Hi Nemos,

I posted this idea a while ago, but I did not see an answer if it was possible.

When switching Algo, there is moment (around 8 second for me) where there is no mining done, GPU utilization is 0 in afterburner (the time the new miner start and connect to the server)

If we would start the new miner first and then 7 or 8 second later close the old miner it should eliminate this lost and make the algo switch even more effective (especially on the non-24h batch file because it change more frequently)

I tried it manually (open new miner wait 8 sec and close the old) and it seem to work the utilization stayed at 100%, I did get an API port error but maybe it can be fix with some modification in the code (I'm no pro for programation). the mining did switch and work fine when I test it on zpool and ahaspool

If it possible, it should make a great gain.

Thank you for this great program.

Hi Nemos,

I posted this idea a while ago, but I did not see an answer if it was possible.

When switching Algo, there is moment (around 8 second for me) where there is no mining done, GPU utilization is 0 in afterburner (the time the new miner start and connect to the server)

If we would start the new miner first and then 7 or 8 second later close the old miner it should eliminate this lost and make the algo switch even more effective (especially on the non-24h batch file because it change more frequently)

I tried it manually (open new miner wait 8 sec and close the old) and it seem to work the utilization stayed at 100%, I did get an API port error but maybe it can be fix with some modification in the code (I'm no pro for programation). the mining did switch and work fine when I test it on zpool and ahaspool

If it possible, it should make a great gain.

Thank you for this great program.

Alexminer give me the best result

To help chose, you can go to Zpool.ca
Chose the scrypt you want on the right and then chose miner at the top.

You will see what everyone use for the scrypt, so the one the most people use should technically be the best for it.

From what I understand from pool transaction, if they do all the payment at the same time, the cost of the transaction is divided among every miner and it become a little individual fee. If it would be "on demand" payment you would pay the normal fee.



Taken from ma790 on reddit:




So I presume this fee is paid by ahashpool and it must be taken in there calculation on how many btc to give you at the start so you pay for it but its hidden and little.
More miner at payout =  less fee

If I mine 0.008 BTC, they pay 0.008 BTC and I receive 0.008 BTC on my wallet, there is no fee at the transaction.
But they chose the payout time (each 24h) so my balance could go up to 0.011 BTC before their payout, they would pay 0.011 BTC and I would receive 0.011 BTC (no fee)
and there doesn't seem to have big delay with the transaction, I only got 2 payout from them but both time I got my BTC in my wallet less than 10 minute after.

Hi,

Great multimer, nice job.

I only find that when the miner switch algo, it close the miner, then start the other one.

There is some delay every-time so the GPU stop working for like 1 sec.

this give some loss.

If the new algo miner would start 1 second before the old miner close, could it solve this ?
The old miner would continue to mine while the new miner start and establish the connection.
Maybe they would work both for 1/2 a second but that should not cause problem.
You could maybe check and stop the old miner once the new one has an connection.

I no programmer but it seem doable.

Just an idea.

DNR...?

I use
CCminerTpruvotcuda9 on ahashpool

184mh/s for 2x 1080ti

I try --submbit-stale  but it always say unknown option, even with the one you linked

You mean those right ?

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

Ok thank you I will remove those, I did not see any gain by adding them and some post on internet report it was better (but it was probably placebo effect ) so I just put them because i did not loss  hashrate either

Hi I have 2 gtx1080ti 
Mining Zcoin (the best coin I found for 1080ti)(around 2050mhz each at 1025mv at 38degree)
on MPH (the best I found for this coin)
I get 6400KH/S on ccminer
But when locking at the Afterburner graph I see random drop in GPU usage.

I read the other post for the best ccminer (djm)  I tried every one I could find and it seem the best but...
Do you all have this,



I use the ccminer_cuda9 from DJM (30 247kb)

With this command line

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
ccminerdjm_cuda9 -a lyra2z -i 19 -o stratum+tcp://us-east.lyra2z-hub.miningpoolhub.com:20581 -u XXX.XXX




Is it normal, I just started so I did not had enough time to calculate if the hash rating from the ccminer match MPH but even if it match, the drop is a % loss of calculation power