Bitcoin Forum

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-fe (https://www.bitnand.com/product-page/nvidia-geforce-rtx-2080ti-fe)

Does 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.

Looks like they probably used the eth-pill rate of a 1080ti at ~55 MH/s, then multiplied by 1.27 to factor in the +27% memory bandwidth gain of GDDR6 in the 2080ti, to get the ~69 MH/s projected value (by comparison, the theoretical zero-latency rate of 616 GB/s bandwidth would be ~75 MH/s, so their estimate seems reasonable assuming "mining-friendly" latencies).

Given that this model (RTX 2080Ti FE) costs $1,225.00 , the increase in hashing rate from 55 to 69 MH/s is not enough to justify the price. The other question important for mining application is how much one can overclock this card.

I think ultimately we need to see how good is the GDDR6 at overclocking. Afterall, it is new technology that is still largely unproven. If GDDR6 turns out to love being overclocked, then RTX 2080Ti will be a BEAST.

No matter how beastly the 2080ti is, it will not beat 2x 2070 or 2x 1080ti.

I don't see how 2080ti will be price competitive for mining. The only exception is rendertoken performance where 11gb may provide advantage over 8gb, and raytracing cores > 1080ti.

Yeah at the current price point, the 2080ti is definitely a NO-GO. This is for both gaming or mining. Like the quoted post says, you'd better off buying 2 cheap 1080tis then linking then with SLI (if you are into gaming). Ray tracing RTX and GDDR6 have yet to have software built for them do so this may change in the future, but at least for now, the 2080tis are complete shit and are just a marketing stunt by NVIDIA.

This product for games is not specifically for miners. if you are looking to mine, please use ASIC or FPGA.

What an absolute scam site.  Why would they be selling the FE card at the same price as Nvidia?  There aren't even drivers available for the 20 series yet.  News sites can't post benchmarks until mid-september, yet this site already knows the mining performance?!

what about other algo i'm interested, but at that price is a stupid buy

considering that mining is UNPROFITABLE RIGHT NOW, i don't see the point of spending a huge amount of money to buy a gpu to mine ETH (because there is no other coin that has not been invaded by ASICS and has been made unprofitable for GPU's) specially when Difficulty bomb is already on the plans of the DEVS as stated in the last stream of 24 Agost on youtube.


The block reward is also going to be made, EIP-1234 would reduce block rewards to 2 ETH and EIP-858 would reduce block rewards to 1 ETH per block.

i'm going to say one important thing: unless ETH goes x2 or x3 in the next month, you´ll not even get close to ROI when difficulty bomb is applied.


you have to be really mad, like really really mad to buy GPU's for mining right now.

Nonsense. Bring down nicehash and you`ll see how profitable everything is again to mine. Nicehash is the doom of home miners, asics not so much.

Nicehash is a huge problem yeah, but if nicehash was down for a while it would just be picked up by all the other profit switching pools

Problem with Nicehash is that it really affects small-cap coins where there isn't a whole lot of hash power. The difficulty adjustment algorithm makes it hugely unprofitable for existing miners when Nicehash switches TH/s worth of power in and out of the network.  :-\

It would be interesting to know the performance of the 20*0 series in other algorithms (verge-scrypt/blake, equihash, CryptoNight and forks).

Now that RTX 2080 series have hit the market go and just do a simple check on Ebay for the Rx Series, the 570/580 all cost less than 200 dollars and we can build a miner with 1000 dollars or even less if we can wait. That would be a 6 card miner which would give us 180 mhs stable compared to single RTX 2080 ti it is clear who is the winner here.

Not worth to use RTX 2080 series for mining when with that price of a RTX 2080 ti can build a big 6 cards miner now.

Yes and no. RTX 2080 series will likely see very strong resale value due to its freshness. I think you can sell your 2nd hand card at near MSRP if you get it first.

absolutely right
or series 10

there could be a surprises with RTX series if mining software devs find a way to utilize the tensor cores, if not the performance bump wont be that impressive considering the price point.   

https://devblogs.nvidia.com/programming-tensor-cores-cuda-9/ (https://devblogs.nvidia.com/programming-tensor-cores-cuda-9/)

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.

As the tensor cores are FLOATING POINT, while mining is an INTEGER function, as I mentioned back when the new Tesla was introduced I don't see them being a factor AT ALL for miners.

Utility of the new features depends on the algorithm, for example:

Tensority algorithm - Uses quad 64-iteration 256x256 matrix multiplications, input matrices are composed of 8-bit quantities, results are compressed from 16-bit back to 8-bits prior to the next matrix multiply operation. This allows use of:

1) DP4A 8-bit vector instructions (present in Pascal GP102/104/106, present in Volta V100, unknown currently for Turing - still waiting for CUDA 10 SDK to be available to verify this feature).

2) Tensor core FP16 multiply with FP32 accumulate. FP16 allows for 8-bit integer input values to be stored with no loss in precision, FP32 accumulate allows for 16-bit integer result from 8-bit multiply to be stored with no loss in precision.

3) Tensor core INT8 mode (new in Turing, absent from Volta). If accumulate is in INT16 (likely as this is a matrix multiply operation with 8-bit inputs, should be verifiable once CUDA 10 SDK is available) this would then allow for use of this mode in this algorithm. Throughput would be twice that of FP16 multiply with FP32 accumulate from #2.

4) Parallel-issue FP32+INT32 CUDA cores (introduced with Volta, also in Turing). Similar to #2, FP32 execution path in cores will allow for 8-bit integer multiply and 16-bit integer accumulate.

Groestl algorithm - (Note: I have not coded a CUDA kernel for this, so I am just speculating based on the specs for this algorithm). Uses matrix multiplication in it's "MixBytes" transformation step, input matrices are 8-bit values, hence the same 4 CUDA features discussed above are possibly applicable to this algorithm as well).

You are staying ETHash algorithm is an integer only function? Are you sure about this?

Some of the divisions in here looks like it utilizes floating point calculations.

https://github.com/ethereum/wiki/wiki/Ethash (https://github.com/ethereum/wiki/wiki/Ethash)

Ethash is all integer (and logical operations), the division operations as well. You can verify this by pulling the source code of Ethminer from Github and checking the kernel files (.CU and .CUH files).

In my own Ethash kernel, it's all integer/logical operations as well.

