https://t.co/NJlqU42pR3@DavidVorick and his Grin ASIC 🔨
ASIC = skin in the game = long term interest alignment = chain security = fair competition of power/ingenuity = path to ultimate decentralization
Wow, look at all the hedging in order for Obelisk to avoid having to deliver on time. I would stay as far away from their miner presales....excuse me...voucher exchangeable for a miner presale as possible.
if they enable me to buy a voucher with multiple coupons that would be great...so far I can only use one coupon for each voucher, so I'll likely sit this one out, especially considering their track record of delivering behind schedule. also, this feels like a desperate attempt to raise cash, possibly in an attempt to fund their defense of multiple lawsuits.
What a shit show, I wish sia never made asics and just focused on peer to peer incentivized file sharing, they might actually have something to show for the last two years of Sia's existence. like the dream always was. Oh well, this is what a failing project looks like I suppose.
Anyone thinking of purchasing should rethink their decision.
Nebulous does not own the Grin blockchain, so if these miners are squashed by innosilicon and Bitmain like the sc1 were, then there is no way to fork away the competition. That should make everyone very scared that they end up buying bricks.
Be careful with your money guys.
This is the main concern. Obelisk has never built a competitive miner in terms of specs. If this miner is 28nm too which I wouldn't put past Obelisk then all hope is lost. They seem to think that they can compete against people with an order of magnitude more money, even going as far to claim they could make a competitive SHA256 asic. Puhleeze
What would particularly worry me is that Bitmain and Inno beat their past miners to market. There's a ton of VC money apparently interested in Grin, so isn't it entirely likely Bitmain and/or Inno are looking at making Grin ASICs too? The ASICs that earn the most are the ones released first.
Your absolutely right, they will be beaten to market and will be absolutely spanked by Innosilicon.
There's no way to hardfork the chains algo like they did to Siacoin to keep their sc1 profitable.
So unless anyone feels like having a $6000 useless brick in their display cabinet, then I would heavily advice against purchasing this box of empty promises.
Yes of course. I'm merely stating that the siacoin developers had made an obelisk asic for sia that got beaten to market by bitmain with their a3 asic.
The Sia devs solution was to Hard fork the chain to a new algo that only their asics could mine on. Bricking bitmains asics...
I'm worried they'll be beaten to market again by innosilicon or someone else who can make asics for grin that are 10x better than the obelisk being advertised now, and this time because they are not in control of the mining algorithm, the Grin devs are, I believe they could sell off these asics and they could effectively be worthless at mining because competitors are much better manufacturers.
That's because you're mining on the 8GB GPU compatible Cuckaroo29 algo.
This ASIC is for mining on the ASIC/11GB+ GPU compatible Cuckatoo31 algo.
If you configure it to do so, then yes. But then it wouldn't compete with this ASIC anyway, and their hashrate comparison wouldn't make sense anymore?
I'm mining with a 2080ti on cuckatoo31 right now, getting 1.7 gps.
I believe that's because (as far as I know) ASICs don't exist yet?
"Eventually we expect C31 ASICs to show up and tip the balance in hashpower."
More details from the algo dev: https://www.grin-forum.org/t/what-to-mine-choosing-between-cuckatoo31-and-cuckaroo29/1732
With a 2080ti you can mine on both c29 (gpu) and c31 (gpu/asic).
They're comparing them because 2080ti cards are being used to mine on the ASIC algo right now.
When these ASICs come online, that won't be economical anymore.
"This can be compared to the current leading graphics card, the NVIDIA RTX 2080 Ti, which performs less than 2 graphs per second.."
if they can't even get that right, why would you trust them with thousands of dollars and making a chip 3 times as complex as a S9, or believe their wild projections of GPS,... we can all see what a 2080ti is getting, and what a 1070 is getting.
anyone who's mined at all would know those numbers are bullshit, or false facts placed to create bias for their yet-to-be product.
It's actually completely correct, not sure what you're talking about?
Remember this is on the ASIC algo Cuckatoo31 where hash rates are lower. I'm mining with a 2080ti right now and getting exactly the same GPS as mentioned here: