phzi
|
|
March 19, 2014, 11:56:24 PM |
|
Does this mean we have a year of gpu mining at best or is there more time? I wasnt mining yet back when bit coin was gpu only so i dont know...
The question with GPU mining, is... what algorithm? I cannot predict the figure, but I doubt profitable SCRYPT mining with GPUs will last much longer. And based on recent developments, I would say a few months at best. So, the question then becomes: do we need more cryptocurrencies with new algorithms? Can an scrypt-adaptive-n or scrypt-chacha or sha3 or etc variant be the next Litecoin? Vertcoin mining is currently hovering near the profitability of Litecoin mining, and Vertcoin is based on scrypt-adpative-n, which cannot be mined with the upcoming scrypt ASICs. Is there enough utility in alternative blockchains that Vertcoin (for example) could maintain it's speculative value and actually support a mass incoming of GPU miners? WAFFLEStats v1.1 - Workers Are Sexy EditionDonate to: 1Pr87ypYdG8eMTNUXR4rT1q2P1VJ7yu1b5 Nice update, Wil. Looking forward to see some workers charts after a day or two. Mar 19, 2014 (partial) 79.09625173 11.42 GH/s 0.00779710
That shut them up with the variance talk!
Looks like we are going to close the day around .075BTC/MH. Cross fingers for a last minute litecoin block to bring up the day's numbers!!! We looked to be having good luck with litecoin blocks (including 2 in quick succession earlier in the day). Definitely helping the profit numbers. --- @PoolWaffle - could you add a few "luck" statistics? Such as "expected" profitability based on switcher estimated profitability and work done (submitted shares); vs real profitability based on work done, blocks found, and block exchange value. (After rejected share info, maybe. heh)
|
|
|
|
elpsycongro
|
|
March 20, 2014, 01:28:19 AM |
|
i can see what you are saying, so as ltc was the asnwer to BTC becoming too high in difficulty due to ASICS we gotta wait and see what the answer to LTC is against the coming ASICS., if there is a such a coin that stands a chance and pools can adapt to it that would be the new standard...
|
|
|
|
Blando
Newbie
Offline
Activity: 9
Merit: 0
|
|
March 20, 2014, 04:33:37 AM |
|
Anyone know what just happened to BTC/MHD over the last hour? Yes! Someone finally used the technically correct term of BTC/MHD! Not very fond of this version either. Strictly technical you are referring to (BTC/MHs)*D ...I was looking at the WaffleStats page>Details>Statistics>Last Hour>BTC/Day/1Mh just now...
I give up. I started typing up a detailed explanation but then realized how stupid it is to waste my time doing so. No matter how incontrovertible my proof, you will all likely keep calling it whatever you want to anyway. So have at it. I will stick to more profitable endeavors. Please explain. I like to read your posts while I'm doing nightly Shake Weight routine.
|
|
|
|
lagster
Newbie
Offline
Activity: 56
Merit: 0
|
|
March 20, 2014, 05:45:21 AM |
|
Anyone know what just happened to BTC/MHD over the last hour? Yes! Someone finally used the technically correct term of BTC/MHD! Not very fond of this version either. Strictly technical you are referring to (BTC/MHs)*D ...I was looking at the WaffleStats page>Details>Statistics>Last Hour>BTC/Day/1Mh just now...
I give up. I started typing up a detailed explanation but then realized how stupid it is to waste my time doing so. No matter how incontrovertible my proof, you will all likely keep calling it whatever you want to anyway. So have at it. I will stick to more profitable endeavors. Please explain. I like to read your posts while I'm doing nightly Shake Weight routine. reading his posts is almost like reading shakespear's posts when he was still blogging a lot. and if some one wanted to multiply his earnings by the D, i guess it could be profitable for few of us..
|
|
|
|
CoinBuzz
|
|
March 20, 2014, 06:06:47 AM |
|
Is Bitcoin-Scrypt enlisted in mining basket?
