The big VTC Vertcoin settings thread

[bitwho]

k nvm. i ran it the same setting again. only this time i removed any additional .bat i had in the folder. i read it somewhere last night that having more then one .bat file in there confuses the mod cgminer

Good to hear you got it fixed - I presume you mean the .bin files cgminer generates?

Yes, modded cgminer for vertcoin should be run in a different folder than regular cgminer, so the bin files from the wrong version don't get picked up.

well, kinda.

i made two .bat files . one solo and one for the pool. the modded cgminer was on a different folder and not in the original cgminer. in its own folder i had those two .bat files together with all the modded cgminer. so i was having problem running it. but once i remove the solo batch away i was able to run the pool batch file. anyways i shut it down since i was too late to the train ride and was gonna collect small payments. good luck to you all

[1715309534]

1715309534

[1715309534]

1715309534

[1715309534]

1715309534

[BorisTheSpider]

Bump to get some more settings - please can someone contribute a working 7950 config.

[AtlasONo]

Can only get one 7950 the work, the other one doesn't like it and just spews HW errors. same with my 5870's

[Sohcahtoa]

I'm running an R9 290, and here's my settings:

Code:

-w 256 -I 16 --thread-concurrency 27400 -g 1

I get 375-380 KH/s, which I suppose is decent, but sometimes when cgminer starts up, my entire system crashes.  It'll say cgminer is starting, then the window goes blank like its about to start, but then my system locks up.  I can move my mouse, but clicking on anything does nothing.  Alt-Tab and CTRL-ALT-DEL do nothing.  After about 15 seconds in this state, my entire screen goes black, and it stays in this state until I manually power off my computer.

But it only sometimes does this.  And when it works, it'll work as long as I need it to.  Once its running, its perfectly stable.  It is only the initial start-up that has an issue.

[Camilio]

sooo, I´m running a Gigabyte r9 280x rev. 2 Catalyst 13.12

My Settings:

Code:

color 02
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
"D:\Program Files (x86)\Vertcoin-CGMinerAleks_NR3\cgminer.exe" --scrypt -o stratum+tcp://us.bitcrush.info:3444 -u -p -w 256 -I 13 -g 1 --lookup-gap 2 --gpu-powertune -20 --gpu-fan 59 --temp-overheat 85 --temp-cutoff 90 --gpu-memclock 1500 --gpu-engine 1100 --thread-concurrency 8212 (marginal difference in range between 8190 to 8250)

I am getting exactly 304.7 kh/s
with the normal cgminer I am getting about 730 kh/s

My problem:

• It spits to many rejects out with an ratio of 12% (!!!). I tried playing with:
  
  thread concurrency, no effect - 70% of 8192 = HW´s on mass
  
  trying with -g 2 or higher was unsuccessful
  
  lowering core clock resulted in lower hash speed (memclock the same)
  
  lowering intensity induced lower hash speed

I really don´t know where the problem could be. Please help me The rejection rate is a catastrophe!

[Crackmacs]

can't seem to get much with my 7970 anyone have a good config to mess around with?

Many thanks

[Camilio]

sooo, I´m running a Gigabyte r9 280x rev. 2 Catalyst 13.12

My Settings:

Code:

color 02
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
"D:\Program Files (x86)\Vertcoin-CGMinerAleks_NR3\cgminer.exe" --scrypt -o stratum+tcp://us.bitcrush.info:3444 -u -p -w 256 -I 13 -g 1 --lookup-gap 2 --gpu-powertune -20 --gpu-fan 59 --temp-overheat 85 --temp-cutoff 90 --gpu-memclock 1500 --gpu-engine 1100 --thread-concurrency 8212 (marginal difference in range between 8190 to 8250)

I am getting exactly 304.7 kh/s
with the normal cgminer I am getting about 730 kh/s

My problem:

• It spits to many rejects out with an ratio of 12% (!!!). I tried playing with:
  
  thread concurrency, no effect - 70% of 8192 = HW´s on mass
  
  trying with -g 2 or higher was unsuccessful
  
  lowering core clock resulted in lower hash speed (memclock the same)
  
  lowering intensity induced lower hash speed

I really don´t know where the problem could be. Please help me The rejection rate is a catastrophe!

Problem solved. I changed the pool, now everything is running smooth!

[gonzzo]

Heya

using asus R9 280X but  i can't  get over around 230kh/s (should get to around 350 kh/s)

can't set --thread-concurrency to more then 4096 or i get the following error:
Error -61: clCreateBuffer (padbuffer8), decrease TC or in crease LG

i'am using VertCoin-CGMinerCPU-Aleks_N on windows 7 64-bit

any help very welcome - thanks!

[gonzzo]

ok around 310kh/s with my asus R9 280X with the following config:

--scrypt -o server -u user -p pw -v 1 -w 256 -I 12 -g 2 --lookup-gap 2 --gpu-powertune -20 --gpu-engine 1100 --gpu-memclock 1600 --thread-concurrency 4096

maybe som1 can use it 2
adios

[BorisTheSpider]

ok around 310kh/s with my asus R9 280X with the following config:

--scrypt -o server -u user -p pw -v 1 -w 256 -I 12 -g 2 --lookup-gap 2 --gpu-powertune -20 --gpu-engine 1100 --gpu-memclock 1600 --thread-concurrency 4096

maybe som1 can use it 2
adios

Thanks for this. Since 280x pretty much == 7970, this should work for 7970 too.