The redeeming thing about getting RTX 2070 or RTX 2080 is that in 6 months time after you're done mining, you still have the fastest GPU on the market. This makes it easier to resell back to the gamers.  ;D

in 6 months time you will not even get 10% of the money you spent on it and will be in sorrow why you used it for mining instead of using it for gaming. It will be a disaster to your brain and especially after amd launches their 7nm hehe

If you buy for gaming then great, if not gaming then mine using it but you need to prioritize gaming over mining and you will be happy with yourself, trust me, this is btc's creator idea,

Imagine if Bitmain gets a hold of these GDDR6 memory chips and uses them in their private ASICs.

I imagine one ASIC would be able to easily achieve 1GHS of Dagger Hashimoto speeds.

Either way, the ETH dev really needs to do something to keep the ASIC resistance, many miners even agree with taking a issuance reduction just to keep ASICs off the network.

Problem is that Vitalik doesn't give a damn about miners. Look at what he says during some of those Ethereum core dev meetings (vids are on YouTube). He thinks ASIC is not a problem up until this day.

And then you look at the relative price of ETH vs BTC. Explains a lot, doesn't it?  :-\

It’s look very interesting  :)

So the overall sentiment is that the price is not worth it?

How do you figure? All AMD needs to do is release a new series of GPUs that would make the RTX 20 series obsolete as well..

My opinion is that it is too late to get into mining at this point in time, whether or not you have free electricity or not. There are simply too many threats just around the corner that could leave you high and dry with computer hardware.

Based on what we can compare, no its not. 
If you want to get into mining, you are better off buying 1080 Tis off ebay.
Though as of recently I am seeing those sell for over $550 each.  Prices are likely to keep climbing until its in cost proportion to the hash rate of the 2080 Ti.

with such a price these video cards are absolutely not interesting for mining, especially against the backdrop of a strong fall in the whole crypto currency market. better to take cards of the past pokrleniya!

The same can be said about Q1 2017. Crypto prices were not attractive back then. Those who started mining at beginning of 2017 are probably millionaires by now.  :P

I purchased 1 EVGA RTX 2080

Cost was 750


It is worth having one of them.

I don’t see that a 1200usd 2080ti is going to be much better then the 2080

The 2080ti has 11gb RAM at a faster mem speed. It'll be a similar difference as between the 1080 and 1080ti.

750 seems good deal. where?

It would be cheaper to buy multiple 1080Ti than the 2080ti.

That is true and it is just like milking the consumers and it is not lime gddr6 is hard to produce like hbm2, also if nvidia say that the tech it brings justify the price it can only br used by gamer also the ray tracing is not yet evident in many games. As such it will still be a good buy for the 1080ti for mining specially if price cuts has been made because of the launch of the 2080ti.

amazon had them yesterday

I agree  but 1200/750 vs 700/500

means  the 2080ti cost 60%  more then the 2080

and the 1080ti cost 40% more then the 1080

so the 2080 is better then the 1080

Assuming you aren't limited to 2-3 cards.

I was talking about performance, not price. Price is set by demand, expense, and market factors, not an accurate reflection of performance.

I mean...if you have the space, then not really, because a cheap mobo setup with bunches of pcie slots isn't very expensive in comparison to the price difference if you're buying in bulk. However, since we don't know what these cards will do, it's all speculative at this point.

Bulk pricing should be vastly cheaper than buying 1 or 2 cards at a time. Bulk discount should be 10-20%.

You guys aren't factoring in the resale value of RTX 20 series. There is a premium because these cards will still be very fresh when you're done mining with them.

Yes so price to price the 2080 is a better deal then the 2080ti only if the performance to performance ratio of 2080ti to 2080. Is the same ratio as the 1080ti to the 1080.

Yes Except for the fact that if you are looking at Dollar spent per Hash Rate,  A used 1080 Ti is still your best investment.
You can buy 1080 Ti on ebay for $525 ( shipped ) all day.  The 2080ti will be lucky if its 30% faster at hashing than a 1080 Ti, yet it still will cost 100%  more.

 I am saying the 2080 is worth it. Kind of maybe.
The 2080ti is not worth it.

A new 1080ti with a three year warranty can be had for about 575

A new 2080 with a three warranty can be had for about 750.

Since I have 12 1080ti’s I purchased 1 2080.  I would not buy a 2080ti for 1200.

If the 2080  = the hash rate of the 1080ti and has low power per hash I will buy some more.

As I pay 750 vs 575 for new with 3 year warranty in both cases.

But do better on power then 1080ti’s

I can’t tell unless I buy 1.

Hmm... please show me where you can find a new 1080Ti for $575????

The question I am wondering: How long under someone programs those new cores to do similar functions as the new ACORN M.2 FPGAs? 

Are you saying program the CUDA cores? I am not sure if that would add any additional mining performance.

price compare to preference is not good.

buy old series is better.

i think miner no need rtx function to help mine

He is probably thinking of RAY cores. I doubt you can use RAY cores for mining.

If it operationally makes sense, mining software devs may even try to shut down the ray tracing cores to lower the power draw.  :D

well here is the deal  evga has an ebay sale store and the were selling them for 650 and ebay had a 15% off coupon  so 650 - 97.50 was 552.50

deal was good for 2 days  on the 27th and 28th of august.

I got a new msi  vega  56 for 399.49  using same 15% off coupon.    sale is done with.

https://i.imgur.com/VpJ9C0I.png

It's speculative at this point.

I'd refrain from jumping to conclusions until actual measurements start coming out. Even after initial benchmarks, drivers and mining software could be better-optimized.

I'm not expecting a giant leap in mining performance either (certainly not enough to gamble on a pre-order), but I'm curious to see actual results.

yeah idk what the rush is, same with the ppl who all felt the need to jump on the fpga bandwagon early
instead of waiting till products were actually shipping.

I guess we never learn in this space lol

That ebay deal was amazing, though honestly it's just a one-off thing, so you can't really consider that as the norm.

have any review on youtube now?

No.

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.

If you're talking about that level of bulk then you may not be factoring in the electric cost. At a certain point the difference in profit trumps the speed of ROI.

(again, speculative of course, since we don't know what the hashrate or draw of the 2080s will be)

To be fair, pre-ordering a gpu with a relatively reliable resale and a warranty is a bit different than plonking down 3-5x the amount on a permanently custom modded, community supported piece of tech. Not saying there are no pros to the latter, of course, but pre-ordering something from Newegg is totally comfortable for me, whereas giving $4k to a stranger in untraceable currency is...well, riskier.

