Bitcoin mining is ruining 280x prices for us

[BallaTheFeared]

Seeing how mining ruined prices for AMD, now I'm wondering, will Nvidia modify their next architectures to totally dominate in mining and wipe the floor with AMD?

With the advances in cudaminer you have to wonder how much kh/s is left on the table for Nvidia.

I did 600kh/s all night @ 350w.

Maybe all Nvidia needs to do is give that guy a bit of help improving the miner code...

Another 200kh/s at that power level and Nvidia will have basically caught AMD.

I just wanted to show that getting a GK110 chip would allow you to pay for a gaming card with side mining. But another thing needs to be said when it comes to 24/7 numbers, it's basically impossible to achieve that up time. Not to mention any time you want to game it's 0 kh/s. This is where some people distort reality when they say a video card will pay for itself before summer they mean you will not game with the card until that time is up, and even using your PC at all will extend that duration as well as any unforeseeable downtime.

 

[blastingcap]

NV's problem is hardware not software, it doesn't have the hardware integer function necessary for efficient mining no matter how much the NV brigade wants to believe otherwise. Similarly, AMD GPUs are a lot less efficient than NV GPUs for other computation like Folding@Home.

Bottom line: NV mining is a lot less efficient than AMD mining, and even AMD mining isn't going to be profitable for much longer. If you buy a 7970 for $310 and pay just 10 cents/kWh under ideal circumstances (never game and always mine 24/7, the rest of your PC eats ZERO watts) then all it takes is ~3.7% increase in difficulty per increment from here out and the 7970 will NEVER pay for itself. Since CPU, RAM, mobo, drives, etc. eat wattage, and it's hard to find 7970s for $310 these days, the real-life percent is probably more like ~2-3%. http://bitcoinwisdom.com/litecoin/calculator For instance, if you pay $400 for a 7970 and the rest of your PC eats only 50 watts, then at 10 cents/kWh and 2% pool fee/stales mining 24/7 with no gaming, the card will never pay for itself if difficulty increases an average of 2.7% or more.

Difficulty is already projected to go up at least 5% next increment and will likely be a lot more than that as availability of AMD cards goes up and power outages resolve themselves in the Northeast. http://bitcoinwisdom.com/litecoin/difficulty

 

[Silverforce11]

NV's problem is hardware not software, it doesn't have the hardware integer function necessary for efficient mining no matter how much the NV brigade wants to believe otherwise. Similarly, AMD GPUs are a lot less efficient than NV GPUs for other computation like Folding@Home.

Bottom line: NV mining is a lot less efficient than AMD mining, and even AMD mining isn't going to be profitable for much longer. If you buy a 7970 for $310 and pay just 10 cents/kWh under ideal circumstances (never game and always mine 24/7, the rest of your PC eats ZERO watts) then all it takes is ~3.7% increase in difficulty per increment from here out and the 7970 will NEVER pay for itself. Since CPU, RAM, mobo, drives, etc. eat wattage, and it's hard to find 7970s for $310 these days, the real-life percent is probably more like ~2-3%. http://bitcoinwisdom.com/litecoin/calculator For instance, if you pay $400 for a 7970 and the rest of your PC eats only 50 watts, then at 10 cents/kWh and 2% pool fee/stales mining 24/7 with no gaming, the card will never pay for itself if difficulty increases an average of 2.7% or more.

Difficulty is already projected to go up at least 5% next increment and will likely be a lot more than that as availability of AMD cards goes up and power outages resolve themselves in the Northeast. http://bitcoinwisdom.com/litecoin/difficulty

How about other altcoins that are anywhere from 50% to 100% more profitable to mine than litecoin at any given time?

Then how about their constant difficulty fluctuations, as miners move from coin to coin, especially multipool types that mine whatever is most profitable.

Litecoin has not been the best thing to mine and sell for a long time now.

 

[blastingcap]

How about other altcoins that are anywhere from 50% to 100% more profitable to mine than litecoin at any given time?

Then how about their constant difficulty fluctuations, as miners move from coin to coin, especially multipool types that mine whatever is most profitable.

Litecoin has not been the best thing to mine and sell for a long time now.

