Alephium mainnet open for early GPU-miners

[sakrecoer]

Greetings Bitcointalk!

I come from Alephium, (which I am affiliated with) to share the news: Alephium’s mainnet was just launched yesterday after over 3years of R&D and we are welcoming miners to the mainnet.

Our initial target hashrate of 550 Gh/s represents ~1000 GPUs, but we’re in the very early stages of the network and to improve decentralization, we want to onboard more miners.

You can find the necessary tools and information on our github: https://github.com/alephium/alephium/wiki/GPU-Miner-Guide

If you have any questions about mining on Alephium, please join our Discord Server: https://discord.gg/JErgRBfRSB

Why mine Alephium?
Alephium is the first operational sharded blockchain bringing scalability, ETH-inspired smart contracts and Dapps capabilities to Bitcoin's proven core technologies (PoW and UTXO) while ensuring better performance and improved energy efficiency.

Technology

BlockFlow - Sharding on BTC’s proven foundations
Alephium is built on the BlockFlow algorithm which delivers efficient and practical UTXO sharding. Inspired by distributed computing models, it combines DAG (Directed Acyclic Graph) and sharding. Much more lightweight than other sharding methods, BlockFlow also natively supports single-step cross shard transaction making Alephium user-experience just as convenient as any single chain platform.
Blockflow Whitepaper

Proof of Less Work (PoLW)
PoLW uses a clever combination of physical work and token economics to dynamically adjust the work required to mine new blocks, ensuring a reduced energy footprint compared to classic Nakamoto PoW mining.
PoLW Whitepaper

Smart contract design
Alephium’s smart contracts have been designed and implemented to be scalable, functional, practical with a focus on security and reduced state usage. Its stateful UTXO model combines the advantages of both eUTXO model and account model.

Novel VM design
Alephium’s novel virtual machine is designed to resolve many of the critical issues of the current dApps platforms.
Alephium’s VM is less IO intensive ensuring unmatched performance. Flash loan is not available by design and the VM eliminates many attack vectors of EVM, including unlimited authorization, double dip issue, reentrancy attack, etc.
Finally, its UTXO style fine-grained execution model reduces risk-free arbitrage.

Miscellaneous

We’re passionate and committed to outstanding quality in everything we do. We took the necessary time to transform theory to actual technologies.

• 16 Shards running on mainnet.
• 100MB of RAM needed by the full node. (Runs on a Raspberry-Pi.)
• Over 3 years of R&D.

Our community is small but really engaged, talented and committed to contribute to the network.

Reach out
Discord : https://discord.gg/JErgRBfRSB
Telegram : https://t.me/alephiumgroup
Website : https://alephium.org
Medium : https://medium.com/@alephium
Twitter : https://twitter.com/alephium
Reddit : https://www.reddit.com/r/Alephium/

[1642866003]

1642866003

[1642866003]

1642866003

[Fascior2000]

GPU count: 2
GPU #0 has #4864 cores
GPU #1 has #1408 cores
cudaError 222 (the provided PTX was compiled with an unsupported toolchain.) calling 'cudaStreamCreate(&(self->stream))' (C:\Users\lbqds\Desktop\alephium\temp\gpu-miner\src\worker.h line 43)

[sakrecoer]

GPU count: 2
GPU #0 has #4864 cores
GPU #1 has #1408 cores
cudaError 222 (the provided PTX was compiled with an unsupported toolchain.) calling 'cudaStreamCreate(&(self->stream))' (C:\Users\lbqds\Desktop\alephium\temp\gpu-miner\src\worker.h line 43)

Thanks for trying and sorry for the late response.

This seems to be solved by installing cuda-toolkit as reported by one of our community members. Unfortunately we do not have access to a lot of windows systems to test that. Without wanting to get into any sort of debate on OS:es, we do however suggest to use Linux (Ubuntu, HiveOs) if possible, which has shown to be more stable for this purpose.

Hope it helps!

(I have enabled notifications now, so I should be able to respond faster next time )

[olegu5]

GPU count: 1
MapSMtoCores for SM 573401430.32762 is undefined.  Default to use 128 Cores/SM
GPU #0 has #0 cores
cudaError 3 (initialization error) calling 'cudaStreamCreate(&(self->stream))' (C:\Users\filipe\mining\gpu-miner\src\worker.h line 43)

?
windows 10

[sakrecoer]

GPU count: 1
MapSMtoCores for SM 573401430.32762 is undefined.  Default to use 128 Cores/SM
GPU #0 has #0 cores
cudaError 3 (initialization error) calling 'cudaStreamCreate(&(self->stream))' (C:\Users\filipe\mining\gpu-miner\src\worker.h line 43)

?
windows 10

Hi thanks for reaching out!
This usually means the GPU-driver is not up to date. Also note that a new version of the GPU-miner was just released: https://github.com/alephium/gpu-miner/releases/tag/v0.4.3

[olegu5]

Ubuntu 20.04 x64
1080ti
One step install
cd $HOME/.alephium-miner-setup; docker-compose logs miner  - 480mH
miningTransferAddress=12QwPTVqujvbrhMbeNrx7hRykCSjXcrZGVvDqXQEaHCJ7

mining works, but coins are not credited.

error on read -104:might be that the full node is not synced.......

what else is required?

[sakrecoer]

Ubuntu 20.04 x64
1080ti
One step install
cd $HOME/.alephium-miner-setup; docker-compose logs miner  - 480mH
miningTransferAddress=12QwPTVqujvbrhMbeNrx7hRykCSjXcrZGVvDqXQEaHCJ7

mining works, but coins are not credited.

error on read -104:might be that the full node is not synced.......

what else is required?

Hi! Sorry for the late response.

Coins are locked for 500 minutes, so it will take some time for the coins to be transferred to your address.
However, this can also happen when there are some temporary network issues. Docker will restart and recover from such temporary issues normally.

Hope it helps!

[mtmtmt]

Cant join Discord, Please make new invite Link

[sakrecoer]

Cant join Discord, Please make new invite Link

Sorry about that, but thanks for the heads up. The link has been corrected in #1. And i'll leave it here for you as well: https://discord.gg/JErgRBfRSB