This may (or may not be) a good time and place to remind others out there (with the intention of helping you) a few things about mining coins. If you are serious about mining you probably allready know everything I am going to state below but many people even those mining a long time may not actually know these things and may find this information useful.
Hash rate measured for a worker on any pool would be based on accepted shares by that worker.
This assumes the software the pool is using is counting things correctly and is normally shares per a given period.
Local hash rate should be measured by what you believe to be submitted shares that were accepted...
Not what your 5 second average reading your gpu says you are running at.
Some believe what a screen shows as your hash rate is your 'real ' hash rate the way they believbe the hash rate of a coin network can actually be somehow measured from information in the rpc console of a coin client ...
not so ...here are details .....
The values returned by the client in the rpc console for any coins network hash rate are always average guesses based upon how quickly prior blocks were found vs. the target parameters of the network for how often they should be found (per minute ,per hour ect...)
The actual mathematics in bitcoin originally was (and I believe still is) done with a taylor series. For alt coins the value returned can vary based on how the developer may have changed the code from bitcoin or which ever coin the source was cloned/based from normally the only figure changed I see is the number of blocks backward used to calculate the average.
here is a link in btc stack exchange with a bit of detailed info since the bitcoin wiki website appears down now
https://bitcoin.stackexchange.com/questions/5556/relationship-between-hash-rate-and-difficultyMy point here is not directed at the individuals below but to remind
everyone who is mining no matter which algorithm that whatever program you are using can say your hash rate is anything.
If you are mining on a pool your hash rate as measured by that pool will always be measured by the number of shares submitted at the target difficulty set by you (or by the pool).
If you are mining solo your hash rate can be best estimated by the WU value which is your average shares work / minute at difficulty one (which also vary a bit due to luck).
Hopefully this will help someone out there understand that many numbers in mining people take as things that are actually measured or 'real measurements' , i.e. any coins given network hashrate at any specific moment in time, are always actually mathematical aproximations and NOT actual 'exact' measurements.
Finally always remember this about mining: solving a block is in cryptography actually a 'partial colission' (not a full one or even close). So if there exists a mathematical probabllity you can solve a block that probability is not ever in any way dependant on past events.
i.e. a pools luck has been bad the past few days , maybe it has failed to hit a block ..... it is human nature to think , well they failed to hit a block for several days so they are 'due' to hit one or more today or more likely to...............
no no no............ this is us, projecting our human experience and view of the world which is not applicable to this branch of mathematics.
Best example I can offer is flipping a coin (let's assume it is a fair coin with 50% chance of heads or tails). If it comes up heads 20 times in a row many people would gamble it has a much better chance of coming up tails on flip number 21............ this is human nature.......
NOW.... If we call a cab and they say 20 minutes , after 20 minutes or so the more time goes on each car that turns the corner in our minds is going to be more likely to be the taxi since it is 'overdue' and this logic is basically correct for most of the world.
The coin that we are flipping does not 'remember' in any way how it 'landed' the last twenty or twenty thousand times it was flipped. On flip number 21 it still has a 50% chance of heads or tails.
This is not exactly related to mining optimizations of sgminer v5 but it is absolutly important to understand these things to have realistic expectations of mining and also learn how to better optimize your mining hardware. If you get a higher 'hashrate' reported on the console of sgminer but are getting many errors and failed submissions you may have better results trimming back your settings to the point where the errors are less frequent so your the 'work' your miner does is 'valid' work.
Finally if using cgminer or sgminer or bfgminer and have never tried this, if you are on the main screen hit the 'd' key twice then the space bar. Now you can see a little more detail of what is actually going on 'inside' your gpu. To go back to your regular view just hit 'd' twice and space again.
Have fun mining and good luck to all.
[/quote]
The results are real.
I also got bad results with 14.6, but decided to try with 15.4 beta, and confirm the results.
without much effort 280x 10,1 m8
gpu/mem 1050/1500 -g 1 -w 64 -X 64
[/quote]
Try xIntensity = 512. It is higher, and the hash rate seen on pool is just slightly lower than local hash rate. If you try xIntensity = 1024 or 2048, the local hash rate is much higher, but the hash rate seen in the pool is similar to xIntensity = 512, but use more electricity.
[/quote]