I ordered 1x 2080 from amazon for 750  I get billed the day it ships I have 30 days no questions asked to return it.

If you are usa based and mine with gpus  to not order 1 would be pretty dumb.

My thoughts are the 2080 will be good  and save power.

I'd love to at that price I'd probably take several, In the UK we get shafted. Your $750 is around £580, The cheapest 2080 on Amazon UK I can find is £800/$1030!

Madness

I might take a flight over when they're released and buy as many as 2 suitcases will allow, maybe about 40  ;D

cheapest on newegg is $800 you can buy newrly two 1070ti or two vegas for that

Prices in the USA are disproportionally cheap. Not fair...  >:(

I think there is no VAT of 20% in the USA so it's cheaper.

Medical bills are insane in the USA  IT EVENS OUT.

i watch in this video  some game is same-same

some game like

BF1 maybe support rtx it faster 11fps

farcry 5 18fps

https://www.youtube.com/watch?v=2B3B3QVOJH8

Just wait until Brexit kicks in! It'll be £2-3K easy...

Do you have anyone you trust in the states? If so get them to buy them here and ship them there (or try ordering from amazon.com, I can usually buy things from Amazon.uk and just pay the extra for shipping).

You can buy lots of things for $800, that doesn't really have anything to do with FPGA vs GPU pre-orders.

Haha you caught a cold and need to stay a night at the hospital?

Sorry bro, that's going to cost you dozens of RTX 2080 graphics cards.

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.

im heard news about ETH mining 2070 - 40MH , 2080 - 40MH , 2080ti - 65MH

RTX 2080 might be the sweet spot for mining. Who need Gigarays anyway? Plus if someone could utilize the Tensor Cores and GDDR6 memory for mining, there would be a huge leap in mining performance versus it's old architecture.

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.

Tensor cores can be used in certain mining algorithms with matrix multiply operations: Tensority and Groestl for example.

See a previous post for details: https://bitcointalk.org/index.php?topic=4948083.msg44769341#msg44769341

i think we will wait for software to make GDDR6 mining faster in some algo like today GDDR5X can mining faster than early

but im not sure it make too much different like 50%up or not

Given current ETH price ($175), we are better off keeping the RTX 2080 for gaming  :'(

i think we need to see in other algo too not only eth

so anyone have tested this gpu? what are the perf on bitcore and lux?

We should see performance tests soon. The Bitnand.com guys say they're getting the first batch for performance testing in around 3-4 weeks time.

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.  :P

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-rig (https://www.bitnand.com/product-page/10-gpu-riserless-mining-rig)

Being 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.

yeah go to buy ETH is better not mine it and HODL until next year

RTX series not profitable for eth. Maybe we can get new miner for RTX series.

Yeah, though TBH, people have been calling Ethereum bottom since $700, $600, $500, $400, $300, $200...

The big breakthrough will be the usage of the huge computing power available in Tensor cores (aka the "AI cores"). AI capabilities is something CPUs, FPGAs, ASICs and other GPUs lack... it's something that will be available to everyone for a "decent" price for the first time. If the devs manage to take advantage of those cores for mining (and supposingly share it with the community) then the RTX20 series has the potential to overcome even ASICs in some algos.
At the very least, a new algo could be created that'll be benefited by AI capabilities... then again there are plans for AI ASICs but at least in this department Bitmain won't be the sole player, Intel is close to a release too.

And who cares about Ethereum anyway? Sooner or later it'll be overcome by ARK or EOS or both. As far as I can Ethereum is already behind in usability and probably technology, at least to ARK, but I'm not an expert so take my word with a grain of salt. My conclusion comes because I'm following every ARK update, most Ethereum updates and some EOS updates.

The tensor cores are already usable in the "Tensority" algorithm (eg. Bytom coin) because this algo uses INT8 value matrix multiplications and the tensor cores (starting with Volta) do FP16 matrix multiplication with FP32 accumulate (ie. FP16 with FP32 accumulate can be used to represent INT8 matmul results with no loss in precision). Turing's tensor cores also support a new INT8 mode with double the throughput of the regular FP16 mode (still waiting for CUDA 10 SDK to verify if this INT8 mode uses INT16 for accumulate).

Aside from "Tensority", "Groestl" is another algo which also uses INT8 matrix multuplications.

In general, if a coin has an algo which uses 8-bit value matrix multiplications, then you can use the tensor cores for this algo.

use them with the current miners or do we need new miner versions to utilize the cores?

Thank you for posting bro, I wasn't aware that the figures for the RTX were released.  If they do meet these expectations, it's definitely going to be a really good card to mine with, although it's agreed that the price of one graphics card could be used to purchase two 1080TI's or similar cards.

mimagneto

New miners needed, as they have to utilize the CUDA WMMA (warp matrix multiply accumulate) API to execute on the tensor cores. WMMA was only made available starting with Volta. So if the current miner was not coded for Volta then it doesn't use WMMA. But If the miner was coded specifically for Volta then it may already utilize WMMA. Alternatively, if the miner code is using CuBLAS, it has to have been modified to set the mode to "tensor op math" mode for it to use Volta's tensor cores.

So in either case, if the miner was not specifically coded to the tensor core APIs for Volta then the miner needs to be modified.

Hey guys , mining ETH now  is like  invest  , mining today  sell it on  top or  near top price.
I was mining ETH on 3 gpu only   at  ETH price 15usd  ,  I was holding it and sold  everything at  340 usd   can you imagine the  profit ??
Mine eth today and sell it  at  500-600 usd  ,   +/- within  3-4 months.

seems like a decent deal.  I think what you said it great if you already have the cards, but even if you have free electricity I think the issue becomes making your investment back before a better performing card comes out at the same price.  I wouldn't recommend buying anything until we see the RTX performance.  Yes they are expensive now, but sooner or later they will be releasing lower end models

That seems a tad late, don't these start shipping later this week?

You can also buy the 2080 ti with BTC!
https://www.alzashop.com/evga-geforce-rtx-2080-ti-xc-ultra-gaming-d5448533.htm?o=1

The price is the same as from the official EVGA shop after you calculate the cheaper shipping charge from AlzaShop.
Limit is 2 pcs / address (while EVGA's is 1 pc) and they still accept preorders (while EVGA is not taking any more preorders)

Not for USA and not all EU countries are supported.
Also note that different cards are available for different countries (I guess it depends on the available stock)

You should take into account that difficulty is still high, you won't get more ETH coins simply because ETH is cheap atm. You should factor in high difficulty as well.

