|
creativex
|
|
December 12, 2012, 02:46:22 AM |
|
Ok... for all the noobie faggots and "ASICS" with SAH will never exist group.
Here is what a REAL progressional group can accomplish as regards to getting SHA into an ASIC. Really DO NOT read this document it will make you VERY depressed.
And you REALLY do not want to know how many weeks it took them, ohhhhhh shit AND they did it in 2011 OVER 1 YEAR ago!!! Your angry rant would have more teeth if you posted something aside from hostility.
|
|
|
|
hardcore-fs (OP)
|
|
December 12, 2012, 02:50:45 AM Last edit: December 12, 2012, 03:03:24 AM by hardcore-fs |
|
Ok... for all the noobie faggots and "ASICS" with SAH will never exist group.
Here is what a REAL progressional group can accomplish as regards to getting SHA into an ASIC. Really DO NOT read this document it will make you VERY depressed.
And you REALLY do not want to know how many weeks it took them, ohhhhhh shit AND they did it in 2011 OVER 1 YEAR ago!!! Your angry rant would have more teeth if you posted something aside from hostility. Yep I thought I would try something new.... I am sick and tired of all the potential ass masters in this group ranting on about ASICS and SHA256 not being possible, so I have released a document I have kept back.(private joke really) and if you really want to get a "hard on" (something you will find anyway) take a look at the second link.... Surprise!!!!! guess who and look at the sizes.
|
BTC:1PCTzvkZUFuUF7DA6aMEVjBUUp35wN5JtF
|
|
|
Jutarul
Donator
Legendary
Offline
Activity: 994
Merit: 1000
|
|
December 12, 2012, 03:12:11 AM |
|
Ok... for all the noobie faggots and "ASICS" with SAH will never exist group.
Here is what a REAL progressional group can accomplish as regards to getting SHA into an ASIC. Really DO NOT read this document it will make you VERY depressed.
And you REALLY do not want to know how many weeks it took them, ohhhhhh shit AND they did it in 2011 OVER 1 YEAR ago!!! Your angry rant would have more teeth if you posted something aside from hostility. Yep I thought I would try something new.... I am sick and tired of all the potential ass masters in this group ranting on about ASICS and SHA256 not being possible, so I have released a document I have kept back.(private joke really) and if you really want to get a "hard on" (something you will find anyway) take a look at the second link.... Surprise!!!!! guess who and look at the sizes. First of all congrats. You seem to have some insight into hardware design. I don't know who said that ASICS should not be possible, but anyone who does that should be considered a troll. What is your opinion on the issue of ASIC heat management?
|
|
|
|
hardcore-fs (OP)
|
|
December 12, 2012, 03:21:15 AM |
|
Ok... for all the noobie faggots and "ASICS" with SAH will never exist group.
Here is what a REAL progressional group can accomplish as regards to getting SHA into an ASIC. Really DO NOT read this document it will make you VERY depressed.
And you REALLY do not want to know how many weeks it took them, ohhhhhh shit AND they did it in 2011 OVER 1 YEAR ago!!! Your angry rant would have more teeth if you posted something aside from hostility. Yep I thought I would try something new.... I am sick and tired of all the potential ass masters in this group ranting on about ASICS and SHA256 not being possible, so I have released a document I have kept back.(private joke really) and if you really want to get a "hard on" (something you will find anyway) take a look at the second link.... Surprise!!!!! guess who and look at the sizes. First of all congrats. You seem to have some insight into hardware design. I don't know who said that ASICS should not be possible, but anyone who does that should be considered a troll. What is your opinion on the issue of ASIC heat management? Sorry, another PM that got my back up and I was off my meds for about 10 minutes. As regards heat, these guys are just proofing the cores for the SHA-3 finalists they happened to implement a Single SHA-256 core to compare results. Looking at the VHDL they have been using, it is designed to go as fast as possible but only with a limited number of rounds, as a result they are not up against any heat limitations, plus it is only a single SHA256 core. But It IS proof that SHA256 in ASIC is firmly achievable, strip out the other cores, take it to SHA256(SHA256(x)), some nonce processing circuitry, and by a strange coincidence (5 other cores), we end up at about 6 single cores per ASIC As you point out.... heat is the killer once it is over-clocked.
|
BTC:1PCTzvkZUFuUF7DA6aMEVjBUUp35wN5JtF
|
|
|
Jutarul
Donator
Legendary
Offline
Activity: 994
Merit: 1000
|
|
December 12, 2012, 03:31:13 AM |
|
Ok... for all the noobie faggots and "ASICS" with SAH will never exist group.