[gonzzo]

ok around 310kh/s with my asus R9 280X with the following config:

--scrypt -o server -u user -p pw -v 1 -w 256 -I 12 -g 2 --lookup-gap 2 --gpu-powertune -20 --gpu-engine 1100 --gpu-memclock 1600 --thread-concurrency 4096

maybe som1 can use it 2
adios

Thanks for this. Since 7970 pretty much == 7970, this should work for 7970 too.

np mate.

i have more then one worker, all with the same gpu's but one one i have to set the gpu-engine to 1070 (default for the asus R9 280x gpu)
if i leave it to 1100 the worker falls to around 240 kh/s after about 30-60 minutes, i have no iedea why (after a restart i will get 310kh/s again for 30-60 minutes).
with 1070 the worker is running stable with 300kh/s. strange .....
so maybe you have to tweak my setting a little

[BorisTheSpider]

i have more then one worker, all with the same gpu's but one one i have to set the gpu-engine to 1070 (default for the asus R9 280x gpu)

Is 1070 really the default clock??? That sounds like quite a big OC to me, my R9 290s default clock is 947 so I would have thought even factory OC'd 280X's would be no more than around 1000mhz

[gonzzo]

i have more then one worker, all with the same gpu's but one one i have to set the gpu-engine to 1070 (default for the asus R9 280x gpu)

Is 1070 really the default clock??? That sounds like quite a big OC to me, my R9 290s default clock is 947 so I would have thought even factory OC'd 280X's would be no more than around 1000mhz

well from asus site:

GPU Boost Clock : 1070 MHz
GPU Base Clock : 970 MHz
http://www.asus.com/Graphics_Cards/R9280XDC2T3GD5/#specifications

but my gpu is running with 1070 after installation on my w7 workers without any tweaks from my side. so i call it default even if asus says boost clock .
2 of 3 workers are running with 1100 now, what i would call oc heheh

the strange thing: the 1100 setting ist good on my old asus mb with an old intel Quad (Q8400) and 4gb ram and a brand new gigabyte mb with a i3 and 4 gb ram.
but on a brand new asus mb with a i5 (same chipset as the gigabyte mb) and 8gb ram i have to use the 1070 setting to avoid falling back to 240 kh/s
i wonder why ? makes no sense for me 

[madjihad]

Have someone successfully setup it for 7870? I've played a week with no luck. Only HW and rejections. Probably bruteforced almost all possible combinations of intensity and TC with step of 256 and a lot of random non x 256 values, still the same. GPU and mem clocks are default.

[Hardwyre]

The custom CGminer is giving me NONCE hardware errors, regardless of changes to my bat file.  I have adjusted thread concurrent, lookup gap, -g, added/remove the --scrypt flag, etc, etc.

The one time I did get any accepts, there would be a NONCE error, then a burst of about six to ten accepts, then nothing for about a minute, then another burst.

I have two water cooled R9 290's running the --no-adl flag, otherwise CGminer just shows a black screen since there is no fan RPM input.

CGMiner - custom from Vertcoin webpage.
AMD APP SDK - 2.9.233.167
GPU DRIVER - 13.250.18.0

I have averaged 830kh per card at intensity 20 on any other coin.


P.S. The six minute post lock is absurd. 

[Camilio]

Heya

using asus R9 280X but  i can't  get over around 230kh/s (should get to around 350 kh/s)

can't set --thread-concurrency to more then 4096 or i get the following error:
Error -61: clCreateBuffer (padbuffer8), decrease TC or in crease LG

i'am using VertCoin-CGMinerCPU-Aleks_N on windows 7 64-bit

any help very welcome - thanks!

try adding

Code:

-g 1

everyone´s posting  configs with -g 2, but i have to set it to 1 otherwise my hashrate drops to ~ 150 kh/s

what values for the thread conncurency di you use. As far as I know no r9 280x is working with <4096. To many HW´s. Did you try to set a value between 8100 and 8300 ?

[gonzzo]

Heya

using asus R9 280X but  i can't  get over around 230kh/s (should get to around 350 kh/s)

can't set --thread-concurrency to more then 4096 or i get the following error:
Error -61: clCreateBuffer (padbuffer8), decrease TC or in crease LG

i'am using VertCoin-CGMinerCPU-Aleks_N on windows 7 64-bit

any help very welcome - thanks!

try adding

Code:

-g 1

everyone´s posting  configs with -g 2, but i have to set it to 1 otherwise my hashrate drops to ~ 150 kh/s

what values for the thread conncurency di you use. As far as I know no r9 280x is working with <4096. To many HW´s. Did you try to set a value between 8100 and 8300 ?

nope the same error with -g 1 if i set tc higher then 4096.
but i'm at 310 kh/s now (check some of my later posts for the config.) and i think this is not to bad for my gpu

[cryfilter]

set GPU_MAX_ALLOC_PERCENT=100
setx GPU_MAX_ALLOC_PERCENT 100
cgminer --scrypt -o stratum+tcp://s.ny.vertco.in:3333 -u user -p pw -w 512 --thread-concurrency 8192 -I 20

This gives me nothing but HW errors with an r9 290 and 4gb RAM but runs fine with 8gb. The only way I can get rid of the errors is by setting the intensity to 13 which only gives 250kh/s. Any suggestions?

[KRStwo]

Good day to all usernames.

Testing my old Sapphire HD6970
got 130 mh/s with this settings:

Code:

"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"kernel-path" : "/usr/local/bin",
"scrypt" : true,
"auto-fan" : true,
"gpu-threads" : "3",
"gpu-powertune" : "20",
"intensity" : "12",
"worksize" : "256",
"lookup-gap" : "2",
"shaders" : "1536",
"thread-concurrency" : "4096"

Is this ok? or not?

[a3860903]

7850 from Rasetsu:

--thread-concurrency 2048 --lookup-gap 2 --gpu-engine 1100 --gpu-memclock 1000 --gpu-powertune 20 -w 256 -I 13 -g 1

Gives 170KH/s

My 6 Platinum Edition 7850 1G card using this set to run only five cards What is the reason?