Asked and answered
Why should I use Crossbelt?
When we first starting mining back in 2017 we found getting started and operating at high efficiency very difficult. Dedicated mining operating systems like ethOS, HiveOS and SMOS simplified much of the initial work, but they fell short in the ease of configuration and rig optimization.
Much of a miner’s work went into constant tuning, bios changes, pool changes, file manipulation, manual calculations using online profit calculators, and scouring the internet for open source (and often outdated) data.
We also noticed that each mining OS created vendor lock in. If a user wanted to switch to a different OS they had to do so at their own peril and learn a whole new user interface.
We designed Crossbelt to allow miners to tune their rigs for the best performance and efficiency. With this ability, one can measure the effectiveness of a custom tune in real time ensuring maximum profit without having to lift a finger. We recognized that creating a cross platform tool was the only way create a consistent UI across all supported platforms and mining operating systems.
On another note, most cloud mining software is a central point of failure. They own your configuration data. So once you stop paying them you are forced to move to something else and start all over again. We at BlockOps didn’t think this was the right choice. We designed Crossbelt to allow the user access to their data even if they stop using Crossbelt. So if you want to transfer ownership of your configuation data or even share with others you have that ability. Furthermore, if we decided to cease development of Crossbelt you can still continue to use the program since you control the data. We wanted to make Crossbelt like an dapp that never dies.
Moreover, Crossbelt is a platform with a library for future expandability and data analysis that go far beyond just mining coins.
What the heck is crossbelt data for?
Crossbelt data is a git based data repository. It is a data repository that you control and decide where it is hosted. This repository contains all the data needed to assemble the configuration files, miner programs, wallets, and other information on every rig. The data repository is meant to be shared so miners can allow others access to their conigurations.
While this data is public there is nothing sensitive about the data. The only sensitive data is wallet addresses which are completly controlled by the user and will never exist in the BlockOps controlled crossbelt data repository (except for the examples).
Today this data is hosted in Gitlab.com but can really be hosted anywhere.
Crossbelt users can decide to fork the repository so they can control their own data and add custom entries of their own or contribute to the shared community repo.
Does Crossbelt replace EthOS, HiveOS, SMOS, …?
No. Crossbelt is designed to be an add-on tool for those systems. Furthermore, those systems often offer a UI or cloud based
interface to control the OS. The data you enter in the their dashboard locks you into their system. This prevents freedom of choice because you don’t own your data. Additionally, even if you did have your data would it be useful on a different management system? Probably not. We call that vendor lock in. We at BlockOps don’t want to control your data. We believe you should be in control.
- Crossbelt is platform you install on other platforms or adhoc operating systems you build that give you the same consistant interface across all of them.
- Crossbelt doesn’t own your data, you do.
- Crossbelt allows you to send your data anywhere you choose or nowhere at all.
- Crossbelt is a platform of tools that let you make choices, create scripts, or make cool plugins.
- Crossbelt will never disappear. It is a distributed app and there is nothing to turn off. While development may stop, the application will still live on.
Is Crossbelt an agent?
No. Currently Crossbelt does not run in the background. Although this may change in the future, furthermore a simple cron job can turn it into an agent if desired.
Crossbelt can and should be used to collect data for monitoring systems. We are currently working on some collectd plugins to dump data to rrd stores. If you come up with your own collectd plugins, please let us know.
Why is Crossbelt so magical?
Crossbelt seems so magical because it just discovers devices, mining software and allows compute unit configurations to travel from rig to rig.
Does Crossbelt support ASICS?
We plan to roll out support for ASICS in the near future since ASICS are becoming the bulk of mining rigs these days. Crossbelt was designed with ASIC support in mind, so adding support won’t be a problem.
Where can Crossbelt be used?
Crossbelt can be used on any supported x86_64 linux operating system with GPUs. While Crossbelt was intended for blockchain mining, there is no reason it could not be used for other purposes such as ML systems that also utilize GPUs.
Is Windows supported?
At this time windows is not supported. We don’t really like windows and there are many other options to choose from when it comes to miner managers.
Can I run this on EthOS?
Crossbelt has been tested thoroughly on EthOS, but you must be running version 1.3.1+
Can I run this on SMOS (Simple Mining)
Crossbelt was designed to run on Linux with kernel 4.15+ so there shouldn’t be any issue running on SMOS. However, we have not tested on SMOS. Please report any issues you find.
Can I run on HiveOS
Crossbelt was designed to run on Linux with kernel 4.15+ so there should not be any issue running on HiveOS. However, we have not tested on HiveOS. Please report any issues you find.
Can I run on
insert linux OS
Crossbelt was designed to run on Linux with kernel 4.15+ so as long as your version of Linux runs kernel 4.15+ we expect smooth operation. However, we have not tested on all kernel versions. Please report any issues you find.
Which GPUs are supported?
Any GPU that uses the amd-gpupro driver and any modern Nvidia GPU is supported. If you have a GPU that is not supported please let us know.
Is flgx driver supported?
No. We know there are still a few folks with these older GPUs around. We may add support if enough people ask, but this is a low priority and we don’t use any ourselves to test with.
Where does the GPU data come from?
We rely on the kernel sysfs files and nvidia-smi to collect our data. The GPU drivers and utilities provide all the data we could ever ask for. We do use a bit of OpenCL to correctly report AMD cards.
What is a Compute Unit?
A compute unit is generic term for a device that performs PoW. This can be a CPU, GPU, or Hashboard. At this time we are only showing GPU data but we expect to expand to hashboards on ASICS at a later date.
Are CPUs supported?
Crossbelt was designed to support any kind of compute unit. However, we disabled CPU detection because we thought they were not relevant anymore. If you would like to see CPUs re-enabled please let us know.
What about amdmeminfo?
Crossbelt provides almost all the same information that
amdmeminfo provides so there is no need to have
amdmeminfo installed. Currently, the only thing missing is the memory type, which we play to add as we get time.
Where can I get support?
Please see our documentation about how to get support. Paid support will always get you a quicker response.