Search content
Sort by
Showing 20 of 77 results by theowalpott
There are 2^32 different possible hashes which give a difficulty 1 share, so on average:
seconds to generate a share = (2^32)/(hashrate in hashes per sec)
shares per minute = 60/(seconds to generate a share)
shares per hour = 60*(shares per minute)
Take this number, multiply by your expected payment per share (e.g. the current PPS rate for whichever pool) and you have BTC/hour.
seconds to generate a share = (2^32)/(hashrate in hashes per sec)
shares per minute = 60/(seconds to generate a share)
shares per hour = 60*(shares per minute)
Take this number, multiply by your expected payment per share (e.g. the current PPS rate for whichever pool) and you have BTC/hour.
Re: [8500 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
The more i look at it, im thinking the shares set up skews towards high powered miners. Still to prove it with some maths and tests. Last exam is on Tuesday. After that I can have a proper look.
I believe the use of "seconds since round start" is the problem, really. It should really be a function of shares since round start, however that would mean a lot of extra computational effort to keep track of. Use of seconds is a decent approximation, and is fair over the long term (assuming perfect uptime).
Last few suggestions I have.. use APP SDK 2.6 or 2.7 - they are apparently the best for 7xxx cards, although I've not noticed any problems with 2.8. Only other thing I can think of is the version of pyopencl? - possible an older version is faster for some reason?
Again, I'm using a fairly recent version (updated within the last month or 2) of pyopencl.. so not sure how much that helps you!
Finally, try 12.8 or 12.11 cats, from memory people had success with those.. I'm just going on forum posts and things I remember though, as I've never really experienced many problems using linux.
Again, I'm using a fairly recent version (updated within the last month or 2) of pyopencl.. so not sure how much that helps you!
Finally, try 12.8 or 12.11 cats, from memory people had success with those.. I'm just going on forum posts and things I remember though, as I've never really experienced many problems using linux.
I'd estimate you'd get ~465MH/s @ 1250 core, based on me getting ~375MH/s at stock (GHz edition). Note the 7870XT has 1536 cores compared to 1280 in the 7870. I've never tried mining in windows, so I can't help much - sorry.
I'd concentrate on your choice of mining kernel (I find diablo and poclbm to be the best) and stick with the 13.1 cats. AFAIK the SDK is built into the latest drivers so you shouldnt need to install it separately unless you want to use a different version.
Maybe you should do a clean install of the AMD drivers (use driver sweeper in safe mode after uninstalling for example), don't reinstall the APP SDK and test the parameters with diablo and pocblem miners. You could also try disabling the windows7 aero stuff.
I'd concentrate on your choice of mining kernel (I find diablo and poclbm to be the best) and stick with the 13.1 cats. AFAIK the SDK is built into the latest drivers so you shouldnt need to install it separately unless you want to use a different version.
Maybe you should do a clean install of the AMD drivers (use driver sweeper in safe mode after uninstalling for example), don't reinstall the APP SDK and test the parameters with diablo and pocblem miners. You could also try disabling the windows7 aero stuff.
I've got both a 7870 and a 7870XT running on 2 different linux systems.
with cgminer (2.11.4) (diablo kernel) or poclbm I get ~420 MH/s @ 1100core/950 mem on the 7870. CGMiner settings are worksize 128 and intensity 12 with 2 threads. Cat 13.1 and SDK 2.8 on ubuntu 12.04 32bit.
with poclbm I get ~500MH/s @ 1100core/950mem on the 7870XT. Poclbm settings are -w128 -f10 (cgminer seemed to crash a lot more than poclbm on the system I have it in). Cat 12.11 and SDK 2.8 on ubuntu 12.10 64 bit.
I found a drop in hashrate when using the 13.3 beta drivers, although I've not tried the 13.4 driver. I also found that the phatk kernel performed worse on both.
Hope that helps
with cgminer (2.11.4) (diablo kernel) or poclbm I get ~420 MH/s @ 1100core/950 mem on the 7870. CGMiner settings are worksize 128 and intensity 12 with 2 threads. Cat 13.1 and SDK 2.8 on ubuntu 12.04 32bit.
with poclbm I get ~500MH/s @ 1100core/950mem on the 7870XT. Poclbm settings are -w128 -f10 (cgminer seemed to crash a lot more than poclbm on the system I have it in). Cat 12.11 and SDK 2.8 on ubuntu 12.10 64 bit.
I found a drop in hashrate when using the 13.3 beta drivers, although I've not tried the 13.4 driver. I also found that the phatk kernel performed worse on both.
