Pages:
Author

Topic: The big VTC Vertcoin settings thread - page 42. (Read 264262 times)

newbie
Activity: 42
Merit: 0
February 04, 2014, 11:31:20 AM
Now my problem is, vertminer is not showing GPU temperatures.

What should i do?

On windows or on Linux/Debian? I had that same issue. You need the AMD ADL_SDK files for it be able to monitor the temperatures.

Make sure the files are in the ADL_SDK folder and recompile it if you're on linux/debian. If that doesn't work; download the ADL_SDK 6.0 and manually place those files in the ADL_SDK folder Smiley

I'm On BAMT linux, where is path of ADL_SDK on linux?

Then you should have a working ADL_SDK in bamt

try this in terminal:

cp /opt/miners/cgminer/ADL_SDK/* /where/you/vgminer-gpu/folder/is/ADL_SDK/

then run buildit.sh again. If that fails, manually download this file: https://dl.dropboxusercontent.com/u/10987558/ADL_SDK/ADL_SDK.zip and extract those files in the ADL_SDK folder and rebuild it again.
sr. member
Activity: 490
Merit: 250
February 04, 2014, 11:18:38 AM
Now my problem is, vertminer is not showing GPU temperatures.

What should i do?

On windows or on Linux/Debian? I had that same issue. You need the AMD ADL_SDK files for it be able to monitor the temperatures.

Make sure the files are in the ADL_SDK folder and recompile it if you're on linux/debian. If that doesn't work; download the ADL_SDK 6.0 and manually place those files in the ADL_SDK folder Smiley

I'm On BAMT linux, where is path of ADL_SDK on linux?
member
Activity: 112
Merit: 10
February 04, 2014, 11:12:51 AM
Hi,

While the extra 10-15 hashes is nice, running I20 on these 290's w/ modified cgminer will literally beat the shit out of your cards. If you are monitoring them, please observe that you probably have constant usage crashes to 0% and if you have the comp on a circuit w/ lights they will be pulsating. You also won't have a responsive desktop. I have a permanent artifact from one of my power surges due to this.

If you value your cards, please consider I17 rather than I20.

I was using afterburner/asus gputweak to monitor the settings so please confirm for yourself. . .mining vertcoin is not like mining litecoin.

Not sure what you mean with usage crashes, GPU usage stays at 100, so does hashrate and yes, lights do pulsate but I think this is due to the power outages in our country due to heavy ice rain tearing electricity cables. Desktop is responsive.

What are you using to monitor your usage? Also, I was using kill a watt, and it was showing fluctuations from 600w to 950+w. . so they were definitely happening. Does your core clock drop at all? What's the rest of your setup like?

When I had symptoms as Yours (power draw fluctuating in my kill-a-watt) it was somethign wrong and soon later I had to rma 2 of my cards..first freezes came, then I couldn't power up windows with them and in the end green horizontal line on screen, and they had just few days (non-reference Windforce Gigabyte 290s)..I think Gigabyte have fucked up something in their PCB, cause vram 1 was 95-100 C, vram ~75..while all other reference cards (which supose to have shitty cooling) have vram 1 and vram 2 around 65-75C.

It only happens for me when using modified cgminer >I17. I can still mine regular scrypt using unmodded cgminer @ I20 with no fluctuations at all. The temps are all stable and within reason. . @70-75% fan speeds and 72-80C w/ 70F ambient temp. I did get one freeze yesterday, but I was still overclocking it too high @ 1075/1470. Now I've gone to 1000/1500 and they're happy for the whole night except sometimes I get a worker banned. I'm assuming it's from the ddosing happening at kilovolt but it could just as easily be due to a card starting to fail. What do you measure vram temps with?

It happened to me on normal scrypt mining. I have elpidia and the reference cards are no problem, I was mining VTC in a pool with I20 like 2 weeks ago for 2-3 days and I don't have problems. I think Your card is damaged permanently and sooner or later it will die (I remember Your post about green line etc. few pages back). Temp measured by Gpu-Z..what card (brand, model ref, non-ref You have?)


All my cards are reference asus. 3 are hynix and 2 are elpida.

The vrm temps are looking pretty normal to me right now no massive differences : listed vrm1/2 : 65/74, 63/72, 60/70, 58/67, 63/73 are what I'm seeing. I also ran @ I20 for 2-3 days, but it was one the 4th or 5th that I started seeing the issues. Honestly I can't remember having the problems for the first 2-3 days, when I had noticed the issue I was over 5 days into mining this @ i20. Actually, it looks like the line has cleared up after them comp ran overnight. .but I haven't reset it for 10 hours so it might very well come back.