|
|
|
|
pRob3
Newbie
Offline
Activity: 26
Merit: 0
|
|
March 20, 2014, 06:15:35 AM |
|
Must have been an sweet update or are we just lucky. It looks good on Poolpicker.eu today.
|
|
|
|
MinerP
|
|
March 20, 2014, 06:32:38 AM |
|
Must have been an sweet update or are we just lucky. It looks good on Poolpicker.eu today. I thought it was the update at first and was thinking pw found the hidden treasure! was over 0.0085 at one point, turns out it was 98% luck 2% update.
|
|
|
|
lagster
Newbie
Offline
Activity: 56
Merit: 0
|
|
March 20, 2014, 06:39:52 AM |
|
Must have been an sweet update or are we just lucky. It looks good on Poolpicker.eu today. I thought it was the update at first and was thinking pw found the hidden treasure! was over 0.0085 at one point, turns out it was 98% luck 2% update. so where are you going to move in 7 hours?
|
|
|
|
tachyon_john
Newbie
Offline
Activity: 28
Merit: 0
|
|
March 20, 2014, 07:00:03 AM |
|
i can see what you are saying, so as ltc was the asnwer to BTC becoming too high in difficulty due to ASICS we gotta wait and see what the answer to LTC is against the coming ASICS., if there is a such a coin that stands a chance and pools can adapt to it that would be the new standard...
An easy way of making ASICs unprofitable is to design algorithms that require large memory buffers and that have performance bound by memory bandwidth rather than arithmetic. ASICs provide the greatest benefits for algorithms that are arithmetic-bound, and they provide the least benefits for algorithms that are bound by memory bandwidth. By combining a large size memory buffer with random access patterns, we would get a level playing field that evolves very slowly. GPUs of today have 200-300GB/s memory bandwidth which has only increased by a small margin generation-to-generation. GPUs are expected to get a nice jump in bandwidth when memory technologies like die-stacked memory show up in a few years, but after that bandwidth growth will be very very slow again. A large part of the complexity and cost in a GPU is the memory system, and this is something that is only feasible to build because millions of GPUs are sold per week. By developing an algorithm that requires a hardware capability that is only cost-feasible in commodity devices that are manufactured in quantities of several million or more, it would push ASICs completely out, and keep them for a very long time, perhaps indefinitely. It's one thing to fab an ASIC chip, it's another thing to couple it to a high-capacity high-bandwidth memory system. If you design an algorithm that uses the "memory wall" as a fundamental feature, it will make ASICs no better than any other hardware approach.
|
|
|
|
MinerP
|
|
March 20, 2014, 07:52:17 AM |
|
Must have been an sweet update or are we just lucky. It looks good on Poolpicker.eu today. I thought it was the update at first and was thinking pw found the hidden treasure! was over 0.0085 at one point, turns out it was 98% luck 2% update. so where are you going to move in 7 hours? nowhere! ill be sleeping in 20 minutes.
|
|
|
|
dexu
|
|
March 20, 2014, 09:54:14 AM |
|
Well, mining about 2 hours and Reject ratio: 0.0% Great job !
|
|
|
|
srSheepdog
Newbie
Offline
Activity: 9
Merit: 0
|
|
March 20, 2014, 10:10:46 AM |
|
Random thought: More than once it has crossed my mind that 'sfire' is Sapphire. Crazy or plausible? Discuss.
|
|
|
|
CrashOD
Newbie
Offline
Activity: 41
Merit: 0
|
|
March 20, 2014, 12:35:40 PM |
|
After over 24 hours mining on the new stratum server, I have <1% rejects down from 6%, but my average hashrate has dropped 17%! I went from an average of 6MH/s before, to 5MH/s over the last 30 hours. At first I thought it was just variance, but it seems to be consistently staying lower. No config has changed on my end. Any clues what could cause such a dramatic change/loss in hashrate?
|
|
|
|
LPCobris
|
|
March 20, 2014, 01:10:32 PM |
|
i can see what you are saying, so as ltc was the asnwer to BTC becoming too high in difficulty due to ASICS we gotta wait and see what the answer to LTC is against the coming ASICS., if there is a such a coin that stands a chance and pools can adapt to it that would be the new standard...
