Ethereum GPU mining?

[Mopetar]

Proof of work wouldn't be as bad if it didn't have everyone competing to complete the same problem at the same time. Redundancy isn't necessarily bad, but the work itself exists mainly for the sake of work. As long as the solutions are easy to verify there isn't any reason you couldn't hand out a different problem (or some set of problems with minimal overlap) to everyone in the network and let the next person to complete forge a block. Anyone who hasn't finished can keep working on their problem (assuming they didn't overlap with someone who just finished it) and forge a subsequent block when they complete their problem. It'd make transactions significantly less expensive or allow a similar rate at considerably less power.

The other idea would be to find some kind of PoW friendly problems that have some use outside work for the sake of work. Would anyone really be griping about BitCoin mining if all of the computations were essentially being used for cancer research or something similar?

 

[The Alias]

Proof of work wouldn't be as bad if it didn't have everyone competing to complete the same problem at the same time. Redundancy isn't necessarily bad, but the work itself exists mainly for the sake of work. As long as the solutions are easy to verify there isn't any reason you couldn't hand out a different problem (or some set of problems with minimal overlap) to everyone in the network and let the next person to complete forge a block. Anyone who hasn't finished can keep working on their problem (assuming they didn't overlap with someone who just finished it) and forge a subsequent block when they complete their problem. It'd make transactions significantly less expensive or allow a similar rate at considerably less power.

The other idea would be to find some kind of PoW friendly problems that have some use outside work for the sake of work. Would anyone really be griping about BitCoin mining if all of the computations were essentially being used for cancer research or something similar?

The security source in blockchain is the fact multiple entities verify the transactions.

 

[Leeea]

The security source in blockchain is the fact multiple entities verify the transactions.

But that does not require for computers to crack the same problem. Reverse verification is a computationally trivial for these proof of work blockchains. Any computer can do it nearly instantly.

ex: ( SHA512 ):
Text: "This is a bunch of stuff."
Number (Hash): 8ffa5045150bbdc16be0c530ab3665e718973b2b501100b61c0a5c6bdc4e0fc498da034813f97c812303deec3cf73bfb9edf02c789c479b3eb13ac7bb7fe3022

These block chains are basically just given the number, and forced to figure out the text. This is done by simply guessing random strings of text and seeing if it generates the correct number.

But if you have the text, it is trivial to generate the number to verify the correct text was found.

The computer that guesses the right text gets the privilege of making the next block in the chain. It usually takes multiple blocks in the chain to finalize a transaction. Since a different computer guesses the right text (or answer) every time, security and safety is provided by having unrelated* computers create the block chain and audit each other.


*when people lose interest in mining a block chain currency ( ex: bitcoin gold ), 51% attacks occur, where one organization/entity is able to make >51% of the guesses, allowing it to both make and audit malicious blocks as valid, and steal from the block chain.

 

[aleader]

While I am not familiar with shake pay, as long as they give you a permanent eth address to send to, I do not see why not .

In my experience it is just addresses. Eth is eth and it does not matter if it comes from mining or a wallet when sending to an address.

Unlucky for you . I am not going to tell you to hold because I never do. Part of me just does not believe in long term viability of proof of work cryptocurrency.

But an Eth crash would be so awesome right now! Well, not for you, but I have very little eth in my name and want to buy a new graphics card.

I *think* it's a permanent address. I copied it down this time. It doesn't change each time I login, so I assume it's permanent.

When I cash out tomorrow, I will have paid for my 5700XT in full already (in a month), so I'm not going to cry about it! Anything after tomorrow is bonus cash to me...to start paying down the 3070. It certainly seems to be in a bit of a freefall at the moment. Looks eerily similar to 2018, but it may be too soon to know.

 

[aleader]

Maybe dumb question, but I can't find an answer for it. If I switch my wallet in the .bat file in Phoenix Miner and lolminer, say hours before the payout from Ethermine, it will still transfer 100% of the unpaid to the new wallet address right? Or does it track how much was 'supposed' to go to the original address?

 

[ozzy702]

Maybe dumb question, but I can't find an answer for it. If I switch my wallet in the .bat file in Phoenix Miner and lolminer, say hours before the payout from Ethermine, it will still transfer 100% of the unpaid to the new wallet address right? Or does it track how much was 'supposed' to go to the original address?

Your unpaid balance from the first address will deposit to the old wallet address. The new address will start to accumulate an unpaid balance and be paid out when it make the threshold.

Etheremine doesn't link addresses, they're all separate.

