[XPM] Primecoin Huge Drop in PPS

[eCoinomist]

Anyone noticed a huge drop by 40% in PPS mining?

I just noticed this thing taking place at exactly 16:24 UTC July 21.

First I thought it was the VPS provider that did something to my account, but upon inspection with other services (I use three different VPS instance providers) and including my local rig, the drop was indeed confirmed to be client related.

I use mikaelh's HP5 client, but this drop has been confirmed on HP4 as well.


As you can see, PPS dropped from 14k to 8k pps.

CAUSE?
I predict it has something to do with new difficulty increase to 9.0?

Anyone got any other findings?

[wibtc]

Same thing happened when the difficulty increased to 8.0.

Anyone noticed a huge drop by 40% in PPS mining?

I just noticed this thing taking place at exactly 16:24 UTC July 21.

First I thought it was the VPS provider that did something to my account, but upon inspection with other services (I use three different VPS instance providers) and including my local rig, the drop was indeed confirmed to be client related.

I use mikaelh's HP5 client, but this drop has been confirmed on HP4 as well.
http://ecoinomist.com/images/pps-drop.png

CAUSE?
I predict it has something to do with new difficulty increase to 9.0?

Anyone got any other findings?

[eCoinomist]

haven't noticed that, as I only started mining it seriously since 8.0, was was the % of decrease back then?

[Carra23]

Don't remember the exact rate, but there was a very noticeable drop too when the difficulty crossed 8.

[dudeami]

The drop was about 30-40% for both difficulty increases, should be similar for 10 and so on.

[gatra]

I would expect PPS to go down when the diff crosses integer boundaries because it's when the sieve is updated. So you filter more numbers, and are left with less primes. But again, since PPS "just for fun", forget about it. It does not mean that the speed or efficiency or your miner went down. It's the other way around actually, since difficulty calculation is not uniform/linear, using the current mining algorithms for looking for blocks with diff 9 is actually easier than finding blocks with diff 8.99.

EDIT: clarification: you are left with less primes, but each one of those has a higher probability of being the first of a chain.