Can I Use 2 Different NVIDIA GPUs on this motherboard?

As per The Fine Manual – it depends on your GPU – and its something I’ve never seen before.

If you have a gen 1 or gen 2 ryzen CPU, the first slot and second slot work at x8 like nearly every other board I’ve seen.

With an 3rd or 4th gen GPU – The card in the first x16 slot will run at x16 (Put your 2080TI there!) and second slot will run at x4 (which is a bit meh, but I don’t think the 1050ti will be bottlenecked that badly).

enter image description here

Its also worth remembering, depending on your CPU option, you may also have a vega adaptor – ‘mixed’ GPUs are supported in windows, and/or using that as the adaptor for your VM host in the other OS may be an option

hash – Since GPUs have gigabytes of memory, does Argon2id need to use gigabytes of memory as well in order to effectively thwart GPU cracking?

The common advice of benchmarking a password hashing algorithm and choosing the slowest acceptable cost factor doesn’t work for algorithms with more than one parameter: adding a lot of iterations at the expense of memory hardness makes the benchmark time go up, but if the attacker can still grab off-the-shelf hardware (like a regular gaming graphics card) then I might as well have used pbkdf2 instead of Argon2id. There must be some minimum amount of memory that makes Argon2id safer than a well-configured algorithm from the 90s, otherwise we could have saved ourselves the effort of developing it.

I would run benchmarks and just see for myself at what point hashing isn’t faster anymore on a GPU than on a CPU, but Hashcat lacks Argon2 support altogether. (Though any software I choose isn’t necessarily going to be the fastest possible implementation to give me a correct answer anyway.)

Even my old graphics card from 2008 had 1GB of video memory and 2020 cards seem to commonly have 4-8GB. Nevertheless, the default Argon2id setting in PHP is to use 64MiB of memory.

If you set the parallelism of Argon2id to your number of CPU cores (say, 4) and use this default memory limit, is either of the following (roughly) true?

  • A GPU with 8192MB of memory can still use 8192/64 = 128 of its cores, getting a 32× speed boost compared to your CPU. The slowdown on GPUs due to increased memory requirements is a linear scale. The only way to thwart GPU cracking effectively, is to make Argon2id use more RAM than a high-end GPU has when divided by the number of parallelism you configure in Argon2id (i.e. in this example with 4 cores, Argon2id needs to be set to use 8192/4 = 2048MB).

or

  • This amount, 64MiB, already makes a common consumer GPU completely ineffective for cracking because it is simply too big to efficiently use from a simple GPU core (because if the core has to access the card’s main memory then it’s not worth it, or something).

Need new GLIBC on Centos 6 to use PyTorch on GPUs

I am using a supercomputer facility which is running Centos 6. The node I want to use has 3 Tesla V100. The problem is that the version of GLIBC installed on that node is not compatible with the latest versions of PyTorch I must use.

I do not have root. Hence, I need a way to use another GLIBC from a user level. I can talk to the sysadmins and do stuff from root (like use Docker, or something like that) but I cannot reinstall OS or GLIBC globally. I have tried to install GLIBC by myself without root, but could not do it right; it did not work. It takes too much time, and cannot find a tutorial to do it right.

I have some ideas in mind, like trying to run a Container that can access the node and use other OS; when my scheduled computing time ends I can leave the node the same way I got it; for the next user on the supercomputer facility.

I was also thinking about chroot; Download an ISO of Centos 8 and chroot into it, but I do not know if it would use the GLIBC used by the Host OS, or Centos 8 GLIBC.

What do you recommend me to do? Do you think Docker would suit my needs? Or other containerization solution? Keep trying with installing other GLIBC?

graphics card – Monitor has a stuck flickering image on the background, it persists through reboots, and it persists through different GPUs

This is very odd behavior I have never seen because it seems the monitor itself, has RAM that has burned in or capacitors keep a buffer up or something for hours, because I have removed all power and tried several reboots and resets and an image of a program since the first time it crashed is still there at the background flickering while you can hover other programs on it but they still get distorted as well.

I would normally consider this a falty panel, but it is clearly a hardware issue in the rest of the monitor itself, because the OSD comes up crystal clear with the artifacted image flickering behind it and windows has the fault after several reboots and on different GPU slots (I have 3 monitors and 2 GPUs).

complexity – Efficiency of different Processors and GPUs

If you can break your operation down to assembly-level, it should be possible to estimate execution time.

