[500 GH/s]HHTT -Selected Diff/Stratum/PPLNS/Paid Stales/High Availability/Tor - page 7.

spiccioli

legendary

Activity: 1378

Merit: 1003

nec sine labore

Quote from: Hippie Tech on June 05, 2013, 02:20:11 AM

Was that miner running 800 ghash/s or was it really only 700 ? 500 ?

I'm very interested to know.

Hippie,

shown hashrate is correct, I can say this because I know my hashrate and the pool reports it correctly.

It fluctuates, though, given that even an Avalon is subject to variance in finding shares and the more higher the difficulty you use the more higher your variance.

spiccioli

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote from: Hippie Tech on June 05, 2013, 02:20:11 AM

Quote from: organofcorti on June 04, 2013, 11:47:31 PM

Diff420 shares are quite high at your ~1.4 Ghps. You'd only be submitting 2 or three shares per hour, and since the only way HHTT can estimate your hashrate is by the shares you've submitted any variance in number of shares will be reflected in your hashrate.

In point of fact and based on the numbers you provide, you could expect your hourly hashrate (as represented by HHTT) to vary between 0% and 215% of 1.4Ghps with 95% confidence.

Edit: You can test this out by submitting Diff1 shares for a while. Your hourly hashrate will vary much less.

I am well aware of how high or low vardiff can affect things. The point was, how the calculated hashrate on the user's stats page can be misleading.

You are correct about the potential 200% swings. Now lets try to correlate that to how bloated the performance may be with one hundred 200 mhash/s chips banging out 128 - 256 diff shares.

Was that miner running 800 ghash/s or was it really only 700 ? 500 ?

I'm very interested to know.

It's impossible to say - the pool only sees the submitted shares. An estimate could be made if averaged over sufficient hours - in your case (1.4 Ghps and Diff420), the 95% confidence interval for a 100 hour average would be +/- 12%.

Hippie Tech

legendary

Activity: 1624

Merit: 1001

All cryptos are FIAT digital currency. Do not use.

Quote from: organofcorti on June 04, 2013, 11:47:31 PM

Quote from: Hippie Tech on June 04, 2013, 08:56:29 PM

Quote from: roy7 on June 04, 2013, 10:58:57 AM

Quote from: silverserpent on June 04, 2013, 10:54:09 AM

Quote from: roy7 on June 04, 2013, 10:37:18 AM

Quote from: Hippie Tech on June 04, 2013, 10:31:48 AM

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Is the hash rate graph not scaled for the difficulty of the submitted shares? The hash rate on graph doesn't seem at all accurate.

nope, just the pure power is charted. I think it might be, just seems like it's 3 Avalons with different allocations over time (with dips showing outages).

Ah ok. I couldn't understand how a 213K miner was getting so many coins.

That miner I linked up was rated over 800 ghash/s, if I'm not mistaken.

The way HHTT calculates that particular graph and ghash rate on the Sucker's stats page, is purely based on, shares per hour * x mhash/s per share. This method can be very misleading because a good/bad luck streak will make a miner appear to be much slower/faster than they really are.

For example... I started out here .. http://hhtt.1209k.com/user-details.php?user=1PijcWRJEXM6bRqv2shxiZscC3Xxm8VoV9 .. with 420 diff shares @ 1.4 ghash/s then dropped it to 780 mhashs/s. Notice how, for at least a few hours, 1.4 appeared to be 2.5 ghash/s ? It reported 2.5 ghash/s because each 420 share adds 500 mhash/s to my hourly 'average'.

Diff420 shares are quite high at your ~1.4 Ghps. You'd only be submitting 2 or three shares per hour, and since the only way HHTT can estimate your hashrate is by the shares you've submitted any variance in number of shares will be reflected in your hashrate.

In point of fact and based on the numbers you provide, you could expect your hourly hashrate (as represented by HHTT) to vary between 0% and 215% of 1.4Ghps with 95% confidence.

Edit: You can test this out by submitting Diff1 shares for a while. Your hourly hashrate will vary much less.