Hope that helps
Re: Lesson to Learn from Alt-Coins? Disruption to Bitcoin?
1. if it takes a block 2 weeks to be solved due to asic jumping, wouldnt that drive the price of btc up, since no coins being mined for two weeks?
2. i was under the impression asic mining isnt possible with litecoin yet.
2. i was under the impression asic mining isnt possible with litecoin yet.
If no coins are being mined, then no transactions are being included in blocks - how do you intend to send the bitcoins to be sold?
Not sure what you're getting at, re-read the topic. I'm verifying it myself if nobody wants to buy it. The account is BARE. I deposited Bitcoins into it, sold them, and couldn't withdraw BTC or USD because it wasn't verified. Because I don't want to wait weeks for 1) my account to be verified and 2) the funds to be transferred to my bank account, I'm selling it. MT. Gox fucked me over by not telling me I would need verification to withdraw funds until I went to do it.
What? - you don't need to be verified to withdraw BTC. You could just buy BTC with the USD, withdraw and then sell on another exchange or use localbitcoins.
Re: [8500 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
Is there something odd about block 17664?
17664 2013-04-24 08:15:04 0:27:03 2742161 99 0.00005715 none 232879 25.28148024 confirmed
Had about 500MH/s on the pool at the time, and I checked on my backup pools that I didn't get any shares submitted around that time.
With those 99 shares I'd expect about 15x that reward.. something gone wrong?
I also had a reduced reward on block 17662
17664 2013-04-24 08:15:04 0:27:03 2742161 99 0.00005715 none 232879 25.28148024 confirmed
Had about 500MH/s on the pool at the time, and I checked on my backup pools that I didn't get any shares submitted around that time.
With those 99 shares I'd expect about 15x that reward.. something gone wrong?
I also had a reduced reward on block 17662
Re: [8500 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
It's possible than when the site does come back online, the attackers could quickly log in as some of us, change our bitcoin address to be theirs, lower our payment thresholds and steal our confirmed coins...
Or am I being paranoid...
Or am I being paranoid...
No, even if they have your login details changing the payout address needs email confirmation, so unless the attacker(s) have access to your email too they can't change your payout address.
Re: [8500 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
Seems stratum is finally back up. Finally got a block pretty much as it came up! Annoyingly I was switching my miner back over from getwork to stratum.. probs down for 2 minutes so I probably missed out on a big chunk of the payout I would've got 
edit: spoke too soon.. lost connection to stratum again
edit: spoke too soon.. lost connection to stratum again
Re: [ANN] cudaMiner - a new litecoin mining application [Windows/Linux]
Getting about 48KH/s out of a Geforce 240GT.. although CPU usage seems pretty high.. using ~300% (4 cpus total).
Thanks for your efforts - shall keep an eye on it
Thanks for your efforts - shall keep an eye on it
Re: [ANN][LTC][BTC][NMC][Pool][PPLNS] - P2Pool.org - P2Pool Multi-Mining Pool
How long does it take for stats to be updated?
Have submitted at least 2 shares and they're not showing up on the page.
EDIT: just noticed you're still implementing it.. didn't read it properly
Have submitted at least 2 shares and they're not showing up on the page.
EDIT: just noticed you're still implementing it.. didn't read it properly
Re: [6000 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
Ah, Thought It Would Be something like that. I was more wondering how to tell but I guess it is on the website somewhere
Two easy options come to mind:
1. keep an eye on blockchain.info and take your miners down when you see Slush get a block.. it updates automatically and there should be a fairly short lag.
2. Keep refreshing the stastistics page until you see "Current round duration" reset.
Since 1 is less likely to cause issues for the pool.. I'd go with that
I find this system very convoluted. Clever but convoluted. Do other pools use systems like this?
The network has gained 50 PetaFLOPS in the last few weeks so I'm having a hard time telling what pool works best for me.
The network has gained 50 PetaFLOPS in the last few weeks so I'm having a hard time telling what pool works best for me.
Since most pools use systems which discourage hopping your best bet is to leave your miners on a single pool (or point each of your individual gpus/whatever at a selection of pools) and leave them alone. If you're that bothered by the variance you should pick a PPS (pays the same per share everytime) pool.. although in most cases you'll pay a higher fee to the pool for the privilege.
Re: [6000 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
So if I needed some planned downtime and I didn't want to mess up my shares, how would I best choose when to shut down the miner?
Right at the start of a round, i.e. when a block has just been found.
Hold the fuck up a minute! If my miners go down for 5 minutes I lose 50% of the work they have done for that block?
