Critical PoW vulnerability closed;
the accidental birth of a new PoW Mining

https://forum.grin.mw/t/critical-pow-vulnerability-closed-the-accidental-birth-of-a-new-pow

Sing with us, exchanges:

No http, no cry
No http, no cry!

I introduced fidelity in this forum post https://forum.grin.mw/t/request-for-all-miner-devs

It's basically the fraction of cycles that your solver finds, and is ideally 1, but in practice more like 0.999

The Network Upgrade to 4.0 proceeded without a hiccup, and everything is running smoothly...

The ASIC friendly Cuckatoo32+ gets nearly 78% of rewards,
with remaining 22% going to ASIC resistant Cuckarooz29.

No ASICs are known to exist for C32+ though...

(there is negligible C31 mining as it's heavily punished by its phaseout)

Not for Cuckarooz29, but it works mighty fine for Cuckatoo32...

lolMiner for Cuckatoo32: https://github.com/Lolliedieb/lolMiner-releases

Less than half an hour to go!

https://github.com/mimblewimble/grin-miner now has a CUDA Cuckarooz29 plugin with source code included.

Hard fork in a few hours!

https://github.com/mimblewimble/grin-miner includes a CUDA Cuckarooz29 plugin with source code.

If all goes well, relative height kernels [1] will be enabled on mainnet.
There's possible interest in allowing duplicate outputs.
These are not intrinsically useful, but forbidding them can cause problem,
such as when a channel settlement tx is blocked because a malicious party has
already created an instance of their payout output.
And possible interest in mandating expiration times on transactions.
An interesting non-breaking change is PIBD, or Parallel Initial Block/Bytes Download,
which will speed up syncing and make it more robust.

[1] https://github.com/mimblewimble/grin-rfcs/blob/master/text/0013-nrd-kernels.md

Grin's 3rd HardFork / Network Upgrade at height 3 * HARD_FORK_INTERVAL = 786240 expected on July 16.

See Grin 4.0 announcement at https://forum.grin.mw/t/grin-grin-wallet-4-0-0-released

Only one more preplanned HardFork / Network Upgrade after that, mid January 2021.

Cuckatoo31 is no longer profitable to mine, and should be replaced by Cuckatoo32.

It shows gaps 2,2,4,2,4,2,4,6,2,6,4,2,4,6,6,2,6,4,2,2.
That last gap ending back at the top is wrong. It should be a 6-gap (from 73 to 79), not a 2-gap.
In fact it's obvious that you can only get consecutive 2-gaps at the start.
You can argue that it was a 6-gap that got truncated by reaching the top, but that would be rather obscure and misleading. It would be better to remove it to indicate the discontinuity.
Or you can argue that it was wrong on purpose just to make it look symmetric at the top. Which is not how we do mathematics.

Grin is now 3/4 ASIC Friendly, with Cuckatoo31+ getting 75% of the rewards.
In next month's 3/4 Hard Fork (3rd out of 4 pre-planned ones every 6 months),
ASIC Resistant Cuckoom29 will be replaced by Cuckarooz29, which will run until HF4.
In August we'll see the last of Cuckatoo31, as its graph weight drops to 0,
leaving Cuckatoo32+ on the ASIC Friendly side.

Haskell would be great for concisely expressing the consensus model, e.g. a function

validWork :: [Block] -> Either String Int

that either returns an error message in case of an invalid history, or the cumulative difficulty in case of a valid history,
written for maximum clarity rather than maximal performance,
could serve as a formal specification of the consensus model.

Sorry for the VERY late reply...

0.5% is arbitrary. We chose to make the emission non-arbitrary. We chose to make it the simplest possible, at 1 grin per second forever.


It start out identical to Bitcoin's inflation rate, for the first 4 years. After that, it just falls much more slowly.


IMO it really doesn't matter. As you noticed above, coins inevitably get lost. At some point the yearly emission roughly balances out against yearly lost coins, defining a softcap on supply, as opposed to Bitcoin's hardcap. This guarantees that the subsidy always provides enough security for the spendable supply. Also see this Hacker News discussion https://news.ycombinator.com/item?id=23147005


There's precious few innovative coins that rely entirely on volunteers and donations for development.
I can't think of too many besides Bitcoin, Monero, and Grin.


IMO the main reasons Grin is a worse store of value are 1) its perpetual emission (i.e. continued dilution), and 2) the fact that the supply is not fully (i.e. transparently) auditable.
These are also the reasons why Grin is not a direct competitor of Bitcoin.


Only 4 hardforks were (pre)planned, at regular 6-months intervals in the first 2 years. The 3rd one is happening next month. After that, consensus model changes should be increasingly rare.


Scriptless scripts (i.e. exploiting the flexibility of Schnorr signatures) allow for multisigs, thresholds sigs, atomic swaps, discreet log contract, and payment channels.

Under what handle did you contribute to Grin?
I never heard of you...

After bullshitting [1] their way out of an earlier scam [2].
[3] shows one of the 48 devfee coinbases that is supposedly unspendable until Sep 29, 2020,
but a full node query [4] on the UTXO set shows that all of the devfees (total of 1M XBG) are in fact spent.
The line of code they point to in [1] is a red herring. The actual code used for maturity verification is at [5] which shows all coinbases treated uniformly with a 1 day maturity.

[1] https://bitgrin.substack.com/p/the-bitgrinio-takeover

[2] https://web.archive.org/web/20190727183718/http://bitgrin.io/

[3] https://explorer.bitgrin.dev/block/64c07b8ec36724ba537523f4adbdaa561c6bf56b1a5bbc0c2c39bfa80e6f49f4

[4] http://localhost:8513/v1/txhashset/outputs?start_index=1 on any machine running a full node with the API enabled

[5] https://github.com/bitgrin/bitgrin/blob/master/chain/src/txhashset/utxo_view.rs#L130

Why is Cuckatoo32 missing?