Vertcoin - 1 | ASIC - 0 | Lyra2RE | Decentralised | GPU Mineable | Open Source

[fastlan]

nevermind working great with 15.7 drivers!
thanks wolf0 you the man!

[fredeq]

Nicehash is reporting 4,7Mhash (constant average) for my 2x750ti while ccminer reports that one card is doing 4,7Mhash.

Can anyone confirm this?

[djmic2k]

Nicehash is reporting 4,7Mhash (constant average) for my 2x750ti while ccminer reports that one card is doing 4,7Mhash.

Can anyone confirm this?

--diff 2

[fredeq]

Nicehash is reporting 4,7Mhash (constant average) for my 2x750ti while ccminer reports that one card is doing 4,7Mhash.

Can anyone confirm this?

--diff 2

I have added that switch to my config, but it didnt change anything (was it supposed to be a switch?)

I have seen those one-line posts before, its very cryptic, could you please elaborate?

[TheCoinFinder]

Please help spread the hash between active vertcoin pools.

[NiceHashSupport]

djm, you should really explain, why you made your sgminer with diff multi 256, but your ccminer with old lyra2 diff multi 128? This has caused a lot of confusion among miners.

[kenshirothefist]

djm, you should really explain, why you made your sgminer with diff multi 256, but your ccminer with old lyra2 diff multi 128? This has caused a lot of confusion among miners.

Yeah, exactly ... since tpruvot already adapted the diff multi 256 for lyra2rev2 for cpuminer-multi I suggest you to change ccminer diff multi to 256, and make new binaries ... and other devs should follow you (sp_, etc ...) ... this should be fixed asap.

[GarlukKY]

Please help spread the hash between active vertcoin pools.

4% pool fee on Verters?  That's enough to get me to move.

[pwpwpw]

Please help spread the hash between active vertcoin pools.

4% pool fee on Verters?  That's enough to get me to move.

moved after they failed to resolve the stratum interrupt issue, that 4% must be a glitch or they want people to use other pools too

[fastlan]

mine on miningpoolhub..it works great

[ReiMomo]

and http://aikapool.com/vtc

[TheCoinFinder]

Please help spread the hash between active vertcoin pools.

4% pool fee on Verters?  That's enough to get me to move.

Yes, its only temporary but verters has around or over 51% at times, which I do not want.

[djm34]

djm, you should really explain, why you made your sgminer with diff multi 256, but your ccminer with old lyra2 diff multi 128? This has caused a lot of confusion among miners.

the never ending story continues...  

For ccminer (the version I maintain at least) they both uses the same setting old and new, didn't touch the difficulty, and as far as I remember I always had to use a diff 2 to get it working on nicehash with the old lyra, even though it wasn't the case on some other pool... so you are probably more at the origin of the confusion than me  )

... and this is the origin of the discrepancies between version:  us (miner devs) trying to accommodate one or the other pool... (so don't expect me to follow that path again)

Also it has always been intended to be 256 (even the old), and for the old,  128 it is a request by someone telling me that it is 128 everywhere for the old lyra.


edit: I have updated github to incorporate Wolf0 change (the pragma unroll), thanks.

to pool owner: if you want specific difficulties, just talk to other pool owner and come up with a setting which can be use everywhere...

[NiceHashSupport]

Regarding old lyra2: we used standard lyra2re diff that was used in all miners (cpu, cc, sgminer). There was no diff parameter needed. No confusion.

REGARDING NEW LYRA2REV2

djm34, you used different diffs multiplicators in your sgminer and in your ccminer. Proof is here:

Your sgminer:
https://github.com/djm34/sgminer/blob/master/algorithm.c#L990

Your ccminer:
https://github.com/djm34/ccminer-lyra/blob/master/ccminer.cpp#L1153

For sgminer, you used 256, while for ccminer you left it at 128. Since your miner was linked in OP of this thread, we considered it as an official miner for Vertcoin. We made our proxy compatible with official miner. We had no idea that you made different multiplicator for ccminer - usually for us is enough to check code of one miner. If that was the case, we would surely bring this to attention sooner, before adding algorithm to our service.

As it stands now, there is huge confusion. Since we followed your sgminer implementation, we used value 256. As a result, miners not using your sgminer gets 50% less speed on our service.

Other ccminer developers followed your ccminer implementation, where multiplicator is set to 128 (as original Lyra2RE). All NVIDIA miners get only 50% speed at NiceHash if not using diff 2 parameter.

This issue is critical and whoever is in charge of Lyra2REv2 algorithm, needs to come forward and state which multiplicator is correct one.

[jamesl22]

Regarding old lyra2: we used standard lyra2re diff that was used in all miners (cpu, cc, sgminer). There was no diff parameter needed. No confusion.

