[XPM] [ANN] Primecoin Release - First Scientific Computing Cryptocurrency

[craslovell]

I have 48 cores in one machine mining but 0 primespersec after 20 mins. My other machines are all working fine

Each thread gets it's own sieve, you're probably running into cache issues.

Then I have no clue what to try next lol, maybe less cores

[1713509096]

1713509096

[1713509096]

1713509096

[1713509096]

1713509096

[craslovell]

I have 48 cores in one machine mining but 0 primespersec after 20 mins. My other machines are all working fine

Each thread gets it's own sieve, you're probably running into cache issues.

Then I have no clue what to try next lol, maybe less cores

Eh not helping either

[Boing7898]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

[Zalfrin]

I was wondering if the prime95 client can be used for pool mining

This is not doing Mersenne prime discovery like prime95, so no.

I'm not convinced there's really any scientific value being created here, but maybe a scientist will step forward with something... The article that was posted about it talked about maybe prime factorization optimizations could be discovered, but that claim seems dubious.

[Buffer Overflow]

Has anybody made a GPU miner yet?

a GPU miner will drive up the difficulty and lower the block reward. We'll be dealing with the same exact shit again. What's the point?

To stop botnets, or they will eventually control the coin.

[nearmiss]

I was wondering if the prime95 client can be used for pool mining

This is not doing Mersenne prime discovery like prime95, so no.

I'm not convinced there's really any scientific value being created here, but maybe a scientist will step forward with something... The article that was posted about it talked about maybe prime factorization optimizations could be discovered, but that claim seems dubious.

May turn out its not scientifically useful I suppose....but still a step in the right direction.

[Buffer Overflow]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

Not as easily. It is very much harder. Plus most users don't have high-end GPUs.

[Boing7898]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

Not as easily. It is very much harder. Plus most users don't have high-end GPUs.

Actually, it isn't. Most "silent miners" sold on forums have support for both CPU and GPU.
Not that hard, I coded myself a silent miner that would mine with GPU and CPU using CGMiner and cudaMiner in autoit.

[mustyoshi]

"We have to stop ec2 instances"
"We have to stop botnets"

And why, exactly?

muh profit

[Buffer Overflow]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

Not as easily. It is very much harder. Plus most users don't have high-end GPUs.

Actually, it isn't. Most "silent miners" sold on forums have support for both CPU and GPU.
Not that hard, I coded myself a silent miner that would mine with GPU and CPU using CGMiner and cudaMiner in autoit.

Then you have just solved the debate if Bitcoin needed ASICs or not.

[Boing7898]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

Not as easily. It is very much harder. Plus most users don't have high-end GPUs.

Actually, it isn't. Most "silent miners" sold on forums have support for both CPU and GPU.
Not that hard, I coded myself a silent miner that would mine with GPU and CPU using CGMiner and cudaMiner in autoit.

Then you have just solved the debate if Bitcoin needed ASICs or not.

Do you mean that Bitcoin needs ASICs to counter those huge GPU-mining botnets?
Yeah, you're probably right.
But still, CPU-only mining or CPU and GPU mining, botnets will still be used to mine.

[mr_random]

I was wondering if the prime95 client can be used for pool mining

This is not doing Mersenne prime discovery like prime95, so no.

I'm not convinced there's really any scientific value being created here, but maybe a scientist will step forward with something... The article that was posted about it talked about maybe prime factorization optimizations could be discovered, but that claim seems dubious.

I have a very strong academic mathematics background. 'scientific value' is a subjective statement. Most mathematicians would regard anything new or original as advancing mathematics. Do you think patterns in integers have scientific value? Depends on the pattern to an extent and if there are any clear consequences but you'll get varied answers to that question. Some mathematicians and especially scientists such as physicists are snobbish and only regard maths that has a 'real world' application as being useful. It's not clear to me whether this project is finding new prime chain sequences within the known prime numbers or actually trying to determine new, unknown prime numbers using the prime chains as a proof of work. Either way, it has value since it will add new data which will either disprove conjectures or strengthen the likelihood of them being true... plus the analysis of the discovered data could in theory lead to new conjectures (note I know next to nothing about these chains and how much data of them is out there, I'm just writing this reply of the top of my head).

You can also look at this another way. This project could provide a financial incentive for individuals to develop more optimised GPU prime finding algorithms, the theory of which could lead to a greater understanding of prime number distribution e.g. a new sieve or similar construct.

Sunny, can you please clarify whether this project is finding new prime chain sequences within the known prime numbers or actually trying to determine new, unknown prime numbers using the prime chains as a proof of work.

[Moebius327]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

Not as easily. It is very much harder. Plus most users don't have high-end GPUs.

It is actually very easy for a botnet to mine on GPU only. These sorts of botnets target only gamer PCs. (binded torrent games, game hacks, etc.)

[Buffer Overflow]

"We have to stop ec2 instances"
"We have to stop botnets"

And why, exactly?

