I have an AMD 8350 that was only about $150, but I need a water-cooler. It will overheat if I try to mine with it very hard. AMD chips do cost less than intel for the same amount of hash rate they can produce but heat is a very serious problem. I was looking at getting this one http://www.newegg.com/Product/Product.aspx?Item=N82E16835181029&cm_re=liquid_cooling_system-_-35-181-029-_-Product which is about $50 additional cost. If for $200 the intel i5 gets 70 khash/sec with the stock fan and the AMD 8350 gets 90 khash/sec with a liquid cooler, the AMD still gets more hash/$ but you definitely have to include that liquid cooler to the price. I also looked at power per khash: AMD 8350: 2.55 W/kh i5 4670: 0.83 W/kh Comparing the two rigs at the same condition at about 72kh/s (i5 between 70-75), AMD 8350: 253W, costs: $0.34/day, $10.2/mo, $61/6 mo i5: 125W: costs: 0.17/day, $5/mo, $30/mo I guess there would be some room to improve i5 hash I didn't try by having efficient cooling. If one can bear with the price difference, intel would be a nice one to go with for the long run imo. |
|
|
|
Hi mate, how using magi is fee tax there, wondering..  |
|
|
|
Hey guys, I'm aware this is a bit off topic...but since we are discussing about mining and so on - I'm looking for a motherboard for AMD CPU, AM3+ socket.
I'm using a AMD FX8320 and have real trouble with oc (now I have really shitty mobo).
Max 100 €.
Been googleing and so on, but I'd like to hear also from you guys, since I'm pretty sure many of you use AMD CPUs.
You can also send me PM if you don't think its appropriate discussing it here.
Thanks,
brgds
This is the one I've come across: ASRock 990FX Extreme9 AM3+ AMD 990FX + SB950 ( http://www.newegg.com/Product/Product.aspx?Item=N82E16813157358). It'll be advantageous along with GPU mining, typically 4 GPU I did. With AMD 8350 on this board, about 80 kh/s, no specific optimization. However, recently I found that AMD is kinda sh*t. After moving 8350 to another 'cheap' board, hash was down to about 50 kh/s. What's made me this feeling is that recently I used my laptop (with AMD) to compile few stuffs, which came out some sh*t, unportable to other machines/VB box (Illegal instructions), just waisted a lot time; while I set up a VB in my intel machine, everything just runs happily in all my home pcs. I'm not a tech in this side, can't figure why. For CPU mining, I came across MSI H81M-P33 V2 LGA 1150 Intel H81 + i5 which produces about 70kh/s, very impressive, importantly producing much less heats. The board is really cheap compared to AMD. Simply load a Linux os in a flash driver, no need of any expensive components except $200 bucks i5; ofc you can run few GPU too. @goldlabel, the game looks impressive, can't wait, I'll play it when I find time.  |
|
|
|
Joe, There's actually an issue with retrieving the version.txt file that's up on coinmagi.org - it's inaccessible via the Qt curl method because of the mod_security permissions for the server. Instead of sending the version information, the site spits back an HTML-based error message. The error is detailed in the latter half of a new pull request. The other half of the pull request makes version checking a little more robust. Since it wasn't immediately obvious to me why the client wasn't reporting that a new version was available, I figured I might as well tidy up some loose ends in the code. Previously, significantly lower program versions could trigger the Update warning. -- If you want to test the success of the version.txt retrieval, I also added a console printout in src/qt/updatecheck.cpp (line 40) for debugging. If the mod_security problem ends up giving you trouble, uncommenting that line as well as the iostream include on line 6 enables the printouts. It appears that the issue can probably be resolved by adding a few lines to ./.htaccess on the server, but that might not be ideal. I'm not entirely sure what the mod_security filtering does, but it might be important. Edit: You can also use http://starflakenights.net/crypto/magiversion/version.txt for an example of what should be going on. Not sure why my web space is more permissive than the official site. Changing the retrieval URL in the code makes everything work as intended. feldenthorne, I saw this also, I'll check out today once I get time or later, thanks. |
