Bitcoin Forum>Alternate cryptocurrencies>Mining (Altcoins)
Pages:
Author

Topic: [Call to Arms] CryptoNight FPGA mining (Read 7960 times)

Wolf0
member
Activity: 81
Merit: 1002
It was only the wind.
[Call to Arms] CryptoNight FPGA mining
September 02, 2015, 08:26:11 AM
#35
Quote from: adaseb on September 02, 2015, 07:23:36 AM
What is so special about FPGA exactly?

Last I heard its only advantage is the low power usage but they are generally very slow. If you want something fast, you need to design an ASIC for that.

Not really true - depending on the task, FPGA can own GPUs using maybe a fourth of the power or less. Keep in mind - FPGA work IS ASIC design; my Verilog could be put to actual chips, for example, if someone had shittons to burn.
Wolf0
member
Activity: 81
Merit: 1002
It was only the wind.
Re: [Call to Arms] CryptoNight FPGA mining
September 02, 2015, 01:01:02 AM
#33
Quote from: bitl0ck on September 01, 2015, 05:29:58 PM
Quote from: Wolf0 on September 01, 2015, 03:39:09 PM
Quote from: kingcolex on September 01, 2015, 10:38:40 AM
Quote from: ?? on ??
Quote from: kingcolex on September 01, 2015, 10:03:11 AM
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.

Sure it is! Stop thinking inside the box. Don't get a Kintex Ultrascale chip for X11, get maybe 8 or 9 Spartan-6 LX150 chips and use a custom board!

X11, if done properly, requires pretty much no memory. Each FPGA chip can be flashed to one algo each, except for maybe one or two that can fit two simpler algos. The algos will even be easy to pipeline - one chip is one step. Board should be custom, for fast communication between FPGA chips, and this will allow you to omit costly extras like RAM.
Hmm, I am not a developer but it sounds very possible and you definitely sound like you know what you are doing. You think you could design one and get one working and start your own FPGA company selling these if you had the right amount of investors or enough coin invested? You are the king of X11 bins and obviously know the technical side of it.

If I could get someone to solder the boards, I could probably do the dev work.

You could order a custom design i guess, but im not sure as im not into fpga's.

You wouldn't need to. I mean "someone" like a person to do the soldering.
Wolf0
member
Activity: 81
Merit: 1002
It was only the wind.
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 03:39:09 PM
#31
Quote from: kingcolex on September 01, 2015, 10:38:40 AM
Quote from: ?? on ??
Quote from: kingcolex on September 01, 2015, 10:03:11 AM
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.

Sure it is! Stop thinking inside the box. Don't get a Kintex Ultrascale chip for X11, get maybe 8 or 9 Spartan-6 LX150 chips and use a custom board!

X11, if done properly, requires pretty much no memory. Each FPGA chip can be flashed to one algo each, except for maybe one or two that can fit two simpler algos. The algos will even be easy to pipeline - one chip is one step. Board should be custom, for fast communication between FPGA chips, and this will allow you to omit costly extras like RAM.
Hmm, I am not a developer but it sounds very possible and you definitely sound like you know what you are doing. You think you could design one and get one working and start your own FPGA company selling these if you had the right amount of investors or enough coin invested? You are the king of X11 bins and obviously know the technical side of it.

If I could get someone to solder the boards, I could probably do the dev work.
kingcolex
sr. member
Activity: 254
Merit: 1258
Re: [Call to Arms] CryptoNight FPGA mining
September 02, 2015, 09:25:37 AM
#29
Quote from: Wolf0 on September 02, 2015, 08:26:11 AM
Quote from: adaseb on September 02, 2015, 07:23:36 AM
What is so special about FPGA exactly?

Last I heard its only advantage is the low power usage but they are generally very slow. If you want something fast, you need to design an ASIC for that.

Not really true - depending on the task, FPGA can own GPUs using maybe a fourth of the power or less. Keep in mind - FPGA work IS ASIC design; my Verilog could be put to actual chips, for example, if someone had shittons to burn.
Yeah which I imagine would happen if the X11 market increases dramatically but this would require BTC and LTC going up since they're somewhat tied in price.  Pretty much if BTC rises back to 1000$ and Ltc to 50$, and Dash at $40+ I expect x11 asics will be in development.
kingcolex
sr. member
Activity: 254
Merit: 1258
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 10:38:40 AM
#24
Quote from: ?? on ??
Quote from: kingcolex on September 01, 2015, 10:03:11 AM
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.