 

[aleader]

Your unpaid balance from the first address will deposit to the old wallet address. The new address will start to accumulate an unpaid balance and be paid out when it make the threshold.

Etheremine doesn't link addresses, they're all separate.

Ok, thanks. So it does 'track' how much has accumulated to the current address then. I would have to time it exactly (switching addresses) then after the payout happens later tonight to make sure that nothing goes to the old address, right?

For instance, right now I have an unpaid balance of roughly 0.07 ETH in Ethermine, with my .bat files pointing to a MetaMask wallet address. This will be paid out to me in about 8 hours (once a week). If I were to shut the miners down right now and change the address to a different wallet, 0.07 would go to the MetaMask wallet, and whatever I accumulate between now and the payout tonight will go to the new address?

 

[MrTeal]

But that does not require for computers to crack the same problem. Reverse verification is a computationally trivial for these proof of work blockchains. Any computer can do it nearly instantly.

ex: ( SHA512 ):
Text: "This is a bunch of stuff."
Number (Hash): 8ffa5045150bbdc16be0c530ab3665e718973b2b501100b61c0a5c6bdc4e0fc498da034813f97c812303deec3cf73bfb9edf02c789c479b3eb13ac7bb7fe3022

These block chains are basically just given the number, and forced to figure out the text. This is done by simply guessing random strings of text and seeing if it generates the correct number.

But if you have the text, it is trivial to generate the number to verify the correct text was found.

The computer that guesses the right text gets the privilege of making the next block in the chain. It usually takes multiple blocks in the chain to finalize a transaction. Since a different computer guesses the right text (or answer) every time, security and safety is provided by having unrelated* computers create the block chain and audit each other.


*when people lose interest in mining a block chain currency ( ex: bitcoin gold ), 51% attacks occur, where one organization/entity is able to make >51% of the guesses, allowing it to both make and audit malicious blocks as valid, and steal from the block chain.

That's not actually how it works. They are not trying to solve for text at all. The mining algorithm takes a bunch of numbers (timestamp, merkle root hash, and nonce) and hashes them. The nonce is the number generated by the miner, it's guess if you will. If the hash of all of those is lower that the target difficulty, the block is valid and included in the block chain. All miners are trying to find a nonce that meets the difficulty criteria, but there isn't just one.

 

[ozzy702]

If I were to shut the miners down right now and change the address to a different wallet, 0.07 would go to the MetaMask wallet, and whatever I accumulate between now and the payout tonight will go to the new address?

Correct.

 

[Leeea]

That's not actually how it works. They are not trying to solve for text at all. The mining algorithm takes a bunch of numbers (timestamp, merkle root hash, and nonce) and hashes them. The nonce is the number generated by the miner, it's guess if you will. If the hash of all of those is lower that the target difficulty, the block is valid and included in the block chain. All miners are trying to find a nonce that meets the difficulty criteria, but there isn't just one.

I was close enough .

ps: I was aware it was not text, I used that to simplify the example. I am surprised though their algorithm is generating multiple nonce. I figured they would just reduce the bit count of the hash to change difficulty.

 

[aleader]

Has anyone heard about this? Is this something that affects the average person mining Eth?

Trading Bots Preying on Ethereum ‘Extracted’ $107M in 30 Days

https://www.coindesk.com/video/trading-bots-preying-on-ethereum-extracted-107m-in-30-days

 

[VirtualLarry]

Has anyone heard about this? Is this something that affects the average person mining Eth?

Trading Bots Preying on Ethereum ‘Extracted’ $107M in 30 Days

https://www.coindesk.com/video/trading-bots-preying-on-ethereum-extracted-107m-in-30-days

It affects mostly people trading ETH, including DeFI users. It's basically a "fundemental flaw" in how transaction-processing works in the Ethereum network.

Pending transactions sit in a globally-visible "mempool", and transactions are NOT processed FIFO, but rather, auction-style, with the highest bidder of transaction fees being included in a block "first".

So what these "bots" do, is spy on the mempool, and inject intermediate transactions based on user-submitted transactions, but they bid higher transaction fees, so these "bot transactions" skip to the head of the line. Very sneaky.

 

[Leeea]

Has anyone heard about this? Is this something that affects the average person mining Eth?

Trading Bots Preying on Ethereum ‘Extracted’ $107M in 30 Days

https://www.coindesk.com/video/trading-bots-preying-on-ethereum-extracted-107m-in-30-days

Yes, it has effectively pushed transaction fees up to $40* for a time. Since there is no way for even a miner to avoid paying the transaction fee at least once, that is a bit negative for the profits also.

