We’re amped to announce the next phase of our Beta Rewards Program, starting on 1st October and running until the Live Network launch on 29th October!
What’s the Goal?
During this phase, we’ll conduct full-scale public tests to validate our internal lab results. We’re focused on refining and optimizing the size of both nodes and chunks to maximize Network capacity and performance. By running these tests, we aim to:
Enable users to make the most of their spare resources (HDD, CPU, RAM, and bandwidth).
Support a wide range of devices for maximum global participation and security, with an eye on future mobile support.
Maximize the potential and capacity of the Network without sacrificing performance or security.
We’ll be using the data gathered from these tests to help us fine-tune the system and find that “sweet-spot” for Network efficiency. Once we wrap up, we’ll provide another update where the community can get hands-on with the ERC20 tokenbefore the big launch!
Incentives & Rewards
To make this phase even more exciting, we’re allocating 500k tokens as rewards over the 4-week period! Here’s how it works:
A single leaderboard will track participants based on Nanos earned—no more individual waves or complicated rankings!
It’s easy to get involved: Join the Discord, spin up some nodes, add your username, and start stacking Nanos. You’ll be earning your share of the 500k token reward pool in no time!
What’s Coming Up?
The current waves and competition will end on Friday 27th September.
On Tuesday October 1st, we’ll launch a fresh network starting with the same chunk size, but nodes with 32GB of storage capacity (+3GB for logs). This will give us crucial data for fine-tuning node sizes.
Each test phase will last one week, and we’ll roll out new tests each Tuesday after that.
As always, your Nanos earned will be tied to your Discord username, and you’ll be able to track your progress using our bot.
So please do dive into this next phase and help shape the future of the Autonomi Network. Thank you all for your ongoing support and participation—it’s because of you that we’re here!
Fantastic! Will be great to see how 32gb nodes affect things, particularly for home nodes where routers handling connections can be a bigger limiting factor than storage.
Will there be a G**gle Docs form to watch for?
Will preview of ERC20 be on some EVM/L2 testnet with faucet for free tokens, or they will have to be earned/farmed?
I know this is just testing, but wanted to re-iterate this
Mobile phones will definitely have trouble with this if you want large scale use of them. Many phones still have 32GB or less and a certain amount is needed for OS and operations (camera for one, and other apps on phone)
Start with the least favourable size (for mobile and small devices) and bandwidth on startup and see how the network fairs, then reduce, test again and see if smaller is better.
We know it’s at the extreme end (without just getting silly) but we don’t anticipate it ending up there. We’ve already been at the other extreme: 2GB; this allows is to validate the observations at two ends of the spectum from the lab, and plot the point to the middle, the sweet spot.
Internal tests continue right through as well, including the chunk size mods. 21 tests, running for 6 hours in various conditions, then narrowing again and repeating to dial in. Validation from public results in paralell, zoning in on our winning combo.
Yeah, we know that, and as I say, 32GB is at the upper bounds of what we think is reasonable (we are obviously not just setting it at 1TB and giving it a go).
But at the same time, it’s a fine balance. We want to be able to hoover up as much of the spare storage capacity of devices as possible, while staying within the limits of other resources, so performance and security isn’t impacted.
And we know that for the furture a large untapped seam of resource will be in mobile — with smaller capacity, patchy connectivity, and different usage patterns (e.g. only running nodes while charging). We know it’s more of a medium to long term goal, but chnaging node size after the fact is tricker, so we are seeing what we can do up front, and in the knowlege that the lifecycle of mobile devices is pretty rapid, and storgae is increasing apace.
But, just to reiterate, 32GB is what we see as the chunkier end of the specturm.
Are you sure? For 2024, there are 343 new phone models, of which 0 are with 16 GB of memory, 1 with 32 GB, 11 with 64 GB and 331 with 128 GB and above.
Realistically how many 32gb nodes could support the number of connections for 1 sim card? In my tests with home 5G internet, the mobile operator has a traffic limit per day (fair use) even with the unlimited plans…
So I doubt even 1 node can run unlimited 24/7 on mobile phone in coming years…
Mobile phones are still going to struggle for battery too, imo. By the time we hit mainstream, i suspect both storage and batteries/power usage will be a better fit.
Tbh, it would be great to give the domestic routers a chance for home host first, imo.
All of the above are true. I believe in design for tomorrow, not today. so disk space is one area I see we set it quite high.
The main issue though is overall simplicity and by definition that should mean power use efficiency as well. This simplicity is paramount and I keep thinking
Cellular autometa
Computational irreducibility (makes design super simple but an Engineers nightmare as simulations and math models are useless. However the number of arguments why the game of life for instance cannot work, is insane).
These 2 areas are critical in my thinking and my drive for simplicity. It makes design very hard, but the outcomes should be mind blowing
Mobile phones, as Jim said, will only be useful when charging due to power draw.
Also, when charging they could be docked with a storage drive with plenty of space for node storage, so 32gb nodes would in no way be an issue for a sensible mobile phone based node setup in my view (also, many phones now have 256/512/1tb storage options, and that will only grow over the years).
Phones are powerful enough to run nodes, but I hope optimisation will be geared towards the kinds of devices that will offer the best performance for the network, i.e. always on computers with good Internet.
The mobile bandwidth price it too high now. So the mobile node is not an efficient node from an economic perspective. The team is focusing on real network, but community is still dreaming of a theoretical and ideal network. Let’s get this network ready to become a reality.
Disrupt a lot the world is not a just funny and enjoyable thing.
At the moment — in the medium term — we are kinda thinking of mobile as another home device you might want to set to work when it’s not in use, so for example when you put it on charge at night, and it’s connected to the home wifi.
So we need to think about:
Storage size
Home routers
Also regular joining/leaving the network at on/off peak times
Global timezone patterns too
etc
Then maybe further down the line as mobile connectivity continues to improve we could be looking at regular background use.
Was uptime not added as an incentive to keep nodes running, (I believe it is now removed)
But the idea of almost encouraging regular switching on and off of nodes like this seems counterproductive to me.
Not saying your statement is encouraging it Jim, but mobile nodes as a whole will.
Is it really practical/useful for small devices to come on for 7 hours at a time in every 24 hour period.
Incentives & Rewards
What’s Coming Up?
