|
1800 pps for $80/mo? That seems...a little high.
|
|
|
|
I tried a few, actually had one hang on boot (which in DigitalOcean terminology means 'Active', yep, they charged me the 2 hours it was stuck...), when finally running the 4 had completely different pps. One got @ 14K, the others between 6k and 10k. None of them hit their full potential. Ran them for a few hours, found a couple blocks but that was back a few days. I think the 8 core droplets are/were the best bang for the buck, you could pretty much get an even pps distribution through those ones.
This...I have easily racked up five bucks in charges for instances that either never worked, or spent hours and hours hung executing processes. Kind of BS for a service that should have 99.9% uptime. |
|
|
|
Anyone rock linode and want me to use their referral code? Thought this might be a good spot to check.
Depends how much you're going to spend  Enough to net you that sweet, sweet $20, at the very least, but alas, you were late to the ball. |
|
|
|
you guys are a funny bunch. I have already noted I can access primecoin.conf
That was the whole point, because I have made the edit, and it still comes up.
The information is all available on ubu's sites, so no I don't need the professors teaching me linux.
all I am asking is for ideas to overcome it
Seems like you may have changed rpcuser/password while daemon was still running, you could try killing it with "killall primecoind" and give it a second to close before running it again Thanks. I have tried that, but it still has the same problem. Says the info is incorrect. I think I was using chemists build before this one. Problem is still open as I am working to find a solution. I don't think daemon is dieing, I am trying to kill the S.O.B! Did you write a script to auto start it? If so, why not remove it and just reboot the instance? |
|
|
|
you guys are a funny bunch. I have already noted I can access primecoin.conf
That was the whole point, because I have made the edit, and it still comes up.
The information is all available on ubu's sites, so no I don't need the professors teaching me linux.
all I am asking is for ideas to overcome it
Seems like you may have changed rpcuser/password while daemon was still running, you could try killing it with "killall primecoind" and give it a second to close before running it again See, that's what someone who know's how to use linux would tell you to do. |
|
|
|
you guys are a funny bunch. I have already noted I can access primecoin.conf
That was the whole point, because I have made the edit, and it still comes up.
But did you edit primecoin.conf while primecoind was running? Could also just hard reboot, but I have no idea which build you'd end up running or how linux delineates that kind of stuff. e: dude, I am making fun of myself. I had never used linux prior to firing up these DO droplets, but now I think I actually know the solution to your problem because I encountered it myself. Old primcoind wants old primecoin.conf. |
|
|
|
Hm, I went for a reinstall for my server and it seems like there is some old data populating back. I have used rm -f primecoin-hp/src and rm --recursive primecoin-hp/src
However, when I reinstall and use these commands, /primecoin-hp/src# primecoind stop
error: incorrect rpcuser or rpcpassword (authorization failed)
primecoind --daemon Error: Cannot obtain a lock on data directory /root/.primecoin. Primecoin is probably already running. I used these commands https://bitbucket.org/mikaelh/primecoin-hpmake -f makefile.unix USE_UPNP=- mkdir ~/.primecoin echo "rpcuser=USER rpcpassword=PASSWORD gen=1" > ~/.primecoin/primecoin.conf sudo mv -f primecoind /usr/local/bin/. primecoind --daemon watch 'primecoind listtransactions & primecoind getmininginfo' Any ideas how to fix the rpcuser problem? I checked sudo nano ~/.primecoin/primecoin.conf and it looks all fine. Edit your primecoin.conf file back to what it was when you started the daemon, then terminate it. Probably old primecoind running. That worked for me when I ran into this. Look guys! You're teaching me linux! Aren't you proud? |
|
|
|
|
Anyone rock linode and want me to use their referral code? Thought this might be a good spot to check.
|
|
|
|
ok, out of curiousity. I'f I were to lease one of their 24core, 96gb ram servers for 24hrs($33.85), would there be a point? I don't know how much processing power it takes to find primes at this point, but I'm thinking it might be an interesting experiment.