Why oh why does this thread even exist.

Seriously, we're now trying to "predict" and then argue with each other over figures that don't even exist yet.

Wow, i think you all need to take a step back in to the real world for a while.
It's no wonder gpu prices sky rocketed last year with this mentality.


ewww. i wonder if it will hash 0.1% more, hmmm, if i buy 1000 of them i wonder if i'll make $1 a day....sheesh, get a grip.

The release date has already been pushed back once. No optimised drivers exist. No updated miners exist. How's about giving stuff time to actually develop.
A better thread would, "now the 2080Ti has been out 3 months, lets review its mining capabilities"

Oh, and when did mining become all about 1 currency....OMG whats its ETH hash...oh, please tell me, i need to know....shit, what about the other 6000 coins and 100 algos ???

I know everyone on here loves mining, but its a gaming gpu, not a specific miners card...as for the new features, heck we havent even got dx12 working properly yet, so tensor cores....whatever they are, maybe 1-2 years and we'll have a clue.

Whilst bi*tchin about the price, and Nvid is the devil himself, funny thing is, all this hype and pre-ordering is only keeping the price high...

Any way rant over, i'll go back to my non-profitable cpu mining, you know the mining where everyone keeps telling me i'm an idiot and cpu mining is dead, and old, and crap, and there;s only cryponight coins u can mine with cpu... where i make more with each Ryzen 1700 than most people make with a 6x1070ti rig....hmmm go figure...

Have a nice day
J

Not even sure what you're mad about.
We got the technical specifications and the Turing whitepaper (https://www.nvidia.com/content/dam/en-zz/Solutions/design-visualization/technologies/turing-architecture/NVIDIA-Turing-Architecture-Whitepaper.pdf).

People with enough technical knowledge who understand the above mentioned specifications AND know how some hashing algos work, can predict the expected hashrates pretty accurately.

This happen in "your" real world?

Lol, depends how much such a rig makes you....
I can earn £5-£7 a day ($6-$9) from each R1700 and I believe a 6 x 1070ti rig makes about $2 a day...I know the 6 x 1080ti rig I was monitoring for 6 weeks made an average of £12 a day over the 6 weeks, that was about 15 weeks ago, and earnings have dropped since then....

So yeah, i'm happy with my earnings, what do your rigs make?

Not really "mad", I just don't get it, the vast majority of post on BTT now are based on bullshit, and speculation, and a seemingly endless need to either spend $1000's to earn virtually nothing (and sucker others into doing the same) or endless posts about how hardware manufacturer are crap and owe us something....its just nonsense, and not real.

"Mining" should be a fun / background hobby as per bitcoins' original plan"
"Professional Mining" - totally different...and should be treat as such. When was the last time anyone posted on an open forum like this about their personal tax returns, or speculated on their finances, or discussed critical info - never -professionals don't do that. They don't bi*ch and whine about every little nothin, they just get on, and share good factual tips where appropriate.

So yeah, i loved BTT when i joined, but its users and spirit seems to be a little strange at the moment. "Slight Desperation" springs to mind.
Having said that, it is also useful, cos everytime i see a "what cpu coin should i mine, or what GPU coin is best" thread, i smile, and think well thats the next set of unprofitable coins to avoid.

As for your "People with enough technical knowledge who understand the above mentioned specifications AND know how some hashing algos work, can predict the expected hashrates pretty accurately." response, is that as accurately as we predict the weather? Cos in the UK our new multi-million pound supercomputer and rooms full of "experts" can't predict anything :) (take in jest please)

Actually i'm quite facinated to see what some bright spark can do with the 2080ti, i just suspect it will take at least 6 months, probably in the form of a private miner for at least another 3-6 months...but hey thats just part of the fun.

As for mining in general, i think FPGA's will dominate everything in another 12-18 months, gpu mining will be irrelevant and gpus will finally become cheap again, miners will be happy, gamers happy, and all will be well in the universe :)

You're on the right track of buying Lambo,... for Your 100-th birthday.  :P

:) not bad for a hobby thou...
Mebbe a toy lambo

I'm in love! <3

https://forums.evga.com/download.axd?file=0;2857005&filename=RTX-Notdjey-Setup.jpg


https://forums.evga.com/download.axd?file=0;2856996&filename=D81BE85C-160A-40E1.AF9-98D2058B668E.jpeg

PS: The above images are not mine, I just found them in the EVGA's forums by some user.

Waiting for numbers 😊
I'm envious..

It's definately got a presence...6 of them would look awesome

J

With 6 of these, what would look even more awesome than your rig would be your bank account statement! :-X

Just for a laff, lets say the 2080 equates to the 1080ti...
up-to-date NiceHash figures for 1 card based on UK domestic power prices


https://www.nicehash.com/profitability-calculator/#nvidia-gtx-1080-ti;GBP;0.17;1;250;0;a5=1427;a7=71.615;a8=1.9;a14=71.09;a20=35;a21=4.77;a22=0.84;a23=0.47;a24=700;a25=1.88;a26=16.2;a27=2.96;a28=6.978;a30=0.84;a32=3.2;a33=15

So x6....and cry
J


BUT, oh so much more important than all that....



Does it have win7 drivers, and
can it play CRYSIS ?
:)

The GeForce 2080 Ti is 45% faster than the GeForce 1080 Ti;
The GeForce 2080 is 15% faster than the GeForce 1080 Ti;
The GeForce 2080 Ti is 84% faster than the GeForce 1080;
The GeForce 2080 is 46% faster than the GeForce 1080.


... in games

Today end NDA  ;) .

https://www.computerbase.de/2018-09/nvidia-geforce-rtx-2080-ti-test/6/#abschnitt_gpucompute_auf_turing_funktioniert_gut

Without any optimisation or oc or "enalargement pills": for eth
RTX 2080 ti : 48.7 mh/s
RTX 2080 35.5
vega : 34.6
GTX 1080 ti : 32.2

far less than titan V.

hahahahahahaha

https://www.cryptominando.it/2018/09/19/nvidia-rtx-2080-ti-test-mining-ethereum/ (https://www.cryptominando.it/2018/09/19/nvidia-rtx-2080-ti-test-mining-ethereum/)

why because you think we shouldn't mine with GPU an kiss the ASIC or FPGA maker ass by buying there shit that is all most use age after 6 month ?.

