|
Folks buying hash from Nicehash - I've had occasional peaks above 10,000 back when it was profitable to do that.
Folks with mid-sized farms (as mentioned, that's only 24 or so L3+).
They may be using LiteCoinPool as a backup "rollover" pool from NiceHash - given how often *my* miners end up giving me short "spikes" on litecoinpool, I'd bet on that being the real explanation.
|
|
|
|
|
If you push the card hard to maximise coin production, and do NOT make an effort to keep it cool, it can shorten the lifetime or kill the card.
If you keep the card under about 70C, you shouldn't have this issue to a significant degree.
This is a "rule of thumb" though, some cards tolerate high temps better (AMD R9 2xx series for example), others want more like 60-65C for good longevity.
Also, TYPE OF FAN MATTERS.
If it is not a "ball bearing" fan, COUNT on it dying far too soon when the seals wear enough for the lube to leak out.
|
|
|
|
Just curious how long some of you run your miners before rebooting, etc.
I finally have mine pretty stable as it is a new rig, and it's been running 24hrs now and still going. 6 1070's at 2771 Sol/s average now. Sometimes it goes over 2800 Sols/s
All 1070's are about 472 Sols average each.
My LINUX rigs usually go for months (if the POWER stays up long enough) without reboots - I'd say years but I haven't had a year straight with no power outages here (had *2* outages last summer due to brushfires burning power lines). Win 7 machines (the few I have left) are lucky to manage A month without problems. |
|
|
|
It depends on WHAT you are mining.
1070 ti is the most efficient card at ZEC mining right now with the 1080 a close second (but worse on hash/$ at any setting and the 1070 and 1080 ti are both fairly close on best efficiency OR hash/$).
For ETH the 1070 matches the 1070 ti hashrate exactly (same memory system on a VERY memory-hard algorithm) and the 1070 blows the 1080 completely out of the water - the 1080 ti beats the 1070 but not by a lot, making the 1070 by far the hash/$ leader of those 4 cards on ETH.
SOME 1070 cards will mine ETH at 32 Mhash/s but many of them do good to get to 30 - but even at stock clocks they pretty much all do more than 28.
But for $400 OR MORE they are not cost effective vs a 1060 3GB card that does 22+ for about half the cost, and they are not even CLOSE to cost effective vs RX 470/480/570/580 cards in the UNDER $250 range that can generally get to 28-30 Mhash (with BIOS mods) and are very close on efficiency.
Then there is Monero, where the Vega 56 is the current king of hash/watt (though some of the NVidia cards can argue there) and *WHEN* you can get one at semi-close to MSRP the hash/$ winner by a LOT.
Vega 56 is even beating those old "open-compute" refurb Intel servers on hash/$....
If you can find a Vega 56 for under $500, it currently has an ROI in the 85 day ballpark, and has been consistantly under 105 days for the last couple weeks.
I fully agree with you. In theory those RX cards should be better than the GTX 1060 but they are either outpriced or not available, GTX 1060 has become the best alternative for ETH. For ZEC or XMR, that's a different story. I tried my GTX 1060 rig on those currency and even if it's not bad, those cards are not high performing cards for those currencies. Somehow my whole rig was consuming way less power with ZEC and XMR than ETH. For some reason, the algorythm behind doesn't get all the juice from the GTX 1060. EVGA GTX 1060 (got a couple of those) just went under 200$ now at Amazon: GTX 1060 on Amazon Must have been a Black Friday / Cyber Monday sale, they're showing as "$211.80" and higher now. On the other hand, Newegg had a few RX 570 cards at $209 for Black Friday - lowest is back up to $239 as of yesterday. |
|
|
|
hey philip - thinking of getting L3 for about USD 3.2k with power supply -
do you think it is still a good time to buy?