But the transaction fees are also what makes mining profitable right now.

Bad news for eth long term I suspect. Turning into another bitcoin, hold able but not really usable.


The interesting thing about eth is it is not like Bitcoin, the developers can just modify it if they want to in order to fix issues. However, this seems to be some sort of oddity in Eth's smart contracts, so perhaps it will not be easy to fix.

*crosses fingers and hopes this will result in a crypto crash

 

[aleader]

Yes, it has effectively pushed transaction fees up to $40* for a time. Since there is no way for even a miner to avoid paying the transaction fee at least once, that is a bit negative for the profits also.

But the transaction fees are also what makes mining profitable right now.

Bad news for eth long term I suspect. Turning into another bitcoin, hold able but not really usable.

The interesting thing about eth is it is not like Bitcoin, the developers can just modify it if they want to in order to fix issues. However, this seems to be some sort of oddity in Eth's smart contracts, so perhaps it will not be easy to fix.

*crosses fingers and hopes this will result in a crypto crash

I'm still trying to figure out the way fees work with this stuff. Actually, I'm not trying that hard...I'm just trying to avoid paying them I just cashed out again yesterday except this time when I transferred from Metamask to Shakepay the 'slow' fee was only $4.55 CAD. Last time it was almost $11. My e-transfer fee was free too, and last time I think it was $2. How did it get cheaper? The amount I transferred this time was double what it was last time. The price shot up at just the right time last night too.

Yah, unless you're cashing it out, I think people are taking a pretty big gamble accumulating this stuff for years hoping that it's one day going to be accepted like fiat. All governments need to do is ban it and your holdings are worthless...vested interests and all.

Also, when I switched wallets Ethermine (in the app) now shows a payout date of 11 days from now, instead of the usual 7? I thought they had a policy that 7 days was automatic? That's what it's been up until now...unless the app is suddenly glitched.

 

[Hitman928]

I'm still trying to figure out the way fees work with this stuff. Actually, I'm not trying that hard...I'm just trying to avoid paying them I just cashed out again yesterday except this time when I transferred from Metamask to Shakepay the 'slow' fee was only $4.55 CAD. Last time it was almost $11. My e-transfer fee was free too, and last time I think it was $2. How did it get cheaper? The amount I transferred this time was double what it was last time. The price shot up at just the right time last night too.

Yah, unless you're cashing it out, I think people are taking a pretty big gamble accumulating this stuff for years hoping that it's one day going to be accepted like fiat. All governments need to do is ban it and your holdings are worthless...vested interests and all.

Also, when I switched wallets Ethermine (in the app) now shows a payout date of 11 days from now, instead of the usual 7? I thought they had a policy that 7 days was automatic? That's what it's been up until now...unless the app is suddenly glitched.

There is an essentially a supply and demand component when it comes to Eth transfer fees. The more people want to move eth and do so within a quick time frame, the higher the demand and the higher the fees. This especially sucks if you put your transfer in with a lower fee and then there's a fee increase at the same time. Your transfer can just sit there waiting for a while until demand comes back down and your fee is accepted.

 

[aleader]

I'm wondering if the wallet address I changed to is 'working'. When I click on the address at the top of the Ethermine dashboard it takes me to the screen where it shows the Eth 'balance'. Is this updated dynamically (i.e. shows the unpaid balance in real-time) or does it only show the balance once the payout happens?

It's sitting at $0 right now and I've been mining to it all night. I've never looked at that page before. How would I know if it doesn't like the address? What's really throwing me is that the payout date is actually getting longer now when I check. It was 11 days (which makes no sense) and now it's over 12 days, even though my daily estimated is actually more than it was at the other address. I am showing an unpaid balance that makes sense.

 

[Hitman928]

I'm wondering if the wallet address I changed to is 'working'. When I click on the address at the top of the Ethermine dashboard it takes me to the screen where it shows the Eth 'balance'. Is this updated dynamically (i.e. shows the unpaid balance in real-time) or does it only show the balance once the payout happens?

It's sitting at $0 right now and I've been mining to it all night. I've never looked at that page before. How would I know if it doesn't like the address? What's really throwing me is that the payout date is actually getting longer now when I check. It was 11 days (which makes no sense) and now it's over 12 days, even though my daily estimated is actually more than it was at the other address. I am showing an unpaid balance that makes sense.

Check your settings and see what your payout threshold is.

