Ethash and PPLNS questions

[Beowulf]

I've done what I think is due diligence hunting through the site help, forums and FAQ and couldn't find answers to the following questions regarding the Ethash algorithm and PPLNS on Prohashing:

1. What is the estimated time or share "window" for PPLNS for Ethash; i.e. when I should expect my hashrate vs. pool hashrate ratio to match my proportional payout? I've been at it for over 12 hours now and have barely hit 25% of where I should be, so just confirming this is expected (i.e. about a 48-hour PPLNS window, which seems quite high).

2. Does PPLNS share history span blocks up to a fixed window of X shares or are the PPLNS stats/ratio reset when a block is found (i.e. more like PPS)? The "Round Height" help for PPNLS statistics seems to indicate shares are only tracked per block found by the pool and reset each time, but I can't determine if the language in this help section is "per miner" or "pool wide".

3. How can we determine block history for Ethereum and whether a block was mined solo or for the pool? With a current TTF of about a day (assuming all the hashrate is pooled, not solo), finding a block is "kind of a big deal". I can see, for example, that miningpoolstats is reporting that Prohashing found an Ethereum block 8 hours ago, but I saw no change in my PPLNS stats or payout of any kind at that time; do I just assume that's a solo miner finding it? The WAMP API doesn't report whether a block was pool or solo mined either, so it's a guessing game.

4. There doesn't appear to be a listed explorer for Ethereum blocks on the website. Is this by design or...? If there isn't an internal one, is there an external one recommended that can be linked for ease of use?

5. Does Prohashing include MEV (or a portion of it) as part of the block+transaction fees reward that is distributed for Ethereum, when it is available?

[Steve Sokolowski]

Hi Beowulf,

1 and 2. PPLNS rounds start when the last block was found and continue until the next block is found. All calculations consider the entire round. Some pools use a "proportional" or "weighted" system, but we don't. In every case except Ethereum and bitcoin, the length of PPLNS rounds is so short that the proportional system doesn't have much of an effect on miner behavior. Our aim is that we will have sufficient manpower to implement a proportional system by the time that Ethereum Classic's price and difficulty skyrocket and it becomes the primary GPU coin.

3. The "Status" page will assist you in determining when PPLNS blocks have been found for any coin. Click on the coin you're looking for and scroll down to "PPLNS stats." This data is updated every 2500 share inserts.

4. The reason we have block explorers for some coins is because we need the data for those coins in our database for some other purpose, or because there are no other reliable block explorers. Since our business is in mining, block explorers don't provide additional revenue on their own. In the case of Ethereum, there are many block explorers available, so we didn't see a reason to offer one. A Google search can point you in the direction of several of them in the first results.

5. Miner extracted value is a broad term that encompasses a large number of tricks to get additional revenue out of Ethereum transactions. They range from private off-chain written contracts to include specific transactions with specific customers, to reordering transactions, to manipulating distributed exchange prices, and more. You would have to be specific about which method you are asking about. In general, we have not focused on this topic because most analyses revealed that the average block could only be supplemented by about $5 in additional value through these MEV tricks in July. As stated above, our aim is to find MEV tricks on Ethereum Classic before that coin starts to rise in price and attract the GPU miners.

[Beowulf]

3. The "Status" page will assist you in determining when PPLNS blocks have been found for any coin. Click on the coin you're looking for and scroll down to "PPLNS stats." This data is updated every 2500 share inserts.

If you mean the Coin Status section under the main Status tab, that only shows global hashrate and no split between hashrate being used for PPS/PPLNS (which I assume is combined) and solo. The block display for PPLNS Stats for Ethereum still says "First Round" which I assume is in context of myself, as the pool has surely found other ETH blocks that weren't solo mined. I'd like to see some historical confirmation of as many historical blocks as possible that the pool finds and how those rewards were distributed (shared or solo). i.e. what happened to that block the pool found 24 hours ago?

Also, not knowing the hashrate split between solo and shared (PPS/PPLNS) makes it harder to make educated decisions on whether to mine here or not, as it can drastically affect projected block TTF. If some monster farm is using Prohashing as its solo mining target, it may be multiple days or weeks before the shared pool finds a block. Some folks can't wait that long and may want to opt for a different pool with lower TTF, depending on their financial responsibilities. It "seems" like an easy fix to simply display the PPS, PPLNS and Solo aggregate hashrates separately, though I am obviously ignorant of how you track hashrate on the backend DB.

[Beowulf]

Another block found by the site 8 hours ago (#13086279, apparently) but no indication on the PPLNS stats of anything. Again, I can only assume this was a solo mined block and have no way to tell.

Based on my current PPLNS round share %, it's clear that at least 25% of the current advertised pool hashrate on the site (81 GH/s-ish) is solo; this is important info to know as it pushes the expected shared pool TTF to multi-day.

[Steve Sokolowski]

3. The "Status" page will assist you in determining when PPLNS blocks have been found for any coin. Click on the coin you're looking for and scroll down to "PPLNS stats." This data is updated every 2500 share inserts.

If you mean the Coin Status section under the main Status tab, that only shows global hashrate and no split between hashrate being used for PPS/PPLNS (which I assume is combined) and solo. The block display for PPLNS Stats for Ethereum still says "First Round" which I assume is in context of myself, as the pool has surely found other ETH blocks that weren't solo mined. I'd like to see some historical confirmation of as many historical blocks as possible that the pool finds and how those rewards were distributed (shared or solo). i.e. what happened to that block the pool found 24 hours ago?