thanks
Does it arrive in a few days at that price. They are earning 100 a day. If you get it soon maybe, At the moment. Allmost every coin has difficulty too low or price too high. The question is how does the market adjust? And how soon. A single L3+ has almost identical hashrate to my entire A2 FARM - which is only pulling in about $20 a day gross (504 "official" hashrate vs about 520 as I have my A2s currently configured). I'm not sure how the reliability of those are compared to the S9 (have seen conflicting info there) but they're definitely not as reliable as my A2s - on the other hand, they also don't eat nearly as much power. |
|
|
|
Is buying from alibaba as safe as buying from ebay? saw something on that site I was interested in buying but not sure how good the buyer protection is or if theres any
Also I'm wondering about aliexpress
Both have a lot of similarity to eBay - some sellers are reputable, some are ripoffs - but LESS potential for consumer protection, and no requirement to use that RIPOFF Paypal garbage. I'd call it a tossup overall. |
|
|
|
So, with that said, there are 2 statements coming from this article (and others) that could be construed as having completely different meanings in regard to the $20k rule. Now granted, I know this is only from 2013-2015, when i didn't trade any BTC, but I am going under the assumption that the same materiality threshold will be put in place for this and subsequent years, which is what has me concerned. http://fortune.com/2017/07/10/bitcoin-irs-coinbase/- excerpt #1: "...only for those accounts that engaged in transactions worth $20,000 or more." to me, this verbiage indicates the $20k threshold refers to a single transaction. Meaning, like you explain, if you have say 15 sells of BTC to USD for $2k each and spread out over time, you would have $30k total but would not be part of the record forwarding since all single transactions were well under that $20k amount. - excerpt #2: "to users with “at least the equivalent of $20,000 in any one transaction type (buy, sell, send, or receive) in any one year during the 2013-15 period.” To me, this could mean that if you sold a SUM TOTAL of $20k+ of BTC for fiat through CB, you would be subject to the record forwarding. So, to me it comes down to the difference between "transaction", which clearly indicates a single action... vs "transaction type", which really seems like it could mean a sum total of that transaction type. It is also a bit strange that they include buy send and receive because none of those inherently indicate a tax liability. Typically, one would only incur a potential liability if they sold BTC to fiat at a gain over their basis in that BTC. Now of course it's different if you are actually mining BTC because all of that would be subject to taxed, but that's gotta be a minor # of people compared to overall users of BTC. Anyone have further insight on this? I suppose I could write to CB support for further clarification. Per Coinbase's statement on their site about this, I am inclined to think it is the "$20,000 of one TYPE of transaction in a year", not "if it had any single transaction over $20,000". Doesn't affect me as I've not done $20,000 business with Coinbase in any given year, and am deliberately working to avoid getting there. What the IRS wants is info that MIGHT indicate a POTENTIAL tax liability, they don't care about ALL the facts being reported or they'd require BROKERAGES to fill out that entire form about stock trades, not just the "sale" part even when the stock was BOUGHT via the same brokerage. It also appears that Coinbase might still be working to narrow the scope of the subpoena even more - and that the JUDGE is taking their "invasion of privacy" concerns seriously. WHY THE HELL did the IRS want LOGIN AND PASSWORD information, for example (which HAS already been struck by the JUDGE as being way excessive).... |
|
|
|
|
GOLEM will not be connected to BOINC at all - and as of the last time I checked it's still in the "testing" phase.
It's also a different concept - it's a "pay to rent computing power" type setup, not a "compensate for work done" setup like GRC is.
|
|
|
|
I see. Noted all the info. Thanks
Wow 3 box fans, thats total of 15,000 cfm?
More like 6000 if they weren't air filtered - I figure I'm probably getting about half that WITH the filters in place, more when the filters are new less when they get close to end-of-life clogged up. Typical Lasko/Galaxy standard low cost box fan is rated a little over 2000 CFM, if I remember the spec sheet I found on them ONCE after a lot of digging. |
|
|
|
|
It's not that I "believe in Gridcoin" - I actually think how they have things set up needs a major overhaul.
It's that I'm a 20(ish) year participant in the Distributed.Net project, and working with DNet through the BOINC Moo Wrapper project instead of running the DNet client directly lets me earn enough coin to pay the electric and sometimes even turn a small profit on doing something I want to do anyway.
|
|
|
|
|
15-20% less power lowers heat a LOT more than changing miner software.
8-P
|
|
|
|
You are pushing those cards too hard with the overclocks, most cards don't run with clocks too high.
I get over 510 sols on my EVGA 1070 Ti with these settings below with the temp at 70c using the zm miner.
Set the memory to -303 and the core to +140 and power to 85 - 90.
Lower your overclocks and keep the fans above 90% to keep them cool.
I was seeing 509 sols/s at 70% power with +200 core +700 memory on my EVGA SC 1070 ti cards using EBWF - I think DSTM was a hair higher at that setting but can't remember for sure. They also run cooler than at 85-90% power with lower fan settings - 56C at 52% fan when I was testing that setting in an 80F ambient room with my standard Afterburner fan profile. I don't have any of the MSI blower models, but I do have a pair of ASUS (one 1080 ti one 1080) that are a very similar design - they DO run quite a bit hotter than any of my "fan-type" designs, but it helps a lot to remove the mounting bracket for better airflow. |
|
|
|
|
Don't insist on running Windows - LINUX supports a lot more cards without having to jump through hoops, and is more stable too.
|
|
|
|
|
Kuroman, since you refuse to bother paying attention to anything I have posted, and refuse to even pay attention to the information YOU YOURSELF HAVE POSTED that refutes your "17 cards for 6k" garbage, I refuse to point out your ERRORS AND MISTAKES AND LIES any more.
