A GTX 1080 ti is VERY cost effective on merged folding - probably the best card to use for that - *IF* you have a folding-optimized rig.
On ZEC, they're very close on hash/watt vs ANY member of the GTX 10xx family when all are operated at "most efficient" settings, and they've usually been very close on a "rig-level" basis on hash/$.
THE most cost effective, no - but not all that far off it and some folks are willing to pay a small premium for the higher rig density they offer (and lower time spent managing those rigs on LARGE farms).
The numbers probably vary if you're not in the US - some areas might charge more of a "premium" for the higher-end cards than we usually see here.
They are cool cards for sure but for mining purposes, the only thing that matters is cost efficiency and they fail pretty hard there. No, they do NOT fail hard on cost efficiency - at the RIG level they are pretty much to a TOSSUP with any other GTX 1070/1070ti/1080 mode - unless you are defining "fail pretty hard" as "more than a percent or two difference". Sounds like you got high rig cost problems. With the plethora of mining specific hardware available these days, anyone can get their slot cost down to at least €25, including power. Its around 15% worse than a 1070 in terms of cost efficiency which in turn is worse than the 1060 and not to mention all the polaris cards. I'd call that failing pretty hard. $25 WITH power is NOT a reasonable number, unless you are using total JUNK for power supplies or used parts - but even the BARE CARD figures on the 1080 ti keep it close to the other GTX high-end cards on cost efficiency unless YOU are getting charged insane amounts for the 1080 ti compared to the others. SHOW ME a system that manages $25 per slot with NEW parts - I really would like to see that. Then show me what kind of crazy rip-off pricing you're seeing on 1080 ti cards vs 1070/1070ti/1080. |
|
|
|
yeah and I can drop any 1080ti down to 125 watts. If I choose. but 135 -145 watt setting is most efficient.
150 was the most efficient I found, but I was only testing at 25 watt intervals. I might go back and revisit that sometime, when me rebuilding project gets around to the planned second "almost all 1080 ti" rig. Right now I'm still trying to get the undervolt tuned on my new XUbuntu-based AMD mining build - went a bit TOO low and it's not been long-term stable quite yet, but the current settings seem to be working so far. I'm not sure if you CAN tune undervolt on a "per card" basis on the current AMDGPU driver stack, though AMD has announced that their next update will add support for undervolt and more sysfs settable stuff. To Gabryrox: IRS I'm pretty sure insists on "FIFO" for cryptomining earned coin, I think I saw that specifically mentioned in one of their announcements, but I'm not CERTAIN. |
|
|
|
Yes it is still profitable if you know what you're doing. Even now my cards bring me equivalent of 30,000-60.000 satoshi per day each. But if you're on Nicehash etc... not sure.
Apparantly nicehash is not the worst place here. What's the worst one then, I wonder. ETH and ZEC adepts? In any case I never understood those who mine and immediately dump, that the dumbest thing IMHO. I mined ETH 3 years ago, did I sell any of it? No, it's still in my wallet  Some of us have bills to pay and no other income (or for a while I had other income but only part of the year). Can't AFFORD to just "hang on" to most of the coin I earn - and what isn't paying the bills has been getting invested into MORE hashrate genration hardware.... |
|
|
|
people heres are die hard anti-sata for some reason. yeah you can have some problems with low quality components, but daisy chaining and using SATA adapters isn't insta-death like some would have you believe.
I'm not so much "die hard anti-sata" as I am "insistent on pointing out that it CAN be a risk" since the SATA power connector is NOT rated to handle the max power that can be drawn from the PCI-E bus and still be in spec on that bus. I recognize that a lot of cards don't pull the full 75 watts (plus voltage conversion inefficiency) the PCI-E bus allows - but unless you can actually measure the power draw of YOUR cards at the settings YOU use, there is no way to be sure that you are staying within the 54 watt max power draw rating for the 12VDC connections on a SATA power connector. Daisy chaining - depends on the specific power supply and cables, but most CABLES for SATA use are not designed to handle 100+ watts that 2 risers commonly pull, much less the over 150 watts they COULD possibly pull. I've seen SOME SATA chains that don't use heavy enough wiring to handle 75 watts - and rare cases of MOLEX chains that have that light of wiring as well. None of the risers I have examined use 5V input at all, much less 3.3V input - ALL of their power comes from the 12VDC lines - but I don't have any "pure SATA" or "pure MOLEX" risers to check, those in theory COULD be different. Note that the PCI-E bus itself only supplies +12VDC and +3.3 VDC - if a SATA-based PCI-E riser used the 3.3v feed from the SATA connector, it wouldn't need any power conversion AT ALL and would only have probably 2 electrolytic caps on it AT MOST - one for 12VDC one for 3.3 VDC. 3.3 VDC on the PCI-E bus is also stated as "standby power", indicating that the load would be very small when it is used at all and that the power conversion circuitry on a riser probably doesn't eat even a watt. |
|
|
|
Couple of people recommended me that i should use the VGA cable i have 4 VGA cables with 6+2 and 6+2 on one string so with the first 6+2 i powered the GPU and then with the second one i powered the riser of course i bought extenders 6pin to 6pin what do you think about that ?