|
|
|
I had a few miners on the pool but as development stopped/stalled and "Mr. Spread" went AWOL the miners left the pool one after the other. (some) Spreadcoin (versions) includes a patch that "rolls" the payout address with every new mined block, so it looks like every block is mined by a new "miner" (or at least by a new address ). This was later also adapted/used by "Pharma" if you remember... Its a good mechanism to hide big solo miners / pools because in all block explorers etc. its not visible that actually one miner is taking all those blocks (however it IS visible if you know a few things which I won't disclose here now). Good to know about that; this space seems filled by floods which are driven by some kind of force. It would be quite understandable that every single new creation needs a thing in support of it's (quick) movement. Likely may people be too much keen on the consequence that matters. |
|
|
|
Hi You All, Very soon there will be a CSS or CSGO battle sponsor by XMG. You can win 1000 XMG with it, need i say its the best crypocoin out there? [img ]https://scontent-ams3-1.xx.fbcdn.net/hphotos-xpa1/v/t1.0-9/s720x720/12107243_879926388743017_501867203921381014_n.png?oh=6b871950d92dd9f5f87be0efc8a9f8ec&oe=56C14FE3[/img] Deadmatch  Really  ?! Please explain more ! Since I've big skills on these games (no, I'm kidding, I'm really bad  ), I can maybe win it ! Teamkill will be on so dont move when your close to me. Don't worry ! I've never money, so I won't be able to buy efficient weapons . Can you tell us more about it ? No sry not yet, but soon there will be more info about the game. I liked your creation in images how XMG / algo merged into the scene; thanks for all the stuffs, mate. I guess we can have all the images you made with a link in the OP. The game looks cool. |
|
|
|
i forgot to translate this important post into spanish, Thanks @111magic for reminding me. sometimes i want to post those important announcements into spanish thread at bitcoin*arden.tk but dunno why sometimes the web seems to be down for me, it happen to you too?
well i will add it tomorrow i think already done Take it at easy. |
|
|
|
NEW Wallets (v1.2.3.1) Official site, Sourceforge, Github release, Windows, Mac OS X, Linux
Thanks to lightsplasher to total balance function on the Qt interface.
Please make sure you have a backup of the wallet before launching the new app. All versions (Win, Mac, Linux) are with db4.8, and this will be the version we should use. There was a mistake in my linux machine which built the linux wallet with db5, and I made sure the linux version is a statically linked version (compiled by linux mint, it'll be fine with all debian based variants, but I'm unsure other linux version). Again please make a copy of wallet.dat before launching the new app.
Regarding the launch time of the current wallet, I believe this is not an issue of the db version, as I have used higher version in my linux box. I have the inclination of migrating latest bitcoin source code to magi, but this would be a long journey to go with.
@lightsplasher and all others, please feel free to make changes and submit merge request; I like to see changes to improve the Qt miner too if some one is interested. Currently I have tighten schedule on my personal stuffs, and I almost spent a week on the stuffs that's so expensive ; this thing is really time consuming. Here is the repo.
https://github.com/magi-project
|
|
|
|
testcc90cc KH/s-----23,657 XMG/Day----- 2,726.157 Wonder what they are testing Now 44,231 KH/s, that's not normal. This is a concern... smells like a gpus on it.. Maybe that explains why the hashrate has been so high recently =(
To get over this concern, two situations:
If it is a botnet, then solution is to go higher hashrate to much exceed botnet's hash, which quite needs GPU mining to come in; this is not what magi intends to be, I meant the high hashrate.
If it is a GPU miner, then solution is to go a different algo. Well this is tough. The said getting a new mechanism will lead to a possibility of hidden facts unless we're 100% sure about the solution and no one can tackle it. After a brief discussion with Wolf0 (mostly I was asking), to resist GPU, a reasonable amount of memory needs to be plugged into the algo, and it simply exhausts the memory of GPU. For this solution to work, CPU must have a significant amount of memory than GPU. However, the nowaday's technology has pushed even higher and cheap memory; huge amount of GPU won't be unlikely in the future, so it isn't absolutely safe with such an algo. There is still another concern, CPU working with higher amount of memory automatically disables the mining capability of those low-end / old pc, at least suppressing their hashing significantly; for example, Raspberry Pi will be out of the scene. This is very much contrast with magi's aim at low cost device mining. What I want is to get magi mining on cheap devices; I don't think we can get to anywhere and I am afraid we're just a regular coin without this purpose. So I'd hesitate to get on an algo like yescript or any others with a large number of memory. But I don't think this is a simple rejection, rather we'll consider once something better comes up.
Reading the above by far, you can see the solutions are very contradictory. You want "a cheap mining" like CPU 100% assurance (well nothing 100%), use a memory enhanced algo (unless there is approach other than memeory), which is actually not "cheap"; also there is the botnet concern. You want "GPU mining", surely no where cheap and no where cheap mining.
Except the above two, what I can see is the idea we have here, the variable PoW block.
http://xmg.makejar.com/xmg-per-day/
Let's take the amount of XMG generated on "17.10.2015" which is quite a normal day, 4 668.85913172 / 444 =10.5 XMG / block. The "18.10.2015" is quite much less than "17.10.2015" (the big miner wasn't online whole day). I might be wrong; what I've seen when the big hashrate online is mostly <1 XMG / block; if it stays a day, the total XMG would be quite less than 444 XMG (taking 444 blocks / a day, same as "17.10.2015"). It's sure that when a big hashrate coming online, the total amount of XMG will be reduced significantly. Let's say the big hash is 40 Mh/s and the rest of the network is 60 Mh/s, then what he gets is 444 * 0.4 * 0.012962 = $ 2.3 in a day (I believed this is quite a reasonable estimate, and actually this miner isn't in the pool for a day in which case the reward may drop down further). So this is $2.3 for a miner with 40 Mh/s.
Speaking of the profits, when the big miner isn't getting profits, the other miners (regular) are receiving loss too. The point is that the big miner has to carry the loss on his own, while the loss is much distributed among the regular miners, so individual miner won't worry too much about that in my opinion (say there are 1000 regular miners over one or few big miners).
This is my philosophy about the idea behind scene as I mentioned; and this reminds me an experiment in physics "Schrodinger's cat" in a sealed box with a flask of poison (https://en.wikipedia.org/wiki/Schr%C3%B6dinger%27s_cat); we don't know the cat is alive or dead unless we observe it by openning up the box; at the moment when we open up the box, we trigger a signal that may cause the poison released to kill the cat, so we don't know the dead cat was killed by the poison or the observer (indirectly). So a miner is wondering if he can catch up the whole block rewards; the difference here is that, once he comes along, he actually pull the trigger, and causes the collapse of the block rewards.
What happens is that this block reward adjustment process must take place real quick, and the regular miners must stay mining as normal (no compromise at all). And there is another condition, the normal miners must outnumber those weird ones significantly.
Another thing is about low hashrate which could cause 51% attack; this seems fine so far for magi, a PoW/PoS hybrid coin, and you know our PoS-II is safe. Either way, this is another point we need to take into consideration in order to improve this coin. Let me know any ideas you could have.
|
|
|
|
To force the mindset change I have an idea that could be benefit (or not) to Magi : why not open a pool that only confirmed Magi member could join ? That could spread the hashes a bit more and if we have a total pool power bigger than big miners, their rewards would be decreased a lot and maybe they will give up.
It seems an idea, however, there are other pools they can register and even they can use different IP. |
|
|
|
i think spreadcoin invented a technology that only allow solo mining
now if u combine that with a check to accept new blocks only for example once every hour from same address
u would make GPU mining senceless he would just mine to many blocks and all beside the first each hour would be rejected by network
nice sideeffect u would have a lot nodes in network because mining would always required to have a node active too
Hi cryptonit, thanks for being around for advices; I've been being out of the time on the things we discussed, really tighten on the things. I'm wondering how solo mining blocks gpu mining, not quite getting this "u would make GPU mining senceless he would just mine to many blocks and all beside the first each hour would be rejected by network". I may read through their page. One thing I now very much concern is that, every time we create a new mechanism that is not working the way we're 100% sure, it creates hidden facts. Actually that "technology" was already circumvented several times, I have (had) a pool up for SPR but closed it due to no interest: https://spr.suprnova.cc I might reopen it if interest comes back. You can mine every block to a new address without problems. Hi ocminer, people have no interest to the pool, is that because they like to sole mining? Also, could provide deail information about "mine every block to a new address"? |
|
|
|
[img ]https://www.filepicker.io/api/file/0VYh3ak6Ther5FZtFwWs[/img]
Nice pic. |
|
|
|
Thanks for the collection of info, mate, I have changed / updated the OP (nearly took me a day lol). I guess some info still missing, let me know that. |
|
|
|
it's not a concern, Just want to make fun... testcc90cc KH/s-----23,657 XMG/Day----- 2,726.157 Wonder what they are testing Now 44,231 KH/s, that's not normal. This is a concern... Please do so. I am very interested to know about how many coins can be mined in a day. That will be the valuable data to me. |
|
|
|
testcc90cc KH/s-----23,657 XMG/Day----- 2,726.157 Wonder what they are testing Now 44,231 KH/s, that's not normal. This is a concern... |
|
|
|
Some post worthwhile should be translated always into more languages, spanish translation thread have this one now HEREofc i will update other spanish threads at other forums too Thanks friend. |
|
|
|
i made a vps miner for fun..
but i cannot understand why people mine this?
i cannot get any profit from this: time to setup miner cost me more than it ever makes coins to me.
vps cost more than magi makes.
cpu uses more power than magi makes you??
explain me please?
pure botnet coin?
It can be easily be profitable. The problem is that the bigger the hashrate is the lower the block reward is. A lot of people don't understand this or simply don't care so they mine with a few thousand kh/s. They could earn more by diminishing their hashrate but they're not really clever so they don't ! This needs a change of mindset. With Bitcoin it was the higher the hashrate the better the reward. With the coins launched after BTC it was almost always the same. This mindset change will take time! yes, dont mine only buy? let else people waste time for mining? I am glad that you asked this question here, as if magi is not profitable, magi is not the only one, obviously. I commented here: https://bitcointalk.org/index.php?topic=735170.msg12702683#msg12702683I'd recall back to 2013, there was a time mining bitcoin is still profitable, but you have to be smart and you got to run faster than the other people. The golden time passed by rather quick. Nowaday's mining is absolutely a side effect of this crypto thing Satoshi never expect. People definitely should blame the mining thing for the global warming up at one day, if it is not significant, it's not a trivial one either. The use of devices exclusively for mining (not supposed to be for this purpose) is another waste; ok, bit off topic. Your question actually is that why people in this space should mine crypto, do you mine bitcoin and no buying? |
|
|
|
Thanks for your help guys.
My miner is running since 12 hours now. My calclulation says it's doing 7-8 XMG / Day at 70 KH/s.
That's not as much as I expected it will be.
That would mean 440 KH/s would only make about 44 XMG / Day which is worth about ~ 0,50 Euro.
Energy consumtion would be 2,50 €.
That would mean that mining XMG is far from beeing profitable.
Even my notebook with 22 nm technology CPU has a power consumtion of 60 Watt at full load. Which means power costs of abou 0,30 Euro / Day. And has got an outcome of ~7,5 XMG which is worth 0,084 Euro.
So how could mining XMG be profitable?
Or do you guys all hope for a rising course?
Speaking of profits, I think all crypto mining is unprofitable. You can compare with any mining including ASIC, GPU, CPU, if you can find a one with nice profits, I'd be interested to know about that, especially CPU coins which are good for profits. The truth is that, if there are profits from mining, people will make it unprofitable finally. Typically people mine coins and sell later. In a case that the strategy of yours that stopping mining is taken, because of the exit of miners, the mining goes towards profitable. You see, this is a contradiction. When it gets profitable, people mining it, and then towards the other way around. How unprofitable it is depends on how strong of confidence in this system (for example price rising in one day); the stronger in it, the less in the profits. I much like casual mining without too much concern on the profits, particularly in a daily base calculation, that would be stressful. |
|
|
|
One more question:
Is there any possibility to increase the intensity of the miner?
At the moment I use cpuminer v2 with the following instructions: minerd-x64-corei7 -a m7mhash -o stratum+tcp://xmg.suprnova.cc:7128 -u username.workername -p password.
It's using all CPU Cores at 100 % but maybe there is a possibility to stress them a bit more?
CPU is a i7 3630QM Ivy Bridge
You can try overclocking (hardware) but that won't be good to the CPU imo. |
|
|
|
As mentioned by fragilefungi, a convenient miner (ideally click & go mining) is a much need, and I believe that will get us more miners without painful compiling (notice if you compile from the source code, you might get a specific optimized miner only for your pc which might be efficient than the following compiled ones). Also, I compiled the miner with "-e" flag to adjust the CPU efficiency from inside the miner. https://github.com/magi-project/m-cpuminer-qthttps://github.com/magi-project/m-cpuminer-v2I have forked the code of ScryptMiner by Matoking ( https://github.com/Matoking/ScryptMiner-GUI), which is based on Qt. This is very advantageous; we can add codes later on to adapt to the special block reward of magi, and design a smart miner with strategy mining. This will be a one in the to-do list. Here are the compiled binaries: http://coinmagi.org/files/m-cpuminer/m-cpuminer-qt-64-win.ziphttp://coinmagi.org/files/m-cpuminer/m-cpuminer-qt-32-win.ziphttp://coinmagi.org/files/m-cpuminer/m-cpuminer-qt-64-ubuntu.tar.gzhttp://coinmagi.org/files/m-cpuminer/m-cpuminer-qt-macosx-10.7.dmgThere are also minerd available here: http://coinmagi.org/files/m-cpuminer: http://coinmagi.org/files/m-cpuminer/m-minerd-64-win.ziphttp://coinmagi.org/files/m-cpuminer/m-minerd-32-win.ziphttp://coinmagi.org/files/m-cpuminer/m-minerd-64-ubuntu.tar.gzhttp://coinmagi.org/files/m-cpuminer/m-minerd-64-macosx.zipThese are how the Qt miner looks like:   Notice "-e 60" in block "Additional parameters" in the above image adjusts the CPU mining usage, e.g., 60%. All the parameters are automatically saved into m-cpuminer.conf, in the same folder. The GUI basically makes use of the miner under folder "miner". If you can compile with a higher efficiency, please let me know. Please make sure you modify the pool parameters, or you will mine through my account. This GUI is kinda similar to NeedIfFindIt's Sweet Spot http://xmg.makejar.com/, which will be better than this I believe; the advantage is that we can make it available for mac os and linux too, and most of other platforms because of Qt base. The linux version is statically linked, if you run into issues, let me know. I very much like the miner being compiled in low end/old system/pc. I believe magi's mining is useful in this aspect. I will get some details for the PoW. I do agree one mining a coin cares about the return, while I like the idea you place the miner with limited CPU usage, you can forget it (say 10% CPU): think about how many computers which run idle in the offices or other places for overnight, at least it's a waste of power. I am not encouraging you use an office PC for mining, but running a limited miner in a daily computer is quite doable, and that, from another point of view, is to save energy as they maintain a coin system provided that magi is useful at some extent and this purpose will be meaningful as well. |
|
|
|
|
Show Posts
