as you said.

Quote
The shortcut is to have a multicore setup and a ton of on-die ram.
A dedicated prng core which does the setup and teardown for the second core.
I don't see the shorcut here. Are you thinking of a two stages pipeline with dual port ram in the middle ?

To conclude, I don't understand why you need funding for your idea because you can test everything with simulation. Altera provides a free web edition of their dev tools that don't allow you to target Stratix but you can target Cyclone V. You should be able to validate your idea with 12 Mb of on-die ram. Then you'll have tangible results to get funds for a dev board which are really expensive Tongue

Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 06:35:55 PM
 #37

I'm new to crypto currencies and I'm looking for a FPGA implementation too.

scrypt differs mostly because it uses an entirely new list so frequently.  
I think the big problem is that you can't unroll salsa mixing because of its recursive form. Thus you can't parallelize calculations as you can do with sha256. The only thing you can do is to have multiple instance of your 'cores' run in parallel. But I don't think Stratix have enough on-die ram (52 Mbit max) to overwhelm a pool as you said.

Quote
The shortcut is to have a multicore setup and a ton of on-die ram.
A dedicated prng core which does the setup and teardown for the second core.
I don't see the shorcut here. Are you thinking of a two stages pipeline with dual port ram in the middle ?

To conclude, I don't understand why you need funding for your idea because you can test everything with simulation. Altera provides a free web edition of their dev tools that don't allow you to target Stratix but you can target Cyclone V. You should be able to validate your idea with 12 Mb of on-die ram. Then you'll have tangible results to get funds for a dev board which are really expensive Tongue

That's a good idea.  I was unaware there was a free tool that wouldn't let me target the chip I'm planning to target.
Is there a compelling reason to choose Cyclone V over Stratix V?  Other than cost I mean?

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
ElGabo
Hero Member
*****
Offline Offline

Activity: 635
Merit: 500


View Profile
May 25, 2013, 06:38:41 PM
 #38

Sent 1 LTC form LXj15uZkCFMecbKcLVMsswTtnWvRqqhuUm

Could you send me some info by PM?

" I'm waiting for my punishment, I know it's on my way
  So cut, cut, cut me up and fuck, fuck, fuck me up"
Puycheval
Member
**
Offline Offline

Activity: 95
Merit: 10


View Profile
May 25, 2013, 06:49:58 PM
 #39

That's a good idea.  I was unaware there was a free tool that wouldn't let me target the chip I'm planning to target.
Is there a compelling reason to choose Cyclone V over Stratix V?  Other than cost I mean?
Cyclones are a smaller, slower version of the Stratix lineup. The biggest Cyclone V is approximatively a third of the biggest Stratix V for LE count.
I suppose it should be enough to validate the shortcut.

Nova! (OP)
Full Member
***
Offline Offline

Activity: 140
Merit: 101


View Profile
May 25, 2013, 07:31:49 PM
 #40

Sent 1 LTC form LXj15uZkCFMecbKcLVMsswTtnWvRqqhuUm

Could you send me some info by PM?


Yes, I'm also finding out some exciting information in this thread and am trying to incorporate the new information in the writeup.

Donate @ 1LE4D5ERPZ4tumNoYe5GMeB5p9CZ1xKb4V
Pages: « 1 [2] 3 4 »  All
  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!