I'm planning to flash my jalapeno , only thing I want to ask will I have any trouble running it with bfgminer or Bitminter clients after the flash because I am having trouble running my jally on CGminer now ?? In other words will doing the flash force me to use CGminer only ?...Thanks
No, nothing should change. But I am curious as to why/how you are having problems running one with one client and not the other. The only thing that would be a difference is making sure you have the correct driver. I used to use my jally on Windows and ran it with bfgminer just fine. I run it now connected to a BAMT rig running cgminer, and again, no problems. On Windows however, if you want to use your Jally with cgminer you will need to change over the drivers to winUSB using zadig. |
|
|
|
How can I disable Intel GPU without using "cgminer -n" at all? I would like to automatically disable Intel GPU without having to manually check up the numbers with -n switch...
If you are talking about the integrated graphics on the board, then you might have to use -d <device number> in order to declare specific GPUs to use so that it will ignore your integrated graphics. If you arent using your integrated graphics at all on your system, then disable it in the device manager and use --remove-disabled. |
|
|
|
|
Two things, 1) Do you have a GPU you are mining with as well? I noticed you have a -g 1 which is an argument you use for GPU threads. But, you also declare -G which means only USB devices. If you are only using your block erupter, then you dont need the -g 1 argument, just run the -G argument. 2) The reason BFG doesnt recognize your erupter (I am assuming because you didnt say it in your post) is because you didnt use zadig to change your winusb driver back to the old ftdi driver that was originally installed. BFG uses the old ftdi driver that is initially installed when you plug it in, so you dont need to use zadig. But, if you were using cgminer and changed it with zadig, you have to change it back in order for BFG to recognize it.
|
|
|
|
Hello, I have problem.  Who can help me? I have mining rig hardware: - 4x XFX7950 FX-795A-TDJC
- ASUS SABERTOOTH 990FXR2.0 motherboard
- 8GB RAM
- CPU AMD FX8320 8x Cores 3.51 Ghz
- PSU Lepa 1600W
Software: - Windows 7 SP1 x64
- AMD Radeo HD drivers 13.4 x64
- AMD SDK 2.7 x64
- CGMINER 3.3.1
My cgminer.conf "api-allow" : "W:127.0.0.1",
"api-listen" : true,
"api-port" : "4028",
"expiry" : "120",
"failover-only" : true,
"hotplug" : "5",
"kernel-path" : "/usr/local/bin",
"log" : "5",
"no-submit-stale" : true,
"queue" : "1",
"rotate" : "10",
"scan-time" : "30",
"scrypt" : true,
"shares" : "0",
"auto-fan" : true,
"gpu-threads" : "1",
"gpu-dyninterval" : "7",
"gpu-engine" : "0-1025",
"gpu-fan" : "0-100",
"gpu-platform" : "0",
"gpu-memclock" : "1250",
"gpu-memdiff" : "0",
"gpu-powertune" : "0",
"gpu-vddc" : "1.087",
"intensity" : "20",
"temp-target" : "75",
"temp-overheat" : "88",
"temp-cutoff" : "95",
"temp-hysteresis" : "3",
"vectors" : "1",
"worksize" : "256",
"lookup-gap" : "2",
"shaders" : "2048",
"thread-concurrency" : "21712",
"no-pool-disable" : true
With this configuration 3x GPU cards running stable. But when I connect 4 card and run cgminer works very bad and turnoff motherboard. But PSU still running. Who can help me with this problem? Maybe not impossible run rig with 4x GPU's? Maybe my configuration is wrong? Or I forget something to install? Than you in advance, TDS With this config you should be able to run those cards, but the problem could be either something with Windows when connecting the 4th card. I believe I have seen posts on this. But, it also could be a heat issue.If the cards are not breathing properly, then they are overheating, and your mining is getting shut down. |
|
|
|
|
Whats your configurations? Whats your drivers version? Most crashes I usually see is due to temps getting to high.
|
|
|
|
|
Would also try SDK 2.4, seems to work pretty well with 6xxx series cards.
|
|
|
|
|
It could be because of the work. Your GPU requests works, gets its, does the work, then submits it. In that mean time between waiting for work to come in, you yours will sometimes throttle down to conserve power, cool off back to your target temp, etc.
