I contacted the company that refurbishes them, he told me if it has the 875watt psu, then its the OEM one it came with from the factory, if it has the lower 550watt units, then its been replaced during the refurbished process, because that is all dell will replace them with currently... I have a new PSU coming in the mail today anyways that i ordered as a backup just to have on hand at all times to minimize downtime.. kinda glad i did... ill just jumper it and run it for the 4 GPU's ill run off the pc...

Question, the 875 said it comes with 2 x 8 pins and 2 x 6 pins, is this true?

I went ahead and ordered from china 250 x 6+2 connectors w/pins and 50 x 4pin molex so i can build my own cables in the future using high quality wire.. been running the amazon risers for 2 weeks now on my other rig and i noticed it had turned off on me when i got home and when i looked into it one of the sata power connections for the riser melted... probably cheap or loose connection... i wrapped all connections with electrical tape to ensure they stay tight.. but guess still just cheap cause the melting is on the back end of the sata connection that came with the riser...

ill be nice, this has taken me hours to compile on my Gigabyte 1080ti cards which are on stock cooling, all benchmarks were done at OEM stock settings. I was forced to run it at stock settings because some algos are incredibly picky with OC, i mean some will crash at the sight of +10 to the core freq.. just boggled me, the stock settings were the only settings i could get even numbers across EVERY algo the 1080ti's would run on all the CCminers i have used to date....... The intensity levels, ignore them they will be different on your machine, these are the highest i could get 1hr stability from each algo... im still performing the test....so all blank intensity boxes are DEFAULT levels with no manual correction

i really should look into this xmr stuff, my ryzen 1700 may make me a few more bucks a day.. its watercooled so heat wouldnt be an issue

wow that is one expensive rig... that mobo is $500+ sigh

Hey guys, i got my T5500 today in the mail... Im not sure if the units ordered now will have the different PSU, but mine only has a silver 80 rated 550watt PSU unit in it, 1x 6 and 1x 8 cables only available for GPU, luckily one of my Gigabyte cards only needed 1 x8 and 1 x6 for power.. so i am about to boot it up for the first time before putting the GPU in to make sure its working... my unit is extremely ruff shape lol, the entire case is covered in what appears to be random scratch art as if children scratched crap into the case out of boredom.. but they included stuff that the listing didnt state, it had 2 keyboards, 2 mouses, and 2 dp to dvi adpaters for monitor.

Ok thanks, just so use to seeing them exchanged like within 30 minutes.. thought maybe my account was jacked up or something and the blake2s was just adding to the issues, because in 3 payouts a payout has not been triggered on my account and im at 1/4 btc unpaid.. last time this happened it took 2 days before it finally payed me

I know yall mentioned that whattomine doesnt have skein coins to determine profitability of the coins on zpool, but what about taking the profitability info from the api from zpool? currently i have been using nemominer and it can detect when skein coins are profitable.. which is about half the time compared to other coins for use Nvidia card owners... just would hate to loose skein mining, because some days i get as much as $19/card on that algo... when the coins are hot, unfortunately lately dgb has been suffering.. so been stuck on blake2s mostly

i am not sure what is up with the immature shares on blake2s, i have immature shares in the couple hundreds now that has been building up since yesterday.. is it backed up?>

Thanks Phillip, i tried running it on stock clocks on Ryzen 7, seems to still freeze up... i wish there was a way to input different core/memory settings for each algo, the only way i can get Lyra to run on my 1080ti's is to turn intensity down to 8, i was unaware it could go that low.. was told 15 was lowest.. but it starts up and says intensity 8... and its been running for about an hour at this intensity with 0 on core clock.. i am just going to disable it because at the hashrates its giving me the profit is terrible.. just trying to get more coins available to be mined.... sigh

Phillip have you ever used Awesome Miner?

Phillip or maybe someone else,

Could one of yall explain to me the difference between the miner intensity that it starts the miner at vs the difficult level that slowly ramps up... Im trying to get some coins i cant seem to mine stable with my 1080ti's to work... it seems some it all comes down to the core clock being to high and then others they will start off stable, mine for a few minutes and if the difficulty level climbs to high during mining it will lock up at that point only..