50-100% is unsustainable and anything way more profitable to mine than LTC gets arbitraged away quickly, sometimes within hours. An exceptionally few number of coins take days or weeks to arbitrage, but in the end everything gets arbitraged away. Also, LTC miners benefit from people who mine altcoins, because that hashing power goes to other stuff and prevents LTC difficulty from rising as much.

If you try to mine every new coin early you may get lucky or you may end up worse off than mining straight LTC since so many coins are pump and dump schemes, pre-mined, or worse. Naive people look at coinwarz or whatever and think omg, that coin is more profitable to mine, I'm gonna mine that today, but they don't realize that a) coinwarz/dustcoin/etc. calculate based on instantaneous price and difficulty which keep changing, and b) by the time you mine enough to be worth anything the price and/or difficulty has adjusted and that top coin may well be no. 25 on the list and way LESS profitable to mine than LTC. This is especially true for low-difficulty coins with fast difficulty retargets and low market cap so that even a little hashing power can affect difficulty a lot, and even a little amount of money on an exchange can affect price a lot. This assumes adequate trading volume, of course; if nobody else wants to buy your coins, then the spreads will be huge and you won't get anywhere near what coinwarz/dustcoin/etc. predict you can get. There are coinhopping pools that may be more profitable at times, but if more people join those pools, any remaining advantage is further eroded because it means arbitrage happens even faster. And from what I've seen their results aren't that much better than mining straight LTC, if you look at long periods of time historically (1+ year). You also eat more pool and exchange fees. But mainly, the more people who know about coinhopping pools, the worse it is for the pools and their users, because arbitrage happens even faster.

 

[BallaTheFeared]

Weird since Titan was doing 300 kh/s, now 600 kh/s... Pretty sure the hardware didn't change.

Why does your forecasting never account for price changes?

When I was mining my 7950s for a few months back in March of last year I was only making $10 a day to start, that dropped to about $5.5 as the diff increases and the coins started to come in slower... Should have been worthless today using your forecasting model, however if I still had them today they'd be making $9.20 a day.

Here is the part you seem to be missing, holding. I didn't sell my LTC as I mined them for $2. I held, I still have over 100 of them but when I sold a few to buy some Christmas presents and my 780 well not really my 780 my 7950s bought that, but still when I sold a few coins recently I got $36 ea for them.

When I was mining I was making about 3 coins per day, those coins I made March 23 thru March 27 made me $432... Or basically $108 per day for those 4 days. Of course so far the rest of the days I made $0, since I'm still sitting but I'm hoping the point I'm making isn't lost.

 

[Silverforce11]

50-100% is unsustainable and anything way more profitable to mine than LTC gets arbitraged away quickly, sometimes within hours. An exceptionally few number of coins take days or weeks to arbitrage, but in the end everything gets arbitraged away. Meanwhile, LTC miners benefit as well from people who mine scamcoins, because that hashing power goes to other stuff and prevents LTC difficulty from rising as much.

It took no effort, I was mining on Hashco.ws last month and netted per day 2 to 4 times more than what i would have got from mining LTC directly. Paid in BTC too.

These days Hashcow.ws and Middlecoin.com which mine the most profitably coin and auto-trade it into BTC for payment are growing stronger, and still yielding ~50% more than direct mining LTC.

It takes a user no effort at all, not like doing it themselves to hunt down the most profitable coin and manually trade it. That is the difference these days with the altcoin explosion. Anytime of day there are opportunities for big mining profits.

Those who stick with LTC are those who a) CBF to do a bit of reading and setup for other multipools, or b) those who believe LTC prices in the long term will go up massively to make it worthwhile.

Generally, even if you are the B type, mining altcoins at higher profitability and trading it for LTC will yield higher than directly mining LTC.

 

[blastingcap]

Weird since Titan was doing 300 kh/s, now 600 kh/s... Pretty sure the hardware didn't change.

Why does your forecasting never account for price changes?

When I was mining my 7950s for a few months back in March of last year I was only making $10 a day to start, that dropped to about $5.5 as the diff increases and the coins started to come in slower... Should have been worthless today using your forecasting model, however if I still had them today they'd be making $9.20 a day.

Here is the part you seem to be missing, holding. I didn't sell my LTC as I mined them for $2. I held, I still have over 100 of them but when I sold a few to buy some Christmas presents and my 780 well not really my 780 my 7950s bought that, but still when I sold a few coins recently I got $36 ea for them.