Here is what a REAL progressional group can accomplish as regards to getting SHA into an ASIC. Really DO NOT read this document it will make you VERY depressed.
And you REALLY do not want to know how many weeks it took them, ohhhhhh shit AND they did it in 2011 OVER 1 YEAR ago!!! Your angry rant would have more teeth if you posted something aside from hostility. Yep I thought I would try something new.... I am sick and tired of all the potential ass masters in this group ranting on about ASICS and SHA256 not being possible, so I have released a document I have kept back.(private joke really) and if you really want to get a "hard on" (something you will find anyway) take a look at the second link.... Surprise!!!!! guess who and look at the sizes. First of all congrats. You seem to have some insight into hardware design. I don't know who said that ASICS should not be possible, but anyone who does that should be considered a troll. What is your opinion on the issue of ASIC heat management? Sorry, another PM that got my back up and I was off my meds for about 10 minutes. As regards heat, these guys are just proofing the cores for the SHA-3 finalists they happened to implement a Single SHA-256 core to compare results. Looking at the VHDL they have been using, it is designed to go as fast as possible but only with a limited number of rounds, as a result they are not up against any heat limitations, plus it is only a single SHA256 core. But It IS proof that SHA256 in ASIC is firmly achievable, strip out the other cores, take it to SHA256(SHA256(x)), some nonce processing circuitry, and by a strange coincidence (5 other cores), we end up at about 6 single cores per ASIC As you point out.... heat is the killer once it is over-clocked. That depends on what you consider as over-clocking. Once you have a simple logic which is constantly in action you need to consider heat management WITHIN the chip. E.g. "Although this approach is still being followed during thermal design for low-power ICs, it is not sufficient for high-performance or power-constrained designs. Increasingly non-uniform power dissipation across the chip leads to local hot spots and elevated temperature gradients across the silicon die. For example, in a 90nm Intel Itanium processor, even after stringent thermal management, local temperature can still be as high as 88◦C, while other parts of the die are relatively cool (61C) [64]" http://www.cs.virginia.edu/~skadron/Papers/wei_huang_dissertation.pdfThat means that the heat gradient WITHIN the chip may be ultimately limiting the possible clock rate. I guess this is what will set the ASIC manufacturers apart in the second generation devices.
|
|
|
|
MrTeal
Legendary
Offline
Activity: 1274
Merit: 1004
|
|
December 12, 2012, 03:50:33 AM |
|
You do know that the VT papers were posted up here and commented on 6 months ago, right? This isn't exactly news that SHA2 ASIC cores are out there in the wild to anyone paying attention.
|
|
|
|
hardcore-fs (OP)
|
|
December 12, 2012, 03:55:30 AM |
|
You do know that the VT papers were posted up here and commented on 6 months ago, right? This isn't exactly news that SHA2 ASIC cores are out there in the wild to anyone paying attention.
Yep......, but it seems , just like goldfish some people have short memories.
|
BTC:1PCTzvkZUFuUF7DA6aMEVjBUUp35wN5JtF
|
|
|
k9quaint
Legendary
Offline
Activity: 1190
Merit: 1000
|
|
December 12, 2012, 04:53:36 AM |
|
You do know that the VT papers were posted up here and commented on 6 months ago, right? This isn't exactly news that SHA2 ASIC cores are out there in the wild to anyone paying attention.
Yep......, but it seems , just like goldfish some people have short memories. Goldfish don't have short memories. Two things you can count on: ASICs will arrive someday. People will be scammed by promises of ASICs. Not necessarily in that order.
|
Bitcoin is backed by the full faith and credit of YouTube comments.
|
|
|
nathanrees19
|
|
December 17, 2012, 03:46:15 AM |
|
Hmmm... Therefore any valid result has to be recovered in a SINGLE clock cycle of the master clock before the result is lost. This is wrong. The nonce is highly predictable, since it is normally just incremented. You can get the nonce a couple clocks late and scan back in the mining software to find the true nonce. What is the point of the article? "Open source project has minor flaws that are easily fixed?" I'm guessing that he/she didn't bother to submit a patch. Also, the hashing ASIC referred to at the end would be useless for mining.
|
|
|
|
MrTeal
Legendary
Offline
Activity: 1274
Merit: 1004
|
|
December 18, 2012, 03:34:41 AM |
|
The author of the code you're talking about has posted a reply. http://pastie.org/5545344
|
|
|
|
Lethos
|
|
December 20, 2012, 01:04:50 PM |
|
He may really ruffle the feathers on the programmers who originally made some of that code, but if he can make big improvements use it. Is he a bit of an arrogant arse about, yes, but is he wrong, doesn't look like it.
|
|
|
|
|