Perfectly viable and safe for any RX 470/480/570/580 model. |
|
|
|
|
They also contacted me by mail, but no phone yet (I was kinda busy today though and might have missed a call).
Phil would probably be best from their point of view as a "near local".
|
|
|
|
A GTX 1080 ti is VERY cost effective on merged folding - probably the best card to use for that - *IF* you have a folding-optimized rig.
On ZEC, they're very close on hash/watt vs ANY member of the GTX 10xx family when all are operated at "most efficient" settings, and they've usually been very close on a "rig-level" basis on hash/$.
THE most cost effective, no - but not all that far off it and some folks are willing to pay a small premium for the higher rig density they offer (and lower time spent managing those rigs on LARGE farms).
The numbers probably vary if you're not in the US - some areas might charge more of a "premium" for the higher-end cards than we usually see here.
They are cool cards for sure but for mining purposes, the only thing that matters is cost efficiency and they fail pretty hard there. No, they do NOT fail hard on cost efficiency - at the RIG level they are pretty much to a TOSSUP with any other GTX 1070/1070ti/1080 mode - unless you are defining "fail pretty hard" as "more than a percent or two difference". |
|
|
|
1080 ti very close on both efficiency and hash/$ on MANY algorithms with anything else even as a bare card - don't ASSUME that all mining is Ethash (ETH/ETC/ECL and such) where the 1080 is a bad choice.
It also offers higher rig density than any other option that is affordable (Titan V is NOT affordable) for which some folks are willing to pay a small premium vs "max efficiency" cards.
Can also depend on what is AVAILABLE the last 3 months - sometimes you buy what you CAN get, not wait a month or two to sneak in a buy on something you would prefer IF you get lucky.
I'd rather wait a month or two instead of waiting for years for ROI. At no point have 1080ti's been anywhere close to being the most cost effective mining card, even with limited options, even considering electricity cost, it just doesn't work out. A GTX 1080 ti is VERY cost effective on merged folding - probably the best card to use for that - *IF* you have a folding-optimized rig. On ZEC, they're very close on hash/watt vs ANY member of the GTX 10xx family when all are operated at "most efficient" settings, and they've usually been very close on a "rig-level" basis on hash/$. THE most cost effective, no - but not all that far off it and some folks are willing to pay a small premium for the higher rig density they offer (and lower time spent managing those rigs on LARGE farms). The numbers probably vary if you're not in the US - some areas might charge more of a "premium" for the higher-end cards than we usually see here. |
|
|
|
Their sales guy said, they have pre-orders up to the hilt already so it was okay for the group to back out from the bulk purchase.
1625$ is for A841s ..... Canaan went to greed mode just like Bitmain - hope this episode taught them a lesson but then again on hind sight they should sell at premium so that they stay in business and compete with Bitmain. They have no incentive to "learn a lesson", other than "price it at what the market will bear and if the market drops drop with it". Folks that payed the crazy "ATH-based" pricing and didn't wait to see if pricing would STAY high deserved what they got. |
|
|
|
"7nm" as currently announced isn't going to be a noteable improvement over "10nm" as currently announced. https://www.semiwiki.com/forum/content/6895-standard-node-trend.htmlIntel and IBM appear to still be standing by their statements that "10nm is the end of the road for pure silicon", though IBM has been working with a mixed Silicon/Germanium wafer setup on their next-gen node work I don't see that buying more than 1 or 2 generations of improvement. |
|
|
|
I have it at 200 watts = 80%
Thanks! Man, that mini is a power hungry SOB!! Nah, it's 1080 ti's in general that want lots of power by current standards, my R9 290s tend to sniff at the power-sipping cool running noob cards. I don't have any 295x2 or 7990 cards or others of the "2 GPU on one board" style though. |
|
|
|
|
1080 ti very close on both efficiency and hash/$ on MANY algorithms with anything else even as a bare card - don't ASSUME that all mining is Ethash (ETH/ETC/ECL and such) where the 1080 is a bad choice.
