ETH is still a viable option for the R9 290 - it pulls more power than a GTX 1070 and always has but it's still quite close on hashrate.
ETH was a good option for the R9 290 but its suffering that DAG bug with the TLB trashing like the R9 280X of our time. Basically under stock clock and memory (947/1250) you could get almost 27MH/s and now you only get like 23.5MH/s. It decreases substancially every new DAG. So you are better off just mining XMR with it because you get a little over 700H/s and you make almost the same daily but use much less electricity. I'm still seeing well over 27 Mhash on ETH on my R9 290s - I suspect it's more the sheer size of the DAG file that is making them slow down from the about 30 they managed 2 years ago, more than TLB thrashing. No, they're not running at stock clocks - though I did have to run them at stock core clock this summer due to the long-running serious heat wave we had for more like 25 Mhash. |
|
|
|
Anyone got the latest RTX series already?
I want to know other algo performance, e.g. Lyra, Equihash (various versions) etc.
Ignore Equihash - they're not going to match 11-12 thousand sol/s on 350ish watts like the Z9 mini can do, much less be in shouting range of the more efficient ASIC units like the Innosilicon. |
|
|
|
The cryptonight V7 algo is good for the really really old AMD GPUs like the Radeon 7970/280X/290.
Basically if you properly undervolt the GPUs you can get them to use the same amount of power as an RX GPU mining ETH, about 130 Watts per GPU.
ETH is still a viable option for the R9 290 - it pulls more power than a GTX 1070 and always has but it's still quite close on hashrate. |
|
|
|
Hi guys, it's Luke from Cudo Miner here. We've just reviewed the latest GEFORCE RTX 2080 and compared its mining performance against the GTX 1080 on 4 different algos. Just wanted to share the video we've produced to show you the results. 📹
Equihash
NVidia GTX 1080 Founders OC: 391 Sol/s
Gigabyte RTX 2080 OC: 566 Sol/s (+45%)
And a Z9 mini can easily pull over 12,000 Sol/s (that's what mine is doing at 550 Mhz clock, *0* HW errors and running pretty cool) on less than 350 watts at the wall with a Seasonic gold PS powering it. Why do people bother testing GPUs on equihash any more? The OTHER results you posted though were useful. |
|
|
|
would be better to have all the algo not just eth, hich is pointless for nvidia to mine anyway...
That is right. The ETH is ASIC mineable now. And the only current ETH ASIC to date barely matches a well-tuned GTX 1070 6-card rig - bit less hash, almost identical power. It DOES cost a fair bit less though, even with the recent price drop on the 1070. |
|
|
|
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. |
|
|
|
Compare to the GTX 1070 at 30-31 Mhash for ETH (speaking of my own cards and production numbers, though testing numbers should normally be the same given how most folks DO their testing). How does the power consumption compare to my cards pulling 104-110 watts (depending on the specific card) for stable hashrate? |
|
|
|
Why is this news? Newegg has accepted BTC for years now. |
|
|
|
hashrate still nothing, however, gaming 2080ti x 1080ti = 35% faster, if 35% is confirmed then is a huge disappointment.
i think if game not support RTX function maybe faster around 10-15% A couple leaked reviews claim 30-45% faster with RTX disabled. Why is that the case? It doesn't make sense RTX function would drag down performance. How many games support the RTX cores so far? Do ANY of them? If you shut them down they don't eat power - leaving more power for the rest of the GPU. |
|
|
|
I have been mining for two years. Slow start with 6 AMD GPUs and Baikal mini miner.
I bought many GPUs in last 10 months.
I have 402 GPUs mixed AMD -- NVIDIA and some Baikal ASICs.
the profit ? i have earned the half of the investment. I need to earn around 50K USD to reach ROI.