Could someone explain to me the difference between intensity and difficulty levels in the miners? I ask because i am trying to dial in my GPU's for the most profit/watt and when i was running skein 24/7 2 weeks ago i managed to dial it in at 6.2Mh/s per watt compared to 2.1Mh/s on OEM settings.. so literally 3x's the efficiency fine tuning it.. but after reading more into nemo's multi-algo when skein took a huge drop in profit last week and i very quickly found that the gpu settings for my stable skein mining was like 80% chance of crashing on other coins, so i have been fighting with them to fine a happy medium as i introduce more coins to the list to mine.. im out to 5 so far stable but i feel like this should be easier... for example i have lyra2v2 set to the lowest intensity setting 8 for ccmineralexa78 and if the difficulty number doesnt climb as high during mining the coin will mine all day 24/7, but when it jumps pass a certain point in difficulty the miner just goes into a GPU Miner error state and crashes...

So i am trying to understand more about the intensity vs difficulty levels as im a new miner still learning...

A tip for the multi-gpu miners, i found that running separate batch files for each card is WAY more stable that trying to group the settings together.. for example my center GPU's in my rig are able to mine skein at intensity of 27 only, but my outer colder ones can do as high as 30 intensity without ever crashing... so right now my setup is at its most stable condition running multiple Nemo-Miners in the background all running separate GPU's on different settings...  it does confuse me when some will mine 1 coin and others mine another coin at the same exact moment... not exactly sure how the whole grabbing which one is most profitable thing works when im literally using the same computer mining these different batch files...

who sells those Y cables you talk about.. also how long are those usb cables your using?

Im still learning, but i was told if you go with 2 PSU's on a rig you have to power the risers from the psu connected to the mobo otherwise the 2nd psu will backfeed through the riser and cause power conflicts and unstable rig... but powering the GPU's directly from a 2nd psu to the 6/6+2 plugs on the GPU doesnt cause any issues...

Im still waiting on some parts for my first riser build.. but one person told me its worth buying connectors that fit whichever psu you use to power the mobo and hard wire the riser to the psu cable directly by soldering the wires to the 4 pins.... was told this is the safest way to go with the molex style risers... or build my own cables from scratch entirely going with oversized wires...

15 on zpool or MPH?

is anyone else getting low earnings on zpool today? my 4xGTX1080tis have been doing like 95% blake2s since the nist5 issue this morning.. ive only make around $6.50 worth of btc since restarting the miner with nist5 disabled at 8am this morning...so $6.50 on 4 cards over 13 hour period....

So do you just create a batch with that command?

idk whats going on.. says i have 132,808 BTC earned... on Nist5

Is there some way to benchmark the algos easier, seen .txt file above that had specific numbers listed as benchmarks for each algo.. how did yall do that? im linking what im talking about

Hey Yun9999,

I just want to thank you for your vast amount of knowledge and tips/tricks you have posted in this post and your part 1. I finally got around to reading the entire thing and i kinda wish i did from the start when i was researching GPU mining 2 weeks ago.... I kinda already made a few mistakes that might cost me some money i didnt need to spend.. but oh well we all learn from our mistakes....

I am currently running 4 GTX 1080ti's mining, have 2 sitting on my desk waiting to mine and 2 more on the way i bought lastnight to fill a rig im building this week....

In the short amount of time mining with them my watercooled pc maintains 38c and has been super stable... literally been mining with it the last 7 days straight on zpool.....my 2nd pc which has 2 air cooled cards in it has been a heart ache, has capacity of 7 cards on mobo, its my old gaming rig i retired in January... but i for the life of me cant get any of the risers to work on the mobo.. just so annoyed with it, sent the first brand back to amazon, ordered brand 2, they didnt work, yesterday i got brand 3 and still no luck.. so its only running 2 GPU's and im already seeing the issues with running it on air i have had to restart the computer and miner a handful of times already in the short amount of time its been in service.. so i have been looking into ways to cool things better because i have been running this computer in my friends garage, since i live in a condo and dont really have the ability to scale more than a few rigs... so i decided to torture test it in his garage in the sunny florida and the temps have been suffer on the hashrates during the day due to thermal throttling even with running it at 65%tdp and less...before actually reading your post i did order a rosewill l4500 4u case like you use.. unfortunate the estimate from newegg on shipping it is the 2nd week of July... i think you bought all of them