Guys we've got a settings topic for this here




Just carrying the discussion to this thread.
newbie
Activity: 42
Merit: 0
February 04, 2014, 11:06:00 AM
I'm gonna say it one more time.

If you mine with a 280x set your GPU_MAX_ALLOC to 200 instead of 100.

You'll do 350ish instead of 270-300.

Doesn't have to be that way mate, my 280x's are doing 355 with the settings i posted above.

Then you're one of the lucky ones. Windows or Linux?
full member
Activity: 322
Merit: 102
February 04, 2014, 10:49:14 AM
I'm gonna say it one more time.

If you mine with a 280x set your GPU_MAX_ALLOC to 200 instead of 100.

You'll do 350ish instead of 270-300.

Doesn't have to be that way mate, my 280x's are doing 355 with the settings i posted above.
newbie
Activity: 42
Merit: 0
February 04, 2014, 10:47:05 AM
I'm gonna say it one more time.

If you mine with a 280x set your GPU_MAX_ALLOC to 200 instead of 100.

You'll do 350ish instead of 270-300.
full member
Activity: 322
Merit: 102
February 04, 2014, 10:25:23 AM
Someone please help me.

I've gotten the modded Vertminer cgminer

But I can't get it to run and mine..

it merely "opens cgminer" and just stays there on the black screen.


i have a r9 280x

this is the settings i have:

setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
C:\Vertminer\cgminer.exe --scrypt -o stratum+tcp://vertcoinheavyindustries.com:4444 -u x.x -p xxx  -I 12 -g 2 -w 256 --thread-concurrency 8212 --shaders 2048 --expiry 1 --scan-time 1 --queue 0 --no-submit-stale --auto-fan



I've tried different values with TC and I etc..  and different pools too.. still get the same thing

"Started cgminer 3.7.2"

and just hangs there..

please any advice is greatly appreciated
Smiley

setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
COLOR 0A

C:\vertminer\cgminer.exe --scrypt -o stratum+tcp://vtc.kilovolt.co.uk:3333 -u Username -p Password-w 256 -I 13 -g 2 --lookup-gap 2 --gpu-engine 1000 --gpu-powertune -20 --gpu-memclock 1490 --thread-concurrency 8193

This should work
member
Activity: 266
Merit: 10
February 04, 2014, 10:22:39 AM
Someone please help me.

I've gotten the modded Vertminer cgminer

But I can't get it to run and mine..

it merely "opens cgminer" and just stays there on the black screen.


i have a r9 280x

this is the settings i have:

setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
C:\Vertminer\cgminer.exe --scrypt -o stratum+tcp://vertcoinheavyindustries.com:4444 -u x.x -p xxx  -I 12 -g 2 -w 256 --thread-concurrency 8212 --shaders 2048 --expiry 1 --scan-time 1 --queue 0 --no-submit-stale --auto-fan



I've tried different values with TC and I etc..  and different pools too.. still get the same thing

"Started cgminer 3.7.2"

and just hangs there..

please any advice is greatly appreciated
Smiley
newbie
Activity: 58
Merit: 0
February 04, 2014, 10:20:13 AM
how much are you getting from a 5870? i get 160kh/s

member
Activity: 99
Merit: 10
February 04, 2014, 09:22:59 AM
hi i am mining this for a bit, what can I earn per Mhs? is there a calculator somewhere? Is mining with a 4720 i7 processor worth the hassle?


I think you you take your KH/s and divide by the difficulty you will get your coins per day. E.g. mining with 120KH/s will get you 3 coins per day worth maybe $3 = $8-9 per day. But difficulty is spiking and so is the price of VTC, so who knows where it will go.
newbie
Activity: 42
Merit: 0
February 04, 2014, 09:19:40 AM
Now my problem is, vertminer is not showing GPU temperatures.

What should i do?

On windows or on Linux/Debian? I had that same issue. You need the AMD ADL_SDK files for it be able to monitor the temperatures.

Make sure the files are in the ADL_SDK folder and recompile it if you're on linux/debian. If that doesn't work; download the ADL_SDK 6.0 and manually place those files in the ADL_SDK folder Smiley
sr. member
Activity: 490
Merit: 250
February 04, 2014, 08:46:31 AM
Now my problem is, vertminer is not showing GPU temperatures.