When I was mining I was making about 3 coins per day, those coins I made March 23 thru March 27 made me $432... Or basically $108 per day for those 4 days. Of course so far the rest of the days I made $0, since I'm still sitting but I'm hoping the point I'm making isn't lost.

Balla, I do talk about price changes all the time in the crypto thread. If you're counting on explosive price growth you are FAR better off buying the coin directly, historically 10-20x better off assuming average USA electric prices. If you look at 2013 as an example, buying litecoins in January and holding them till December would be 10x more profitable than buying video cards and mining with them at 10 cents/kWh. If you think prices will merely increase somewhat, like say double for 2014, you are still better off buying-and-holding than mining at 10 cents/kWh. If you mine-and-sell-as-you-go to cover power bills, it's even worse for mining and better to buy coins directly; this is because you would be paying lots of power bills in less-valuable coins so that by the time the coins gain in value, you have a lot fewer of them than someone who mine-and-held, and LOT lot fewer of them than someone who bought-and-held.

I was mining ... last month

Silverforce, please read my post, the part about historical profit. You talk about the last month as if it were normal. Last month was very abnormal because of Dogecoin. Dogecoin was the biggest reason why people made more, but now it's back down to near-parity with Litecoin. Coinhopping without dogecoin hasn't been great historically (1+ year long periods, not cherrypicking a month here or there), and the more people who attempt to arbitrage directly or who join coinhopping pools, the less profit there is for everybody on those pools, especially with the higher pool fees, stale shares, and rejected shares that coinhopping pools have--in fact it's possible to make less money on coinhopping pools than mining straight litecoins. Furthermore, as you alluded to, the coinhopping pools made LESS money than people who mined dogecoin directly and traded it directly because they must cash out every day even at times where it makes no sense. If you are banking on dogecoins showing up every month going forward to boost profits, good luck. And the more people who join coinhopping pools, the faster the arbitrage, and the less profitable it is for users of those coinhopping pools.

P.S. In just the last few hours time, the entire LTC mining network gained another 5 GH/s. WOW. Thousands of 290/290X, 79xx, 280X, and even older cards are being plugged in as we speak. Even if you already have a CPU/mobo/RAM/PSU/OS/etc. and just need to buy a video card, a $400 7970 or 280X will never pay for itself if difficulty adjustment rises average a mere 2.65%. And if you pay more than 10 cents/kWh, it's even worse. For instance, at 15 cents/kWh, a 7970 or 280X will NEVER pay for itself if difficulty increases 1.4% per difficulty readjustment period. The next projected difficulty readjustment is +5.7% and rising.

 

[KingFatty]

Similarly, AMD GPUs are a lot less efficient than NV GPUs for other computation like Folding@Home.

Now hold up, is there any conceivable way to create a cryto coin that is based on the same encryption algorithm used in Folding@Home?

I think there is a huge market for a coin that would be super-easy to mine on NVidia hardware, and not AMD. I mean, I hate to use the term, but I can see this as a remedy for the ah, butt-hurtedness (my apologies, but I'm at a loss to describe the feeling when one video card has an unfair strangehold on something, where there is an itch to sort of get even, and there may be some pressure build-up in the Nvidia card owners who could really dominate on this new coin)?

 

[96Firebird]

The only people who are "butthurt" are the ones who are looking to buy AMD at MSRP prices for gaming, no?

 

[itsmydamnation]

F@H isn't a hardware problem for AMD. that one is a software problem. Also F@H has nothing to do with crypto, it folds proteins. you get a job which is a predetermined set of proteins, and you hardware runs simulations and then returns the completed simulation back to the server. You will get the same job over and over again.

Coin and F@H are nothing alike.

The only people who are "butthurt" are the ones who are looking to buy AMD at MSRP prices for gaming, no?

there are the usual suspects who are, they are quite easy to pick.