A quick possible low costly mistake i made 2 days ago was order a refurbished dell t-series server workstation for $160. I was doing research and found a posting of a guy who mined with 300+ GPU's and the way he cheaply scaled his setup was with a bunch of the T-series rigs at first, because they either come with dual xeon cpu, 8gb ram, 4 PCIe mobo or the 5PCIe mobo and a 875watt power supply with a copy of windows 10pro.... i ordered it because literally every combination of cheap mobo/cpu i would lookup on newegg was either shipping out like a month from now or out of stock.. so i kinda purchased it out of not wanting to wait and it will be here tuesday, since ordering it i have found out that the 4 x 8 plugs it uses on the PSU for pcie expansion are PEG 8 pin (sigh).. which is kinda annoying as i need to either buy adapters or cut them off and re-pin them for 6+2 connectors... currently the power reported used on my 4 gtx1080tis with the tdp settings im using is around 550watts, but just to make sure id be covered i went ahead and bought a EVGA 1300 watt PSU as a just in case if needed PSU.. this purchase was to get my 2 spare cards up and running making me money, and since i will have 2 extra slots or possibly 3, i went ahead and bought 2 more cards last night to fill the rig to 4 to bring my total mining to 8 once its up and running...

Im seriously doing a lot of research into the possibility of scaling up from 8 and im debating between 3 ways and i wanted to get your input on cooling them...

idea 1: would be a similar idea to your shed build, but the setup would be slightly altered to use closed off spaces.. im really liking the idea of using the server cases to keep hot air from recirculating, so i was thinking of building a shed with a hollow wall inside on the side the exhaust fans are at, this would act as a giant duct/space to direct the hot air out of the shed... i could build the wall out the full length of the one side or build it in sections and as i build 4U case builds i can basically unscrew the plywood section for that chamber leading up and cut off sections as i add more rigs and just slide the case into its spot and screw back the wall plywood to seal off the area, this would keep the hot air in the back of the server case trapped in a small area and the fans you posted would act as a vacuum source to suck it out of the shed, then fresh air would enter the front of the case... I could build the entire setup out of cheaper wood instead of buying the sexy server frame boxes like you have and just paint them with fire resistant paint..

Idea 2: This idea is kinda an odd one, when i searching the internet i came across a small 18 gpu setup a guy designed that sounds logical, but at the same time im kinda unsure on.. i may buy some stuff to test it on a small scale with a 2 card rig for concept... basically the guy used like i think 4/5 inch pvc pipe that ran above each shelf of his shelving above the motherboard, basically he cut small holes big enough for him to slide the gpu and riser card inside sitting it on its side with the fans facing upwards inside this pvc pipe and then just taped up the hole sealing the card off inside the pvc pipe.. He had 6 GPU's running the length of the pipe. The pipe with the GPU's inside had a stack of 2 like 140mm or 180mm fans at the end of the pipe blowing air into the pipe across all the GPU's, then the other end of the pipe ran into a T fitting into a vertical pipe which ran up and over to a window in his garage to exhaust the air... the guy had 18 GPU's running in this system claiming to be able to maintain around 60c with ambient temp air... but the poster never added anymore info after his initial build since 2014... the concept seems logical in a way to be honest...

Idea 3: This one is the most expensive to do upfront, but could be rewarding if it allows me to mine with the GPU's for much long period of time... Since playing around with my gaming rig with gpu minig, i have greatly loved the silence and amazingly low GPU temps my loop can maintain cooling 2 x gtx1080ti's with an external Phobia 1080mm radiator mounted on the wall under my desk on a french cleat with spacing for air flow for fans... the system just seems sooooo stable and runs so much smoother due to the low temps on the actual GPU's... The costly part is buying a waterblock for every single card, but im unsure how far i could go on a single water loop before the amount of GPU's causes the loop temps to be to high... and im unsure if its worth the extra expense if as a miner you rotate your cards every year or 2.. being new these are questions im unaware of... since im sticking to the GTX1080ti theme right now, i have noticed i can purchase 5 different 1080ti's that come already watercooled from the manufacture for around $800/ea, which is cheaper than buying the waterblocks seperately, but there is the possibility of getting a bulk discount if i can talk ekwb into selling them to me at a volume rate... the pump and radiator could be the easy thing... i can either go water to air and use something like a car/semi radiator outside the shed or do water to water like the btc miner in California in the news who stuck his water cooling radiators in his pool, the pool water soaked up the heat