Since each machines assembly-code can be easily translated to its respective machine instructions, you can simply “count” the number of clock-cycles a piece of software would take to run. This approach is sometimes used on simple processors running simple programs (e.g. micro-controllers), when precise timing is essential.

However keep in mind:

  1. Every processor architecture has it’s very unique way of handling high-level operations. This means the amount of work, required to “estimate” the runtime of your high-level code on different machines would be insane.

  2. That modern computing systems usually run your application on top of an OS. This additional layer of indirection makes it impossible to “guess” when your code will be executed and thus how quickly it will finish.

TL;DR No, you probably won’t be able to know, when your software will be finished rendering something without testing it.

drivers – ubuntu install failed detect the available gpus and deal with any system changes. dell 5820 i9

I have a new Dell 5820 i9 workstation I am trying to load ubuntu-20.04.1-desktop-amd64.iso with usb drive.

The install fails at:
“ubuntu install failed detect the available gpus and deal with any system changes”

a radeon pro wx 5100 card is used.

any help would be great.

Thanks in advance
Phil
enter image description here

enter image description here

enter image description here

enter image description here

mining pools – OVMiner v1.7 for AMD GPUs

OVMiner v1.7- fast miner with good hashrate and high performance for AMD GPUs

Download:

Supported algorithms:

  • Cuckaroo29 (Grin) (Amd)
  • Cuckatoo31 (Grin) (AMD only)
  • Cuckoo29 (Aeternity) (Amd)
  • Cuckaroo29s (Swap) (Amd)
  • Equihash 96.5 (MinexCoin) (AMD only)
  • Equihash + Scrypt (Vollar) (AMD only)
  • Equihash 125.4 (ZelCash) (AMD only)
  • Equihash 144.5 (Bitcoin Gold, BitcoinZ, SnowGem, ZelCash) (Amd)
  • Beam Hash (BEAM) (Amd)
  • Equihash 192.7 (Zero, Genesis) (Amd)
  • Equihash 210.9 (Aion) (AMD only)

Performance in GPU settings:

  • Cuckaroo29 / Cuckaroo29s:
  • Palit GTX 1060 JetStream 6GB ~ 3.2 G / s
  • Gigabyte GTX 1660 Ti OC 6GB ~ 4.55 G / s
  • Palit GTX 1070 Dual 8GB ~ 4.65 G / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 5.25 G / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 5.5 G / s
  • Gigabyte GTX 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 8 G / s
  • Palit RTX 2060 Dual 6GB ~ 6 G / s
  • MSI RTX 2070 Armor 8GB ~ 7.6 G / s
  • MSI RTX 2080 DUKE 8GB ~ 8.8 G / s
  • PowerColor RX570 8GB ~ 1.6 G / s
  • PowerColor RX580 Red Devil 8GB ~ 1.8 G / s
  • Sapphire VEGA 56 8GB ~ 3.2 G / s
  • MSI VEGA 64 8GB ~ 3.85 G / s

Cuckatoo31:

  • Palit GTX 1070 Ti JetStream 8GB ~ 0.47 G / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 0.48 G / s
  • Gigabyte GTX 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 0.70 G / s
  • MSI RTX 2070 Armor 8GB ~ 0.94 G / s
  • MSI RTX 2080 DUKE 8GB ~ 1.13 G / s

Cuckoo29:

  • Palit GTX 1060 JetStream 6GB ~ 3.18 G / s
  • Gigabyte GTX 1660 Ti OC 6GB ~ 4.63 G / s
  • Palit GTX 1070 Dual 8GB ~ 4.65 G / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 5.23 G / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 5.5 G / s
  • Gigabyte GTX 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 8 G / s
  • Palit RTX 2060 Dual 6GB ~ 6 G / s
  • MSI RTX 2070 Armor 8GB ~ 7.54 G / s
  • MSI RTX 2080 DUKE 8GB ~ 8.8 G / s
  • PowerColor RX570 8GB ~ 1.88 G / s
  • PowerColor RX580 Red Devil 8GB ~ 2.07 G / s
  • Sapphire VEGA 56 8GB ~ 3.2 G / s
  • MSI VEGA 64 8GB ~ 4.4 G / s

Equihash + Scrypt:

