can anyone (kenshirothefist: *poke*) summarize the problem with cgminer 3.7.2 and NiceHash's scrypt pool, point out the breaking commit (between cgminer 3.1.1 and 3.7.2) or the fix (in sgminer) relating to it?
btw, great job on this service so far! it's great to see some innovative new pool ideas this year
cgminer 3.7.2 has many many bugs that have since been patched in newer versions and branches such as sgminer. There's really no good reason to be using an old version of cgminer anymore - switch to sgminer.
---
Everytime the pool switches jobs, it seems to disconnect all my miners briefly. Lost mining time appears to be 1-5 seconds everytime, but any work that was performed immediately prior is lost as well. And sometimes when there are a bunch of small jobs, or several new jobs are posted in short succession, this is a pretty significant loss of hashrate. It's super obvious for me, because the R9 290 fans spin up slightly when the GPU isn't under load (noise change is very noticeable). With NiceHash, I hear the rig in my living room drop for a a few seconds several times in a minute sometimes.
I see this regularly on my consoles:
[21:50:14] Stratum connection to NiceHash interrupted
[21:50:15] NiceHash difficulty changed to 512
[21:50:14] Stratum connection to NiceHash interrupted
[21:50:14] NiceHash stratum share submission failure
[21:50:15] NiceHash difficulty changed to 512
[21:50:19] Network diff set to 84.7M
[21:50:19] New block detected on network before pool notification
[21:50:19] NiceHash communication resumed, submitting work
[21:50:20] Rejected 01529678 Diff 49.5K/512 GPU 1 NiceHash (Job not found.)
[22:08:30] Stratum connection to NiceHash interrupted
[22:08:31] NiceHash stratum share submission failure
[22:08:31] NiceHash difficulty changed to 32
[22:08:32] NiceHash difficulty changed to 64
[22:08:32] Network diff set to 10.4M
[22:08:32] Stratum from NiceHash detected new block
[22:08:32] Stratum from NiceHash requested work restart
[22:08:36] NiceHash communication resumed, submitting work
[22:08:36] Rejected cf84d7c0 Diff 316/32 GPU 4 NiceHash (Job not found.)
[22:08:30] Stratum connection to NiceHash interrupted
[22:08:30] Lost 2 shares due to stratum disconnect on NiceHash
[22:08:31] NiceHash stratum share submission failure
[22:08:31] NiceHash difficulty changed to 32
[22:08:32] NiceHash difficulty changed to 64
[22:08:32] Network diff set to 10.4M
[22:08:32] Stratum from NiceHash detected new block
[22:08:32] Stratum from NiceHash requested work restart
[22:08:36] NiceHash communication resumed, submitting work
[22:08:36] Rejected 04c785be Diff 54/32 GPU 3 NiceHash (Job not found.)
[22:08:36] Rejected 06e87f69 Diff 37/32 GPU 0 NiceHash (Job not found.)
[22:08:30] Stratum connection to NiceHash interrupted
[22:08:30] Lost 1 shares due to stratum disconnect on NiceHash
[22:08:30] NiceHash stratum share submission failure
[22:08:31] NiceHash difficulty changed to 32
[22:08:32] NiceHash difficulty changed to 64
[22:08:32] Network diff set to 10.4M
[22:08:32] Stratum from NiceHash detected new block
[22:08:32] Stratum from NiceHash requested work restart
[22:08:35] NiceHash communication resumed, submitting work
[22:08:36] Rejected 02350f7d Diff 116/32 GPU 2 NiceHash (Job not found.)
[22:08:36] Rejected 07b180d4 Diff 33/32 GPU 4 NiceHash (Job not found.)
There's no reason for the stratum server to be disconnecting miners like this when switching jobs. Please fix.
Also, the server has been dropping me to down to low-diff on reconnects sometimes, even thought I have p=5.0;d=512 set as my password.
---
Edit: just observed 3 connection drops within 20 seconds.
) 
That leaves only two options: encourage luke-jr to incorporate GridSeed support into latest BFGminer or fork cgminer-3.7.2+upstream patches with GridSeed support and create "gsgminer" 