I am well aware of how high or low vardiff can affect things. The point was, how the calculated hashrate on the user's stats page can be misleading.

You are correct about the potential 200% swings. Now lets try to correlate that to how bloated the performance may be with one hundred 200 mhash/s chips banging out 128 - 256 diff shares.

Was that miner running 800 ghash/s or was it really only 700 ? 500 ?

I'm very interested to know.

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote from: Hippie Tech on June 04, 2013, 08:56:29 PM

Quote from: roy7 on June 04, 2013, 10:58:57 AM

Quote from: silverserpent on June 04, 2013, 10:54:09 AM

Quote from: roy7 on June 04, 2013, 10:37:18 AM

Quote from: Hippie Tech on June 04, 2013, 10:31:48 AM

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Is the hash rate graph not scaled for the difficulty of the submitted shares? The hash rate on graph doesn't seem at all accurate.

nope, just the pure power is charted. I think it might be, just seems like it's 3 Avalons with different allocations over time (with dips showing outages).

Ah ok. I couldn't understand how a 213K miner was getting so many coins.

That miner I linked up was rated over 800 ghash/s, if I'm not mistaken.

The way HHTT calculates that particular graph and ghash rate on the Sucker's stats page, is purely based on, shares per hour * x mhash/s per share. This method can be very misleading because a good/bad luck streak will make a miner appear to be much slower/faster than they really are.

For example... I started out here .. http://hhtt.1209k.com/user-details.php?user=1PijcWRJEXM6bRqv2shxiZscC3Xxm8VoV9 .. with 420 diff shares @ 1.4 ghash/s then dropped it to 780 mhashs/s. Notice how, for at least a few hours, 1.4 appeared to be 2.5 ghash/s ? It reported 2.5 ghash/s because each 420 share adds 500 mhash/s to my hourly 'average'.

Diff420 shares are quite high at your ~1.4 Ghps. You'd only be submitting 2 or three shares per hour, and since the only way HHTT can estimate your hashrate is by the shares you've submitted any variance in number of shares will be reflected in your hashrate.

In point of fact and based on the numbers you provide, you could expect your hourly hashrate (as represented by HHTT) to vary between 0% and 215% of 1.4Ghps with 95% confidence.

Edit: You can test this out by submitting Diff1 shares for a while. Your hourly hashrate will vary much less.

roy7

sr. member

Activity: 434

Merit: 250

Yeah, he could multiply the shares by the difficulty of the share submitted so the graph will show actual hash rates. It'd be easy enough, I do that on my TRC pool and I run his SockThing backend myself. Although maybe the goal of that graph isn't actually to show hash rate but something else.

Hippie Tech

legendary

Activity: 1624

Merit: 1001

All cryptos are FIAT digital currency. Do not use.

Quote from: roy7 on June 04, 2013, 10:58:57 AM

Quote from: silverserpent on June 04, 2013, 10:54:09 AM

Quote from: roy7 on June 04, 2013, 10:37:18 AM

Quote from: Hippie Tech on June 04, 2013, 10:31:48 AM

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Is the hash rate graph not scaled for the difficulty of the submitted shares? The hash rate on graph doesn't seem at all accurate.

nope, just the pure power is charted. I think it might be, just seems like it's 3 Avalons with different allocations over time (with dips showing outages).

Ah ok. I couldn't understand how a 213K miner was getting so many coins.

That miner I linked up was rated over 800 ghash/s, if I'm not mistaken.

The way HHTT calculates that particular graph and ghash rate on the Sucker's stats page, is purely based on, shares per hour * x mhash/s per share. This method can be very misleading because a good/bad luck streak will make a miner appear to be much slower/faster than they really are.

For example... I started out here .. http://hhtt.1209k.com/user-details.php?user=1PijcWRJEXM6bRqv2shxiZscC3Xxm8VoV9 .. with 420 diff shares @ 1.4 ghash/s then dropped it to 780 mhashs/s. Notice how, for at least a few hours, 1.4 appeared to be 2.5 ghash/s ? It reported 2.5 ghash/s because each 420 share adds 500 mhash/s to my hourly 'average'.

spiccioli

legendary

Activity: 1378

Merit: 1003

nec sine labore

Quote from: silverserpent on June 04, 2013, 11:03:30 AM

Quote from: roy7 on June 04, 2013, 10:59:47 AM

Quote from: silverserpent on June 04, 2013, 10:58:23 AM

I'm considering doing a split between here and another PPS pool (best of both worlds?), I was on BTC Guild but those fees are high...any ideas?