EC2 virtual machines are performing poorly with this coin. I have a hunch it's something to do with their entropy. Only guessing at the moment mind.

[paulthetafy]

"We have to stop ec2 instances"
"We have to stop botnets"

And why, exactly?

EC2 virtual machines are performing poorly with this coin. I have a hunch it's something to do with their entropy. Only guessing at the moment mind.

+1  I have tried ubuntu varieties and had zero primespersec and the windows AMI produces5 primeapersec.  So don't bother trying

[laughingbear]

I would also love to know...  are we discovering new primes?

[Boing7898]

GPU mining is absolutely pointless. A public GPU miner will only drive up difficulty, and we'll be right back at ground zero again. Pointless.



Anyways, I have now mined 140 XPM! (ON a single computer) Seems I'm quite lucky.

There are two reasons why I ask ... one is "selfish" : I have 3 strong GPUs while only 1 mediocre quad CPU. ... the other is, I think it's better  to have GPU mining in place IMO is more resistant to botnet-mining. Or am I mistaken to believe this?

Mistaken, imo.
Botnets can use GPU to mine as much as they can use CPU to mine.

Not as easily. It is very much harder. Plus most users don't have high-end GPUs.

It is actually very easy for a botnet to mine on GPU only. These sorts of botnets target only gamer PCs. (binded torrent games, game hacks, etc.)

Exactly, you just have to code a simple downloader (or if you want something more professional code something that injects into memory) that downloads CGminer and runs it at startup with certain parameters hidden.

[nmersulypnem]

I was wondering if the prime95 client can be used for pool mining

This is not doing Mersenne prime discovery like prime95, so no.

I'm not convinced there's really any scientific value being created here, but maybe a scientist will step forward with something... The article that was posted about it talked about maybe prime factorization optimizations could be discovered, but that claim seems dubious.

I have a very strong academic mathematics background. 'scientific value' is a subjective statement. Most mathematicians would regard anything new or original as advancing mathematics. Do you think patterns in integers have scientific value? Depends on the pattern to an extent and if there are any clear consequences but you'll get varied answers to that question. Some mathematicians and especially scientists such as physicists are snobbish and only regard maths that has a 'real world' application as being useful. It's not clear to me whether this project is finding new prime chain sequences within the known prime numbers or actually trying to determine new, unknown prime numbers using the prime chains as a proof of work. Either way, it has value since it will add new data which will either disprove conjectures or strengthen the likelihood of them being true... plus the analysis of the discovered data could in theory lead to new conjectures (note I know next to nothing about these chains and how much data of them is out there, I'm just writing this reply of the top of my head).

You can also look at this another way. This project could provide a financial incentive for individuals to develop more optimised GPU prime finding algorithms, the theory of which could lead to a greater understanding of prime number distribution e.g. a new sieve or similar construct.

Sunny, can you please clarify whether this project is finding new prime chain sequences within the known prime numbers or actually trying to determine new, unknown prime numbers using the prime chains as a proof of work.

There are websites where I can download the first few MILLION primes, and the largest know prime is millions of DIGITS long.  Handling primes of millions of digits doesn't seem to be in the code, so I have a hard time understanding what new information this project can add to the mathematics community...

[Buffer Overflow]

EC2 virtual machines are performing poorly

They're always performing very poorly compared to their cost. (Unless what you're looking for is to be a hipster.)

The entropy is an old issue of virtual machines any noob should be able to fix.

I meant on a hardware primespersec basis, not cost basis.

Easy to fill with prng entropy, but that didn't make a difference. Of course that's not truly random.

What's your theory?

[coinerd]

So far this coin has been pretty funny.

I don't feel like typing this out again, so....

just gotta figure out how to convert it into a prime number

There is no "prime number". It appears the proof of work certificate is evidence that you identified a "prime chain" of acceptable type which is divisible by the block header hash.

How we extract the new primes from the Primecoin Blockchain?

Someone understands how this research the coin does can be obtained, it's not clear for me from the Primecoin whitepaper.

This is not research.  It's just busy work.

I hope Sunny is releasing soon an updated whitepaper or more details, I don't understand why so much silence.

The real explanation for this coin is on the PPCoin forums ...

In a few years that chart would have many different types of proof-of-work, bitcoin's mining share would be further reduced. This means the relative security against 51% attack is  going to be weakened for each pure proof-of-work cryptocurrency.

Primecoin may help speed up this process, taking a significant piece of mining market while attracting a larger user base to both primecoin and ppcoin. I believe it could strengthen our positioning against litecoin and other potential competitions.

http://www.ppcointalk.org/index.php?topic=288.msg1557#msg1557

The confusion surrounding the coin is very helpful to the goal, I think. Acting mysterious doesn't hurt any either.

If there was scientific value to it he would be proud and careful to explain it clearly...

Clicky Clicky go read the link.  That quote from sunny is just as "good" in it's full context.