Also, not knowing the hashrate split between solo and shared (PPS/PPLNS) makes it harder to make educated decisions on whether to mine here or not, as it can drastically affect projected block TTF. If some monster farm is using Prohashing as its solo mining target, it may be multiple days or weeks before the shared pool finds a block. Some folks can't wait that long and may want to opt for a different pool with lower TTF, depending on their financial responsibilities. It "seems" like an easy fix to simply display the PPS, PPLNS and Solo aggregate hashrates separately, though I am obviously ignorant of how you track hashrate on the backend DB.

The PPLNS statistics show "first round" because all blocks found for that coin have been through one of the other two mining modes. FPPS is by far the most popular mining mode here, because in all algorithms except the high memory version of ethash, switching between coins provides more profit for customers than the lower fees for PPLNS provide. You're correct that there haven't been any Ethereum blocks found for PPLNS yet.

The hashrates displayed on the website always are for FPPS data, as that was the first mining mode we supported. While it is possible, as you pointed out, to display PPLNS hashrates, we don't have sufficient manpower at this time to add them. Right now, our primary focuses are on integrating with Salad for FPPS ethash mining, adding ACH payouts, deploying Chia mining, reducing the seat license cost and improving the speed and number of methods of our customer service by changing to chatwoot, training two new employees, and rebranding and SEO-optimizing the website.

We would definitely like to add the hashrate you requested in the future, but our manpower is limited by the speed at which we can train employees. We need to get our current two employees fully up to speed first before hiring more, because otherwise we would progress even more slowly due to the drag from the training. Perhaps Vance and Michael can look into adding this field in October or November.

[Beowulf]

The PPLNS statistics show "first round" because all blocks found for that coin have been through one of the other two mining modes. FPPS is by far the most popular mining mode here, because in all algorithms except the high memory version of ethash, switching between coins provides more profit for customers than the lower fees for PPLNS provide. You're correct that there haven't been any Ethereum blocks found for PPLNS yet.

To be clear, are you saying the hashrate through FPPS and PPLNS are not combined into a single pool of hashrate with shared rewards? I had assumed that the combined hash would be used in a round and a block reward would be split amongst all the contributors (except SOLO ones, obviously). For rewards, the PPLNS miners would get their calculated % share based on share contribution first. The remaining block reward (equal to the aggregate % shares supplied by all the FPPS miners) would go back into Prohashing coffers to cover the pre-payments to the FPPS miners.

[Steve Sokolowski]

The PPLNS statistics show "first round" because all blocks found for that coin have been through one of the other two mining modes. FPPS is by far the most popular mining mode here, because in all algorithms except the high memory version of ethash, switching between coins provides more profit for customers than the lower fees for PPLNS provide. You're correct that there haven't been any Ethereum blocks found for PPLNS yet.

To be clear, are you saying the hashrate through FPPS and PPLNS are not combined into a single pool of hashrate with shared rewards? I had assumed that the combined hash would be used in a round and a block reward would be split amongst all the contributors (except SOLO ones, obviously). For rewards, the PPLNS miners would get their calculated % share based on share contribution first. The remaining block reward (equal to the aggregate % shares supplied by all the FPPS miners) would go back into Prohashing coffers to cover the pre-payments to the FPPS miners.

No. The mining modes are handled completely separately.

Your idea, however, is a great one, and nobody had suggested it at the time we implemented the feature. If we had known about it, I would have done that.

[Alynn]

So If I'm understanding this....

FPPS becomes a combined Hashrate that is applied to the PPLNS with all those PPLNS miners. When a block is found either by PPS or PPLNS the reward is split between the PPS combined pool (to the pool) and everyone in pplns. So, essentially everything is PPLNS with all PPS miners counted as a single "user".

I think this would be a much better concentrated use of the pool hashrate where Solo is going it alone, but the other two work together which on average would decrease the time between found blocks, correct?

[Murr808]

I have found the payouts at Prohashing are significantly lower than Ethermine. I will keep mining with Ethermine for now.

[Beowulf]

FPPS becomes a combined Hashrate that is applied to the PPLNS with all those PPLNS miners. When a block is found either by PPS or PPLNS the reward is split between the PPS combined pool (to the pool) and everyone in pplns. So, essentially everything is PPLNS with all PPS miners counted as a single "user".

I think this would be a much better concentrated use of the pool hashrate where Solo is going it alone, but the other two work together which on average would decrease the time between found blocks, correct?

Yup...essentially correct as my proposed future state; that's obviously not how it works right now with FPPS and PPLNS as completely different pools of hashrate.

FPPS users are paid as they mine, regardless of block discovery, at a rate proportional to their hashrate and expected return value. So, Prohashing is covering for them in advance and needs to be reimbursed which is why they would get the "remainder" of the block reward once PPLNS users are paid for their % cut. On blocks that take a long time to find, they may advance FPPS miners a greater amount than they get reimbursed for with their "remainder" cut once a block is find, but this would even out over time with blocks that are found "early". Prohashing also takes a larger fee for FPPS miners to help offset the risk they take in their advance payment of those miners.

For single-coin algos or those with really tough blocks that require a lot of hash (e.g. BTC), it makes sense to amalgamate the hash of FPPS and PPLNS miners as both Prohashing and PPLNS miners benefit from more frequent block discovery. FPPS miners don't really care, as they just keep getting paid regardless. Though, I'm not sure why anyone would want to FPPS mine Ethereum unless they really didn't like getting paid in ETH and were willing to pay extra in fees to get auto-exchange.