There is no point in my beating my head against the stone wall of your STUPIDITY any more to try to break through it with the FACTS.
Don't bother spewing at me any more, you are going on ignore.
|
|
|
|
|
I suspect the Extreme model is a waste of money on the Aorus - but I do have a few of the non-extreme 1080 ti models, one more on order, and I figure on ending up with about 5 of them in the medium term (one in my gaming machine, 4 to go with my EVGA SC 1080 ti and Gigabyte Windforce 1080 ti models in the B250 Pro short-term build).
Good cards, BALL BEARING fans unlike the rest of the Gigabyte recent line, and very nice cooling.
Even at best efficiency though, they're only about 4.1-4.2 range as opposed to the 1070 ti routinely exceeding 4.4 at best efficiency settings - works out a LITTLE closer at total system draw, but still loses by a bit.
I am considering building a couple rigs with 2 of them and 2 1070 ti cards though, for use on the EVGA G2 850 power supply since it doesn't have as many connectors as the Seasonc X-850 or SS-860 but those power supplies are getting to be VERY hard to find and expen$ive.
They'll still work well in my new "shelf-rack" setup as I can space the risers out more with 1 less card to split the space between - or I put 2 cards on the MB and only 2 on risers and place the MB between the PS and the 2 riser-mounted cards should also work, though it would be tighter.
1080 even if you push the hash does not do 10% more than the 1070 ti - but it tends to COST 10% more.
If you run them at their most efficient point, the 1080 is a waste - I've tested a few of them now and only the Zotac and EVGA SC models of the 1080 can match hashrate with my EVGA and Zotac 1070 ti cards - the Gigabytes are all 20-30 sol/s LOWER at their max efficiency point.
1070 ti and 1080 ti max efficiency power point is pretty much identical in my testing - so for a MAX EFFICIENCY build the 1080 is a total waste of money vs the 1070 ti.
500 sol/s out of a Zotac 1080 was running 70-75% TDP for good efficiency, NOT pushing for high hashrate - which he's mentioned he aims for with his limits on electric.
Where is there any link to the P104-100 actually showing up at all?
Last I heard it had died a quiet "nobody interested" death due to gouge pricing dying down.
|
|
|
|
One other note: on rig start up sometimes you may forget to run whatever S/W you are using to setup the TDP. This is especially important if you are operating 1080ti's and accidentally forget this step and start at 100% TDP and your rig is laid out for say 65%.
Easy to set up LINUX to automatically and reliably set your TDP as part of the bootup process. Windows - well, I mostly like Afterburner but it DOES occasionally lose track of settings for one or more cards.... |
|
|
|
The room with my miners currently has a humidity level of %12. The air is very dry, I am not sure at what point it will become a problem....or what I could do about it? I suspect a humidifier wont accomplish much as the hot air is being exhausted outside, so any humidifying I do will just be exhausted out as well I think.
don't mess with it, 12% is low but better than 70%+ in fact ideal range is between 30-60. Just exhaust the hot air out and pull in outside air with filters, that should balance the humidity Nope, it doesn't. Keep in mind that the heat generated by the miners causes the room air temp to go up which drops the RH. I had *2* big Brisa units AND a small "misting" system running in my mining room (which is perhaps 500 square feet) and even when it was RAINING outside (which was bloody RARE last summer, we went about 2 months at one point with ZERO rain and 3+ months with a single rainy day) I never saw 50% RH - mid-40s was as high as it ever got. I'm currently in the process of revamping the air system to push more air through the room, as the temps were getting pretty marginal at times during the major heat wave we had. Electronics is generally specified to work between 20-80 % RH, sometimes wider, as long as it's "non-condensing" (which will NOT be a problem in a miner due to it's heat generation) that's the critical part. Also, higher humidity levels help keep dust count down some. |
|
|
|
Thank you for the feedback, yes i have even new version of this one for farms where it only reports "ONLINE/OFFLINE" because all data is monitored on site screen sections. This software was ment to only report, that's why it has this kind of glitches in it currently.  And this new version is where? |
|
|
|
|
Where are you at that RX 570 and 580 cards are "not available?
Outpriced a bit - still, yes, but that's dropped a LOT the last month.
At current pricing I'm seeing on Newegg for them, you'd have to find 1060 cards a fair bit UNDER $200 to have a prayer of being competative on hash/$ right now.
|
|
|
|
Why do you want to make a room filled with electronic equipment more humid?
I think that's the dumbest thing I've ever read...
No it's NOT stupid. Humidity levels that are too low (and 12% is DEFINITELY too low) allow static electricity buildup - which KILLS electronics. The data centers I have worked in generally aimed for ballpark 40% RH - and at LEAST 30% - for that reason. An evap cooler would probably be the best option. |
|
|
|
|
Show Posts