I'm a small GPU miner and I mine on my own TRC pool right now, but the 3 pools I suggest people look at for BTC are HHTT, Bitparking, and Eligius.

thanks. are Bitparking and Eligius 'Avalon compatible' - I've heard good things about p2pool but that Avalons have issues?

p2pool is not Avalon compatible at the moment.

spiccioli

spiccioli

legendary

Activity: 1378

Merit: 1003

nec sine labore

Quote from: silverserpent on June 04, 2013, 11:41:23 AM

are Transaction fees split, 50/50? cant seem to find any, payments are for the PPLNS credit amount.

Yes, when you find a block you get two payments, one for the PPLNS Share and one which is half the block fees

spiccioli

silverserpent

member

Activity: 83

Merit: 10

Quote from: roy7 on June 04, 2013, 11:06:02 AM

Quote from: silverserpent on June 04, 2013, 11:03:30 AM

thanks, are Bitparking and Eligius 'Avalon compatible' - I've heard good things about p2pool but that Avalons have issues?

Yes they both are. Eligius uses CPPSRB and Bitparking uses DGM. So you can't apples to apples another PPLNS pools with them though.

fireduck hasn't been around much recently but I trust him and I like his pool design. I use his software for my own tiny pool. (If you ever want to mine some TRC, hit up my pool (with a high difficulty setting). Cheesy

)

gr8. may take you up the TRC soon. thanks again.

silverserpent

member

Activity: 83

Merit: 10

are Transaction fees split, 50/50? cant seem to find any, payments are for the PPLNS credit amount.

spiccioli

legendary

Activity: 1378

Merit: 1003

nec sine labore

Quote from: BenTuras on June 04, 2013, 08:21:27 AM

Quote from: spiccioli on June 04, 2013, 07:30:21 AM

Pool stabilized (as if I did something...

) and two blocks found as of now for today.

spiccioli

ps. so, stop wondering and point your hashing machines here!

Sorry spiccioli, but that's too simple.
I want to know how a pool handles its shares and payout and for this pool it's not clear to me, the information on the website is outdated and fireduck is missing in action.

Yeah, I'd like a word from fireduck as well, but from his profile he has been away since May the 30th....

spiccioli

roy7

sr. member

Activity: 434

Merit: 250

Quote from: silverserpent on June 04, 2013, 11:03:30 AM

thanks, are Bitparking and Eligius 'Avalon compatible' - I've heard good things about p2pool but that Avalons have issues?

Yes they both are. Eligius uses CPPSRB and Bitparking uses DGM. So you can't apples to apples another PPLNS pools with them though.

fireduck hasn't been around much recently but I trust him and I like his pool design. I use his software for my own tiny pool. (If you ever want to mine some TRC, hit up my pool (with a high difficulty setting). Cheesy

)

silverserpent

member

Activity: 83

Merit: 10

Quote from: roy7 on June 04, 2013, 10:59:47 AM

Quote from: silverserpent on June 04, 2013, 10:58:23 AM

I'm considering doing a split between here and another PPS pool (best of both worlds?), I was on BTC Guild but those fees are high...any ideas?

I'm a small GPU miner and I mine on my own TRC pool right now, but the 3 pools I suggest people look at for BTC are HHTT, Bitparking, and Eligius.

thanks. are Bitparking and Eligius 'Avalon compatible' - I've heard good things about p2pool but that Avalons have issues?

roy7

sr. member

Activity: 434

Merit: 250

Quote from: silverserpent on June 04, 2013, 10:58:23 AM

I'm considering doing a split between here and another PPS pool (best of both worlds?), I was on BTC Guild but those fees are high...any ideas?

