How to use TeamRedMiner for AMD GPUs? (Script Optimization)

Installation and Prerequisites

1. Verify that your AMD GPU model is supported by TeamRedMiner, including RX 400/500 series, Vega, and RDNA-based cards such as RX 5700 XT and RX 6000 series.

2. Install the latest compatible AMD Adrenalin driver—versions 22.5.1 through 23.12.1 are confirmed stable for mining workloads without excessive thermal throttling.

3. Download the official TeamRedMiner release archive from the GitHub repository, ensuring the binary matches your operating system (Windows or Linux).

4. Extract the archive into a dedicated directory with no spaces or special characters in the path to avoid parsing errors during script execution.

5. Confirm OpenCL runtime libraries are present; on Linux, install opencl-amd and ocl-icd-libopencl1; on Windows, these ship with the Adrenalin driver package.

Basic Configuration via Batch or Shell Script

1. Create a launch script named mine.bat (Windows) or mine.sh (Linux) in the same folder as the miner binary.

2. Define pool connection parameters using --pool, --user, and --pass flags, specifying stratum URL, wallet address, and worker name respectively.

3. Set algorithm explicitly with --algo, for example ethash for Ethereum Classic or kawpow for Ravencoin.

4. Assign GPU devices using --devices, listing zero-indexed IDs like 0,1,2 to target specific cards while excluding integrated graphics.

5. Enable persistent mode on Linux with sudo amdconfig --set-pmode=1 -d before launching to prevent GPU clock downshifts during idle intervals.

Advanced Tuning Parameters

1. Adjust memory tuning using --mt followed by an integer value between 10 and 30 to control memory timing offsets for improved stability under high load.

2. Apply core clock reduction with --cclock to lower power draw—for instance, --cclock -50 reduces base frequency by 50 MHz across all selected GPUs.

3. Use --mclock to fine-tune VRAM frequency, typically ranging from +100 to +300 for GDDR6-based cards when paired with adequate cooling.

4. Enable auto-tuning with --tune to let TeamRedMiner run internal benchmarks and select optimal memory timings per device during startup.

5. Add --log-file miner.log to capture real-time hashrate, temperature, and error messages for later diagnostics without console clutter.

Monitoring and Stability Checks

1. Monitor GPU utilization and junction temperature using rocm-smi on Linux or GPU-Z on Windows to detect thermal saturation above 95°C.

2. Observe rejected share rates in the miner’s output—if rejections exceed 0.5%, verify pool connectivity, adjust --retry-pause, or reduce overclock aggressiveness.

3. Check for OpenCL initialization failures in logs; these often indicate driver mismatch or insufficient permissions on Linux systems running under non-root users.

4. Validate memory integrity by running memtest_cl prior to extended mining sessions, especially after applying aggressive --mclock values.

5. Rotate log files daily using external cron jobs or Windows Task Scheduler to prevent unbounded disk growth when --log-file is active.

Frequently Asked Questions

Q: Why does TeamRedMiner report “No OpenCL devices found” despite having AMD drivers installed?A: This occurs when the OpenCL ICD loader cannot locate the AMD vendor JSON file. On Linux, ensure /etc/OpenCL/vendors/amd.icd exists and points to the correct libamdocl64.so. On Windows, reinstall Adrenalin with “OpenCL Support” enabled during custom setup.

Q: Can I mine different algorithms simultaneously across multiple AMD GPUs using one TeamRedMiner instance?A: No. TeamRedMiner enforces uniform algorithm selection across all specified devices. To run heterogeneous algorithms, launch separate instances with distinct --devices and --algo configurations.

Q: Is it safe to use --tune on production rigs with 24/7 operation?A: Yes. The tuning process executes only once at startup and writes optimized parameters to memory. It does not modify firmware or persist beyond the current session unless manually saved to a config file.

Q: What causes intermittent “Connection timed out” errors even with low network latency?A: These stem from pool-side rate limiting or firewall interference. Increase --retry-pause 10 and disable UPnP on consumer routers to prevent NAT table exhaustion during rapid reconnect attempts.