I  felt i had to feed the troll.

good call that's bs :

there is no way a Ryzen 1700 makes more then 6x1070ti rig  LMAO 

Oh yes there is.
And it's very, very sad.

Why not actually disclose some figures from your 1070 / 1080 rig.
I'm not saying everyone who is mining with 6 1070tis is doing crap, just most, and that i make more with my R1700....
There's no point at all in doing the same as everyone else if it aint working....
but see for yourself
https://bitcointalk.org/index.php?topic=5016425.0

look on ebay / amazon....there are literally 1000's of cards being dumped at the moment becuase so few are making the sum they expected.
As that all passes thru, then the diff on some major coins....may drop, then the remaining miners might make a decent return.
Anyway, mine was just a real life example, i'm not here to have to prove myself, believe my figures or dont...i'm happy with them.
What i will say thou, is next year, when they get released i'll be queing up to buy an EPYC2 with 64 cores rather than a 2080ti rig - but thats just my preference.
I will be buying a 2080 for gaming thou....will be a great upgrade from my 1060 3gb :)

Bunch of reviews out there, however i haven't seen mining performance test so far

OKay, first one: https://www.cryptominando.it/2018/09/19/nvidia-rtx-2080-ti-test-mining-ethereum/
 (https://www.cryptominando.it/2018/09/19/nvidia-rtx-2080-ti-test-mining-ethereum/)
Looks like 2080TI is 48 MH /s with Claymore

2080Ti does 48MH. It costs [at least] $1200

1080Ti does 50-55 MH and can be obtained for less than $500.

You do the math. It’s like buying a Titan for mining. Makes no sense. And all reports so far shows the 2080ti uses MORE power than a 1080ti. So it’s not any more power efficient.

What i’d Like to know is what memory speed was used to get 48MH. Is that stock or with a memory overclock?

The 1080ti has the edge because of the ETHLargement Pill. Will a pill be created and offered for the GDDR6 RTX cards? We don’t know yet. And all of this can change if a pill does pop up for it. But I’m guessing no, not for free.

Looks like they probably did not use ethlargement enhancement for reporting 1080ti hashrate

Here's the sauce
https://www.computerbase.de/2018-09/nvidia-geforce-rtx-2080-ti-test/6/#abschnitt_ethereummining_ist_nicht_ueberragend


https://i.imgur.com/B7puwKU.jpg
https://i.imgur.com/2MkYmWt.jpg

Here's the sauce
https://www.computerbase.de/2018-09/nvidia-geforce-rtx-2080-ti-test/6/#abschnitt_ethereummining_ist_nicht_ueberragend


https://i.imgur.com/B7puwKU.jpg
https://i.imgur.com/2MkYmWt.jpg

If the ethlargement enhancement also works for the 20xx series, then it may not be particularly relevant (depending on the final numbers, of course).

Since the pill was created and written specifically for gddr5x, it wont just work out of the box on a different type of memory, lets use some common sense here.

I'm getting around 45-50MH on the 1080ti with the ethlargement enhancement.  The folks that created ethlargement enhancement will need to update it for the 2080tis, perhaps 60 MH after the update and I'm thinking at least 280-300W on electricity.  

9x1060 3G at 630W crunch 208 Mh/s in my hands (those 20xx cards cannot been maxed at all) and cost less than 3x2080ti

ProgPoW and other algos relevant for the future might be more interesting, and the max rates from the 2080ti where it's barley stable (undervolted etc). If they overclock 10 or 20% more than Pascal have a big impact.

But the prices Nvidia charge is completely uninteresting.

AMD Navi would be the next thing to wait for, and will probably arrive before the profits rise again.

New winning cards 20x series are unlikely to be in demand with the miners in the current situation on the market.
Support for new technologies, such as raytracing and tensor cores, will not be used in mining.

https://msi.com/Graphics-card/GeForce-RTX-2080-TI-GAMING-X-TRIO/Specification
POWER CONNECTORS  8-pin x 2, 6-pin x 1 ;D :) ;D  250 watt or more

FPGA's are highly indemand now in the market and is already beginning to grow an optimized even more powerful and profitable than 6 gpu setup. Turing will likely be for gaming only. And i believe the price will tank so much. They anticipated that miners will using turing cards but for the current cost? Well no.

the biggest problem with FPGAs is cost to entry.

and after you empty your wallet for the hardware, there isn't much in the way of software for them. everyone talks about "bitstream this" and "bitstream that", but most miners dont even know what this means. and the developers aren't freely releasing it, rather keeping it all very close to the chest. seems like the business model is to double charge customers. pay for the hardware, and pay for the software.

or, pay for the hardware, and you're on your own to make your own software.

for FPGAs to take off, there needs to be freely available software that can be downloaded and relatively easily configurable for whatever algorithm you want to mine. something akin to Claymore's mining software. miners dont mind the devfee model (within reason), but not many people are going to adopt FPGAs if they have to straight up pay fiat for the software to use the hardware.

hmmm, I suspect the last thing on their minds when developing raytracing was miners....

Here are the latest game tests of new video cards

https://hexus.net/tech/reviews/graphics/122270-nvidia-geforce-rtx-2080-ti-rtx-2080/
https://www.guru3d.com/articles-pages/geforce-rtx-2080-ti-founders-review,1.html
https://www.overclock3d.net/reviews/gpu_displays/nvidia_rtx_2080_and_rtx_2080_ti_review/1

I do not see huge benefits in performance.
1070-1080ti have not lost their relevance.

We did the same, but it will cost 1.5 times more expensive  ;D

a NIB 1080ti is $699 at the cheapest , maybe $649 , where do these $500s come from, and don’t say used , only
a fool would buy a 3 year old card about to come out of waranty for $500

Ethereum mining is not interesting. We need other algorithms - neoscrypt, lyra2rev2, x16r/s, x17, cryptonight variants, progpow, equihash variants?

The developers of the NVIDIA stated that they did not prioritize the use of video cards for mining. I think it's very good that the new generation of cards will not displace the old generation from the mining market.

Yes, that is lucky for us. And maybe old cards won't be so expensive any more!