Honestly, I have flipped my views on DO the more I play around with their service. It sort of blows. It seems like every time you spawn an instance, it's a total crapshoot as to whether you will get what is promised. I've also had a bunch of instances hang for ridiculous lengths of time on power off commands, and had two freeze altogether that I just told their engineer to destroy (shared wallet ftw). They have been super chill about this onslaught of miners, and they deserve maaaaaaad credit for that, but it seems fucked to me that you could open two of the same instances, even in the same locale, and easily have one running ten times the speed of the other. Amsterdam is running about double the speed of the US servers right now for me. I know PPS is not the end-all indicator, but it's something, and I suspect something is seriously amiss with an 8-core instance runs slower than single cores for prolonged periods. I have seen it time and time again while playing with builds and resizing. I'd say maybe 20% of my droplets will just be fucked and hobbled upon creation, and I will have to try again and hope for a better roll of the die. I finally got some multi-core instances in Amsterdam that aren't totally crippled and showed performance increases that corresponded to the increase in cores. Hopefully it's a sign of improvement. Maybe we are just totally killing them, but then why the hell wouldn't they just kick some of the heavier abusers and stop letting their respectable customers suffer at our hands? I think it's likely they are revelling in the numbers to flash at their investors and suit dummies, and all the while just pushing lackluster service to all their users. Whatever. It's cheap and making us unsustainable money, I guess. edit, forgot to answer the question: As far as finding blocks...still seems to be a combination of computational power and luck. My PPS has increased over tenfold in the past maybe five days, and my rate of block finding hasn't, if that's an indicator. Some people are running way more than me and have been for a week and are still high and dry. Day two, I hopped on with only my laptop as I was out of town and was snatching a block every few hours. Slowly increased with the improved miners and snagging better instances, then I went from micro to 2 cpu instances in amsterdam, now sitting at like, 20k PPS and have stayed above a block/24hrs thus far. If the 24-core instance ran as promised/equal to 24 micro instances, you could get maybe 30k pps...but who knows what that even means? Also, 24 micro instances would cost $.17/hr. Depends on how much you like to gamble, and how much you think XPM will be worth when you want to cash out. It would be a safer bet to buy some. Block reward is like, 12.2 something, my first block was 22. something. |
|
|
|
On testnet, I found that a sievesize=6144000 outperformed every other sievesize I tried (incl. 6000000 & 6246400) by 30% using an i7.
However, when i try to use this sievesize mining live, h5 crashes as soon as mining starts (every time). It's unfortunate because it showed such promise on testnet.
Blocks were coming in 4-5 chunks at a time steadily, even hit 6 blocks in 1 minute (no other setting hit above 4 blocks/min).
same here, 6144000 crashed h5 as soon as it attemps to mine on a i7 3770k with 16gb of ram. Did you get better results on testnet with 4M than 1M ? +1 Works on hp4, but no clue as far as efficiency. I have my miners running all over the place for sievesize, figuring I will play the board until consensus is reached. I'd have to figure out how to get the logs, and then figure out what they were running when they found the blocks. Sounds like a lot of work. Testnet once VPS mining is a total waste of time. I guess I am maintaining my lucky streak, because the VPS thing is still disgustingly profitable, at least at current prices. I think I am going to sit on these and either ride them to the ground or get me a bitchin' jon boat with a surface drive motor. We shall see what the future holds. |
|
|
|
|
Hello good sirs and madams of bitcointalk,
I am signing up for linode, and figured I'd see if anyone has a referral code they want my to use. Get yourself some free shit!
|
|
|
|
What was the magnitude and range of the primes being enlisted? (not the chain but the consecutive primes) The primes should have been too big, because Dirichlet theorem estimates a much bigger number than 11... And I suspect that the primes being found for the coin shouldn't be too big or too small for the blockchain efficiency sake, am I correct? (and if so, how it will help prime studies in continuation?!) And if it won't stop finding primes and will continue forever, applying Moore's law, finding new blocks and coins, to compensate mining expenses, minimize transaction fees and inflation, then tell me, who is ready enough to invest on some stock with infinite shares?!  The algorithm doesn't search for new primes, it is finding prime chains. Numbers are in the order of ~200-300 digits currently, if my understanding is correct. They've already discovered quite a few new records for both chain length and integer length. There is a dynamic block reward that corresponds to difficulty, so the notion is that it will reach a balance between inflation and network strength without blowing out the stability. Whether that actually works remains to be seen. Either way, I think it is a novel take on cryptocurrency and I am willing to keep an eye on it. These blocks I stacked don't hurt my interest levels, either. Read the whitepaper, since you seem like you understand some math. It might help your B-list trolling. |
|
|
|
|
Goddamn...I finally snagged some Amsterdam servers. They blow SF and NY out of the water!
|
|
|
|
If you are a paying customer and are interested in them opening a new tier for high CPU, low disk and memory, do send them a note. I mentioned it to them, but I'm only one person.