 

[blackened23]

So the OP is a usual suspect? He's butthurt? That's so ridiculous. He's the only one that both expressed a strong desire for an AMD card for gaming and also concern over the pricing of the 280X. Don't think anyone else in the thread did. Most buyers (including myself if I were on the market) would just get the cheaper, same-performing alternative and call it a day without caring. No F's given. If they're gamers, anyway. This of course doesn't apply to miners. At 100-150$ over MSRP, how much time does that add to a single GPU in terms of mining time to break even? Another 3 months or so? This entire subtopic is over "butthurt" is pretty stupid anyway. Who the heck cares. It's not like there's a single choice on the market if you're a PC gamer. I don't know of anyone that has been butthurt over silicon myself. Shrug.

 

[BallaTheFeared]

Balla, I do talk about price changes all the time in the crypto thread. If you're counting on explosive price growth you are FAR better off buying the coin directly, historically 10-20x better off assuming average USA electric prices. All the money people put into expensive cards, if they had simply bought LTC for a couple of bucks in March, they would be waaaay better off than the guy who bought cards and mined and paid monthly power bills. If you think prices will merely increase somewhat, like say double for 2014, you could go either way; if you think prices will more than double, it's better to buy coins directly assuming you pay 10 cents/kWh. If you pay more than that for electricity, it's more profitable to buy coins directly.

I think you're missing the part where most of our community is made up of kids who post mostly from school and don't pay power bills.

Upfront cash to invest requires a job and disposable income, neither of which are common when it comes to miners. Not saying all, just that isn't isn't common for someone with wealth and a disposable income to waste their time with petty cash.

Pretty easy for a kid to talk their parents into a 7950 for gaming, it's another story asking for $300 to invest in this thing called concurrency.

 

[AnandThenMan]

I think you're missing the part where most of our community is made up of kids who post mostly from school and don't pay power bills.

Hey I did miss a link to this data, please provide.

 

[IEC]

I think you're missing the part where most of our community is made up of kids who post mostly from school and don't pay power bills.

Upfront cash to invest requires a job and disposable income, neither of which are common when it comes to miners. Not saying all, just that isn't isn't common for someone with wealth and a disposable income to waste their time with petty cash.

Pretty easy for a kid to talk their parents into a 7950 for gaming, it's another story asking for $300 to invest in this thing called concurrency.

Yup, petty cash:
http://www.zerohedge.com/news/2013-12-25/trip-through-bitcoin-mines

 

[BallaTheFeared]

Your confusing the conversation.

 

[HeavyClocker]

As a AMD graphics fan/user I am going to buy 780 because of these prices.

 

[nenforcer]

Seeing how mining ruined prices for AMD, now I'm wondering, will Nvidia modify their next architectures to totally dominate in mining and wipe the floor with AMD?

Highly unlikely nVidia would want to cannabalize their consumer grade gaming graphics cards. Those cards are purposely gimped for compute purposes as as not to interfere with their Quado line of professional workstation / engineering cards with higher double precision performance.

However, I haven't heard that the Quadro cards perform any better for mining purposes and they already have a huge premium over the Geforce counterparts.

What AMD should do is move their mining performance advantage over to their FireGL Pro cards to give them a push and so not to have the prices jacked up on their consumer Radeon graphics cards.

That is the only reason the miners love the AMD Radeon cards because they are cheap and perform better than any other card on the market.

 

[BallaTheFeared]

You can't gimp mining directly, it's SP and cutting SP would cut gaming performance.

Likewise it wouldn't make any sense to make it a feature of FirePro since Firepro is simply a different support structure and DP now that AMD has gimped DP on non Firepro cards.

 

[SPBHM]

what about the 270x? it's pretty easy to find for the MSRP, which means, half the 280x price at the moment, is it a bad card for mining?

 

[nwo]

what about the 270x? it's pretty easy to find for the MSRP, which means, half the 280x price at the moment, is it a bad card for mining?

A regular 270 can easily get 450 or higher which is pretty good if you look at hashes/$. If you ask me, it is a great bang for the buck mining card. But you can only have so many of them before you have to build a new rig. That's where 280x or higher cards shine. They give you more hashes without taking up PCI-e slots.

 

[Bubbleawsome]

what about the 270x? it's pretty easy to find for the MSRP, which means, half the 280x price at the moment, is it a bad card for mining?

MSRP is $199 and most reference are running $219 with overclocked models ranging from $229 to $259. The miners finally used all the 280x's and are eating the 270 series now. :/
Sucks for me, I needed a new 270x and now they are all going OOS and up.