An easy way of making ASICs unprofitable is to design algorithms that require large memory buffers and that have performance bound by memory bandwidth rather than arithmetic. ASICs provide the greatest benefits for algorithms that are arithmetic-bound, and they provide the least benefits for algorithms that are bound by memory bandwidth. By combining a large size memory buffer with random access patterns, we would get a level playing field that evolves very slowly. GPUs of today have 200-300GB/s memory bandwidth which has only increased by a small margin generation-to-generation. GPUs are expected to get a nice jump in bandwidth when memory technologies like die-stacked memory show up in a few years, but after that bandwidth growth will be very very slow again. A large part of the complexity and cost in a GPU is the memory system, and this is something that is only feasible to build because millions of GPUs are sold per week. By developing an algorithm that requires a hardware capability that is only cost-feasible in commodity devices that are manufactured in quantities of several million or more, it would push ASICs completely out, and keep them for a very long time, perhaps indefinitely. It's one thing to fab an ASIC chip, it's another thing to couple it to a high-capacity high-bandwidth memory system. If you design an algorithm that uses the "memory wall" as a fundamental feature, it will make ASICs no better than any other hardware approach. Great Post and so true... If they want a leveled plane of mining, that should be the way... Best Regards, LPC
|
|
|
|
whiteslashasian
Newbie
Offline
Activity: 2
Merit: 0
|
|
March 20, 2014, 01:17:46 PM |
|
After over 24 hours mining on the new stratum server, I have <1% rejects down from 6%, but my average hashrate has dropped 17%! I went from an average of 6MH/s before, to 5MH/s over the last 30 hours. At first I thought it was just variance, but it seems to be consistently staying lower. No config has changed on my end. Any clues what could cause such a dramatic change/loss in hashrate?
I noticed something similar as well. I lost close to 10% of my reported hashrate according to wafflestats. I went from 10.2 MH/s over the last few weeks down to 9.25 MH/s since the update. Quite concerning.
|
|
|
|
Thirtybird
|
|
March 20, 2014, 01:32:53 PM |
|
i can see what you are saying, so as ltc was the asnwer to BTC becoming too high in difficulty due to ASICS we gotta wait and see what the answer to LTC is against the coming ASICS., if there is a such a coin that stands a chance and pools can adapt to it that would be the new standard...
An easy way of making ASICs unprofitable is to design algorithms that require large memory buffers and that have performance bound by memory bandwidth rather than arithmetic. ASICs provide the greatest benefits for algorithms that are arithmetic-bound, and they provide the least benefits for algorithms that are bound by memory bandwidth. By combining a large size memory buffer with random access patterns, we would get a level playing field that evolves very slowly. GPUs of today have 200-300GB/s memory bandwidth which has only increased by a small margin generation-to-generation. GPUs are expected to get a nice jump in bandwidth when memory technologies like die-stacked memory show up in a few years, but after that bandwidth growth will be very very slow again. A large part of the complexity and cost in a GPU is the memory system, and this is something that is only feasible to build because millions of GPUs are sold per week. By developing an algorithm that requires a hardware capability that is only cost-feasible in commodity devices that are manufactured in quantities of several million or more, it would push ASICs completely out, and keep them for a very long time, perhaps indefinitely. It's one thing to fab an ASIC chip, it's another thing to couple it to a high-capacity high-bandwidth memory system. If you design an algorithm that uses the "memory wall" as a fundamental feature, it will make ASICs no better than any other hardware approach. Great Post and so true... If they want a leveled plane of mining, that should be the way... Best Regards, LPC Ya, so there's already coins that do this. YACoin was the first, and currently takes 4 MB per thread to complete a calculation. That will be 8 MB on May 31st. All the other scrypt-chacha coins will get there eventually, but YAC is the trailblazer
|
|
|
|
bigted
Newbie
Offline
Activity: 1
Merit: 0
|
|
March 20, 2014, 01:33:50 PM |
|
nevermind
|
|
|
|
madian
Newbie
Offline
Activity: 57
Merit: 0
|
|
March 20, 2014, 01:47:24 PM |
|
author=comeonalready
And it is payout per MHD!