That's crazy, in this market you're barely able to break even on electricity used. All the miners in china are just mining and selling with their cheap electricity, they have absolutely no interest in holding anything, they're just selling selling selling and pushing all the prices down. There is no financial intelligence, just make a dollar and buy raman noodles.. I'm doing a lot better than breakeven - but the only low-cost electric rate areas in the US are pretty much ALL getting anti-crypto and implamenting new HIGHER rates for cryptocoin mining operations, sometimes a LOT higher than their actual cost to generate that electric. |
|
|
|
This is because you can put the cards closer together without having anything between the cards. The V008 risers are nice with multiple connection types, but having the SATA front facing and the 6-pin on the side is pointless. When using the 6-pin connector on V008, it causes clearance issues between cards when you want to push the cards tightly together.
Personally I’m most cases, I use a splitter to split a single 8-pin PCIe connector to 2x 6+2 pin connectors. I then plug one into the riser and one into the cards PCIe connector. Allowing me to reduce cable clutter and power both the card and the riser from the same single 8-pin PCIe. Works great.
If you are pushing the cards THAT close together, they're not getting air and overheating. The triple-power risers are NOT all that wide. Most multi-card rigs need all the PCI-E power connections to power CARDS with, it's uncommon for there to be enough "spare" PCI-E connections to power more than 1 or 2 risers with them. Splitting an 8-pin to a pair of 6-pin connectors shouldn't be an issue though, if you use a high quality splitter. |
|
|
|
My doubt is the color of the riser leds, they are permanet red since the beggining, shouldn t be blue, im seeing rigs that is blue for normal functionality ..  Depends on the riser - all of the ones I've used to date that have a "power good" indicator use a red LED, probably because those are more common and less expensive than blue. |
|
|
|
When you choose a riser, consider:
4. solid capacitors not electrolytic
Solid capacitors are A TYPE OF electrolytic. |
|
|
|
Your technology itself looks brilliant BUT how may I ensure that it is all fore real and your Power Station really works??  The demonstrating prototype of the Power Station will be showcasing how the technology works. The streaming is scheduled to start on October 1st, 2018. Okay, not bad. But is there any chance to meet team in personal or just ensure the work of Power Station by myself? Yes, investors with 10,000 Tokens or more are welcome to come to our European lab to see the prototype at work and meet us in person. Why exactly 10k tokens are needed? Why not everyone who wants to teake a look?? Our lab is not so huge to allow hundreds of visitors to take a look  Person with 10000 tokens may be considered as a true project's supporter and that's why it has a sense to allow such a person to visit our lab and this way to say "Thank You for your support" In other words, "we want you to spend thousands before we ALLOW you to do your due dilegence". BLOODY PASS. |
|
|
|
The best vega 56 are the original blower fan type,as you can flash them with vega 64 bios and have almost same perf > 2000H/s
Capable of 1950 more or less with NO bios flash, for that matter, in the case of my Sapphire. |
|
|
|
|
Tensor cores are FLOATING POINT - 100% useless to mining.
GDDR6 is a given, since that's the RAM - the question is "will it out-perform GDDR5 or will it be more like GDDR 5x for mining" there.
Given the PRICING on all of the 20xx cards announced to date though, I don't see any of them being GREAT performers on a cost/benefit basis vs the 10xx series - and so far it's looking like they might not MATCH the 10xx series for cost/performance.
|
|
|
|
Anyone know version with X11 algorithm and available parameter "--coinbase-addr=Wallet_Adress"
Can you add this parameter in the latest versions, to do solo mining to do specified static address?
Why are you trying to use a GPU to mine X11 with ballpark 2 YEARS after the first ASIC showed up for X11? Total waste of time and money. |
|
|
|
My thoughts are the 2080 will be good and save power.
I'd estimate it won't match the performance of the 1080 ti on mining - probably more efficient than the 10xx series due to the newer process they're made on, but for the cost it would probably take YEARS for the efficiency to pay off. |
|
|
|
|
Waste of money unless you have the cards already for some other reason - they're nothing more than standard Polaris-class cards with INFERIOR clocks and cooling to the consumer versions at a MUCH MUCH higher price point.
|
|
|
|
|
Given that the only patents listed involve "cold fusion", which has been DEBUNKED a couple decades ago as an experimental error, I have to wonder about the apparent VERY HIGH SCAM LEVEL probability of this.
|
|
|
|
|
Show Posts
