Topic: [ANN] Grin | PoW Mining | Electronic transactions for all. Community driven. - page 10. (Read 73510 times)

We are ready for the upcoming hardfork.


GRIN Mining Pool

Grin v5.0.0 network upgrade (Hard Fork 4) will happen at block height #1048320 (Approximately January, 15)

What will change with the new Hardfork?

1. Only slatepack addresses will be used. http(s) addresses are now disabled.
2. C29 algorithm will be terminated. Only Cuckatoo algorithm (C32) family remains.

https://grinnews.substack.com/p/129-the-final-scheduled-fork-

Hard Fork Incoming!

Node changes:
Improve transaction fee calculation, as outlined in https://github.com/mimblewimble/grin-rfcs/blob/master/text/0017-fix-fees.md
Improve difficulty adjustment algorithm, as outlined in https://github.com/mimblewimble/grin-rfcs/blob/master/text/0018-fix-daa.md
(Partial) Parallel IBD: Preparation for support of an improved process for initial sync and download of blockchain data, as outlined in https://github.com/mimblewimble/grin-rfcs/blob/master/text/0020-pibd-messages.md

Wallet changes:
Late locking of outputs: Removing the need for outputs to be locked during transaction building before the final stage.
Remove http listener for external addresses: As per the deprecation announcement with v4.0.0, remove support for unsecured http listeners in favour of the slatepack 43 standard.

Countdowns:
https://www.nicehash.com/countdown/GRIN-forking-2021-01-17-12:00
https://grinnode.live/

I wrote 3 short articles that try to explain Grin without assuming any knowledge in cryptography. One article is about Grin's emission model. Here's the link to the post on hackernews https://news.ycombinator.com/item?id=25598571
Let me know if some things are not clear and I'll try to explain them here

Im interest to see grin coin movements. Not yet get point or reason why GRIN coin went for infinite- linear.
Could anyone want to share what’s purpose for that things?
Been analyzing before, the prices valued on bitcoin is lowest currently, so have look to buy now.

Well my order got filled within a few hours so I'd put your money where your mouth is if you're serious  

I guess they want to pretend every PoW is a hashcash, or are too lazy to properly account for different PoW.


No; a 1080Ti does only 0.46 gps on Cuckatoo32, according to https://whattomine.com/

I still don't understand why those calculator saying hashrate in h/s instead of GPS.

So do you think if I just put them same as what GPS speed they are all the same?

Incase if that miner is running at 1.4gps well the speed is much lower than gtx 1080ti. I mean gtx 1080ti is way more faster than this miner.

I'll buy a day of Grin at that price....

(funny that a whole day of Grin now costs less than 1 BTC...)

I'm looking to buy grin by the 1000s @ 1k sats each. The price is subject to change of course.

I'm looking for at least 5000 maybe 10,000. DM me if interested. You send first or we can use a trusted forum escrow.

1.4 graphs per second (gps) is the graph rate, not the hashrate.

Grin's Pow is not like most others that just compute a hash function (hashcash).
Instead it's looking for a 42-cycle in a random graph.

See https://cryptorials.io/beyond-hashcash-proof-work-theres-mining-hashing
for more explanation.

Once a 42-cycle is found, Grin does compute a final cyclehash to see if the solution difficulty is high enough. But on average only 1 out of every 42 Cuckatoo graphs has a 42-cycle. Assuming that this ASIC finds every 42-cycle that exists (fidelity 1), the solution rate of this ASIC is 1.4/42 = 1/30 sol/s, or two solutions per minute. Since Grin has a one minute blocktime, this miner gives you on average two chances to win each block. That's quite different from a bitcoin miner like the S19pro, that gives you 600s * 110Th/s = 66,000,000,000,000,000 chances to win each block!

Note that for Equihash performance, people often specify the solution rate.
Since a single Equihash instance has on average 1.89 solutions, this makes perhaps more sense than using the instance rate. Also, there is not a good name for a single Equihash instance. You'd end up using Equihashes per second.

But for Cuckatoo, with solutions being so rare, it's much easier to quantify the graphrate, with a graph being a single Cuckatoo problem instance.

Btw, a C32 graph is itself described by 2^32 * 2  = 8,589,934,592 siphashes, but an ASIC could avoid computing some fraction of these, and could be computing others several times, so siphash rate is not a very suitable performance measure either.

In summary: hashrate is for hashcash PoW. For non-hashcash PoW like Grin's Cuckatoo or ZCash's Equihash, other measures like instance rate or solution rate are appropriate.

What is the specs of this miner?

I found on google but it shows that the hashrate is around 1.4Gps can you convert GPS in hashrate? Just want to know the exact hashrate of this for calculation.

Well well, now we see the reason for the pump.

https://grinnews.substack.com/p/128-launch-of-the-first-grin-asic

https://john-tromp.medium.com/a-case-for-using-soft-total-supply-1169a188d153

https://grinnews.substack.com/p/127-winter-bug-bash-challenge-

https://grinnews.substack.com/p/126-more-grin-g1-mini-asic-details

https://grinnews.substack.com/p/125-v500-beta-released-

https://twitter.com/ipolloglobal/status/1333254924071694336?s=21

Grin ASic official.

The attacker amassed more than 50% of total graphrate to perform a reorg.
No network rules were changed. What changed (drastically) was the most-worked branch.


Yes. the longest chain rule worked as intended.

What makes it a 51% attack if 51% of the network decides to change the network rules?

Isn't it working as intended?