Sure it is! Stop thinking inside the box. Don't get a Kintex Ultrascale chip for X11, get maybe 8 or 9 Spartan-6 LX150 chips and use a custom board!

X11, if done properly, requires pretty much no memory. Each FPGA chip can be flashed to one algo each, except for maybe one or two that can fit two simpler algos. The algos will even be easy to pipeline - one chip is one step. Board should be custom, for fast communication between FPGA chips, and this will allow you to omit costly extras like RAM.
Hmm, I am not a developer but it sounds very possible and you definitely sound like you know what you are doing. You think you could design one and get one working and start your own FPGA company selling these if you had the right amount of investors or enough coin invested? You are the king of X11 bins and obviously know the technical side of it.
kingcolex
sr. member
Activity: 254
Merit: 1258
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 10:03:11 AM
#23
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.
TunerDude007
sr. member
Activity: 465
Merit: 252
Re: [Call to Arms] CryptoNight FPGA mining
January 18, 2017, 04:55:18 PM
#20
So since this thread was originally started we can see the proliferation of x11 asics and fpga's (I still think Baikal's seem more fpga then asic).

But my question has anyone made any progress towards cryptonight fpga's/asics?
adaseb
legendary
Activity: 3808
Merit: 1723
Re: [Call to Arms] CryptoNight FPGA mining
September 02, 2015, 07:23:36 AM
#19
What is so special about FPGA exactly?

Last I heard its only advantage is the low power usage but they are generally very slow. If you want something fast, you need to design an ASIC for that.
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 05:29:58 PM
#18
Quote from: Wolf0 on September 01, 2015, 03:39:09 PM
Quote from: kingcolex on September 01, 2015, 10:38:40 AM
Quote from: ?? on ??
Quote from: kingcolex on September 01, 2015, 10:03:11 AM
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.

Sure it is! Stop thinking inside the box. Don't get a Kintex Ultrascale chip for X11, get maybe 8 or 9 Spartan-6 LX150 chips and use a custom board!

X11, if done properly, requires pretty much no memory. Each FPGA chip can be flashed to one algo each, except for maybe one or two that can fit two simpler algos. The algos will even be easy to pipeline - one chip is one step. Board should be custom, for fast communication between FPGA chips, and this will allow you to omit costly extras like RAM.
Hmm, I am not a developer but it sounds very possible and you definitely sound like you know what you are doing. You think you could design one and get one working and start your own FPGA company selling these if you had the right amount of investors or enough coin invested? You are the king of X11 bins and obviously know the technical side of it.

If I could get someone to solder the boards, I could probably do the dev work.

You could order a custom design i guess, but im not sure as im not into fpga's.
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 01:00:51 PM
#17
Quote from: ?? on ??
Quote from: kingcolex on September 01, 2015, 10:03:11 AM
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.

Sure it is! Stop thinking inside the box. Don't get a Kintex Ultrascale chip for X11, get maybe 8 or 9 Spartan-6 LX150 chips and use a custom board!

X11, if done properly, requires pretty much no memory. Each FPGA chip can be flashed to one algo each, except for maybe one or two that can fit two simpler algos. The algos will even be easy to pipeline - one chip is one step. Board should be custom, for fast communication between FPGA chips, and this will allow you to omit costly extras like RAM.

I only started this thread cause i heavily support cryptonote, the issue here is the fund, XMR has a development fund im not sure if that happens with any X11 coins so this can become public and not only for some users.
djm34
legendary
Activity: 1400
Merit: 1050
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 10:44:34 AM
#16
Quote from: kingcolex on September 01, 2015, 10:38:40 AM
Quote from: ?? on ??
Quote from: kingcolex on September 01, 2015, 10:03:11 AM
Quote from: Blawpaw on September 01, 2015, 07:38:34 AM
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
Would be the only logical solution and then just sell the coins and buy the cryptonight coin you are interested in but still I don't think x11 is worth enough to warrant one yet.

Sure it is! Stop thinking inside the box. Don't get a Kintex Ultrascale chip for X11, get maybe 8 or 9 Spartan-6 LX150 chips and use a custom board!

