Author

Topic: [ANN] [HVC] Heavycoin - Ultra-secure, Decentralized Block Reward Voting, Fast - page 195. (Read 542277 times)

full member
Activity: 210
Merit: 100

I was laughing from the beginning on about this coin. Taken from their website:

Quote
The problem is that, due to Quarkcoin's simple use of function compositions, if BLAKE512(x) has collisions, then so does BMW512(BLAKE512(x)) and SKEIN512(KECCAK512(... and so on, until we reach Quark(x), which also has collisions. Similarly, if SKEIN-512 or Grøestl-512 have collisions, then so does Quark(x). Simply put, if there's a collision attack or second-preimage attack for BLAKE-512(x), then Quark(x) is cracked.

That's already complete nonsense. The developers are just full of crap. Let's say there is a second-preimage attack for BLAKE-512. So am able to compute an input value X, so that BLAKE-512(X) = Y, where Y matches the difficulty and finds a block. If Keccac runs before Blake, I still need to find a l need to find an input Z, so that Keccak(Z) = X.

We never claimed Quark is bad for second-preimage resistance.  It's pretty good.  We claim Quark is no better for collisions than BLAKE-512, and we have proposed an alternative solution that does not rely solely on BLAKE-512.  With regard to second-preimage on Blake, the Blake hash is the first function in the chain, so Keccak cannot come before Blake.  (We should probably get the webdev to fix the confusion on the website: "if there's a collision attack or second-preimage attack for..")

Furthermore, they claim:

Quote
Heavycoin takes 64 bits from the output of each of 4 well-known cryptographic hash functions (SHA-256, Keccak-512, Grøestl-512 and BLAKE-512) and interleaves these bits into a combined 256-bit hash that is more resistant against collisions and second-preimage attacks.

LMFAO. These kids have no clue what they are talking about. In fact, the opposite is true. If one hash function is broken, it's possible to freely choose the corresponding 64 bit in the output. Oh guys, just quit this coin. It's pathetic and an insult to anyone capable of adding 1 and 1.

In Heavycoin if one of the 4 cryptographic hashes is completely broken, then you lose 64 bits out of 256 bits.  In Bitcoin if SHA-256 is completely broken, then you lose 256 bits.  Both scenarios are unlikely, but in one of them you lose fewer bits.
sr. member
Activity: 1112
Merit: 256

I was laughing from the beginning on about this coin. Taken from their website:

Quote
The problem is that, due to Quarkcoin's simple use of function compositions, if BLAKE512(x) has collisions, then so does BMW512(BLAKE512(x)) and SKEIN512(KECCAK512(... and so on, until we reach Quark(x), which also has collisions. Similarly, if SKEIN-512 or Grøestl-512 have collisions, then so does Quark(x). Simply put, if there's a collision attack or second-preimage attack for BLAKE-512(x), then Quark(x) is cracked.

That's already complete nonsense. The developers are just full of crap. Let's say there is a second-preimage attack for BLAKE-512. So am able to compute an input value X, so that BLAKE-512(X) = Y, where Y matches the difficulty and finds a block. If Keccac runs before Blake, I still need to find a l need to find an input Z, so that Keccak(Z) = X.

Furthermore, they claim:

Quote
Heavycoin takes 64 bits from the output of each of 4 well-known cryptographic hash functions (SHA-256, Keccak-512, Grøestl-512 and BLAKE-512) and interleaves these bits into a combined 256-bit hash that is more resistant against collisions and second-preimage attacks.

LMFAO. These kids have no clue what they are talking about. In fact, the opposite is true. If one hash function is broken, it's possible to freely choose the corresponding 64 bit in the output. Oh guys, just quit this coin. It's pathetic and an insult to anyone capable of adding 1 and 1.



sr. member
Activity: 1106
Merit: 255
Betking.io - Best Bitcoin Casino
Today is the official men´s day "Steak & blowjob day"

And all this time I thought it was Pi Day. Sheesh, have I got it all wrong!
sr. member
Activity: 288
Merit: 253
God I'm so horny right now

Good for you.

Today is the official men´s day "Steak & blowjob day"
full member
Activity: 205
Merit: 100
newbie
Activity: 11
Merit: 0
God I'm so horny right now
member
Activity: 112
Merit: 10
    
Re: [ANN] cudaMiner - a new litecoin mining application [Windows/Linux]
Today at 12:36:02 PM
   
Reply with quote  #8701
Quote from: phm on Today at 12:33:06 PM

What hash rates are you getting on your hardware? Just want to compare... something.  Roll Eyes

AMD Phenom II X6 1055T - CPU standalone ca 95 kHash/s on 6 threads

AMD Phenom II X6 1055T using the first 3 GPUs in the machine and 6 CPU threads: close to 800 kHash/s   (Groestl remains on CPU)

cannot get GPU use readings, as GPU-z has decided to always crash on start after the previous driver update (revision 334 drivers I think). Currently running R 335 drivers from nvidia.
   

ok it doesnt seem so bad actually, hes only getting that improvement with the cpu and 3 gpus, can't see it being a massive gain overall and would require most rig owners to buy better cpus. the difficulty might not even make it viable.
sr. member
Activity: 349
Merit: 250


http://heavy.stablehash.com is up on RC2. OPTIMIZED MINER! (20% FASTER!!)