Exactly that's the point. Mining is about the average Joe being able to loan his/her hashpower profitably.
Enter ASICs and FPGA and he suddenly can't, ultimately centralizing the network(s) and thus killing the point of PoW to begin with.

Dedicated miners would be the doom of this industry. Mining *has to* be as "inefficient" as possible, else you do not have decentralization. Instead you get professionals who control the networks.

Basically any network that tries to throw FPGAs and ASICs out, get it, the rest will be centralized sooner rather than later. And we already have a far more efficient centralized coin, it's called the US Dollar.

So really dedicated miners are really just inefficient validators compared to human/machine validators used by national currencies. There is no point to them.

GPUs/CPUs are even more inefficient but they also decentralize the network so those networks don't have to compete with the Dollar (and consequently be crashed by it) to begin with.

Hi,
What on earth are you talking about?
Since when did it become an issue for a home miner to have an ASIC?
ASICS and FPGA boards have nothing to do with centralisation...
As a home pc user, if I was investing say $500 in mining "scrypt coins",
would I
a) Buy as many pc's as possible  and achieve say 300KH/s using about 250 watt
b) Buy a gpu or 2, and possibly achieve say 2-3 MH/s using about 250-350 watt
C) BUY an L3 getting around 500MH/s using 500 watt
hmm, give me a second to think about that....
Does that mean I am centralising the coin, just cos I bought 1 ASIC - I think not

Stop bashing ASICS / FPGA boards, when the real issue is with professional organisation who will buy 10,000 of cpus/gpu/asic/fpga boards or whatever helps them make the most profit...they are the ones centralising crypto, they're also the ones affecting mining hardware prices, not you and I buying 1 or 2 mining devices...

The price/performance ratio looks horrible on these cards... it's going to take quite some time to break even off the investment.  I think FPGA is the future, or even ASICs when we know the algo will not change for a long period of time.  I can see FPGA killing off GPU mining once more development comes for it.

I'm a bit ridiculous when people in our time talk about decentralization.
All large crypto-currencies are centralized no more than a dollar.
Decentralization was in the days of Satoshi nakomoto, when he communicated in this forum
https://bitcointalk.org/index.php?action=profile;u=3
Now all the ideas are distorted. Centralization and decentralization of the network depends not only on the ASICs, the GPU or the FPGA.

Please keep this thread focused on RTX performance! I'm in mining world from 2013.. I don't want to read stupid and boring discussion about asic vs fpga vs gpu. You can do that in many other topics! thanks  ;)

The Russians have a saying: "if the topic is flooded, then there is no new information :)"

Well, here's the first test RTX 2080 Ti

https://s8.hostingkartinok.com/uploads/images/2018/09/c198c38c808d9b64ff6d46b5bb57866c.jpg (https://hostingkartinok.com/show-image.php?id=c198c38c808d9b64ff6d46b5bb57866c)
https://www.youtube.com/watch?v=Uz5tVTldWXw
1min 51 sec

As expected, the results are weak. We are waiting for software update

Ye I've seen these results but I didn't even bother to post them here because they're sooooo unrealistic. In some algos the hashrate is lower from the 1080ti. We'll have to wait for good miners to be released.

RTX 2080 Ti ETH test

56 MH/s

http://www.legitreviews.com/wp-content/uploads/2018/09/claymore-oc2.jpg

So I have to ask again since it didnt seem to get any response last time.

Is there any potential to program the RTX Tracer Ray cores to be able to perform a similar functionality to the "Acorn" FPGA?

It seems the Acorn takes processes that would normally slow down the GPU, and computes them elsewhere.  Could the unused RTX cores be made to do similar?

ACORN Thread:  https://bitcointalk.org/index.php?topic=4391318.0

I'm not a mining software developer so I can't answer with 100% accuracy, but then again I don't think a mining s/w dev will bother answering in this thread anyway! :P

The Ray Tracing cores are more like an ASIC, by that I mean they are crafted to perform a certain faction very fast. Therefore they can't be programmed to support the GPU in anything else than Ray Tracing. The only possibility would be to make a hashing algorithm which supports the factions of these cores, but that's very unlikely... why would anyone do that?

The AI cores (aka Tensor cores) though might have some use. Those are also for a specific faction as well, I think it's called "4x4 matrix computation", but at least those computations have some general uses outside games and therefore they might somehow, somewhen be used in mining... maybe...

I can use my 1080Ti for mining a bit longer.

Centralized network means "mining hardware producers", when you think you are buying NEW s9 from Bitmain - you are wrong, that s9 was working already at the top of currency price and then utilizing to another ppl when rentability goes to 50% from the beginning

Just did a livestream last night

Here is the recap with a pair of RTX 2080 OC's in 5min

https://youtu.be/i7tlNwl0FoQ (https://youtu.be/i7tlNwl0FoQ)

Thanks for the video.
You also look included XDNA and I'm impressed that the 2080 was faster than the 1080ti prior to OC and with obviously not an optimized miner! I wanted to see the XDNA's results after you've overclock the card though... but +1000 OC in the memory and it kept going steady?! That was impressive!

Oh, and the Enemy miner has been updated to 1.20.

https://i.postimg.cc/9XJ5zRBN/Cattura.jpg

.

power usage is 20 percent less across all algos on the 2080 rtx. , for lomg term
miners who already are in the green and are looking to future proof thier farms the rtx is definitely an
option.  The key is not to fomo buy.  Wait a few weeks or months for prices to at least be $700 per card no more

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.

New cards dont show good results but i think it is too early to conclude
The miner software not yet optimized for this gpu and memory

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. 📹

You can view the video here: https://www.cudominer.com/rtx2080benchmark/

[Too long; didn't watch😉] Here are the results:

Hashrate

Ethash

NVidia GTX 1080 Founders OC: 20.2 Mh/s (28.7 with Ethpill)   

Gigabyte RTX 2080 OC: 35.3 Mh/s (+75%)

Cryptonight Variant 1

NVidia GTX 1080 Founders OC: 507 h/s   

Gigabyte RTX 2080 OC: 769 h/s (+50%)

Lyra2REv2

NVidia GTX 1080 Founders OC: 42.1 Mh/s   

Gigabyte RTX 2080 OC: 72.1 Mh/s (+70%)

Equihash

NVidia GTX 1080 Founders OC: 391 Sol/s   

Gigabyte RTX 2080 OC: 566 Sol/s (+45%)


See the following images for tables and graphs: https://imgur.com/a/ZvPdYcohttps://imgur.com/a/iQmZkdT


What are your thoughts on the results? Has anyone else tested the RTX 2080 on their rig? What hashrate did you get?Cheers!