  • Palit GTX 1060 JetStream 6GB ~ 14 KSol / s
  • Gigabyte GTX 1660 Ti OC 6GB ~ 19.5 KSol / s
  • Palit GTX 1070 Dual 8GB ~ 19.7 KSol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 23.1 KSol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 27 KSol / s
  • Gigabyte GTX 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 37.3 KSol / s
  • Palit RTX 2060 Dual 6GB ~ 24 KSol / s
  • MSI RTX 2070 Armor 8GB ~ 28.5 KSol / s
  • MSI RTX 2080 DUKE 8GB ~ 36.5 KSol / s

Equihash 96.5:

  • Palit GTX 1060 JetStream 6GB ~ 15.3 KSol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 24.7 KSol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 28 KSol / s
  • EVGA GTX 1080 Ti Founders Edition 11GB ~ 37.7 Sol / s
  • Gigabyte 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 39.5 Sol / s

Equihash 125.4:

  • Palit GTX 1060 JetStream 6GB ~ 22.3 Sol / s
  • Gigabyte GTX 1660 Ti OC 6GB ~ 26.1 Sol / s
  • Palit GTX 1070 Dual 8GB ~ 32.9 Sol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 40.1 Sol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 42.3 Sol / s
  • Gigabyte 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 56.9Sol / s
  • Palit RTX 2060 Dual 6GB ~ 35.5 Sol / s
  • MSI RTX 2070 Armor 8GB ~ 45.3 Sol / s
  • MSI RTX 2080 DUKE 8GB ~ 58.9 Sol / s

Equihash 144.5:

  • Palit GTX 1060 JetStream 6GB ~ 37.5 Sol / s
  • Palit GTX 1070 Dual 8GB ~ 55.5 Sol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 65 Sol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 69 Sol / s
  • Gigabyte 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 96 Sol / s
  • MSI RTX 2070 Armor 8GB ~ 65.6 Sol / s
  • MSI RTX 2080 DUKE 8GB ~ 68 Sol / s
  • PowerColor RX570 8GB ~ 24 Sol / s
  • PowerColor RX580 Red Devil 8GB ~ 27 Sol / s
  • MSI VEGA 64 8GB ~ 43 Sol / s

Beam Hash:

  • Palit GTX 1060 JetStream 3GB ~ 13.9 Sol / s
  • Palit GTX 1060 JetStream 6GB ~ 14.8 Sol / s
  • Gygabyte GTX 1660 Ti OC 6GB ~ 16.7 Sol / s
  • Palit GTX 1070 Dual 8GB ~ 22.7 Sol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 28.5 Sol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 31 Sol / s
  • Gigabyte GTX 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 41 Sol / s
  • Palit RTX 2060 Dual 6GB ~ 25.8 Sol / s
  • MSI RTX 2070 Armor 8GB ~ 32 Sol / s
  • MSI RTX 2080 DUKE 8GB ~ 41.5 Sol / s
  • PowerColor RX570 8GB ~ 11 Sol / s
  • PowerColor RX580 Red Devil 8GB ~ 13 Sol / s
  • Sapphire VEGA 56 8GB ~ 18 Sol / s
  • MSI VEGA 64 8GB ~ 21 Sol / s

Equihash 192.7:

  • Palit GTX 1060 JetStream 6GB ~ 21 Sol / s
  • Palit GTX 1070 Dual 8GB ~ 30 Sol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 37 Sol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 39 Sol / s
  • Gigabyte 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 54 Sol / s
  • MSI RTX 2070 Armor 8GB ~ 36 Sol / s
  • MSI RTX 2080 DUKE 8GB ~ 52 Sol / s
  • PowerColor RX570 8GB ~ 14 Sol / s
  • PowerColor RX580 Red Devil 8GB ~ 17 Sol / s
  • MSI VEGA 64 8GB ~ 26 Sol / s

Equihash 210.9:

  • Palit GTX 1060 JetStream 6GB ~ 147 Sol / s
  • Palit GTX 1070 Ti JetStream 8GB ~ 209 Sol / s
  • MSI GTX 1080 SEA HAWK EK X 8GB ~ 227 Sol / s
  • EVGA GTX 1080 Ti Founders Edition 11GB ~ 335 Sol / s
  • Gigabyte 1080 Ti AORUS Waterforce WB Xtreme Edition 11GB ~ 347 Sol / s

GPU tuning:
BTM + ETH

The appropriate “secondary rate” depends on the kernel performance / memory bandwidth ratio.
GPUs with relatively low memory bandwidth, for example. 1070ti, can customize “di”. In pr

mining pools – Bminer: a fast Equihash/Ethash/Cuckaroo29m miner for AMD/NVIDIA GPUs 16.2.9

Bminer is a highly optimized cryptocurrency miner that runs on modern AMD / NVIDIA GPUs (Maxwell and Pascal, i.e. GPUs with a computing power of 5.0 or higher). Bminer is one of the fastest miners available today – we use a variety of techniques, including tiling and pipelining, to realize the full potential of the hardware.

