i understand, thing is my rig has higher average hashrates on mintpond using older 2.1.19 build with intesnity 20 and gridsize 4096 on all 5 GPU, with later builds it has intensity 15 and grid sizes are 7000+ for GTX1080Ti and 5000+ for GTX1080 but i don's see any hashrates improvements in 24 hour period, difference is around 1mh/sec daily average 16mh.sec+ VS 15mh.sec on newer builds, so maybe better to use older build at the moment with lower default gridsize, miner hashrates itself are more or less same with both builds
Can you try to use grid size settings instead of intensity? If 4096 gives you good result try -gs 4096. Please do not use -i in that case. Please let me know if that changes something. Please also add -luck. This shows you your 'session' luck. This gives you an ide, if the miner sends as much shares as is should do on the 'LONG RUN'. This value should be close to 100%. Values over 100 means that the miner doesn't find enough shares for the diff coming from the pool. values below 100 means that the miner finds more shares than expected - so you have some luck. After a day or two this values should be quite constant on 100 - 102.
The miner show the hashrate it gets from the gpu algo. if two versions shows the same hashrate, then the hashrate is the same. But the hashrate @ pool is calculated. So if the miner doesn't find 'eough' shares, thr pool will show a lower hashrate. Not sure how long you made the test, but you might check the hashrate a 3 hours - or so. Then calculate the average - as a starting point for the hashrate @ pool.
Thanks for reporting.