I'm a small GPU miner and I mine on my own TRC pool right now, but the 3 pools I suggest people look at for BTC are HHTT, Bitparking, and Eligius.

roy7

sr. member

Activity: 434

Merit: 250

Quote from: silverserpent on June 04, 2013, 10:54:09 AM

Quote from: roy7 on June 04, 2013, 10:37:18 AM

Quote from: Hippie Tech on June 04, 2013, 10:31:48 AM

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Is the hash rate graph not scaled for the difficulty of the submitted shares? The hash rate on graph doesn't seem at all accurate.

nope, just the pure power is charted. I think it might be, just seems like it's 3 Avalons with different allocations over time (with dips showing outages).

Ah ok. I couldn't understand how a 213K miner was getting so many coins.

silverserpent

member

Activity: 83

Merit: 10

Quote from: BenTuras on June 03, 2013, 05:21:28 AM

Thanks for your answer spiccioli,
So how this pool calculates what we earn is not written out on the webpage...
I have looked at a few miner pages, on all pages earned equals paid.
This sounds wrong to me, after finding a share, earned should increase.
Too bad fireduck is 800 BTC in the negative. And of course no PPS 'sucker' felt the need to partly refund.

Too bad two big miners decided to leave today, hashing power of 5 Avalons. Or they had a problem, looks like they're back.

I will stay on this pool for a few more days.
I would like to get some more info though.

I'm considering doing a split between here and another PPS pool (best of both worlds?), I was on BTC Guild but those fees are high...any ideas?

silverserpent

member

Activity: 83

Merit: 10

Quote from: roy7 on June 04, 2013, 10:37:18 AM

Quote from: Hippie Tech on June 04, 2013, 10:31:48 AM

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Is the hash rate graph not scaled for the difficulty of the submitted shares? The hash rate on graph doesn't seem at all accurate.

nope, just the pure power is charted. I think it might be, just seems like it's 3 Avalons with different allocations over time (with dips showing outages).

roy7

sr. member

Activity: 434

Merit: 250

Quote from: Hippie Tech on June 04, 2013, 10:31:48 AM

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Is the hash rate graph not scaled for the difficulty of the submitted shares? The hash rate on graph doesn't seem at all accurate.

roy7

sr. member

Activity: 434

Merit: 250

Quote from: silverserpent on June 04, 2013, 10:29:26 AM

Quote from: roy7 on June 04, 2013, 09:45:40 AM

I can only speculate, but if your shares in the look back window dropped in half in 4 hours and you mine constantly, then I'm guessing it'd only take 8 hours for your score to go to zero. Meaning the full N shares only take 8 hours for the pool to mine. It's only your shares within the last N shares that count for payment.

actually, the PPLNS value dwindled down over time but the outage itself was about 4 hrs. Perhaps the fact that the history is so limited had a significant impact (only started on HHTT about 5-31)

Right, I just mean if you add shares at a steady rate and after 4 hours half are gone, then the window probably moved N/2 shares in 4 hours. At that rate you'd be at zero in 8 hours.

When you stop mining your score would slowly go down each time one of your shares was pushed out of the N share window by someone else's newer share.

Hippie Tech

legendary

Activity: 1624

Merit: 1001

All cryptos are FIAT digital currency. Do not use.

Quote from: silverserpent on June 04, 2013, 09:34:38 AM

Quote from: Hippie Tech on June 04, 2013, 09:18:15 AM

I'd like to know why the PPLNS credit on these asic miners last so long.

My credits were gone in less than a day, yet the top miner still has more than 60% of theirs after the same amount of idle time.

yeah, its pretty bad...lost internet access for 4 hrs was knocked down to .96, from 1.84; as a result lost more than half of what I should have rec'd

I'm sure this guy isn't complaining.
http://hhtt.1209k.com/user-details.php?user=19ihxgbVYHv5P9uAWhPJYcPEqJfqkrGy9n

Topic: [500 GH/s]HHTT -Selected Diff/Stratum/PPLNS/Paid Stales/High Availability/Tor - page 7. (Read 56345 times)