Source: http://heavy.stablehash.com/heavycoin-rc2-FAST.zip

Somebody can build for windows.
member
Activity: 112
Merit: 10
Thank you massiveman - and you are correct that I missed your point somewhat. Your argument makes sense in light of the current state of the GPU miner requiring to offload one of the five algos onto the CPU. However, it's only because the Christians haven't yet managed to accelerate the fifth algo - Groestl. Once that gets sorted, then HVC can be offloaded from the CPU entirely.

Note the distinction between porting and accelerating - they are already able to offload Groestl to GPU just fine. They just haven't been able to accelerate it yet. So mining HVC without significant CPU load is in the works, hence miners will be able to hang onto their beloved Sempies.

You are probably right although I hope they aren't able to offload HVC from CPU altogether because I happen to have a good graphics card in my rig so if I could get the 8x increase I would essentially be running on 192 cores without many others being able to do similar Smiley. Regardless though it wont be the death of this coin, being gpu mineable might actually increase the coins popularity. Dark is GPU mineable and that is still doing well.
sr. member
Activity: 1106
Merit: 255
Betking.io - Best Bitcoin Casino
Thank you massiveman - and you are correct that I missed your point somewhat. Your argument makes sense in light of the current state of the GPU miner requiring to offload one of the five algos onto the CPU. However, it's only because the Christians haven't yet managed to accelerate the fifth algo - Groestl. Once that gets sorted, then HVC can be offloaded from the CPU entirely.

Note the distinction between porting and accelerating - they are already able to offload Groestl to GPU just fine. They just haven't been able to accelerate it yet. So mining HVC without significant CPU load is in the works, hence miners will be able to hang onto their beloved Sempies.
member
Activity: 112
Merit: 10

I think 8x improvement is too shabby when it means sacrificing the ability to gpu mine also. whos gonna really benefit? those with decent cpus and decent gpus in the same rig. most gpu miners have shitty single core processors which in this instance will be turned into 8 threads but again to do this they lose the ability to gpu mine other coins.

High end AMD cpus are pretty cheap and it`s not that changing mobo and cpu on mining rig is hard.

this is a good point but again it will take a monetary investment and leave them unable to gpu mine. say they buy quad core amd chips, this will give them essentially 32 cores. o not be able to GPU mine the sacrifice isn't worth it, I have 2x12 core xeons and only get around 280hvc per day from them and the difficulty is low now. Like I said the ones who will benefit are those with good cpus a single nice GPU as well, basically gaming rigs.
member
Activity: 112
Merit: 10
I think 8x improvement is too shabby when it means sacrificing the ability to gpu mine also. whos gonna really benefit? those with decent cpus and decent gpus in the same rig. most gpu miners have shitty single core processors which in this instance will be turned into 8 threads but again to do this they lose the ability to gpu mine other coins.
That's a constant though - no matter what coin you ARE mining, you're NOT mining all the others. If you think of mining as lost opportunity on every coin that you're not mining, then you, my friend, are a pessimist of considerable capacity!

no my point(which you have clearly missed) is that if you have 4x gpus in a mining rig with a single core processor(this is standard) then are you going to want to continue gpu mining or choose to in essence convert that single core processor into a quad core(8 thread) processor. That one is a no brainer, with an 8 thread processor I could mine very few HVCs per day but in monetary terms it would be much more profitable to just use the rig as intended, as an out and out GPU miner. Thus this 8x increase only really benefits someone who say has a quad core already with a decent gpu which can essential convert it into a 32 core processor for minign HVC. not many in the crypto world have this, they either are a cpu miner with loads of cores and shitty gpus or gpu miners with loads of gpus and shitty single core processors.
newbie
Activity: 15
Merit: 0

I think 8x improvement is too shabby when it means sacrificing the ability to gpu mine also. whos gonna really benefit? those with decent cpus and decent gpus in the same rig. most gpu miners have shitty single core processors which in this instance will be turned into 8 threads but again to do this they lose the ability to gpu mine other coins.