Thank you Luke.

Too bad the Lyra2REv2 will be ASIC mined now...
From another benchmark, I noticed that the HEX algo (XDNA coin) using the Enemy miner 1.20 (newer version is 1.21) offered a big boost with the 2080ti over the 1080ti, you might want to try this algo too.
As for the Ethash, which miner did you use? Claymore or something else?

Read this:

https://www.phoronix.com/scan.php?page=article&item=nvidia-rtx2080ti-compute&num=1

.

I have 30mhs on eth with claymore on nvidia gtx 1070 (ethpills dont work on gddr5 and have samsung mem ) with PL 55 clock +150 Mem +600 .

There's not a single post here that isn't comparing apples to oranges currently.

To be 100% useful, you'd have to compare a 1080 with pill vs a 2080 with pill on ETH (not possible as the latter pill hasn't been released)
You'd have to compare FE cards vs FE cards. An EVGA 1080 SC does 37MH/s on ethash with the pill but the FE does only 28? Extrapolating this, why would one compare a Gigabyte 2080 vs a 1080 FE?

So yeah. These "comparisons" are all tosh.

Personally I think the best comparison would be 1080/1080ti w/ ETHLargement pill to 2080/2080ti without pill.

This compares what is possible RIGHT NOW. If you’re trying to run a 1080/1080ti on memory dependent algos without the pill you’re just leaving performance on the table for no reason.

You shouldn’t assume that a GDDR6 pill will materialize. At least not for free. The current pill works for GDDR5X only, and it has been reported that a GDDR5 version exists, but available only for some amount of money to farm owners. I suspect that IF (huge IF) a GDDR6 version of the pill comes out, it will be a private for sale version only. Another point leaning to no pill for GDDR6 would be that the devs seem to already have their hands full with other more pressing projects. Mineority and FPGAs. My bet is that it just never happens.

A.  I have never been able to get the pill to work on my 1080 cards.  Yes I know it works for most people but using a 3rd party program doesn't show true performance comparison.  It would be like comparing a 2018 Corvette with a 2019 Corvette, except the 2018 has been modified. 

B.  The 2080 FE cards are already overclocked.  The performance of the Gigabyte is probably similar.  I don't think its possible to even buy a stock clock 2080 right now.

Thanks for the video, well done. There are ASICs for both Lyra2REv2 and Equihash and hashrate for Lyra2REv2-based coins is going through the roof thanks to Dayun Zig Z1 so we might forget about those two algos completely. To be honest I do not see RTX line as a good fit for mining given its price.

price for this GPU is to expensive, and in other side,
eth price was fall into deep.
so this GPU not giving solution about profit for mining ETH.
I think can use for mining other coin with high ROI

https://cdn.videocardz.com/1/2018/09/NVIDIA-GeForce-RTX-2070-Launch-date-768x1139.jpg

.

If this card is not as equal as the performance of 1080, then i'm afraid to say this is the end of GPU mining. Also this card can be good for VR and 1440p gsync.

Given the huge price point of these GPUs, I am wondering if Nvidia will have issues getting them sold for gaming uses.

Basically there are tons of 1070,1080,1080ti on the market right now. Probably more will hit the market in one month when ETH reduces its block reward from 3 to 2 ETH.

I am pretty sure a gamer is better off buying 2x of the 1080 second hand rather than a new 2080ti and just SLI them together. It would use more power sure, but since its a gaming rig it doesn't get used constantely so power shouldn't be an issue.

For mining, its definately not worth it. Since there are tons of second hand RX 470/570 going for cheap on eBay right now.

ppl are not buying nvidia gear to mine eth , Ive not mined it even with my 1070s for like 5 months and
don’t use that dumb eth has so many gpus on it arguments either.  50 to 60 percent of eth and eth forks hash
is asic based , it barely afffected us when eth dropped to $170 cuz we haven’t been mining it anyways

stop fucking comparing the 2080 to the 1080 , no one who is fucking sane would buy a 1080 vs a 1080ti

when the 1080ti costs LESS than the 2080 and performs better than the 1080 by alot.


where are these idoit newbies coming from benching a 1080 , no one gives a fuck about a 1080 if they are even considering a 2080... not GAMERS or MINERS. NOONE

stop it with these bullshit 1080 benches

Ok, my 2080 is up and running. In Ethos, with no tinkering, it's running 39.8 @ 207W. Will be playing with settings to drop that wattage.

With ethpill and tweaks, I get between 49-51 with my 1080ti at about 175W.

2080ti is, of course, delayed, so no numbers there yet.

I decided to cancel my other 2080 (which is also delayed) and will be picking up a 2070 for comparison, but I would say that once miners are coded for these they may be truly competitive. Until then I'd rather spend the money on another 1080ti unless I was building a gaming rig/media center.

It's like everybody's retarded.

EThash is a memory-bound algo that's never favoured nVidia until sombody managed to artificially lower mem timings for a small selection of GPU models.

Surely if you wanted to compare raw hash power of new generations of nVidia cards you WOULD NOT use Ethash as battleground? Yet everybody here in this thread is doing just that.

Use a core-intensive algo if you're gonna compare, plonkers.

It's probably because the 2080 is easier to get at the moment, so people think that the best comparison would be with the 1080. However your point is valid, I'm just saying what I think people are thinking! :P

Um no in the usa at least you can buy 1080tis all day for $729 or less
https://m.newegg.com/ProductList/keywordsearch?keyword=1080ti

cheaper used if you go on amazon, there are plenty of 1080tis available so its not that

I really don’t understand the logic @dragonmike exactly , its like the forum has gone full retard? Ethhash? on Nividia ? Even 1080tis with the pill are not even mining ethash algos eith ethereum being under $300

There is no point of comparing the 2080 to anything in a "controlled" test. We all know what the Nvidia cards we have do in our own hands. The interesting thing is what can you squeeze out from a 2080.

Values of < 25 Mh/s on Ethash (e.g. as mentioned for a 1080) is at or lower than my 1060 3G (9x makes 210Mh/s dead stable), i.e. worthless piece of crap.