I see you insist... Ok. Let me show you my reasoning for why IT IS NOT MHD and then you can just slam down your irrefutable proof and I'll stand rebuked. let's say that your hashrate is 1 MHs. Meaning that your computational power can "solve" 1 million hashes in a second. Analogous to active power being measured in kW, the energy consumption is measured based on the consumption of power in a unit of time thus the kWh unit of measure. When you want to see how much energy you consumed in a DAY you just count the kWh which means you consumed 24 kWh/day (for 1 kWh). when you use BTC/MHD you are implying that your hash rate is 1 "MH per day" which is false. your hash rate is an average of 1 MHs over a day's time and you contributed with 3600* 1 MH in a day's time. WP states [0.01 btc/(average MHs) in a day] and not the amount of hashes in a day. And lastly, the unit of measure is MHs. you can twist that figure to show a day or a year but it's still MHs as reported by your miners.
|
|
|
|
tachyon_john
Newbie
Offline
Activity: 28
Merit: 0
|
|
March 20, 2014, 02:05:36 PM |
|
i can see what you are saying, so as ltc was the asnwer to BTC becoming too high in difficulty due to ASICS we gotta wait and see what the answer to LTC is against the coming ASICS., if there is a such a coin that stands a chance and pools can adapt to it that would be the new standard...
An easy way of making ASICs unprofitable is to design algorithms that require large memory buffers and that have performance bound by memory bandwidth rather than arithmetic. ASICs provide the greatest benefits for algorithms that are arithmetic-bound, and they provide the least benefits for algorithms that are bound by memory bandwidth. By combining a large size memory buffer with random access patterns, we would get a level playing field that evolves very slowly. GPUs of today have 200-300GB/s memory bandwidth which has only increased by a small margin generation-to-generation. GPUs are expected to get a nice jump in bandwidth when memory technologies like die-stacked memory show up in a few years, but after that bandwidth growth will be very very slow again. A large part of the complexity and cost in a GPU is the memory system, and this is something that is only feasible to build because millions of GPUs are sold per week. By developing an algorithm that requires a hardware capability that is only cost-feasible in commodity devices that are manufactured in quantities of several million or more, it would push ASICs completely out, and keep them for a very long time, perhaps indefinitely. It's one thing to fab an ASIC chip, it's another thing to couple it to a high-capacity high-bandwidth memory system. If you design an algorithm that uses the "memory wall" as a fundamental feature, it will make ASICs no better than any other hardware approach. Great Post and so true... If they want a leveled plane of mining, that should be the way... Best Regards, LPC Ya, so there's already coins that do this. YACoin was the first, and currently takes 4 MB per thread to complete a calculation. That will be 8 MB on May 31st. All the other scrypt-chacha coins will get there eventually, but YAC is the trailblazer Sorry, but 4MB isn't a lot of memory. 1GB or more would start to be the size of memory I'm talking about. Anything that's just a few megabytes in size is small enough that someone that wanted it badly enough could just put SRAM on-die. CPUs and GPUs already have aggregate on-chip cache sizes that are 10 times that size, so 4MB is nowhere near large enough. The data size has to be large enough so that the on-chip caches are useless, and remain useless over at least a 10 year period. I would put that at something over 1GB.
|
|
|
|
dspair
|
|
March 20, 2014, 02:13:40 PM |
|
Hashrate: 2.63 GH/s I think something is wrong.
|
|
|
|
|