High end AMD cpus are pretty cheap and it`s not that changing mobo and cpu on mining rig is hard.
sr. member
Activity: 1106
Merit: 255
Betking.io - Best Bitcoin Casino
I think 8x improvement is too shabby when it means sacrificing the ability to gpu mine also. whos gonna really benefit? those with decent cpus and decent gpus in the same rig. most gpu miners have shitty single core processors which in this instance will be turned into 8 threads but again to do this they lose the ability to gpu mine other coins.
That's a constant though - no matter what coin you ARE mining, you're NOT mining all the others. If you think of mining as lost opportunity on every coin that you're not mining, then you, my friend, are a pessimist of considerable capacity!
member
Activity: 112
Merit: 10
devs when you get a change could you please let us know about this gpu support for cpu mining which has been linked above?

To the best of our knowledge it's not possible.  It sounds like he's achieved a speedup for some of the underlying hash functions (to be expected), but the overall hash still requires a CPU.  Later on he says "provided that we achieve the necessary speed-ups for this to be worthwhile", which implies he's unsure of the final results.

For the coins sake, I hope he/she's unsuccessful. 
The Christians will be successful - make no mistake about that. GPU mining is coming and it's coming soon. And yes, while they may be unsure of the final results, 8x performance ain't too shabby for alpha-stage code.

I think 8x improvement is too shabby when it means sacrificing the ability to gpu mine also. whos gonna really benefit? those with decent cpus and decent gpus in the same rig. most gpu miners have shitty single core processors which in this instance will be turned into 8 threads but again to do this they lose the ability to gpu mine other coins.
sr. member
Activity: 1106
Merit: 255
Betking.io - Best Bitcoin Casino
devs when you get a change could you please let us know about this gpu support for cpu mining which has been linked above?

To the best of our knowledge it's not possible.  It sounds like he's achieved a speedup for some of the underlying hash functions (to be expected), but the overall hash still requires a CPU.  Later on he says "provided that we achieve the necessary speed-ups for this to be worthwhile", which implies he's unsure of the final results.

For the coins sake, I hope he/she's unsuccessful. 
The Christians will be successful - make no mistake about that. GPU mining is coming and it's coming soon. And yes, while they may be unsure of the final results, 8x performance ain't too shabby for alpha-stage code.
member
Activity: 112
Merit: 10
devs when you get a change could you please let us know about this gpu support for cpu mining which has been linked above?

To the best of our knowledge it's not possible.  It sounds like he's achieved a speedup for some of the underlying hash functions (to be expected), but the overall hash still requires a CPU.  Later on he says "provided that we achieve the necessary speed-ups for this to be worthwhile", which implies he's unsure of the final results.

Thanks Dev, that's what I thought in fairness, plus logically most gpu miners only have single core processors so it wont give them a huge spike in performance and any improvement will make it impossible to gpu mine at the same time.
sr. member
Activity: 382
Merit: 256
devs when you get a change could you please let us know about this gpu support for cpu mining which has been linked above?

To the best of our knowledge it's not possible.  It sounds like he's achieved a speedup for some of the underlying hash functions (to be expected), but the overall hash still requires a CPU.  Later on he says "provided that we achieve the necessary speed-ups for this to be worthwhile", which implies he's unsure of the final results.

For the coins sake, I hope he/she's unsuccessful. 
member
Activity: 112
Merit: 10

Thanks for the replies guys.

I do not want to get into ASIC because it's already low-power; I wouldn't take advantage of my free power. Building some basic PC cases is the way to go ı believe,with relatively modern CPUs.



I've seen a blade server with 40 cores, but it needs 8000W to run it, is that really worth it? That's costing around £1 and hour on energy.

that seems silly, i have 24 cores in my pc which is 750w, I'd say it isn't worth it.

Did you build your PC yourself, or is it one that you can buy off the shelf?

had mine built to spec at www.scan.co.uk, 6k pc though.

wow, you paid £6k for a PC? Do you have any other use for it apart from mining? That's quite a lot of CPU power.

yeah i basically mine with gpu on it also but mainly I just wanted a monster that will be pretty future proof and so far I love it, does exactly what it says on the tin. I can game to the max if i want while still using 67% of the cpus to mine. In fact for use as a normal pc its fine when using 47 of the 48 threads to mine. don't get me wrong i don't have buckets of money but I had an ebay business over the xmas period and used the profits to buy this thing, boom!. if you want to be a CPU miner just buy second hand servers from ebay.
full member
Activity: 210
Merit: 100
devs when you get a change could you please let us know about this gpu support for cpu mining which has been linked above?

To the best of our knowledge it's not possible.  It sounds like he's achieved a speedup for some of the underlying hash functions (to be expected), but the overall hash still requires a CPU.  Later on he says "provided that we achieve the necessary speed-ups for this to be worthwhile", which implies he's unsure of the final results.
Jump to: