 Show Posts
|
|
Pages: [1] 2 3 4 5 »
|
proccessor is too weak, ram can probaby be sweapped out but an i3 or higer is best for rigs now due to algos like mtp
The processor does not matter as the GPUS are the main reason for this rig, you could probably still mine on the CPU, one extra hash is one extra hash, right? It only has to do one job, is not a media centre or gaming PC... Nice looking unit anyway, looks well made too. Would be good if there were a choice of PSUs if you did not want to utilize or need all that power. You didn't read what Marvell2 wrote. Yes, this CPU is too weak for algos like mtp (and its very likely soon-to-come offsprings). The box does look good indeed, but nowadays if you want something future proof, you'll need to get more than a Celery and at least 8GB of ram. The memory is upgradable, so making this rig future proof is not a problem. I would say the only potential limiting factor is the PSU, but then again that can also be swapped out. |
|
|
|
I've personally designed miner boxes from cooler boxes for the various ASICs I operate (L3+, etc.), and I've found they all work well as long as you have a way to ventilate the air the miners create, and you monitor temperatures as needed so you can add fans to increase intake or exhaust or up fan RPM so the miners don't overheat. I've seen people design miner boxes that twist around apparently to trap sound and to allow air to flow, but in reality those designs restrict airflow too much to be useful. Always make sure to test what you make over an extended period of time to ensure your design can work 24/7.
What are your thoughts on designing immersion cooling boxes ? The 3M immersion cooling liquid is a major cost component. It's very expensive per liter. |
|
|
|
Does Bitnand offer international shipping? I like this 4U case.
Hey! OCTOMINER here. We have 4U cases in stock also and we offer international shipping. We also have a new 12 GPU rig that has a 1151 socketed CPU, it can take up to 6 core i7 CPUs. Usually people spec it with a G3930. We actually manufacture our own hardware to improve the reliability and functionality. For example our OCTOMINER VENTUS fans in our cases are even more efficient than Delta fans. We put extra effort in to get the fans as efficient as possible, the use double thick PCB boards, thicker copper coiling and imported IC chips and Japanese NMB double ball bearings. This helps bring the power draw down lower per rig and will also improve the lifespan of the fans.   I think Octominer is a scammy company. I've heard terrible things about them in the past from my friends. The quality of their product is also questionable... last I heard it's made by some tier-4 manufacturer. I would not suggest getting Octominer products.  |
|
|
|
If you want to maximize your GPU's life, and hash rates of your cards, build an open air rack. My 8 GPU racks have at least 2 inches of spacing between the cards. in that case the cards are going to be right next to each other meaning each card is going to be sucking in the heat from the neighboring card. The result is once the system warms up. all your cards will be hitting max Temps, and hash rates will drop as the card try to keep temps under control. With something like that you would have to use blower style cards and they don't cool as well or hash as fast as traditional cards.
If you want to be serious about this. Build open air rigs, and keep them at floor level where the cooler air is.
Boy you are totally wrong. I have over 700 machines and at first I thought like you. Testing and research tells you the exact opposite. Forcing air through at a rapid rate with enclosed units will cool the card. I have both still but only like 40-50 open air left to convert. Temp on enclosed rigs are normally 10C cooler IF THE FANS ARE GOOD. Also on the ground is incorrect, they need to be about 2 ft above ground for optimal cooling. The reason is at ground level the rotating air (vortex) is present. At a level higher that is eliminated. This looks nice but I hope those fans are fast. Honestly I like Octominer solution best with fans on both entry and exit. Agreed, the enclosed casing forces air to ram through the GPU. Surprisingly, enclosed case like this got much better cooling performance than open-air rigs. |
|
|
|
Now, the 8-slot case looks like really sweet deal.
My rigs are 7 Vega cards with homemade case, and the costs are:
Motherboard: 150$
Ram: 8gb sodimm: 40$
SSD: 128gb mini sata: $50
Power supply 1200W, breakout board and 17x power cables: 100$
Wooden and aluminum parts for case: $20
total: 360$ + time effort to make the case.
So this nice looking case is just a little bit more than my DIY solution, having much bigger PSU but a little bit less ram and ssd. I wish there was such a deal when I was building the rigs, but now it doesn't make sense for me, only if i was going to build new rigs.
I think ram can be swapped out. Though tbh, if you're mining on Linux, 4GB ram is plenty enough. You don't need anything more than that. |
|
|
|
|
Everything - business, sport, culture has its ups and downs. The fact that it's dead right now doesn't mean it will not be profitable in the future. Stay open minded.
|
|
|
|
considering that mining is UNPROFITABLE RIGHT NOW
Interesting theory, but I'm still making enough to live on from mostly GPU mining. Profitability IS down to where only folks with very cheap electric are making any money though. With that said, I don't see the announced RTX 2070, 2080, or 2080 ti being good mining cards vs the 10xx series at their VERY HIGH PRICE, especially with the older cards starting to drop to pricing at or below where they were at 2 years ago before the big cryptocoin price runup happened - and the move of Bitmain, Innosilicon and others to put ASIC into most of what used to be the more popular GPU mined coins hasn't helped that any. It will be interesting to see how these newer cards perform on Folding (which CAN be mined indirectly), but I suspect they're not going to be price/performance competitive even there. I bet you that there will be a crazy new ETH pill available for RTX series. With the pill, I expect some crazy hashrates from 2080Ti. |
|
|
|
you can buy a 1080 for $300 used and you can get 40MH with the ETH pill.
why would you buy a $400 card, for the same performance, and less resale value? that P104-100 can only mine and doesn't have display outputs, a 1080 can do everything else, making it much more appealing to potential buyers.
hell even the 1080ti is a better buy, less than $500 and does over 50MH with the pill with very low power use.
keep crying about your poor investment, its ok.
..i was bought 2x P104-100 one for 277€ (321$) and i have stable 40+MH/s on ethash i was really happy becouse my electricy is only 300W for both (and this is the reason - electricity) POWER limit: 70 Core Clock: 100 Memory Clock: 845  cool story. my 1080tis use less power than that. it's not very impressive. you could have bought a 1080 for less. they go for 300 all day used. sometimes even less What you are saying makes no sense. How can you compare a new card price with old card price? The P104-100 is $400 brand new, that's a MUCH BETTER deal than buying some used old 1080 with unknown quality. does a new card perform better than a used card? NOPE. why buy a new card with a 90 day warranty, when you can buy a used card with 2+ years of transferable warranty left? i've never bought a used video card that was damaged. its VERY hard to damage a video card from end user use. be it mining, or overclocking, or gaming, or whatever. the cards have built in protections from damage. if a card gets damaged, it was probably defective anyway and if you buy a used card from a manufacturer that carries a transferable warranty (like EVGA), if anything goes wrong you just RMA it no big deal. again, no reason to buy a p104, unless you get them for like $100. then the risk of a failed card might be worth it. It's about negative selection. People dump used graphics cards that are broken. You will encounter a high percentage of broken cards in 2nd hand market. |
|
|
|
you can buy a 1080 for $300 used and you can get 40MH with the ETH pill.
why would you buy a $400 card, for the same performance, and less resale value? that P104-100 can only mine and doesn't have display outputs, a 1080 can do everything else, making it much more appealing to potential buyers.
hell even the 1080ti is a better buy, less than $500 and does over 50MH with the pill with very low power use.
keep crying about your poor investment, its ok.
..i was bought 2x P104-100 one for 277€ (321$) and i have stable 40+MH/s on ethash i was really happy becouse my electricy is only 300W for both (and this is the reason - electricity) POWER limit: 70 Core Clock: 100 Memory Clock: 845 cool story. my 1080tis use less power than that. it's not very impressive. you could have bought a 1080 for less. they go for 300 all day used. sometimes even less What you are saying makes no sense. How can you compare a new card price with old card price? The P104-100 is $400 brand new, that's a MUCH BETTER deal than buying some used old 1080 with unknown quality. |
|
|
|
|
Why fix something if it is not broken?
|
|
|
|
After seeing the performance of RTX series, I am more convinced that the P104 is still the best mining card on the market. 40MH/s+ for around $4xx with the P104-100 is a really good deal imho. I don't think it's good to compare to AMD Radeon because per card hashrate AMD is much lower, and consumes much more power. https://www.bitnand.com/product-page/nvidia-p104-100-mining-cardAgreed. P104 has surprisingly stood up to the test of time. The MH/$ and MH/Watt figure is still the best. I just wish Nvidia would release a new mining series card using the RTX series. It would be sick to see a P-series card based on RTX 2070 pulling in 80MH/s  No. 1080s are cheaper. Get the same hashrate. And have resale value. P104-100 is a terrible buy. Ugh, not this shit again. You can go out and find second hand P104, and you'll see how WELL these cards have held-up in value. They are WAY more efficient than GTX 1080 for mining Ethereum. Furthermore, P104 is DEFINITELY cheaper than 1080. |
|
|
|
After seeing the performance of RTX series, I am more convinced that the P104 is still the best mining card on the market. 40MH/s+ for around $4xx with the P104-100 is a really good deal imho. I don't think it's good to compare to AMD Radeon because per card hashrate AMD is much lower, and consumes much more power. https://www.bitnand.com/product-page/nvidia-p104-100-mining-cardAgreed. P104 has surprisingly stood up to the test of time. The MH/$ and MH/Watt figure is still the best. I just wish Nvidia would release a new mining series card using the RTX series. It would be sick to see a P-series card based on RTX 2070 pulling in 80MH/s |
|
|
|
|
This post needs to go up to the top.
IDCM is truly a scam exchange site that is known for stealing customer deposits.
DO NOT DEPOSIT YOUR MONEY ON THE SITE
이 사이트는 사기입니다!
|
|
|
|
So what is the best performance algo at the moment? |
|
|
|
Ethereum mining is not interesting. We need other algorithms - neoscrypt, lyra2rev2, x16r/s, x17, cryptonight variants, progpow, equihash variants?
Lyra2rev2 definitely, but forget equihash it's ASIC owned. Do you think we will get ASIC or FPGA invasion on Lyra2rev2 algo? |
|
|
|
I've been long Monero since early 2017. Made small amounts of money on this |
|
|
|
I am seeing estimated mining performance figures of RTX 2080Ti at 69MH/s. Please see source here: https://www.bitnand.com/product-page/nvidia-geforce-rtx-2080ti-feDoes that make sense to you? Let's compare GTX 1080Ti vs RTX 2080Ti memory performance here. GTX 1080Ti vs RTX 2080Ti Memory type: GDDR5X vs GDDR6 Memory speed: 11Gbps vs 14Gbps (+27%) Size: 11GB vs 11GB (same) Interface width: 352-bit vs 352-bit (same) Memory bandwidth: 484GB/s vs 616GB/s (+27%) Given ETHash is a memory intensive algorithm that does not depend much on the GPU itself, the difference in memory performance should be a good predictor of RTX 2080 series performance. Also, just noticed a decent deal on the 10-GPU rig on the site. I'm tempted to go for it given GPU prices are also low right now. https://www.bitnand.com/product-page/10-gpu-riserless-mining-rigBeing from somewhere near a hydro power source, my electricity cost is only 0.03c / kWh. This means I still make money monthly despite low ETH prices. |
|
|
|
Given current ETH price ($175), we are better off keeping the RTX 2080 for gaming  The ETH price is $200 now, but it is still better not to mine with the RTX 2080. Yes i agree. The ROI takes upto more than a year if ever the speculation is true. It is such a waste of money and time. I'd rather plant some vegetables. Or just straight up buy ETH at the current prices.  |
|
|
|
it's the other way around. it is dangerous to connect multiple PSUs to a single ASIC. you can connect as many asics you want to a PSU as long as both your PSU and the 6pin connected can handle the power needed. from a theoretical point of view you can have 4 z9 minis connected to 1 APW 1600w psu ( each z9 mini takes about 350w if not overclocked" ). how ever i would not do that until i verify that the 6 pin connectors which are going to be split can actually handle the load. You don't want to run your engine on max RPM to loong. It's the same with PSU's. this is completely based on your own or someone else's speculations and it has no solid proof what so ever, the max you see on the label is the safe point which a good psu should handle all day long. I have been pulling about 970w of a 1000w psus on my rigs for years and zero problems, it all comes down to the room ventilation and the quality of your PSU. bitmain psu is a solid psu. i have some A3 and S9s slightly over clocked consuming close to 1500 for ages and never a problem with the PSU. you can test it out yourself if you like. Yep, agreed. I haven't had any problems with my Z9 running at max settings. Bitmain left lots of room for tuning in these things. |
|
|
|
Given current ETH price ($175), we are better off keeping the RTX 2080 for gaming |
|
|
|
|