It also offers higher rig density than any other option that is affordable (Titan V is NOT affordable) for which some folks are willing to pay a small premium vs "max efficiency" cards.
Can also depend on what is AVAILABLE the last 3 months - sometimes you buy what you CAN get, not wait a month or two to sneak in a buy on something you would prefer IF you get lucky.
|
|
|
|
The only way you will build confidence and prove this isn’t a fake is by inviting some of the high profile members here to your facility and showing them the hardware or sending them a demo model.
*Volunteers myself HagssFIN and PhilipMa1957 as demo model reviewers* |
|
|
|
well, the times for good profits, when a miner could earn 5-7$ per day from gtx 1080ti is long over
gladly I was able to ROI all may rigs, so I'm really OK with today's profits 1-1.5$ per card
will wait when profits will start to cone back again ( 5- 6 month)
I get about 3.5-4 dollars per 1080 TI right now. That's not what it used to be but again it is a hell of a lot better than most. It almost impossible You could not get more then $1,5 per GPU in a day now. I mean now. Not in old good times. 1080 ti is capable RIGHT NOW of over $3/day - via merged folding. If I could get my issues with Bittrex straightened out I'd be considering that option. |
|
|
|
well, the times for good profits, when a miner could earn 5-7$ per day from gtx 1080ti is long over
gladly I was able to ROI all may rigs, so I'm really OK with today's profits 1-1.5$ per card
will wait when profits will start to cone back again ( 5- 6 month)
I get about 3.5-4 dollars per 1080 TI right now. That's not what it used to be but again it is a hell of a lot better than most. Yeah, still 1080Ti is the best! Titan V > 1080 ti for income. We won't talk about the ROI though.... 9-) |
|
|
|
I expect a big drop in overall monero hashpower just after the fork. Should we try to get a few extra coins from it?
Definitely going to have the rigs at the ready haha. Think Nicehash will have the algo? Hope not lol. Nicehash is using either XMRig or Stak-XMR - both have announced upgrades, XMRig already had the new code available next week and it's trivial to install the new code manually. I wouldn't count on Nicehash having no miners prepared for the fork when it happens.... |
|
|
|
for those of you still chasing XMR profits, I just swapped from claymore to xmr-stak-win64 and am seeing better results; more consistent poolside hashrate reports, etc....
Monero fork is due on 28.03.18 After the fork monero will be mined with modified algo (so called V7). At the moment other cryptonight coins do not follow the fork. So we have question - to keep mining coins with old algo with ASIC threatening any use of GPU for mining or to switch to new algo. BTW XMR-STAK will release updated version of miner that will automatically switch to new monero algo and Claymore do not announce any intensions to update cryptonight miner to new algo. I expect a big drop in overall monero hashpower just after the fork. Should we try to get a few extra coins from it? You are spreading wrong information... Monero fork is due on 6.04.2018: https://github.com/monero-project/moneroClaymore did announce a new version for both CPU and GPU miners: https://bitcointalk.org/index.php?topic=647251.msg32659476#msg32659476The Monero fork was originally announced for March 28, but it looks like it may have been pushed back a bit for additional testing. XMRig already had a new miner update out last week, xmr-stak had support for the changed algorithm announced but not sure if their update is published yet. I figure hashrate will dump bigtime if only due to the existing botnets probably not being able to upgrade for a while (if ever), plus some folks WON'T get the word for a while. |
|
|
|
The one thing I would worry about is that solar and car alternator systems are NOT in fact +12VDC, but 13.6 "nominal" and commonly up to 15 VDC when charging.
That is some great advice, to be completely honest with you, we didnt even think about the voltage being that high when we ran it off the diesel engine. The GPU seemed to not have any issues with it, atleast the one we are testing it out on. Im shocked we didn't damage the GPU trying to run it this way, because the alternator ramped up to 14.7v under load of mining.... In regards to hooking up the GPU, the gpu is plugged directly into the mobo of the test pc currently... no riser at all being used for this setup. The GPU itself is getting its 12v and grnd from outside of the atx psu. Im pretty sure the solar controller is capable of adjusting the max voltage down closer to 12v, just need to look further into it. It can't go TOO low or it doesn't charge the batteries. Standard lead-acid batteries (which are still the norm in solar/wind power systems 'cause they're CHEAP for the capacity, and the weight isn't an issue) require a good bit more than 12 VDC to charge them - and tend to deliver a bit more than 12 VDC under light-to-medium load when fully charged. |
|
|
|
Test 1:
since true sinewave inverters are so expensive and the PCIE connections are just +12dc and GND, we ran power cables directly from the post his solar controller is wired into and i soldered them to a PCIE pigtail and plugged it into the GPU in a DC to DC configuration, ran the PC off normal 120v, but the GPU would be powered by the solar controller/battery circuit. PC booted up fine and it mined without a single hiccup the 20ish minutes we ran it for, no issues at all, just the solar controller adjusted to the load.
