Author

Topic: 3 x sapphire 7950 low speed or freezing (Read 718 times)

newbie
Activity: 49
Merit: 0
February 23, 2014, 02:08:15 PM
#13
and still big difference in hashrate?
http://s13.postimg.org/5z6odycav/Untitled.png
newbie
Activity: 49
Merit: 0
February 23, 2014, 01:56:14 PM
#12
uninstalled in safe mode all drivers with Display Driver Uninstaller
installed 13.4 + 2.9 SDK (windows 7 64bit)

using this config

"intensity" : "19",
"worksize" : "256",
"kernel" : "scrypt",
"lookup-gap" : "2",
"thread-concurrency" : "21712",
"shaders" : "0",
"gpu-engine" : "1000",
"gpu-fan" : "0-65",
"gpu-memclock" : "1500",
"gpu-memdiff" : "0",
"gpu-powertune" : "20",
"gpu-vddc" : "0.000",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "75",
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "120",
"gpu-dyninterval" : "7",
"gpu-platform" : "1",
"gpu-threads" : "1",
"hotplug" : "5",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "30",
"scrypt" : true,
"temp-hysteresis" : "3",
"shares" : "0",
"failover-only" : true,
"kernel-path" : "/usr/local/bin"

and after 5 mins freezes whole system?
newbie
Activity: 49
Merit: 0
February 23, 2014, 12:23:34 PM
#11
hi
windows 7 74bit
I just used the driver cd that came with it
where I can see what version it is?

and where I can download this 13.4s + 2.9 SDK ?

I found this one which I am downloading now
AMD-APP-SDK-v2.9-Windows-64.exe
is this the correct one?

however I cannot find 13.4s?
legendary
Activity: 854
Merit: 1000
February 23, 2014, 12:15:20 PM
#10
I would dare say you may have a driver issue due to that type of spread in hashrates.  Which driver set are you using and what OS?  

Personally, i've had excellent success with the 13.4s + 2.9 SDK for all my 7950s in Win7/Win8.

If you do try that combo out, make sure you uninstall your old set using Display Driver Uninstaller in Safe Mode first.
newbie
Activity: 49
Merit: 0
February 23, 2014, 12:07:30 PM
#9
here the current state
to gpu 0 I have the monitor connected, would it be because of this so much lower?
and why is there such a big difference between gpu1 and gpu2?
http://s8.postimg.org/vbx9ncfkl/Untitled.png
newbie
Activity: 49
Merit: 0
February 23, 2014, 12:00:54 PM
#8
ok changed powertune to 20

howverb I see directly already all cards still have different hashrate?

could 1 reason be because on 1 card I have the monitor connected?

is it possible to use the onboard vga for the monitor? and the cards only for mining?
I tried, but cgminer didn't see my cards anymore then?
legendary
Activity: 854
Merit: 1000
February 23, 2014, 11:52:45 AM
#7
Looking good, but change your --gpu-powertune to "20". 
newbie
Activity: 49
Merit: 0
February 23, 2014, 11:47:06 AM
#6
changed it and currently testing like this

"intensity" : "19",
"worksize" : "256",
"kernel" : "scrypt",
"lookup-gap" : "2",
"thread-concurrency" : "21712",
"shaders" : "0",
"gpu-engine" : "1000",
"gpu-fan" : "0-65",
"gpu-memclock" : "1500",
"gpu-memdiff" : "0",
"gpu-powertune" : "0",
"gpu-vddc" : "0.000",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "75",
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "120",
"gpu-dyninterval" : "7",
"gpu-platform" : "1",
"gpu-threads" : "1",
"hotplug" : "5",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "30",
"scrypt" : true,
"temp-hysteresis" : "3",
"shares" : "0",
"failover-only" : true,
"kernel-path" : "/usr/local/bin"
legendary
Activity: 854
Merit: 1000
February 22, 2014, 11:55:06 PM
#5
OP, You do not need to triple up on the parameters if you are running the same config for all 3 cards. If you have not already, define GPU_MAX_ALLOC_PERCENT as 100 in your environmental variables either through System properties or the setx command at the cmd prompt.

Note the difference in Wipeout's config vs the one you are currently running: Vectors is not needed for scrypt mining, -g = 1, -i = 19, and the -thread-concurrency = 21712.  
sr. member
Activity: 840
Merit: 255
SportsIcon - Connect With Your Sports Heroes
February 22, 2014, 11:28:18 PM
#4
Maybe because either "gpu-memclock" is too high or the VRMs are at absurd temperatures.

Use gpu-z "sensors" to check the latter, while mining.

"intensity" : "19",
"lookup-gap" : "2",
"thread-concurrency" : "21712",
"gpu-threads" : "1",
"gpu-engine" : "1000",
"gpu-fan" : "60-90",
"gpu-memclock" : "1500",
"temp-cutoff" : "80",
"temp-overheat" : "75",
"temp-target" : "68",
newbie
Activity: 182
Merit: 0
February 22, 2014, 08:32:36 PM
#3
no one?

You are better with 1thread and tc of 40960 is whati use. I get 630 at 1050, 1250 with sgminer At 1.12vddc.
newbie
Activity: 49
Merit: 0
February 22, 2014, 02:24:07 PM
#2
no one?
newbie
Activity: 49
Merit: 0
February 19, 2014, 04:21:45 PM
#1
hi
I have running 3 x sapphire 7950
I tried many configs
I can get it on around 600kh/s but then allways freezes
the fastest a littlebit stable I get it is this

http://s27.postimg.org/f198dv9oz/Untitled.jpg

I currently use this config

]
,
"intensity" : "13,13,13",
"vectors" : "1,1,1",
"worksize" : "256,256,256",
"kernel" : "scrypt,scrypt,scrypt",
"lookup-gap" : "0,0,0",
"thread-concurrency" : "14336,14336,14336",
"shaders" : "0,0,0",
"gpu-engine" : "1020,1020,1020",
"gpu-fan" : "0-65,0-65,0-65",
"gpu-memclock" : "1550,1550,1550",
"gpu-memdiff" : "0,0,0",
"gpu-powertune" : "0,0,0",
"gpu-vddc" : "0.000,0.000,0.000",
"temp-cutoff" : "95,95,95",
"temp-overheat" : "85,85,85",
"temp-target" : "75,75,75",
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "120",
"gpu-dyninterval" : "7",
"gpu-platform" : "1",
"gpu-threads" : "2",
"hotplug" : "5",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "30",
"scrypt" : true,
"temp-hysteresis" : "3",
"shares" : "0",
"failover-only" : true,
"kernel-path" : "/usr/local/bin"
}

also 2 cards are way slower?
any ideas?
Jump to: