We need to point out that Linux VEGA support is pretty limited right now. Getting things working requires some linux knowledge, such as configuring GRUB, setting up unique environments, and some special BIOS qualifications. Also requires on VEGA in a platform, and use of at least x8 PCI slots. If you have a VEGA board, hunt me down on Discord and we can chat through the requirements. AMD has published a roadmap that brings better support for VEGA and Ravenridge product over the next few months. We are dependant on AMD to complete this work though, as those parts are closed source and can only be used by us after release from AMD. Expect full support soon! As for 19 GPUs on a single B250 mining expert.. That also needs support from the vendors, mining applications, and drivers. When the B250 BIOS code supports all 19 GPUs online of just one make and model, we will be a lot closer. But please follow the board makers directions for laying out GPUs please, we cannot change that. I personally do not understand the fascination with using just one MB to drive so many GPUs. To get the density, they use lots of little bridges. And when I use the same bridges, it causes me lots of issues. Memory transfer rates drop drastically on some memcopy intensive algorithms. For example, Cryptonight drops by about 1/5th the speed when the GPUs mining it are behind a 4:1 PCIe bridge. Another issue with such GPU density on a single MB, is the power requirements for a single rig. If you have 19 VEGAS say, into one board.. you could be pushing 7,000 watts of power. The average hobby miner would never want todo that. Besides the power, you have the fault management. If one GPU fails out of 19, and you need to reboot. You have to take all of them offline to handle that. All mining stops on all cards when you take an outage for anything. Regardless, the answer for both of these questions is simply that we need better support from the vendors to insure these cards and mb works right.

I would love to have All The Miners in miner.farm! If these devices are USB attached to your rig, and use a miner program running on PIMP, it should work as expected. If it does not, I will need to integrate the mining program API into the agent - easy to do if we have access to a device, but we have a limited budget for the PIMP lab so help from miners in the field is appreciated! If you have a device running on PIMP that doesnt report and are willing to provide developer access for integration, please let us know. If these devices are standalone, using their own OS and mining program, it is a much different prospect. Depending on the OS and what it will allow, it may be possible to install the standalone agent and have it report to miner.farm, but again we would need access to a running device to properly test and integrate. Thank you, Lily

@Anjin, I am only setting the clock, core and power in the gputool config file. I used the same settings as I did in Ethos. I will need to put the SSD back in and do the "helpme", but the same SSD in an Nvidia only rig works fine, except I noticed that the hashrate per card using ethminer is also approx. 1MH/s per card lower than in ethos with the same overclock settings. So...clearly I must be doing something wrong.

If bin file not found, pimpup the miner. It may not be installed.

Are the two cards identical (same maker, brand, etc?)

Thanks for the info !

How's 2.5.1 running?

In the mean time you can use ccminer-cryptonight: ccminer-cryptonight is available, use pimpup to install ccminer-cryptonight then use pimp --add 2121 to add the profile. You must use del ccminer.xmr.pcfg or change ALGORITHM=cryptonight to ALGORITHM=monero If you get rejects, then this is the fix. 20180415 [M] Added ccminer-cryptonight (nvidia 2121) for nvidia xmr v7 support (must install with pimpup!!) Thank you for your patience, we wanted the ccminer devs to merge the update in the correct way to the main branch but they did not, so we provided this instead.

ccminer-cryptonight is available, use pimpup to install ccminer-cryptonight then use pimp --add 2121 to add the profile. You must use del ccminer.xmr.pcfg or change ALGORITHM=cryptonight to ALGORITHM=monero If you get rejects, then this is the fix. 20180415 [M] Added ccminer-cryptonight (nvidia 2121) for nvidia xmr v7 support (must install with pimpup!!) Thank you for your patience, we wanted the ccminer devs to merge the update in the correct way to the main branch but they did not, so we provided this instead.