|
|
|
|
So, I took down the post I have and am just going to post what to do here since all of the other steps are here, and only really need one step changed. During the nano update.sh process, just need to change the ./configure flags. This will enable scrypt mining, Block Erupters, Avalon and any BFL device. export DISPLAY=:0
export GPU_USE_SYNC_OBJECTS=1
rm *.bin
make clean
cd /
tar xf /opt/AMD-APP-SDK-v2.4-lnx32/icd-registration.tgz
ln -s /opt/AMD-APP-SDK-v2.4-lnx32/include/CL /usr/include
ln -s /opt/AMD-APP-SDK-v2.4-lnx32/lib/x86/* /usr/lib/
ldconfig
cd /opt/miners/cgminer
./autogen.sh
ATISTREAMSDKROOT="/opt/AMD-APP-SDK-v2.4-lnx32/" CFLAGS="-O2 -Wall -march=native" ./configure --enable-bflsc --enable-icarus --enable-scrypt --enable-avalon --bindir="/opt/miners/cgminer" --prefix="/opt/miners/cgminer"
make -j3
sync
./cgminer -n |
|
|
|
|
The issue I see with this is your are mining scrypt, which using higher memory clocks. And from what I know, in either cgminer, or bfgminer, there is now way to throttle memclock. You set it at one setting and go. If you were mining regular sha256 you would be worried about your gpu engine which you could throttle. You would then have your system throttle it between a range to hit a target temp (or try to).
Only thing I can tell you to do from reading below is increase your fan speeds:
--gpu-fan 90
This will make your fans spin at 90% of max speed (can go 100, but, will wear them out faster and burn out your fans).
Seems the best solution would be to get better airflow across those cards.
|
|
|
|
|
Got BE myself today. Have 8 of them (soon to be 12, as soon as I can get another plug near here), a Jally, and 2 GPUs all running on BAMT rig. Now to order more.
|
|
|
|
|
SDK numbers it is referring to is for Linux systems, if you are running Windows then are worried about Catalyst version.
The issue you will have with running two different series cards that require minimal driver versions is that its been seen that sometimes skipping up to the next series gives your older card a downgrade in performance. For example, 5xxx series cards work best with 11.x series drivers, specifically 11.11 or 11.12. If you step up to 12.x series drivers which are needed for 6xxx series cards, your 5xxx cards take performance hits.
Now I could be wrong and someone could suggest something else that would work, but I have always gone off the idea that you want to stick to the same series cards inside the same rig.
|
|
|
|
|
Should be able to get a little more than that. What are your settings for it?
|
|
|
|
I won't be touching the 1x slots yet, the board I'm using has 4 16x slots. I'll be using risers with the 1x slots later to expand of course.
I have a feeling the powered risers are still recommended even in this case, though?
The 2 PSUs in question are a SolyTech SL-8600EPS which is 80%+ rated and a Thermaltake Toughpower W0103RU 600W, rated at 85%+.
I was also planning on testing the overall system with a kill-a-watt once it's put together.
Dont know if they are still needed, never done any 16x to 16x extending, but its better to be safe than sorry, right? Having them isnt going to hurt so might as well go the safe route. |
|
|
|
...
What about if you're not using risers? for mining LTC on multiple cards should I get some powered risers regardless if I need risers or not?
...
Then good luck keeping the cards cool. If you have three cards plugged directly into a mobo then the gaps between the cards will not be enough to allow the fans to move enough air to cool the cards when they are mining LTC. I would image that depends entirely on the mobo in question. But I will probably end up using risers. I'd just rather get as much info as possible. If you are mining LTC, you WILL need powered risers, regardless of MB and spacing. LTC mining taxes your VRAM, where as BTC mining is GPU heavy. Your GPU gets power from the PCI-E power connectors from your PSU, while your VRAM draws its power from the PCI-E bus on the MB. If you are trying to draw PCI-E 16x power, overclocked power, out of a 1x slot, bye bye board. Do any search around here for the numerous boards that have fried, caught fire, etc because of this. |
|
|
|
|
Firstly, the question about powered vs non-powered risers depends on what kind of coin you are mining. BTC, you dont need powered risers, LTC you absolutely do.
As for as dual PSU's, couldnt tell you, but I will tell you what I would do. Its just electricity to them from what I know of, so I would have my normal PSU setup, but my second one, in order to get it running, will need an ATX connection. You can pick up cheap PSU testers that you can connect to the second PSU that will turn it on. Then I would have one PSU feeding its PCI-E connectors to one card, and the other PSU to the other card. All in some open air care or configuration. And thats it. Each card would be getting enough power from the PSU's and can mine just fine. Nothing else fancy would need done I dont think.
|
|
|
|
what 5870 do you have? my 5870 gets 12.5 Mkey/s
Reference. I haven't actually tested it yet, waiting on a SATA to MOLEX connector. I believe there are settings available to get a 5870 to 30 MKeys though. Ok. I probably have a faulty 5870 because when I push it higher than 64 threads in the software, which pulls 12.5 Mkey, it causes the drivers or system to fail and reboot. Would a 7870 or 7970 be better though? What driver version are you using? As proven by my earlier post, I have no clue about vanity keys. I know when they first came about I messed around with it for like a day but then left it alone as I didnt want my GPUs not to be mining. Now, I could probably do it. A couple of 5830's and a 5870, would let them do their work and make a key or two. Anyway, back to the topic at hand. The reason I ask the driver question is because if its anything like mining, driver versions are a big part of it. I know 5xxx series cards like 11.11 or 11.12 drivers. Anything above that it actually starts taking performance hits. |
|
|
|
My problem is I have 2 video cards, and a Jalapeno all running on the same system. 1 video card is giving me problems and I want to disable it and still use the other video card AND my Jalapeno (the cards are hooked up in a watercooling rig and I was trying to prevent ripping it completely apart in order to remove the bad card from the system altogether). All of the things I tried would either let me run just the Jalapeno, or just one the one video card, not both.
Yes I understand that. So to disable the bad GPU you need to know which device it is and then use the remove disable in conjunction with that device. In my suggested example "-d 0 --remove-disabled" That would completely remove device 0 from CGMiner's knowledge and would just mine your other devices. If your bad GPU is a different device then substitute the 0 with the correct number. Sam Well, thanks for your help. I wasnt getting that result when I was trying, I behavior I was getting was both my Jalapeno and my GPU 0 were disabled (and once removed disabled was used they both disappeared). In the mean time I shut the rig down for an hour, started it up, and set the bad card to stock settings. Its been running fine for nearly 24hrs so Im keeping my fingers crossed on it. If it starts acting up again I will just have to pull it and call it a day. Thanks everyone for the help. |
|
|
|
Hey guys, need a little help myself this morning and my Google-fu isnt working well (lack of food probably isnt helping). I have a rig running 2 GPUs and a Jala. One of my GPUs is giving me fits for some reason today, so I want to shut it down and mine with just the one card and my Jala. Its a watercooled unit and I am in the middle of work so dont have time to just rip it apart before anyone just asks why dont I take the card out.
I am trying to shut down the one card and mine with just the other and Jala, I can use -d 1 to use the one card, or I can use -G to shut them down and mine with the Jala, but I cant get them both working with the combination of -d 1 and --usb. I think my problem is the --usb and the syntax, but for the life of me I cant seem to figure out what it is (I've tried --usb 0, --usb BAJ, --usb baj, --usb BAJ0, and --usb baj0).
Is there something I am obviously missing, or is it just not wanting to play nice with me today?
--remove-disabled Remove disabled devices entirely, as if they didn't exist --device|-d <arg> Select device to use, one value, range and/or comma separated (e.g. 0-2,4) default: all So it could look something like this -d 0 --remove-disabled Correct me if I am wrong, removing disabled will just not show the disabled item on my cgminer screen --remove-disabled will completely remove that GPU/device from CGMiner's knowledge. So preface it with -d 0, -d 1 or -d 2 or whatever device the bad GPU is. I thought that is what you wanted to do? If so that is how you do it. Sam My problem is I have 2 video cards, and a Jalapeno all running on the same system. 1 video card is giving me problems and I want to disable it and still use the other video card AND my Jalapeno (the cards are hooked up in a watercooling rig and I was trying to prevent ripping it completely apart in order to remove the bad card from the system altogether). All of the things I tried would either let me run just the Jalapeno, or just one the one video card, not both. |
|
|
|
|
What FPGA and ASIC devices are you using specifically?
|
|
|
|
|
Show Posts