X11, if done properly, requires pretty much no memory. Each FPGA chip can be flashed to one algo each, except for maybe one or two that can fit two simpler algos. The algos will even be easy to pipeline - one chip is one step. Board should be custom, for fast communication between FPGA chips, and this will allow you to omit costly extras like RAM.
Hmm, I am not a developer but it sounds very possible and you definitely sound like you know what you are doing. You think you could design one and get one working and start your own FPGA company selling these if you had the right amount of investors or enough coin invested? You are the king of X11 bins and obviously know the technical side of it.
then a cloud mining company 
then Bahamas,
then prison  (low chance though) Grin
Blawpaw
legendary
Activity: 1596
Merit: 1027
Re: [Call to Arms] CryptoNight FPGA mining
September 01, 2015, 07:38:34 AM
#15
If I had to pay someone to code me some FPGA for mining I would choose another algorithm, at least a more profitable one Wink

You probably should consider paying this coder to work on FPGAs for X11, X13 and X15 coins Wink
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 07:49:37 PM
#14
Quote from: ?? on ??
Quote from: ?? on ??
Quote from: bitl0ck on August 31, 2015, 04:32:50 PM
People can have the option of mining at non profitability and profit later...anyways

"All we are asking from the coder is the hashrate/ price of the fpga is equal or better than the gpu's / cpu's."

If it can be done, why not? Would be the same as a gpu/cpu. ( at least)
Because it isn't as simple as you are making it out to be, the coder will need dev boards, time and money to make it work. Dev boards are very expensive when they are one of a kind and when you throw 8 bitcoins at this project and he does get a code and set up worthwhile who is going to fabricate them? Who is going to purchase the MOQ to actually have the fabrication worthwhile?


The idea of these donations going anywhere but inside someone's pocket is extremely small.

Let him learn the hard way. Sure, it's possible, but it's not simple, and it'll likely be more expensive than GPU/CPU.


Thats why we have you Wolf, you can always do your performance tweaks :p
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 04:39:08 PM
#13
Quote from: ?? on ??
Quote from: bitl0ck on August 31, 2015, 04:32:50 PM
People can have the option of mining at non profitability and profit later...anyways

"All we are asking from the coder is the hashrate/ price of the fpga is equal or better than the gpu's / cpu's."

If it can be done, why not? Would be the same as a gpu/cpu. ( at least)
Because it isn't as simple as you are making it out to be, the coder will need dev boards, time and money to make it work. Dev boards are very expensive when they are one of a kind and when you throw 8 bitcoins at this project and he does get a code and set up worthwhile who is going to fabricate them? Who is going to purchase the MOQ to actually have the fabrication worthwhile?


The idea of these donations going anywhere but inside someone's pocket is extremely small.

This is still being discussed so i cant give any further details.
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 04:32:50 PM
#12
People can have the option of mining at non profitability and profit later...anyways

"All we are asking from the coder is the hashrate/ price of the fpga is equal or better than the gpu's / cpu's."

If it can be done, why not? Would be the same as a gpu/cpu. ( at least)
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 04:29:20 PM
#11
Quote from: ?? on ??
Quote from: bitl0ck on August 31, 2015, 04:22:59 PM
Quote from: ?? on ??
Quote from: bitl0ck on August 31, 2015, 04:15:46 PM
Quote from: ?? on ??
Quote from: traumschiff on August 31, 2015, 11:43:21 AM
Quote from: ?? on ??
Why would he make a FPGA for that algorithm? It would be a lot more profitable to do it for X11 and mine dash if he can get these FPGA to be efficient enough. I think i am smelling scam on this one.

Also I can't imagine custom built FPGA miners to be cheap enough for a decent enough ROI timeframe. If the guy could build custom FPGAs and mining software which would be profitable on XMR or whatever cryptonote fork is the most profitable he would have done it already and sold it for private investors.

Vanillacoin has FPGA mining because the developer did it by himself (some community members took part in the dev process) to support the project.
Exactly if it could be done profitably it would have already been done by now and we would have seen it on the more profitable algorithms first.

VanillaCoin has nothing to do with this besides having fpga mining.

Maybe no one tried to build custom fpgas yet, or maybe the "botnets" are fpgas or asics.

Also "if it could be done profitably it would have already been done by now " this argument isnt really valid, did you ever imagine something like bitcoin in 2000?
Listen if it was easily done it would have been done, were talking money now and when there is money to be made people are paying attention. The same people who made the FPGA's back for bitcoin before asic I am sure looked at it but I also am sure they looked at X11 more. When a coin has enough cap and people to want to buy FPGA's constantly they will be made that is why Litecoin led to Asics and if Dash stays up or picks up a larger market cap we will see an X11 fpga or an asic. If any Cryptonight coin starts being huge and having a very large market cap then yeah we will see one then because it will be profitable to do so.

