Quote from: MrJay on October 01, 2013, 09:03:01 PM

Quote from: MoDu on October 01, 2013, 03:49:29 PM

Quote from: MrJay on October 01, 2013, 09:01:52 AM

Are you running these cards in 1 rig or 3? What mining software are you running? What are the settings for your cards?

2 rigs actually, on has 2 GPUs. Running cgminer through GUIMiner-scrypt. Maybe I'll update it when the rigs are back up ( rigs are down due to power failure in the area).

Settings (from memory), are close to this:

-g 1 --thread-concurrency 7500 -I 18 -w 256
as someone suggested it in this thread, I also tried adding
--expiry 5 --queue 0 --scan-time 5

Seems to make no difference.

Quote from: atp1916 on October 01, 2013, 11:38:41 AM

I an running multiple 7xxx cards and my reject rates have always been low on HC.

Must be your config.

That may be possible, but other pools don't report such high reject rates, including middlecoin, give-me-ltc, etc. Most report under 5% rejects.

Quote from: swiftshoot on October 01, 2013, 03:34:08 PM

You probably did not use --no-submit-stale with cgminer... This is why the high reject rates.
Turn that option on and try again.

But that will only mask the rejects, then I'll only see the efective hash-rate at the pool and never figure out why there are so many rejects in the first place!

I can't comment on guiminer-scrypt as I use cgminer myself with cgwatcher on top. One of my rigs has 2x5870's and 2x5850's and runs with a 5% reject rate. I haven't included my mem and clock speeds but my cgminer command line reads:

--thread-concurrency 8000 --intensity 18 --lookup-gap 2 --worksize 256 -g 1 --expiry 5 --queue 0 --scan-time 5

Some people say that expiry, queue and scan-time have no effect, but I find if I don't have these in for the multipools I get a higher reject and stale rate. If I mine LTC directly I do not put those settings in and get 1-2% rejects mining on wemineltc. I don't have no-submit-stale enabled as half the time with the above settings a stale share that is submitted it is accepted.

[2013-10-01 22:00:33] Pool 0 stale share detected, submitting as user requested
[2013-10-01 22:00:33] Accepted 25ecd518 Diff 532/37 GPU 0 pool 0
[2013-10-01 22:00:33] Accepted 55759c15 Diff 122/37 GPU 2 pool 0

Try cgminer, try playing with your thread-concurrency, try changing your intensity. I find that running at anything over 18 again gives myself more stales on multipools, I run everything on 20 mining LTC. Might take some playing but there will be a sweet spot.

Additionally, if you are using Windows, Intensity is max 13. On Linux is 20.