Can anyone enlighten me to the discrepancy between local calculated hash-rate and accepted hash-rate by the pools? I was not able to find anything that makes sense to determ if the current mining setup waste is at least bearable with the used pool?

My current conlusio is mining monero with a pool from someone else and public ccminer Versions:

• wastes best case average 40% of my hash rate with all tested ccminer version for nothing as the pools try to minimize their network payload. (big ones are the worst)

• there is no pool server feedback implemented to the miners how much was wasted. Yeah constantly check by hand how many submitted hashes from the server are logged against the hash rate. ^^

Things I could try:

• give up monero as its mining development is currently based on wasting resources

• set up a server client system to distribute the crazy pools to something at least partial working

• use my own pool just for me^^

• find out how to tell the server that I want max 30 sec. till the next accepted(yeah) which hopefully results in wasting only ~15-20%.(Example below is that best case and thats a server with the worst latency as it's also farthest away)

I've tested over the last months several versions based on the ccminer

Each 20 devices from 50H/s up to 800H/s (no over or underclocking/voltage etc.)

Connected to up to 3 different internet connections to check if latency has any cause for this effect. -> Nope, doesn't also seem to be no limiting by providers.

The average of 1 calculation device drops from the beginning with >=95% to around 70% with big pools even down to 35%. If I add more Devices to the same pool the percentage drops from the 70% to 50% and sometimes even lower.

The only thing that seems to trigger the drop is the difficulty In some way difficulty seems to be connected to an accepted(Yeah) but that seems to be handled by each pool differently.

Example with a small pool that needs 15 Minutes to drop:

So starting is 10k or 20k for most servers below 1kH/s

Local 1.7KH/s(~50% of the whole pool)

8 sub-devices, 2 sub-devices@each 1 ccminer;

2x 600H/s + 1x 500H/s

-> more or less accepted fully with starting 10k Difficulty (same with 20k difficulty option)

After not even 15 Minutes difficulty rises step by step to over 50k

Pool accepts average shown with 50k Difficulty is only 1.36Khs - 1.42KHs

After again 10 Minutes it starts constantly switching between 15k and 50k and oh wonder my best result in months of average 1.47kH/s

with around 30 seconds for an yeah -> I'm possibly further testing that pool and possibly drop another 4-5kH/s but the interest without any further background is on a low.

At The same time I checked the average of an identical single unit connected to an big pool where I removed the above mentioned.

612Hs ended as ever changing between average 257Hs and 450Hs during no change in difficulty and that for hours.

The time till the next accepted(yeah) is best case 50 seconds

Thats an additional knowledge if you remove some of your miners you can start all left anew otherwise you losse even more.