Support referred me here to request features: http://digitalocean.uservoice.com/forums/136585-digital-oceanLet's descend on the DO forums and make our demands heard! |
|
|
|
Just wanted to say great miner! I feel like this is probably the most optimized miner on the market right now. Looking over the source code of the original miner and yours I think I see a way to make the miner (potentially much) faster from a mathematical standpoint rather than a code-optimization standpoint. My optimization centers around the Primorial and the loop that occurs in main.cpp on line 4622. Before getting into the code, it is important to realize why a primorial is helpful (if you already understand this, skip this paragraph). With numbers on the order of 2 256 there is about a 1 in 177 chance that a random number is prime. If you select 8 random numbers, then, the odds of all of them being prime is about 1 in 1 quintillion--impractically low. If you limit your search to only odd numbers, though, the odds shoot up tremendously. Further limiting your search to numbers not divisible by 3, 5, 7, and so on can cause the odds of finding a prime number to become much, much better. If the hash value is divisible by lots of these small numbers then multiples of that hash ± 1 will not be divisible by any of those numbers. Thus, it is convenient to use a hash that is already a multiple of 2 * 3 * 5 * 7 * ... as it will produce far more primes than another hash. As I understand the aforementioned loop, the program is searching for a hash that is divisible by a primorial. Each iteration of this loop requires a hash to be generated as it increments the nonce value. In the present form the primorail is of degree 7 (line 4579: static const unsigned int nPrimorialHashFactor = 7). I suspect that this value is carefully chosen and it's probably ideal with the way that it is written. However, I think there's an additional step that can be added to make the process much faster. Increasing the degree of the primorial is incredibly expensive as it gets larger, since adding the 8th prime requires 19 times as many hashes to be checked; the 9th prime requires 23 times as many, and so on. There is another way, though. Prime origins are required to be in the form O = H * N where O is the origin, H is the hash, and N is any integer; H is selected to be divisible by p#7 (the primorial shown above). If we extend this to O = H * N * P 2 where P 2 is 19 * 23 * 29 * ... * 51--a product of primes starting after the last prime used in the existing search--then the checked origin is still valid (an integer times an integer is still an integer). This grants the benefits of searching with a higher degree primorial while not requiring a longer search for a working hash. Nothing is free, though, as this method inflates the size of the primes to be checked. If the fast modular exponentiation is implemented through the same method as is used on the Fermat Primality Test Wikipedia page then the algorithmic efficiency is O(log 2(n) * log(log(n)) * log(log(log(n))) ). There should be some sweet spot of how large of a primorial to use where the increased frequency of primes more than offsets the extra time required for the fast modular exponentiation. It's possible that the sweet spot is 0 extra primes, but I think it's worth looking into. Well, sir, you've managed to impress me. I am a biologist...math is one fish two fish, red fish blue fish. |
|
|
|
How much did you gain rethaw 18000 with your referal ? oO At 10$ the referal that made your day ! I wish. Trying to resolve some issues regarding this still, but I can tell you the majority of my referrals haven't gone past the $10 promo. I deposited $20 last night so you could get yo' scrilla after reading their referral program page. I don't know what the fuck I will do with it since they aren't raising my limit further, but I just don't want them thinking the whole crypto community is full of douchebags and I have already made a pile of loot, so I'll support them. You guys should do the same. Thanks, but not really necessary. I have suggested they open up a high CPU instance with modest memory for Primecoin miners. If a pooled mining protocol comes along I expect they will get a lot of new customers. Eh, I figure it gives you a nice little bonus and I am more than happy to support them. I will just have to bone up my linux skills and figure out a good use for them! |
|
|
|
How much did you gain rethaw 18000 with your referal ? oO At 10$ the referal that made your day ! I wish. Trying to resolve some issues regarding this still, but I can tell you the majority of my referrals haven't gone past the $10 promo. I deposited $20 last night so you could get yo' scrilla after reading their referral program page. I don't know what the fuck I will do with it since they aren't raising my limit further, but I just don't want them thinking the whole crypto community is full of douchebags and I have already made a pile of loot, so I'll support them. You guys should do the same. |
|
|
|
With the price now 400% up on the time that I wrote that I would buy anything under .003, and that you were all stupid for selling them at that price, I thought it would be a good time to re-quote this post  I didn't buy much at that price (a few BTC), and what I did buy I ended up selling off at .01 for a happy 3-4x profit. I'll be keeping what I have mined for a little longer. Thank you - PTT Nice I bought some at $0.32 and sold for $0.56 but kept the profit in coins which are now worth double I'm gona keep them till I see 0.05 or never might use them hopefully. what if they hit 0.1? they will if no gpu miner apear or big bug in prime codes... I mean it's only logical. Like, ten minutes ago, I wasn't sneezing. I sneezed just now. That is infinite growth of a series right there, so I imagine that I will only exist as a sneeze in the form of pure energy travelling at the speed of light in another ten minutes. Extrapolation! |
|
|
|
The mining itself seems very likely to turn into a speculative bubble. lets speculate as to whether leaving our machines on another minute, another hour, another day etc etc will end up profitable once we really do happen upon another block...
The exchange rate increasing will feed that...
Intermittent reinforcement is so insidious...
-MarkM-
I don't care how much the network hits me, as long as it gives me a treat every once in a while. I will still love it. People keep telling me to leave the network, but they don't understand how stressed the network gets with proof of work, and then there are all the little instances screaming at him and clamoring for validation. How would they take it if I left the network? How would I take care of them without the network? Really, it's my fault for burning the roast anyways. I don't even deserve the measly blocks it does give me, but it's a good, kind network that cares about me. |
|
|
|
if this carries on the price of XPM is really going up.....
Oh no, if that happens ordinary PCs will be able to go how many days/weeks without finding a block and still turn a profit?!?! -MarkM- Right? I hopped in on day two or three and pulled down five on my ultrabook that pulls 45w max. I am pretty sure I am in the black for a long while, and that's not sustainable. Coin has merit, but this is a speculative bubble. |
|
|
|
|
Show Posts