Bminer also comes with REST APIs to simplify product deployments (like mining farms).
Bminer supports Equihash based coin mining (like Zcash) with 2% devfee.
Bminer supports Zhash / Equihash 144.5 based coin mining (e.g. BitcoinGold, BitcoinZ) with 2% from devfee.
Bminer supports Ethash based coin mining (like Ethereum) with 0.65% devfee.
Bminer also supports dual mining mode – simultaneous mining of Ethash based coins (like Ethereum) and Blake14r based coins (like Decred) / Blake2s (like Verge). Devfee for dual mining mode is 1.3%, and the second coin (e.g. Decred / Verge) is mined without devfee.
Bminer supports Tensority based coin mining (like Bytom (BTM)) with 2% devfee.
Bminer supports Grin (GRIN) mining with 1% devfee.

Download

Specifications

Fast

Grin31 mining on stock settings
2.60 G / s on RTX 2080Ti
1.65 G / s on RTX 2080
1.55 G / s on GTX 1080Ti
0.95 G / s on GTX 1070

Bminer 16.0.6 Grin29m mining on stock settings
8.32 G / s on GTX 2080Ti
5.18 G / s on GTX 2070
3.96 G / s on 2060
2.18 G / s on 1060
3.34 G / s on 1070
5.03 G / s on 1080ti
3.00 G / s on P104-4G
3.56 G / s on P104-8G
4.95 G / s on P102-10G

AE mining on stock settings:
11.8 Sol / s on GTX 2080Ti
8.90 Sol / s on GTX 2080
7.40 Sol / s on GTX 1080Ti
4.7 Sol / s on GTX 1070
3.4 Sol / s on GTX 1060 6G

Beam mining on stock settings
30 Sol / s on GTX 1080Ti
21 Sol / s on GTX 1070
12 Sol / s on GTX 1060 6G

Equihash mining on stock settings
735-745 Sol / s on GTX 1080Ti
450-460 Sol / s on GTX 1070
315-325 Sol / s on GTX 1060

Equihash 144.5 (Zhash) mining on stock settings
61 Sol / s on GTX 1080Ti
25 Sol / s on GTX 1060
Ethash mining on GTX 1080Ti stock settings (power: 250 W)

With OhGodAnETHlargementPill: 46.7 MH / s
Without OhGodAnETHlargementPill: 32.2 MH / s

Dual mining using automatic tuning (default) on GTX 1080Ti stock settings (power: 250 W)

With OhGodAnETHlargementPill:
ETH 46 MH / s and DCR 1000 MH / s
ETH 46 MH / s and XVG 1770 MH / s

Without OhGodAnETHlargementPill:
ETH 32 MH / s and DCR 2200 MH / s
ETH 32 MH / s and XVG 3750 MH / s

Bytom mining on stock settings:
4650 H / s on GTX 1080Ti
2850 H / s on GTX 1070
1800 H / s on GTX 1060 6G
Safe and reliable

SSL support
Automatic reconnection to recover from network failures.
Automatic restart when GPU freezes

Operation friendly

Comes with REST API to simplify deployment

Fast start

Depending on the coins you want to mine, find the appropriate script in the folder. For example, the corresponding script is mine_grin29.bat (on Windows) or mine_grin29.sh (on Linux) when mining Grin using the Cuckaroo29 algorithm.
Change the address and account information in the script.
Run the script and enjoy mining clear.png
Release Notes

16.2.9 (current)
Improve the performance of the Cuckaroo29z miner.

linux – Mounting virtual GPUs on different machines

I have multiple computers connected over a VPN, each computer having one or more Nvidia GPUs. All computers run Linux. I’m wondering if it’s possible to mount them over network such that they appear to be attached on a single machine.

For example, if I have a machine (say MA) with one GPU and another machine (say MB) with two GPUs. Running nvidia-smi on MA would list the one GPU and running it on MB would list two GPUs.

I want to mount MB’s GPUs on MA as virtual GPUs such that running nvidia-smi on MA will now list three GPUs. Is this possible?

I realize this will have high latency. But in my case, that is not a concern.