REGARDING NEW LYRA2REV2

djm34, you used different diffs multiplicators in your sgminer and in your ccminer. Proof is here:

Your sgminer:
https://github.com/djm34/sgminer/blob/master/algorithm.c#L990

Your ccminer:
https://github.com/djm34/ccminer-lyra/blob/master/ccminer.cpp#L1153

For sgminer, you used 256, while for ccminer you left it at 128. Since your miner was linked in OP of this thread, we considered it as an official miner for Vertcoin. We made our proxy compatible with official miner. We had no idea that you made different multiplicator for ccminer - usually for us is enough to check code of one miner. If that was the case, we would surely bring this to attention sooner, before adding algorithm to our service.

As it stands now, there is huge confusion. Since we followed your sgminer implementation, we used value 256. As a result, miners not using your sgminer gets 50% less speed on our service.

Other ccminer developers followed your ccminer implementation, where multiplicator is set to 128 (as original Lyra2RE). All NVIDIA miners get only 50% speed at NiceHash if not using diff 2 parameter.

This issue is critical and whoever is in charge of Lyra2REv2 algorithm, needs to come forward and state which multiplicator is correct one.

I think 256 is correct. Its used on P2Pool, by yourselves and also Verters. @djm34, can this be updated in ccminer so that its correct?

[djm34]

Regarding old lyra2: we used standard lyra2re diff that was used in all miners (cpu, cc, sgminer). There was no diff parameter needed. No confusion.

REGARDING NEW LYRA2REV2

djm34, you used different diffs multiplicators in your sgminer and in your ccminer. Proof is here:

Your sgminer:
https://github.com/djm34/sgminer/blob/master/algorithm.c#L990

Your ccminer:
https://github.com/djm34/ccminer-lyra/blob/master/ccminer.cpp#L1153

For sgminer, you used 256, while for ccminer you left it at 128. Since your miner was linked in OP of this thread, we considered it as an official miner for Vertcoin. We made our proxy compatible with official miner. We had no idea that you made different multiplicator for ccminer - usually for us is enough to check code of one miner. If that was the case, we would surely bring this to attention sooner, before adding algorithm to our service.

As it stands now, there is huge confusion. Since we followed your sgminer implementation, we used value 256. As a result, miners not using your sgminer gets 50% less speed on our service.

Other ccminer developers followed your ccminer implementation, where multiplicator is set to 128 (as original Lyra2RE). All NVIDIA miners get only 50% speed at NiceHash if not using diff 2 parameter.

This issue is critical and whoever is in charge of Lyra2REv2 algorithm, needs to come forward and state which multiplicator is correct one.

I think 256 is correct. Its used on P2Pool, by yourselves and also Verters. @djm34, can this be updated in ccminer so that its correct?

Sure I can, but just need to be sure that it is the case for all pool (coinotron is also using it), there were report that trying to use 256 as default on ccminer caused people not getting their proper hashrate reported (and I am pretty sure it was nicehash...   )
That's why I didn't change  the difficulty (and other developpers, who tried got burned) so far on ccminer (and use --diff 2 if needed ).
But if nicehash finally agrees to use the same difficulty (256) as other pools, I can do it in no time.

[nicehash]

Sure I can, but just need to be sure that it is the case for all pool, there were report that trying to use 256 as default on ccminer caused people not getting their proper hashrate reported (and I am pretty sure it was nicehash...   )
That's why I didn't change  the difficulty (and other developpers, who tried got burned) so far on ccminer (and use --diff 2 if needed ).
But if nicehash finally agrees to use the same difficulty (256) as other pools, I can do it in no time.

At NiceHash we are using difficulty 256 from day one since we based our implementation on your sgminer release, therefore anyone who is using a miner with diff 256 (for example sgminer, cpuminer-multi) never had an issue, but those who were using miners with diff 128 of course had issues in terms of half reported speed (without using --diff 2 param). Nevertheless ... yes, please change it to 256 in ccminer and make new binaries ... and please also pass this info to sp_, tpruvot and other popular ccminer developers. Thanks!

[djm34]

Sure I can, but just need to be sure that it is the case for all pool, there were report that trying to use 256 as default on ccminer caused people not getting their proper hashrate reported (and I am pretty sure it was nicehash...  )
That's why I didn't change  the difficulty (and other developpers, who tried got burned) so far on ccminer (and use --diff 2 if needed ).
But if nicehash finally agrees to use the same difficulty (256) as other pools, I can do it in no time.

At NiceHash we are using difficulty 256 from day one since we based our implementation on your sgminer release, therefore anyone who is using a miner with diff 256 (for example sgminer, cpuminer-multi) never had an issue, but those who were using miners with diff 128 of course had issues in terms of half reported speed (without using --diff 2 param). Nevertheless ... yes, please change it to 256 in ccminer and make new binaries ... and please also pass this info to sp_, tpruvot and other popular ccminer developers. Thanks!

ok done, here the new release: https://github.com/djm34/ccminer-lyra/releases/tag/v0.5

[pallas]

I see djm34 already pushed diff 256 to ccminer.
Now, what should we use on pools that didn't need --diff 2 like give me coins?

[nicehash]

I see djm34 already pushed diff 256 to ccminer.
Now, what should we use on pools that didn't need --diff 2 like give me coins?

All of the pools should adapt diff 256, you should notify pool operators about this.