What should i do?
full member
Activity: 350
Merit: 100
February 04, 2014, 08:41:33 AM
hi i am mining this for a bit, what can I earn per Mhs? is there a calculator somewhere? Is mining with a 4720 i7 processor worth the hassle?
newbie
Activity: 42
Merit: 0
February 04, 2014, 07:57:12 AM
set GPU_MAX_ALLOX to 200 when using 280x or7970's Smiley
sr. member
Activity: 490
Merit: 250
February 04, 2014, 07:55:10 AM
I Could solve the HW errors with running these two settings before run the vertminer. other things are normal and i did not specify thread concurrency at all !

export GPU_MAX_ALLOC_PERCENT=100
export DISPLAY=:0


Currently my R9 280x giving 303KH/s, 7950 Dual-x at 256 and 7950 Reference at 253.
newbie
Activity: 2
Merit: 0
February 04, 2014, 07:40:18 AM
My sapphire 7950 on a computer with 4gb ddr2 cannot go past -i 13 without getting hardware errors, I've tried thread-concurrencies for hours. Currently it's only doing 160kh/s

Yes! Same problem here, what the hell is this.. Anything over 13 gives me loads of HW errors on Sapphire 7950 Boost.

need 8G ram
You are wrong. I have another rig with 4x 280X, 4 GB RAM, and it works fine. This one was just 2x 7950 and 4 GB RAM, and had the issue.

The solution is to run these cards with gpu-threads 2 and intensity 13. Getting stable 300 KH/s per card and no errors.

He's not wrong, 7950's need high TCC's to perform well. High TCC's need more RAM. 8GB recommended. You can't compare to 280x's because they work on 2 threads with low TCC's.
newbie
Activity: 42
Merit: 0
February 04, 2014, 07:24:33 AM
Code:
root@bamt-miner:/tmp/vertminer-gpu# ./builtit.sh
bash: ./builtit.sh: No such file or directory
I used /tmp instead of downloads but that shouldn't make any diff right?

Maybe it's because i'm using Litecoin BAMT 1.3 and not the normal BAMT
Try to read this pages:
http://www.spellzone.com/unit02/page1.cfm
http://www.oxforddictionaries.com/definition/english/build

yeah that was my mistake, typo on my end. Smiley
legendary
Activity: 1639
Merit: 1006
February 04, 2014, 07:20:13 AM

R9 290:


"gpu-engine" : "1000",
"gpu-memclock" : "1500",
"gpu-powertune" : "20",
"intensity" : "13",
"temp-overheat" : "80",
"temp-cutoff" : "85",
"temp-hysteresis" : "3",
"vectors" : "1",
"worksize" : "512",
"lookup-gap" : "2",
"thread-concurrency" : "24550"

have a sapphire 290x and anything higher than 8166 thread-concurrency gives me an out of memory error.

running 6 GB of ram on my PC and no other miner running.

the most I can get out of this 290x is 80 khs running intensity 18 with thread-concurrency 8166.

I am a total novice miner so any help appreciated.....
full member
Activity: 144
Merit: 100
February 04, 2014, 07:17:03 AM
Code:
root@bamt-miner:/tmp/vertminer-gpu# ./builtit.sh
bash: ./builtit.sh: No such file or directory
I used /tmp instead of downloads but that shouldn't make any diff right?

Maybe it's because i'm using Litecoin BAMT 1.3 and not the normal BAMT
Try to read this pages:
http://www.spellzone.com/unit02/page1.cfm
http://www.oxforddictionaries.com/definition/english/build
full member
Activity: 144
Merit: 100
February 04, 2014, 06:42:11 AM
My sapphire 7950 on a computer with 4gb ddr2 cannot go past -i 13 without getting hardware errors, I've tried thread-concurrencies for hours. Currently it's only doing 160kh/s

Yes! Same problem here, what the hell is this.. Anything over 13 gives me loads of HW errors on Sapphire 7950 Boost.

need 8G ram
You are wrong. I have another rig with 4x 280X, 4 GB RAM, and it works fine. This one was just 2x 7950 and 4 GB RAM, and had the issue.

The solution is to run these cards with gpu-threads 2 and intensity 13. Getting stable 300 KH/s per card and no errors.
Pages:
Jump to: