Bitcoin Forum
November 09, 2024, 02:31:16 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: What next to ASIC'ize?  (Read 963 times)
Reclaim3r (OP)
Full Member
***
Offline Offline

Activity: 139
Merit: 100


View Profile
July 29, 2014, 12:01:03 AM
 #1

So far, I've seen SHA256 and Scrypt ASIC'd.
What's next?

I know, it seems like a newbie question but I'm just curious  Wink
xstr8guy
Hero Member
*****
Offline Offline

Activity: 784
Merit: 1004


Glow Stick Dance!


View Profile
July 29, 2014, 01:46:25 AM
 #2

Nothing. The other altcoin algorithms aren't worth the bother to develop an ASIC for them. I'm going to go out on a limb here but it appears that developing ASICs for Scrypt was a horrible idea and they will not have a profitable period in this next generation.
jonnybravo0311
Legendary
*
Offline Offline

Activity: 1344
Merit: 1024


Mine at Jonny's Pool


View Profile WWW
July 29, 2014, 02:03:08 AM
 #3

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Jonny's Pool - Mine with us and help us grow!  Support a pool that supports Bitcoin, not a hardware manufacturer's pockets!  No SPV cheats.  No empty blocks.
Reclaim3r (OP)
Full Member
***
Offline Offline

Activity: 139
Merit: 100


View Profile
July 29, 2014, 04:35:20 AM
 #4

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Thank you for that answer  Smiley
notlist3d
Legendary
*
Offline Offline

Activity: 1456
Merit: 1000



View Profile
July 29, 2014, 05:46:28 AM
 #5

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.
Reclaim3r (OP)
Full Member
***
Offline Offline

Activity: 139
Merit: 100


View Profile
July 29, 2014, 05:57:34 AM
 #6

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.

In fact, an X11 ASIC was a project I wanted to take on. I'm well aware of the capital required to produce ASIC's but if the cost can be justified and an FPGA implementation followed by a few sample batches produced ( well under 500,000 USD I hope, but if costs get out of whack this is gonna be one hell of a project  Cheesy ) I think the garnered interest could be used to develop one. Of course, I'm talking in very loose terms.

Making my own ASIC company has been a long time dream of mine. I've heard plenty of "you're not going to make it" or "somebody get rid of this noob" but I honestly don't care. Just gotta pull through. I'm just gathering some useful data for that venture  Cheesy Smiley

monsterasic
Newbie
*
Offline Offline

Activity: 16
Merit: 0


View Profile WWW
July 29, 2014, 10:04:10 AM
 #7

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.

In fact, an X11 ASIC was a project I wanted to take on. I'm well aware of the capital required to produce ASIC's but if the cost can be justified and an FPGA implementation followed by a few sample batches produced ( well under 500,000 USD I hope, but if costs get out of whack this is gonna be one hell of a project  Cheesy ) I think the garnered interest could be used to develop one. Of course, I'm talking in very loose terms.

Making my own ASIC company has been a long time dream of mine. I've heard plenty of "you're not going to make it" or "somebody get rid of this noob" but I honestly don't care. Just gotta pull through. I'm just gathering some useful data for that venture  Cheesy Smiley




I would buy a few G-Blades from our website monster-asic.com, they are currently at sale for 0.5 BTC per unit...
TookDk
Legendary
*
Offline Offline

Activity: 1960
Merit: 1062


One coin to rule them all


View Profile WWW
July 29, 2014, 10:08:37 AM
 #8

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.

In fact, an X11 ASIC was a project I wanted to take on. I'm well aware of the capital required to produce ASIC's but if the cost can be justified and an FPGA implementation followed by a few sample batches produced ( well under 500,000 USD I hope, but if costs get out of whack this is gonna be one hell of a project  Cheesy ) I think the garnered interest could be used to develop one. Of course, I'm talking in very loose terms.

Making my own ASIC company has been a long time dream of mine. I've heard plenty of "you're not going to make it" or "somebody get rid of this noob" but I honestly don't care. Just gotta pull through. I'm just gathering some useful data for that venture  Cheesy Smiley

If you can make a functional X11 FPGA miner, then is the step to a ASIC really not that big. 

Cryptography is one of the few things you can truly trust.
Klubknuckle
Sr. Member
****
Offline Offline

Activity: 285
Merit: 250


View Profile
July 29, 2014, 12:20:11 PM
 #9

So far, I've seen SHA256 and Scrypt ASIC'd.
What's next?

I know, it seems like a newbie question but I'm just curious  Wink

x11 should be next since its the next most profitable..

Reclaim3r (OP)
Full Member
***
Offline Offline

Activity: 139
Merit: 100


View Profile
July 29, 2014, 02:29:42 PM
 #10

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.

In fact, an X11 ASIC was a project I wanted to take on. I'm well aware of the capital required to produce ASIC's but if the cost can be justified and an FPGA implementation followed by a few sample batches produced ( well under 500,000 USD I hope, but if costs get out of whack this is gonna be one hell of a project  Cheesy ) I think the garnered interest could be used to develop one. Of course, I'm talking in very loose terms.

Making my own ASIC company has been a long time dream of mine. I've heard plenty of "you're not going to make it" or "somebody get rid of this noob" but I honestly don't care. Just gotta pull through. I'm just gathering some useful data for that venture  Cheesy Smiley

If you can make a functional X11 FPGA miner, then is the step to a ASIC really not that big. 

Well there's my first piece of optimistic news  Cheesy
I know X11 uses 11 hashing algos ((blake, bmw, groestl, jh, keccak, skein, luffa, cubehash, shavite, simd, echo),
Some of these look like SHA-3's but the rest, I'm not so sure. In fact, I'm not even sure how X11 puts this all together.
Is data hashed through all of them sequentially or do certain algos get performed multiple times?
It feels like it could be paralleled although I think the GPU miner might have already done that.

Is there any way of converting the GPU CUDA/OpenGL code to FPGA or do you have to rough it and start from scratch?
blockjoe
Member
**
Offline Offline

Activity: 115
Merit: 10


View Profile
July 29, 2014, 03:28:08 PM
 #11

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.

In fact, an X11 ASIC was a project I wanted to take on. I'm well aware of the capital required to produce ASIC's but if the cost can be justified and an FPGA implementation followed by a few sample batches produced ( well under 500,000 USD I hope, but if costs get out of whack this is gonna be one hell of a project  Cheesy ) I think the garnered interest could be used to develop one. Of course, I'm talking in very loose terms.

Making my own ASIC company has been a long time dream of mine. I've heard plenty of "you're not going to make it" or "somebody get rid of this noob" but I honestly don't care. Just gotta pull through. I'm just gathering some useful data for that venture  Cheesy Smiley

If you can make a functional X11 FPGA miner, then is the step to a ASIC really not that big. 

Well there's my first piece of optimistic news  Cheesy
I know X11 uses 11 hashing algos ((blake, bmw, groestl, jh, keccak, skein, luffa, cubehash, shavite, simd, echo),
Some of these look like SHA-3's but the rest, I'm not so sure. In fact, I'm not even sure how X11 puts this all together.
Is data hashed through all of them sequentially or do certain algos get performed multiple times?
It feels like it could be paralleled although I think the GPU miner might have already done that.

Is there any way of converting the GPU CUDA/OpenGL code to FPGA or do you have to rough it and start from scratch?

X11 uses a lot of algorithms, but they are all FPGA/ASIC friendly, they may develop a semi-programmable system, where hash units can be chained together to be compatible with new variants (X13, ...).
Reclaim3r (OP)
Full Member
***
Offline Offline

Activity: 139
Merit: 100


View Profile
July 29, 2014, 08:02:24 PM
 #12

I won't go with "nothing".  My answer is, "whatever is profitable enough to invest in ASIC design and development".  So, if X11, or BobsMagicHashingFormula becomes the next big thing, then somebody will develop an ASIC for it.  If there's money to be made, somebody's going to make it Wink.

Perfect answer.  You will notice coins say asic resistant.  They do not uses (or should not) proof.  If there is a big enough market share there will be a product for it. 

A n/x11 would be nice but I don't know if or timeline we will see that.   Scrypt and of course SHA-256 are controlling the game right now.

In fact, an X11 ASIC was a project I wanted to take on. I'm well aware of the capital required to produce ASIC's but if the cost can be justified and an FPGA implementation followed by a few sample batches produced ( well under 500,000 USD I hope, but if costs get out of whack this is gonna be one hell of a project  Cheesy ) I think the garnered interest could be used to develop one. Of course, I'm talking in very loose terms.

Making my own ASIC company has been a long time dream of mine. I've heard plenty of "you're not going to make it" or "somebody get rid of this noob" but I honestly don't care. Just gotta pull through. I'm just gathering some useful data for that venture  Cheesy Smiley

If you can make a functional X11 FPGA miner, then is the step to a ASIC really not that big. 

Well there's my first piece of optimistic news  Cheesy
I know X11 uses 11 hashing algos ((blake, bmw, groestl, jh, keccak, skein, luffa, cubehash, shavite, simd, echo),
Some of these look like SHA-3's but the rest, I'm not so sure. In fact, I'm not even sure how X11 puts this all together.
Is data hashed through all of them sequentially or do certain algos get performed multiple times?
It feels like it could be paralleled although I think the GPU miner might have already done that.

Is there any way of converting the GPU CUDA/OpenGL code to FPGA or do you have to rough it and start from scratch?

X11 uses a lot of algorithms, but they are all FPGA/ASIC friendly, they may develop a semi-programmable system, where hash units can be chained together to be compatible with new variants (X13, ...).

I was thinking of an ASIC FPGA hybrid with X11 in ASIC and FPGA allowing further implementations. Just need to check the FPGA implementations of each alto and chain them somehow.
Reclaim3r (OP)
Full Member
***
Offline Offline

Activity: 139
Merit: 100


View Profile
July 29, 2014, 10:57:19 PM
 #13

UPDATE: I found some useful info here regarding a SHA-3 ASIC developed by Virginia Tech.
It has all the finalist SHA-3 algos on one chip except it's pretty low-performance by current standards.
I will attempt to contact them regarding further info. It appears they halted altogether around 2012 on project statuses so it would be pretty interesting to see what I/the community could do with it:
http://rijndael.ece.vt.edu/sha3/

NOTE: X11 seems to use the 512-bit variation of these algos so some augmentation may be necessary.

Second of all, there was a co-op project that was supposed to get an FPGA running an X11 miner but it appears deserted until today.
Some disappointing news seems to have killed the hype until some forum member dug it out:

https://darkcointalk.org/threads/darkcoin-fpga-mining-co-op.836/


I'd be really interested in getting everything up and running.
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!