The one thing I would worry about is that solar and car alternator systems are NOT in fact +12VDC, but 13.6 "nominal" and commonly up to 15 VDC when charging. If you're just feeding the risers and the GPU with that, it PROBABLY won't be an issue as they use on-board power conversion circuitry to regulate it down to the voltages they actually use, and their caps are probably rated for 16 VDC - marginal but as long as you don't get voltage spikes probably OK. I'd be inclined to put a pair of 15 amp silicon diodes in series on each riser feed, and the same on each +12VDC line to a PCI-E power connector, to drop the voltage a bit just to be sure you don't overvolt the GPUs and risers. Might be able to get away with 10 amp diodes instead, if you leave the diodes out in the air enough for them to get decent cooling. |
|
|
|
Just finished up initial work on a project. Been wanting to move my Polaris cards over to LINUX but the high power consumption was a drawback - so I did some digging and found that AMD DOES in fact allow for undervolting and underclocking on them under LINUX with a bit of work. Apparently one of their employees has written a "ROCm-smi" utility (also called ROC-smi in some places) that is similar in concept to the long-running Nvidia utility - but it's still pretty basic and limited from what I've seen of it, and I have not tried it out yet. Instead, I went the route pioneered by Phoronix in their https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/amd-linux/918649-underclocking-undervolting-the-rx-470-with-amdgpu-pro-success article, with some additional input from https://github.com/ktsol/dwarfing/blob/master/amdgpu-mod.sh, and did a bit of patching work on the applicable AMDGPU module that is in charge of voltage and clock control. Just getting the drivers installed and working was a bit of a nightmare, but the key there proved to be https://math.dartmouth.edu/~sarunas/amdgpu.html - which I THINK also offers the option to make NVidia and AMD COEXIST ON LINUX for the first time in a few years (but this time around the Nvidia card has to be the "primary" display-running card) Test system was one of my FM2 motherboards with A10-7860K APU (never did figure out how to get X working on the iGPU - yet - but haven't tried VERY hard yet), using my pair of Sapphire RX 470 reference cards, a Gigabyte 480 4GB I just got on a "refurb" sale, and a MSI 480 8GB that was also on sale recently. Cards are STOCK at this point - I plan to BIOS mod them eventually for memory straps - so the hashrates aren't impressive. Before I did my patching to the kernel module, the system was pulling about 770 watts (the iGPU DOES work on this setup, but can't do ETH on it so it's doing D.Net RC5-72 work - which adds about 30 watts to that total). After kernel mods (I went with 830 mv instead of the Phoronix 818 just to be conservative, and a "fixed" 20% underclock at this point instead of the dwarfing "configurable" option or the Phoronix 13%), I ended up with the SAME hashrate at about 540 watts. At this point, I can easily add a 5'th card and STILL come in under my 700 watt "standard rig" target (motherboard only has 5 slots anyway). A low wattage motherboard/CPU setup would probably be able to run 6 cards on a 700ish watt rig, though I might have to do a bit more "fine tuning" to get it to that point. I'm not sure how far this could be pushed, might play with it some more over the rest of the week to find out - and DEFINITELY will be adding a 5'th RX card before I call it "done". I'm also thinking about cleaning up the setup, setting it up with "generic" admin/password type access, and pushing it out as a "basic" mining OS for contributions, as AMD-capable mining OSs at this point are either kinda limited (rxOC) or CHARGE way too bloody much (EthOS, the MONTHLY charge for SimpleMining, don't get me started on PIMP). One thing I DO want to figure out is how to get X to run the display on the iGPU instead of the first discrete GPU - I THINK I know how to do it, and I THINK it will work without causing stuff like fan/clocks control to stop working (like FGLRX had major issues with). I did specifically size it to fit on a 16GB USB drive, as those are pretty much the "entry level" any more, but I am sure I could get it down to 8 without a lot of work - there's stuff from the standard XUbuntu installation that is NOT needed that I've not uninstalled yet. (edit) 5'th card up and running, now for more fine tuning like strap mods. |
|
|
|
|
Show Posts
