Could we get a block mined on stream? Is there an issue with keeping the stream up for a few more hours?
They were trying to have everything live to show a few blocks being mined, but the phone trying to do the stream kept crashing. I believe they may try it again once they get a more stable method.
And no, the speed up will not be 20x or 30x, scroll up through this thread and check the data, it will likely be less than 10x
Ok awesome thank you for your response I appreciate it. I am looking forward to seeing more data when it is published for the tests.
Regarding the live stream attempt, my apologies, my phone kept crapping out and I don’t think you guys could see the print on the screen. At least you got to see Tim’s mining software running a bit. Believe it or not, just 10 seconds after we stopped recording the miner validated a block! Here is the picture: Microsoft OneDrive - Access files anywhere. Create docs with free Office Online.. We had a big laugh when that happened… Right, who’s going to believe that? So if you got time to burn, watch it here: Zeropond Zcash testnet validation, Live Stream - YouTube
In any case, the real test is when Tim runs through the ‘Austin-Williams’ test.
There is only 3 clips worth showing, I removed the others that had no content. Go easy on me, this was my first live stream.
… That’s a non-starter.
Thanks to @farallon for taking time to do this!
Blocks 6771 and 6818 last night appear to have been legitimately solved in 10 and 8 seconds (block times before and after were about 150 seconds). These are the first in beta with these characteristics.
Please keep us informed of total time mining and total blocks acquired.
I think I made an error above: you should have only 2.9% blocks of your blocks below 30 seconds when D/10 = 8 and you’re at 1.2 H/s. This is 2.9% of 168 x 1.2/8, not the actual blocks on an D/7 actual network hash rate. Which is about 1 block instead of 3 on 168 x 1.2/8. But there were about 4-too-many < 30 s blocks that night. So I’m not convinced my previous math is wrong.
Anyway, this makes your network-measured hash rate (18-1)/168 x D/7 = 1.24 H/s. D=86 that evening.
Zeropond needs to delay a day to get everything ready. We’ve been working nonstop to get this live but we are on a very aggressive schedule and need one more day before launching sales. We will open sales at 17:00 US Pacific Daylight time (UTC-7) on Thursday (tomorrow)
We performed the Austin Williams testcase for the miner. At the moment we are swamped trying to start our business so we deviated/improved a little bit from the original idea by Austin.
-
Set “s” equal to BitcoinBlock0000000000000000008647f2bb1a5e8defacfbe3e3d6507c1c2115e84930efa6blockHeight430842ZeropondProof. We used bitcoin block 430842 (https://blockchain.info/nl/block-height/430842) here as proof of starting point. We stopped the miner at around bitcoin block 430895 which gives a total running time of around 7.2 hours.
-
Instead of using H( I || V || xi ) (I: blockheader, V: nonce, xi: solution index) for the hashing function, we now use H( s || V || xi ). Instead of computing the difficulty as H( I || V || x1 || x2 || . . . || x2^k ), we now compute it as H( s || V || x1 || x2 || . . . || x2^k ).
-
We are not convinced that giving the maximum achievable difficulty is a good measure for our hashrate. Instead we dediced to give you the complete output of the miner on this block. The file is 166mb and contains all the output for all nonces during the time which the miner ran. The file is structured as follows:
Nonce BlockHash Difficulty Solution (Truncated and separated by commas) -
Together with “s” you can check the validity of all data in the file. We figured that the difficulty of our miner would be somewhere around 9.5 for it to find a block every 2.5 minutes. You can find this number by plotting the data. If you want an easy way to check this you could execute
awk ' $3 > 9.5 ' ./blocksmined | wc -l
. This would give you the number of blocks the miner would have found if the difficulty were 9.5. At this difficulty, the miner would have mined around 173 blocks (mined 175), which is consistent with the time the miner actually ran. If we use @zawy’s formula, you would see that the miner did a bit better compared to his formula. However, we are skeptical about the number “7” in his formulas. The error bars on this number would in our opinion be at least 3, so it could be either 4 or it could be 11. A better measurement is nessasary from his side to support his formula.
A TLDR on the results for us that don’t want to read through an entire nonce file?
If you’re willing to take someone else’s word for it, why all the fuss and jumping through hoops to prove legitimacy?
It’s best for a 3rd party like @zawy to do that.
I agree with (3) and was thinking about that over the last few days.
Good stuff here.
@tim_olson, in the event that you sell a certain amount of hashrate but are unable to scale to that hashrate in time for the launch, do you commit to refunding the affected customers their complete purchase price immediately (within a day or two of the launch)?
We are capping sales to something we are comfortable installing before Genesis Block. If we are ahead of schedule, we might reopen sales, but no promises.
I deeply regret to announce that we have decided to block USA residents from being customers of Zeropond. Although our mission is to make Zcash mining available to everyone, after considering the council of our talented and experienced attorneys at Perkins Coie, we feel forced to exclude residents of the US from this offering. Accredited US investors may still apply for contracts by emailing info@zeropond.com.
It’s truly a shame that everywhere in the world is free to be involved in the cryptocurrency revolution, except this country which purports to be egalitarian and capitalist. In fact the opposite is true here.
We will work hard with our attorneys to craft a business model which is compatible with US Law, so we can offer this product in the future to crypto enthusiasts here in our homeland.
Any plans to start a new business somewhere that the local mafia keep their stink to themselves?
tim: 5pm pacific is in the middle of the night in Europa and Asia. Not a good time to start the sale.
Can you please tell me, are you selling GPU hashing power or CPU hashing power? The homepage announces CPU-equivalents which might be confusing.
Also, what about offering options for hashrate resellers outside the US. They could sell hashes to the US residents
They’re selling hashes. So long as they deliver, why would it matter how they’re generated?