I got these things on Wifi Every time a big truck drives by they go out for 20 to 30 sec. What measurement of time triggers the "oh shit he's a pool hopper fuck him over" device?
I got these things on Wifi Every time a big truck drives by they go out for 20 to 30 sec. What measurement of time triggers the "oh shit he's a pool hopper fuck him over" device?
On a longer round, yes - the score is based on an exponential function with the round time as the exponent. It's scaled by a factor "1/C", but on longer rounds this will dominate - check your score under my account. More info found in the link given above ( https://bitcointalk.org/index.php?topic=1976.msg50002#msg50002 for your convenience). The longer the round, the more important staying connected (and submitting a share close to the point a block is found) becomes.
Re: [6000 GH/s] Slush's Pool (mining.bitcoin.cz); TX FEES + UserDiff; ASIC tested
The problem for me is: There was a block which took a long time to resolve, and my mining rig just self-restarted during the process due to my ISP problem, and the reward for that block became 1/7 of normal reward, this has never happened with other pools 
Have had a low reward on a long block from the same issue, 5 minutes downtime = half (or less) the reward I should've got.. always seems to be the case on a 3+ hour round as well :E.
The payment system treats you the same whether you were trying to pool hop or you just happened to disconnect at a bad time. In the long term it averages out because everytime it happens to someone who isn't you, you technically benefit from their lower reward in that instance.
I still don't understand the "Suggested difficulty" field. Dose it have some thing to do with long polling? If my connection is bad do I set it higher or lower?
It means that you aim for a share with more worth. E.g. a share at difficulty 1 is worth 1 share, 2 is worth 2 shares.. and so on. The difficulty refers to the target set for your hashes. At difficulty 2 there are half as many possible hashes which would contribute a share when compared to difficulty 1.
The reason it's useful is that with a very high hashrate (>2GH/s) you'll find so many difficulty 1 shares that it'll use up a lot of communication and it floods the server with requests.. so if you look for more difficult shares, you still (theoretically) get the same rewards, but the communication with the pool server is lower.
The assumption of a bubble is generally based on Paul Krugman's neo-keynesian fantasy world where people don't spend deflationary currencies. But people do - when their need for something in the immediate present surpasses the potential future value of the currency they will spend, and this is why we see record numbers of merchants signing up with bitpay. They're not signing up based on speculation, they're signing up because regular customers are emailing them and requesting that they accept bitcoins.
We don't need to pop the bitcoin bubble, we need to pop the fantasy bubble in these peoples' heads that convince them that bitcoin is a fad based on neo-keynesian "spend spend spend" bullshit. Bitcoin (or at least cryptocurrencies in general) are here to stay. And yes, it can be both a deflationary store of value and a currency. Perhaps it doesn't encourage unsustainable levels of consumption like the USD - the dollar is like a hot potato. I get rid of USD as quick as I can, either into deflationary alternatives or goods/services that I can't yet buy with deflationary decentralized currencies.
We don't need to pop the bitcoin bubble, we need to pop the fantasy bubble in these peoples' heads that convince them that bitcoin is a fad based on neo-keynesian "spend spend spend" bullshit. Bitcoin (or at least cryptocurrencies in general) are here to stay. And yes, it can be both a deflationary store of value and a currency. Perhaps it doesn't encourage unsustainable levels of consumption like the USD - the dollar is like a hot potato. I get rid of USD as quick as I can, either into deflationary alternatives or goods/services that I can't yet buy with deflationary decentralized currencies.
To play devil's advocate, how many of these are genuine requests because people want to spend bitcoin? How many are those with a vested interest trying to sustain the price growth with no real interest in actually spending their bitcoins?
Re: [22000 GH] BTC Guild - PPS/PPLNS with TxFees, Stratum+Vardiff ASIC Tested
Hi, I have a coupole of questions...
- Do we have an IRC for the Guild?
- How is the Stratum Address for Merged Mining with Namecoins? mergedmining.btcguild.com 8332 sounds like getwork.. ?
- Why isn't merged mining default, are there any shortcoming if I mine merged?
TY
- Do we have an IRC for the Guild?
- How is the Stratum Address for Merged Mining with Namecoins? mergedmining.btcguild.com 8332 sounds like getwork.. ?
- Why isn't merged mining default, are there any shortcoming if I mine merged?
TY
Since eleuthria might be sleeping (sounds like he needs it):
- #btcguild on irc.freenode.net
- There is no option of merged mining/namecoins using stratum, only via the getwork server.
- As far as I understand, because namecoins are considered fairly dead/out of fashion (the project hasn't been updated in a long time) and so there is little incentive to implement merged mining into the stratum protocol.