If a 2080 cannot beat 3x1060 (micron) in any algo, it's only useful for compact rigs or bragging rights.
70 Mh/s Ethash (210w)
20 Mh/s ProgPow (195w)
70 Sol/s Equihash 144 (180w)
3 card values average over 9 cards.

Nvidia have plenty of Pascal chips in stock, why would they have any interest in selling the 20#0 cards, it's mostly marketing to get some attention (and priced to not disturb the Pascal sales).

Hi,

not sure if I got your table correct, but do you get 20 Mh/s ProgPoW out of a single 1060??? Do you mind to share your power and OC settings? I run 1070 @ 50% power and hardly get more than 12 Mh/s.

Sorry!, that was for all 9... so it should be 20 for 3.

I get about 7.5 out of my EVGA 1060 with Samsung. Maybe you want to have a look at my miner which should give you some more hashes?

Typical corporate-greed-hole move. They release a newer model a month after everyone else has already bought the more expensive model.

Anyway, the reviews of the 2080 hashrate are disappointing so far. Hope VoskCoin does a proper review of this GPU soon. I'm interested to confirm if mining is truly dead for GPU lovers. Hoping it isn't but reality is starting to kick in..

It's not a newer model. It's a consumer level model, and pretty much every industry has always done that--the pro, expensive models come out first, the cheaper consumer grade ones follow.

Why wait for Vosk, there's loads of mining reviews out there already

try
RTX 2080 cryptocurrency performance! We test 2 at same time! Quick Results Video! - bits be trippin
https://www.youtube.com/watch?v=i7tlNwl0FoQ

Why does the release of a new "gaming card" signal the death of "gpu mining"? I think it's a step forward, it just needs gen 2 to bring it to the masses, which will be in about 6-9 months
Oh you mean the reality of everyone buying stupidly expensive rigs cos BTC was gonna hit 50K....well guess what, never going to happen...

Here is some info I got, but still testing.... check out the Reddit post. Just sharing my info, I welcome insight, coaching and recommendations to get the best info for the community.

I done some more testing, per other miners request on specific algos, and below the link to reddit is today's testing.


https://www.reddit.com/r/gpumining/comments/9ixng5/rtx_2080_mining_hashrateperformance/

I will continue to update the threads as soon as I can.

Thank you,

SerpentXSF
CMVJAX

Great review and video MAN! good work thanks! .

great review finally

2080 looks like a bust, power draw is barely better than a 70% powlimit 1080ti. 

I suspect the 2080ti is the only card worth buying in that class but not even for that price

X16r - 30mh
X16s - 14mh
X17 - 14mh
X16 algos should be benched for long time. But +/- they should be equal x17.

You have to be very careful.
Now scammers will begin selling fake miners with an overestimated hashed or distribute virus miners.
The current results of new video cards do not justify their performance in the mining.

ETH rewards going to 2 coins per block is going to drive so many miners off the coin and onto others. 

I think the difficulty of other coins will rise a lot.

if ETH price going up maybe miner will mine ETH not go to other coin

Look at the XDNA project and HEX algo. It seems to do very well with 2080ti and Enemy miner, also the block reward is scaling with the hashrate.
The emission starts from very low while there are few miners and when the people's interest and the miners increase the hashrate, then the block reward increases automatically. That way there is no sudden profit drop because of many miners.

Ask Zimbabwe how that worked out...

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.

Why is this news?
Newegg has accepted BTC for years now.

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?

Lyra2rev2 definitely, but forget equihash it's ASIC owned.

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.

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.

Do you think we will get ASIC or FPGA invasion on Lyra2rev2 algo?

There already is. (https://bitcointalk.org/index.php?topic=5038441.0)

Your link is empty.

The forum doesn't like quotes in the bbcode. Fixed now.
https://bitcointalk.org/index.php?topic=5038441.0

Anyone got the latest RTX series already?

I want to know other algo performance, e.g. Lyra, Equihash (various versions) etc.

I feel like there has got to be something that RTX cards excel at. We must be missing something.

I have both cards, RTX 2080 and RTX 2080Ti.
I have shared some of my finding in here.
 
https://bitcointalk.org/index.php?topic=5042514.0

So what is the best performance algo at the moment?

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.

i did some tests on gigabyte rtx 2080
all tests with
core +100
mem -500
power 60%

almost all agos are taking 150w from wall. idle pc - 50w form wall, with mining started arround 200w from wall - so that means card is drawing arround 150w.

heres the table
https://docs.google.com/spreadsheets/d/17IvMXquTycjr2nBC4H4qvWmsO-66qEGzHCTLIE3Ipvg/edit?usp=sharing

         

You could request from us, might help ROI some. ti's should be available to.

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.

I would not assume that at all as you have no good explanation .No eth pill for  1070, 1070 ti . So their is a risk of it never happening.

Even if it doubles the hashrate of RTX, still it won't be profitable to mine. Maybe you can generate a minimum of $2 a day. With a constant increase in hashrate every month, and ASIC's invading other algos, there won't be any reason to mine in GPU. There will be few coins. Try looking at X16r algo RVN coin.

The free Eth pill might work on the upcoming 1060Ti (DDR5x). It owuld be fun if that card would beat the RTX 2080 on Ethash :) , but in Feb it might be irrelevant, if ETH goes ProgPow.

thats the first interesting post this thread has seen in a long time.

but whether the pill works or not will depend on if the pill checks the memory type of the cards, or if it checks the model name of the card. if it's the former, it might not work at all when it sees a 1060 and assumes it can't be used. since the pill was made before this GDDR5X 1060 existed.

Are you sure ETH will go ProgPow?

Seems at current price levels RTX series is unprofitable at all. Maybe later we will see the using of tensor cores in mining and corresponding increase of hashrate. But now GTX1080ti is a better choice of Nvidia GPUs for mining.

In this hobby/business filled with scammers and other type of lo-lifes you cannot be sure of anything (until you have FIAT in your hand). That's why I used "might" and "if".
ProgPow rewards core clock and power a lot more that ethash so power draw will increase.

It's not that the RTX series is UNprofitable, it's that it has a lower profit-to-cost ratio than the older 10xx series.

I don't understand why Nvidia went crazy-high on the pricing on the 20xx series, seems like they're shooting themselves in the foot to a large degree with that decision for such a small improvement in performance to their MAIN market (Gaming).

If your electricity price is over $0.15/kWh, it is not profitable to mine ETH with 1080ti.