 

[aleader]

Check your settings and see what your payout threshold is.

It's 0.1. It's the lowest that it can be set at now. Maybe that's what it's estimating now? Those rules changed mid-payout for me. If I'm showing an unpaid balance though, it means that I'm getting credit for the shares doesn't it?

Should this balance be zero:

 

[Hitman928]

It's 0.1. It's the lowest that it can be set at now. Maybe that's what it's estimating now? Those rules changed mid-payout for me. If I'm showing an unpaid balance though, it means that I'm getting credit for the shares doesn't it?

Should this balance be zero:

View attachment 40222

Yes, if your unpaid balance is accruing, you should be fine. If your unpaid balance stops increasing (if you're still mining) or decreases without a payment to you wallet, that's when you should be concerned. What is your unpaid balance at now? Is it over the threshold and if so, for how long has it been over?

 

[aleader]

Yes, if your unpaid balance is accruing, you should be fine. If your unpaid balance stops increasing (if you're still mining) or decreases without a payment to you wallet, that's when you should be concerned. What is your unpaid balance at now? Is it over the threshold and if so, for how long has it been over?

Ok, thanks. My unpaid balance is increasing like it should, pretty consistently. I'm at about 0.005 right now. I was just paid out last night (roughly 0.07 Eth) as it has been paying out every 7 days (if your balance is over 0.05, as per their policy). Right after that I changed the wallet address so it goes directly to Shakepay now.

 

[aleader]

Will they soon nerf our other Nvidia cards too? It certainly does hurt those of us mining just to pay our cards off, but it maybe will prevent the bot-snatching of 30 series cards that seem to be going to every miner on youtube...who proudly displays (rubs it in the faces of gamers) every single card in a tedious unboxing ceremony:

https://www.pcgamer.com/nvidia-hints-that-further-cryptocurrency-mining-blocks-could-be-incoming/

 

[VirtualLarry]

Will they soon nerf our other Nvidia cards too?

Yes, they will. At least, going forward. The 3060 ti, 3070, 3080, and 3090 are getting their PCI IDs retired, and the "new edition, anti-mining" will be getting new PCI IDs.

It remains to be seen, if the "VBIOS flash" necessary for NVidia Ampere cards to utilize "Resizable-BAR support", will also include a cryptocurrency hashing limited embedded into it, and a mysteriously-vague EULA included with the flashing, that will purport to "make it OK" (for Nvidia, so that they don't get sued).

 

[aleader]

Yes, they will. At least, going forward. The 3060 ti, 3070, 3080, and 3090 are getting their PCI IDs retired, and the "new edition, anti-mining" will be getting new PCI IDs.

It remains to be seen, if the "VBIOS flash" necessary for NVidia Ampere cards to utilize "Resizable-BAR support", will also include a cryptocurrency hashing limited embedded into it, and a mysteriously-vague EULA included with the flashing, that will purport to "make it OK" (for Nvidia, so that they don't get sued).

I guess we can avoid it by just not updating. You're an old vet at his, have you ever had to change thermal pads on your mining cards? Ever had a card fail because of the memory overclocks?

The Son of a Tech guy recommends changing thermal pads on the 3070 for example every 3 to 6 months. I don't want to have to do that as he says the pads are 2mm (he measures with calipers), and I can't find a 2mm thermal pad anywhere online.

 

[MrTeal]

I guess we can avoid it by just not updating. You're an old vet at his, have you ever had to change thermal pads on your mining cards? Ever had a card fail because of the memory overclocks?

The Son of a Tech guy recommends changing thermal pads on the 3070 for example every 3 to 6 months. I don't want to have to do that as he says the pads are 2mm (he measures with calipers), and I can't find a 2mm thermal pad anywhere online.

Fans, Thermal Management | Thermal - Pads, Sheets | DigiKey

Fans, Thermal Management - Thermal - Pads, Sheets are in stock at Digikey. Order Now! Fans, Thermal Management ship same day

www.digikey.ca

 

[ozzy702]

I guess we can avoid it by just not updating. You're an old vet at his, have you ever had to change thermal pads on your mining cards? Ever had a card fail because of the memory overclocks?

The Son of a Tech guy recommends changing thermal pads on the 3070 for example every 3 to 6 months. I don't want to have to do that as he says the pads are 2mm (he measures with calipers), and I can't find a 2mm thermal pad anywhere online.

No clue on what the 3070s will need, but I have 480s still chugging away from 2016... and all I did was change the thermal paste a year ago. I've never messed with thermal pads.