You are just stating an theoretical evolution which might not be wrong at all, have you seen bytecoin marketcap?
Yeah and look at litecoin the lowest cap that has an asic, right around 120million, so Bytecoin has 8% of the total value of litecoin. When it hits 50 million then I would expect some custom hardware coming out for it.

People make the evolution, its not directly related to the marketcap. If its profitable why not?
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 04:22:59 PM
#10
Quote from: ?? on ??
Quote from: bitl0ck on August 31, 2015, 04:15:46 PM
Quote from: ?? on ??
Quote from: traumschiff on August 31, 2015, 11:43:21 AM
Quote from: ?? on ??
Why would he make a FPGA for that algorithm? It would be a lot more profitable to do it for X11 and mine dash if he can get these FPGA to be efficient enough. I think i am smelling scam on this one.

Also I can't imagine custom built FPGA miners to be cheap enough for a decent enough ROI timeframe. If the guy could build custom FPGAs and mining software which would be profitable on XMR or whatever cryptonote fork is the most profitable he would have done it already and sold it for private investors.

Vanillacoin has FPGA mining because the developer did it by himself (some community members took part in the dev process) to support the project.
Exactly if it could be done profitably it would have already been done by now and we would have seen it on the more profitable algorithms first.

VanillaCoin has nothing to do with this besides having fpga mining.

Maybe no one tried to build custom fpgas yet, or maybe the "botnets" are fpgas or asics.

Also "if it could be done profitably it would have already been done by now " this argument isnt really valid, did you ever imagine something like bitcoin in 2000?
Listen if it was easily done it would have been done, were talking money now and when there is money to be made people are paying attention. The same people who made the FPGA's back for bitcoin before asic I am sure looked at it but I also am sure they looked at X11 more. When a coin has enough cap and people to want to buy FPGA's constantly they will be made that is why Litecoin led to Asics and if Dash stays up or picks up a larger market cap we will see an X11 fpga or an asic. If any Cryptonight coin starts being huge and having a very large market cap then yeah we will see one then because it will be profitable to do so.

You are just stating an theoretical evolution which might not be wrong at all, have you seen bytecoin marketcap?
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 04:15:46 PM
#9
Quote from: ?? on ??
Quote from: traumschiff on August 31, 2015, 11:43:21 AM
Quote from: ?? on ??
Why would he make a FPGA for that algorithm? It would be a lot more profitable to do it for X11 and mine dash if he can get these FPGA to be efficient enough. I think i am smelling scam on this one.

Also I can't imagine custom built FPGA miners to be cheap enough for a decent enough ROI timeframe. If the guy could build custom FPGAs and mining software which would be profitable on XMR or whatever cryptonote fork is the most profitable he would have done it already and sold it for private investors.

Vanillacoin has FPGA mining because the developer did it by himself (some community members took part in the dev process) to support the project.
Exactly if it could be done profitably it would have already been done by now and we would have seen it on the more profitable algorithms first.

VanillaCoin has nothing to do with this besides having fpga mining.

Maybe no one tried to build custom fpgas yet, or maybe the "botnets" are fpgas or asics.

Also "if it could be done profitably it would have already been done by now " this argument isnt really valid, did you ever imagine something like bitcoin in 2000?
traumschiff
legendary
Activity: 1498
Merit: 1001
180 BPM
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 11:43:21 AM
#8
Quote from: ?? on ??
Why would he make a FPGA for that algorithm? It would be a lot more profitable to do it for X11 and mine dash if he can get these FPGA to be efficient enough. I think i am smelling scam on this one.

Also I can't imagine custom built FPGA miners to be cheap enough for a decent enough ROI timeframe. If the guy could build custom FPGAs and mining software which would be profitable on XMR or whatever cryptonote fork is the most profitable he would have done it already and sold it for private investors.

Vanillacoin has FPGA mining because the developer did it by himself (some community members took part in the dev process) to support the project.
bitl0ck
sr. member
Activity: 240
Merit: 250
Re: [Call to Arms] CryptoNight FPGA mining
August 31, 2015, 09:56:23 AM
#7
Quote from: ?? on ??
Why would he make a FPGA for that algorithm? It would be a lot more profitable to do it for X11 and mine dash if he can get these FPGA to be efficient enough. I think i am smelling scam on this one.

Cause i support cryptonight coins, thats the why.
Pages:
Bitcoin Forum>Alternate cryptocurrencies>Mining (Altcoins)
Jump to: