Ethereum GPU mining?

[Madpacket]

Don't they use the same distribution model and both have s 1% fee if you're using stratum?



I bought with the intention of selling the cards if prices don't move, after rebates and Hitman the cards came out to $245 or so a pop, so I should be okay on that end. I anticipate them moving personally.

They both use PPLNS but there are zero fees at ethermine as long as you let the payouts stay at one Ether. Also pay for orphaned blocks which I'm not sure if Nanopool does.

Also that's the right attitude about mining. Buying cards isn't that risky if you can get them at a good price. We still have a few solid months to mine before Polaris but I'll give you all a tip. Once Polaris is about to be released almost everyone unloads their cards driving the price down to stupid levels. You're better off keeping most of your cards through the dip and selling 6 months after. This may seem counter intuitive but the used prices will eventually shoot up and in the mean-time you could be mining provided your energy costs are reasonable.

 

[EightySix Four]

They both use PPLNS but there are zero fees at ethermine as long as you let the payouts stay at one Ether. Also pay for orphaned blocks which I'm not sure if Nanopool does.

Good call, missed the payout fees.

Also that's the right attitude about mining. Buying cards isn't that risky if you can get them at a good price. We still have a few solid months to mine before Polaris but I'll give you all a tip. Once Polaris is about to be released almost everyone unloads their cards driving the price down to stupid levels. You're better off keeping most of your cards through the dip and selling 6 months after. This may seem counter intuitive but the used prices will eventually shoot up and in the mean-time you could be mining provided your energy costs are reasonable.

Thanks for the tip, I probably would have dumped around Polaris' release but nice to know the prices rebound. Depending on the Polaris cards availability and pricing, I may add a few of those and remove these when the time comes... Or pick-up an HTC Vive and actually use the 380's

 

[DrMrLordX]

Re difficulty, without the price going up in lockstep with difficulty we have maybe a 2 months left of profitable mining. This is why I warned people about cutting off buying additional cards as it becomes self defeating. We're all reducing profits for each other by adding more cards. This is why mining is always a temporary endeavor.

Ether has gone up about $1.35 since late March so it is improving somewhat.

 

[SimianR]

You must be calling the wrong conf file from within the batch file. At least that's what it looks like.


Re difficulty, without the price going up in lockstep with difficulty we have maybe a 2 months left of profitable mining. This is why I warned people about cutting off buying additional cards as it becomes self defeating. We're all reducing profits for each other by adding more cards. This is why mining is always a temporary endeavor.

I agree with this to some extent. I get that for some people it's a hobby and such, but I feel like going overboard and setting up multiple rigs and rushing out to buy as many GPU's as you can get your hands on just results in the difficulty going up, the next guy doing the same thing and a vicious cycle of less profits for everyone (besides the few with absolutely ridiculous setups). But I get if you stumble across a deal on some GPU's then it might make sense financially. Of course no one can see the future, and we're only looking at the current price when figuring out "profitability", I'm curious what will happen to the price when it eventually switches over to Proof of Stake. I suppose there's no guarantee that the price will go up. There were more than a few coins (BlackCoin) for example that everyone promised would be very valuable once they switched to Proof of Stake but that never really panned out besides a large bump in price before a quick crash.

 

[DrMrLordX]

In theory, Proof of Stake should drive up the cost if people want to increase their stakes to participate in the PoS "mining". If there's no interest in the currency then it will crash like anything else. All the alt coins will benefit if Bitcoin takes a dump (moreso than it has already) due to the problems it is just starting to experience now.

 

[wege12]

Would you guys consider buying two R9 290's at $250 a piece, a good investment at this time if electricity is free?

 

[DrMrLordX]

What's the hash rate on those? I've heard people say the 290x is good for 25 MH/s, so I wouldn't think the 290 would be too far behind.

At that price, I'm not sure it would be a great deal.

 

[thilanliyan]

What's the hash rate on those? I've heard people say the 290x is good for 25 MH/s, so I wouldn't think the 290 would be too far behind.

At that price, I'm not sure it would be a great deal.

I'm getting 30mhs on a 290 at 1100/1250.

 

[DrMrLordX]

Really? That's pretty good. Okay, then maybe $250 would be a good price for em. What brand 290 do you have, and can you undervolt it at those speeds?

 

[beginner99]

Would you guys consider buying two R9 290's at $250 a piece, a good investment at this time if electricity is free?

Just for mining no. And since you already have a fury x, that would be the case. Assuming you had a old GPU and needed a GPU upgrade anyway it would be a different story. Plus 850w psu is probably too little for a fury x and 2 290s.

My personal opinion is if you want to cash out big with ethereum, just buy some (for 1k-10k $) and wait couple of years.

 

[3DVagabond]

Just for mining no. And since you already have a fury x, that would be the case. Assuming you had a old GPU and needed a GPU upgrade anyway it would be a different story. Plus 850w psu is probably too little for a fury x and 2 290s.

My personal opinion is if you want to cash out big with ethereum, just buy some (for 1k-10k $) and wait couple of years.

If a $250 card can earn you $500 worth of coins, which you can still hold on to a couple of years, then how is that a bad investment? Especially since you can still sell the cards and buy more Ethereum, if you are so inclined.

 

[thilanliyan]

Really? That's pretty good. Okay, then maybe $250 would be a good price for em. What brand 290 do you have, and can you undervolt it at those speeds?

I have 2 reference 290s and no they run at stock volts...can't undervolt at those speeds.

 

[DrMrLordX]

I have 2 reference 290s and no they run at stock volts...can't undervolt at those speeds.

Okay. What kind of power consumption are you getting at those speeds?

Also, has anyone here monitored bandwidth usage by geth.exe and ethminer.exe? According to Win10, my periodic mining over the last few weeks (started March 21st) has produced 5.34 Gb of traffic for geth.exe and 2.19 Gb of traffic for ethminer.exe . There's no breakdown of whether that's upstream, downstream, or combined (I'm thinking combined).

I haven't mined out that many ether yet, so that seems like a heck of a lot of traffic per ether generated. Maybe that's a consequence of downloading/redownloading the DAG?

 

[Madpacket]

Okay. What kind of power consumption are you getting at those speeds?

Also, has anyone here monitored bandwidth usage by geth.exe and ethminer.exe? According to Win10, my periodic mining over the last few weeks (started March 21st) has produced 5.34 Gb of traffic for geth.exe and 2.19 Gb of traffic for ethminer.exe . There's no breakdown of whether that's upstream, downstream, or combined (I'm thinking combined).



I haven't mined out that many ether yet, so that seems like a heck of a lot of traffic per ether generated. Maybe that's a consequence of downloading/redownloading the DAG?

It's the DAG files..

For proof look at the download locations and you'll see multiple copies. Delete the old ones to free up space. The more you tweak and start/stop your miners the more often DAG files are downloaded wasting bandwidth.

290's with Hynix memory are within spitting distance of a 390 when clock speeds are matched. The reference cards have excellent Volterra VRM's which allow ample undervolting and because they dump heat out the back you can stack them without too much worry. Only issue is noise but those blower fans last forever at high speeds as long as you air dust them every 3 months or so.

I know people around here hate reference 290's but they're arguably better than 390's for mining with weak fans if you can get them for a good price. Only concern is running multiple cards on the same host, the memory requirements go up but I think this can be solved by running multiple instances of miners.

 

[DrMrLordX]

For point-of-reference, it looks like ethminer.exe has used 40 Mb of bandwidth over the last 2.5 hours on my little 50 MH/s machine. Geth.exe reports no usage.

So, there is some traffic, but it isn't that major.

Only concern is running multiple cards on the same host, the memory requirements go up but I think this can be solved by running multiple instances of miners.

System memory requirements, or card memory requirements?

edit: it looks like a lot of 290s are voltage-locked. Of course, that's not a big issue if you are willing to flash the card BIOS anyway, which is . . . more-or-less necessary for using them in Linux land, what with there being so few software tools to modify card voltage/clockspeed/etc.

In comparison, these MSI 390s are unlocked in every which way, making it very easy to deal with them under Windows. But they're a good bit more expensive.

 

[VeryCharBroiled]

For point-of-reference, it looks like ethminer.exe has used 40 Mb of bandwidth over the last 2.5 hours on my little 50 MH/s machine. Geth.exe reports no usage.

So, there is some traffic, but it isn't that major.

is there any way to minimize bandwidth? run a stratum proxy, command line options, a particular pool etc?

single r9 390 here ~30 mhs, but on wireless internet (out in the boonies, no dsl/cable).

 

[DrMrLordX]

I have no idea, though I too wouldn't mind bringing down the bandwidth usage. The easiest way to reduce bandwidth usage is to minimize redownloading the DAG.

 

[VeryCharBroiled]

I have no idea, though I too wouldn't mind bringing down the bandwidth usage. The easiest way to reduce bandwidth usage is to minimize redownloading the DAG.

i thought the dag file was generated by the miner based on the block height? also when restarting within a couple minutes it seems to start hashing right away as opposed to the initial run when it took a few minutes to build the dag.

lots to learn i see. but learning and tweaking is the fun part anyway.

minimizing restarts seems easy enough though.

 

[Madpacket]

I haven't done any comparisons between stratum proxy vs GET mining bandwidth wise. Outside of that and repeated DAG file downloads I imagine the number of milliseconds set to pull down work units of data may play a role in bandwidth usage. I doubt it's a lot though.

Oh and no reference 290/290X's are voltage locked. The only gamble is getting a card with Elpedia memory which will lose you a Mh or two at the same clock speeds with a card with Hynix. All 390's I've tested so far are voltage unlocked and come with Hynix but that extra RAM needs more wattage to run so it's a bit of a tradeoff.

Also reference 290's (especially the early ones) have a good chance of unlocking extra shaders which would make up for cards with Elpedia and be a bonus for Hynix boards. All that being said I wouldn't spend more than 200 on a used 290, especially a non reference model.

 

[EightySix Four]

Over ~24 hours now I am seeing 3-5MH/s lower effective rates using Ethermine than I was on Nanopool. Going to give it another day and see if it continues.

 

[DrMrLordX]

i thought the dag file was generated by the miner based on the block height? also when restarting within a couple minutes it seems to start hashing right away as opposed to the initial run when it took a few minutes to build the dag.

It depends on how it restarts. If you delete the DAG file(s) then it has to go through the process of rebuilding the DAG. Or at least, that's what I've observed.

Near as I can tell, ethminer doesn't report the bandwidth usage under Win10 for DAG generation, but geth.exe does. I've maybe rebuilt 4-5 times and that has resulted in ~5.7 Gb in traffic.

ethminer itself is producing ~360 Mb/day in traffic on my 50 MH/s machine. I was mining at around 2.5 MH/s for awhile (lulz), then I went up to ~18 MH/s, and now I'm up to 50 or so. I've been mining for 13 days, so if it had been 360 MB/day throughout the entire run, then the total traffic would have been ~4.7 Gb, rather than the 2.31 Gb reported by Win10 (downtime notwithstanding). It's pretty clear that increasing my hash rate is raising the bandwidth usage. Doing this on a metered connection could get kinda ouchy. It could also hurt people with bandwidth caps.

 

[VeryCharBroiled]

ethminer itself is producing ~360 Mb/day in traffic on my 50 MH/s machine.

thanks.

Ill have to grab something to monitor the connection. Im used to cgminer and its stratum connection using only a few kb/s bandwidth on btc mining, so this is way different.

Ill setup a stratum proxy at some point and see what happens. new to eth mining.. mining it mainly to offset the price of the r9 390

 

[beginner99]

If a $250 card can earn you $500 worth of coins, which you can still hold on to a couple of years, then how is that a bad investment? Especially since you can still sell the cards and buy more Ethereum, if you are so inclined.

Because he would probably also need an new PSU and maybe even mobo. he already has a card so he needs 3 Pcie 16x slots (at least 16x size, not speed).

Mining is fine to pay off a card you own anyway. Else if you want to make bank, just invest some money ($1k+) and wait.

 

[thilanliyan]

Okay. What kind of power consumption are you getting at those speeds?

I am using about 575w at the wall. That is with 4 drives, 6 fans, 1 water pump, and the system listed in my sig, so I'm guessing I'm pulling about 500w for the two cards at the wall.
If I go -50mV (at stock 947/1250) I use about 100w less total.

 

[Madpacket]

Over ~24 hours now I am seeing 3-5MH/s lower effective rates using Ethermine than I was on Nanopool. Going to give it another day and see if it continues.

I switched all miners over the weekened (except one 380 card) and appear to hashing around the same rates given the variability of PPLNS. It's still a little early to tell but so far it seems kind of a wash.

I like that I'm paying 0 fees so I'll stick with Ethermine for a while longer and continue to monitor it. It's easy enough to switch back and forth.

Perhaps in your case it's a latency to server thing? Maybe ping both servers from each pool (whatever server is closest to your area). I'm not sure how much latency affects hashrates but when I switched over to a closer server at Nanopool it appeared my hashrates were more stable. Again with PPLNS it's difficult to gauge and compare.