[ANN] Kryptohash | Brand new PoW algo | 320bit hash | ed25519 | PID algo for dif

[KeyserSozeMC]

-address- ^_^still mining @ 6khs

Hey man, you better put that addy in your signature, instead here in the topic.
You might get banned and topic can be closed.

It's against forum rules to "spam" with addresses.

Peace

[1715180020]

1715180020

[slashbinslashnoname]

Did not know the rules, thanks for the advice Its to have some transactions on the network, did'nt expected it could be considerated as spam. I'll put it in the sig ** or not

[Atomicat]

You can only mine on the first card until dev works on the miner.  Everyone with multi card rigs has this problem.

I'll try starting separate instances, see if that works.

Has been tried, you may do better if you install 12x/13x drivers.  I didn't go that far because of the increased performance offered with the 14.9 drivers.

Nope!  HW errors still there and a ton of rejects on the 7950.  Dug up the Stilt 290 bios mods, picked a middle-ground one and it's much more agreeable shall we say.  Now I can overclock that properly for Diamond, check out this algo with the 7950, of course 7 cores for the Magi and my dance card is full!  *sigh*  Still won't really know if miner's working until I pick up a block and I'm still completely um... indescribable... Now I stopped programming a long time ago, 386-486 long time ago, but I programmed a lot for a long long time before that, back past the Apple/Commodore 6502 can't mult or div back to punchcards and Fortran... and I can't imagine programing a set of tools like the qt wallet software, cgminer, minerd etc etc, and having NO FEEDBACK to see how they're communicating!  Why hasn't someone just added that in?  The wallet is the pipeline, no?  It coordinates, hands out work etc etc, so, well you know what I'm saying.  It doesn't seem to be a problem to add a network bandwidth monitor, why not a simple readout to show connected miners?

And just now I notice two Found Blocks for pool 0!   And then I notice that the wallet hasn't synched and is 22 hours behind on block 5654 for some reason.  *sigh*  Time for something else for a while.

[WORE]

I think we have another fork guys ..

What makes you say that?  Aside from the first fork the rest were hard forks by dev to sort out the diff algo.

[kondiomir]

Sorry I'm wrong. 

[WORE]

You can only mine on the first card until dev works on the miner.  Everyone with multi card rigs has this problem.

I'll try starting separate instances, see if that works.

Has been tried, you may do better if you install 12x/13x drivers.  I didn't go that far because of the increased performance offered with the 14.9 drivers.

Nope!  HW errors still there and a ton of rejects on the 7950.  Dug up the Stilt 290 bios mods, picked a middle-ground one and it's much more agreeable shall we say.  Now I can overclock that properly for Diamond, check out this algo with the 7950, of course 7 cores for the Magi and my dance card is full!  *sigh*  Still won't really know if miner's working until I pick up a block and I'm still completely um... indescribable... Now I stopped programming a long time ago, 386-486 long time ago, but I programmed a lot for a long long time before that, back past the Apple/Commodore 6502 can't mult or div back to punchcards and Fortran... and I can't imagine programing a set of tools like the qt wallet software, cgminer, minerd etc etc, and having NO FEEDBACK to see how they're communicating!  Why hasn't someone just added that in?  The wallet is the pipeline, no?  It coordinates, hands out work etc etc, so, well you know what I'm saying.  It doesn't seem to be a problem to add a network bandwidth monitor, why not a simple readout to show connected miners?

And just now I notice two Found Blocks for pool 0!   And then I notice that the wallet hasn't synched and is 22 hours behind on block 5654 for some reason.  *sigh*  Time for something else for a while.

If you are trying to test the 290 with this coin, you'll need to set it up as device 0.

[wr104]

I think we have another fork guys ..

What makes you say that?  Aside from the first fork the rest were hard forks by dev to sort out the diff algo.

Sorry I'm wrong.  

Dude, you just made me spill my coffee.  

need some help to compile wallet

current wallet gives some erros when compiling it.
 want to push random blockreward to maximum, i need custom wallet!

What OS?

[berbip]

Blocks are very slow, my miner waits a lot for work.
Is this normal ?
Any plans to enable multi gpu mining soon ?

[WORE]

Blocks are very slow, my miner waits a lot for work.
Is this normal ?
Any plans to enable multi gpu mining soon ?

Yes is normal, there is no 'dummy' work with this coin.

[TheCoinFinder]

If someone can make a kryptohash python hash I don't mind putting up a pool (and seeing how to modify it)

[karimdr2]

Dev can you post instructions or flags for compiling with gcc, I have boost linking error on linux and osx. clang refuse to configure with c++11 flag btw.

[wr104]

There seems to be another problem with the Diff algo.

I think I'm done with this algo.  I'm going to dump it and replace it with something known to work

[antonio8]

There seems to be another problem with the Diff algo.

I think I'm done with this algo.  I'm going to dump it and replace it with something known to work

You mean the difficulty algo? Not the hashing algo, hopefully?

[shojayxt]

I encrypted a wallet and when I launched kryptohash after it closed down I got the following error and kryptohash wallet would not open.  The wallet was created in version 3.2 so I thought that might be the problem but I created a new wallet in V 3.6 and got the same error after encrypting the wallet.  I had backed up the wallet prior to encrypting so I didn't lose anything.  I am running Windows 8 if that matters.

[TheCoinFinder]

For misquoting fun...

There seems to be another problem with the algo.

I think I'm done with this.  I'm going to dump it [the premine] and replace [my stash] with something known to work [ie bitcoin]

Just kidding :p

[wr104]

There seems to be another problem with the Diff algo.

I think I'm done with this algo.  I'm going to dump it and replace it with something known to work

You mean the difficulty algo? Not the hashing algo, hopefully?

There is nothing wrong with KSHAKE320 Proof of Work. It is the PID algo that calculates Diff the one that has been nothing but problems since day one.  

The hashing algo was thoroughly tested for months and even fine tuned for OpenCL using CodeXL but, unfortunately I cannot say the same thing about the Diff algo because, testing Diff algos require real world environments. A Testnet with a couple of GPUs at home isn't going to simulate the real conditions of the network so, I relied on simulations using models which clearly were incorrect.
  

[jox]

There seems to be another problem with the Diff algo.

I think I'm done with this algo.  I'm going to dump it and replace it with something known to work

So we can stop mining now and wait for next wallet with new hardfork?

[z38630610]

There seems to be another problem with the Diff algo.

I think I'm done with this algo.  I'm going to dump it and replace it with something known to work

You mean the difficulty algo? Not the hashing algo, hopefully?

There is nothing wrong with KSHAKE320 Proof of Work. It is the PID algo that calculates Diff the one that has been nothing but problems since day one.  

The hashing algo was thoroughly tested for months and even fine tuned for OpenCL using CodeXL but, unfortunately I cannot say the same thing about the Diff algo because, testing Diff algos require real world environments. A Testnet with a couple of GPUs at home isn't going to simulate the real conditions of the network so, I relied on simulations using models which clearly were incorrect.
  

wallet not sync ? now stop mining ?

[wr104]

There seems to be another problem with the Diff algo.

I think I'm done with this algo.  I'm going to dump it and replace it with something known to work

So we can stop mining now and wait for next wallet with new hardfork?

Yes. Please stop while I investigate what went wrong at retarget

[wr104]

Ok, this is going to take a while.

Please be patient.   I'm debugging why some wallets ended up calculating a different Diff.