Moved to hardware troubleshooting - not a software issue

@techno_express Hmm, sorry your having trouble getting on discord, not sure why that is - tell you want I leave myself a note to update you via this thread on future developments as the arise :) Best Louis

There are some other bits in the image as well. And actually, the average SSD is 16GB. There are a couple cheap brands, kingdian as an example, that are not. The difference is what we used to call marketing GB vs real GB... centered around someone in engineering used 1000 instead of 1024.. shrug.. hard for us to control. An easy work around is just buy a bigger unit, price difference for 32GB is not much more. Or just use an old HDD already have laying around.

those are already set in your environment. [/root]:# env | grep GPU GPU_USE_SYNC_OBJECTS=1 GPU_MAX_ALLOC_PERCENT=100 GPU_FORCE_64BIT_PTR=1 GPU_MAX_HEAP_SIZE=100 And in your sample script, you say 'cleanbins' then in the script, you use 'clearbins'. As for the startup.sh, try changing the wait time a bit. Make it wait just a bit longer so the cards have a chance to fully boot strap and get started. Use 'clinfo' to give you an idea how the AMD drivers see the card. And in the miner, run a device list to see what platforms are what, to insure the opencl-platform aligns with what the miner finds.

Im having the same issue, just got PiMP too but havent been able to get my cards going. I initially got that error msg but if i unplug the cards and just run my main vga I can load PiMP OS but if I plug even one back in I get this error message, sometimes it will load to the point of showing my IP for rig after connecting to network and validating key but freezes at IP line and have to shut down. I have a H110 Pro Motherboard and the only instructions I couldn't follow was where it says to plug in first card into PCIE2, which is my main VGA and I have onboard disabled and mining edition cards dont have video adapter so couldn't really follow that one, Im using Sapphire RX470 8gb. Im thinking its either because I didn't follow that step or maybe the firmware BIOS since I did run them on another comp to test them out when I got them.

I'm running this board with 8 MSI GTX 1050 2GB OC cards without any problems on Pimp v2.5-all However, when card 9 and 10 come back from MSI RMA (due to physical damage unrelated to mining), I may run into some problems... From what I've heard, this board has limits to how many nVidia cards it can run at one time. Asus support shows that this board is designed to run a combination of nVidia and AMD... something like 8 NV and 11 AMD... or something close to that. Some have said that the latest BIOS fixed this? It doesn't seem that way to me, but then again, I don't have any AMD cards to test with and I won't be able to go over 8 NV cards until the other two come back. Check the FAQ on this board from the Asus support page. They provide diagrams for known good video card populations of the PCIe slots. What I have been able to test is that unless my nVidia cards are in the upper half of PCIe slots, weird things begin to happen when Pimp loads up (if it even loads at all). Good luck.

Anyone has any suggestion for me? I'm using 1102 profile => https://forum.getpimp.org/topic/10/setting-up-miner-profiles with basically default configuration default core and memory clocks values for my VGA (tried with default values and OC values) http://xfxforce.com/en-us/Promotions/2017/oc XFX Part Number and Edition True Clock (Default) OC+ Core Clock OC+ Memory Clock RX-580P8DBD6 - GTS Black Edition 1405 1425 2025 Using Bios2 now Adapter 0 (BN=01, DN=00, PCIID=67DF1002, SSID=C5801682) Asic Family : Ellesmere Flash Type : M25P40/c (512 KB) Product Name : C940 Polaris20 XTX A1 GDDR5 256Mx32 8GB 300e/300m Bios Config File : 5808SHD1.M80 Bios P/N : 113-58085SHD1-M80 Bios Version : 015.050.002.001.000000 Bios Date : 04/10/17 23:15 ROM Image Type : Hybrid Images ROM Image Details : Image[0]: Size(59904 Bytes), Type(Legacy Image) Image[1]: Size(58368 Bytes), Type(EFI Image)

No. This rig are all AMD RX 480's

No it is not. I have tried reinstalling OS on both machines and creating new API keys without any success.