Topic: There will be lots of cheap GPUs for sale eventually (Read 4540 times)

Those of us who are mining for a profit are obviously decent with money and aren't likely to be struggling financially.

tl;dr - I have a job too. Enough to support a wife and 4 kids.

Who exactly are you talking to when you say "people like you"?  Because I can guarantee you there aren't 50TH/s worth of people who will mine at a loss, which is about what it would take to push out those of us who are mining for the profit and have efficient GPU's and decent electric costs.

I went out and bought a whole new rig to mine BTC.  It will probably take me a year or more to pay it off with mining.  However, I was going to buy a new machine anyway and I have a job, so whether or not it is a good investment in your eyes means nothing to me.

Admittedly, I wouldn't have bought 2 outrageous video cards if it weren't for mining (I would have just used the onboard GPU as I'm a sound guy, not a graphics guy).  But still, I am doing it to support the project and because I wanted a new machine.  Not everyone is mining for profitability.  Hell, I don't even sell my coins that I mine...  You greedy profit hungry buy every GPU available to make a profit miners are going to get squeezed by hoarders like me who will mine at any price or difficulty with the expectation of higher BTC prices several years from now.  The funny part is, the cheap GPUs that will eventually be for sale, will come from people like you.  Not the people like me who are buying overpriced GPUs that won't return a profit.     We have jobs and will outlast you.

I disagree regarding FPGAs.  They are insanely expensive for the MH/s, and if you're only looking at the difference in electric costs, it would take years to pay back the difference.  Not to mention, they're not going to be as easy to resell...  Very few people will make significant investments in FPGAs.  The only people who will mine on FPGAs are those who do not care if they are wasting money, or those who already have access to clusters.  It's too risky of an investment otherwise, counting on being able to mine for years and years.

ASICs might have possibilities, but again, it would require significant up front capital that would take months or years to pay off.  But, I could definitely see someone interested in technology who just might give it a try.  They make a whole heck of a lot more sense than FPGAs, at least if the estimated performance figures are in the ballpark.

This is true, but it is a matter of how many people you can find who will mine bitcoins at a loss.  It's not a "feel good" project for most people.  For most miners, it's all about the money, and they will leave when the money leaves.  Sure, there will always be a core group of people who will mine no matter what, but when you're looking at the overall difficulty level, the vast majority of those megahashes comprising the difficulty level will be out the door as soon as it becomes unprofitable.  Which means, it will become profitable again.

Like I said, the only way we'll see close margins for some of the better MH/s/watt video cards is if bitcoins are reduced in value (somewhere on the order of 1/5th of what they are currently worth), or if they somehow, miraculously, stabilize at the exact same value for the next several years.

I remember the times when people (and me as well) distributed some calculation power for SETI@Home, BOINC, Folding@Home, or wanted to be the number one in a benchmark rating chart. This all for free and never got some money back for the wasted energy, invested time, patience, and for the whole equipment used for.
Now, with Bitcoins you get a little bit back for using a computer, and the opportunity to win a lottery is also much fun.
There will be always people who want to participate in a race.

Especially when graphic cards are rare for miners then prices increase.
When Bitcoins becoming seldom the same will happen.

The future of mining is in FPGAs.  CPU mining was replaced by GPUs.  GPUs will be replaced by FPGAs.  FPGAs will be replaced by ASICs.  The latter will only happen if bitcoins manage to keep their value long term and more real world uses are found for bitcoins in general.  ASICs require a very large investment. 

If, by then, many people have quit, how would the difficulty continue to rise?   

I never said the 5970 would be worthless -- though it will diminish to the point of not being worth it for mining, for most people.

Most people will not run a PC with 1 -- or even 2 -- 5970's full bore for a week for $1 or $2 -- and if BTC price wavered just a bit, they'd be losing money!  That's when it would be practically worthless for mining.

I know, by then many people will have quit. But I wouldn't be surprised if mining got a lot less profitable than it is today.

As you point out, there's a large gap between 15% profit and 300%. I think some people would mine even for 15% profit -- because it's fun, they don't care about accounting, etc. Heck, how many people carry credit card debt? I would never carry a credit card balance, but I'm pretty sure a lot of people do! Are they kind of foolish for paying 20% more for everything by living in debt?  Objectively speaking, yes. But a TON of people do it.

130 days is a *long time*.  Even with 10% difficulty increases every 14 days, it's going to be hard to reach $350 with a card purchased today.  And the card is not exactly coveted by gamers. When not one, but TWO graphics card series have been released after it -- both of which excel the 5XXX for gaming -- the card is not going to find a welcome market.

The 6xxx series out out, and look at the prices of the 5xxx series.  Nothing is guaranteed.

It is seriously gonna happen, one way or another. Even it trends won't go as bad as I thought, once 7xxx are on market, there WOULD be lots of cheap GPUs for sale.

Ok, I will conceded that point.

I suppose then, my point is that as long as bitcoin prices continue to rise over the long term, we will not see anywhere close to a 15% margin on mining, and we will continue to see 300%+.  If prices were to stay consistent, that would reduce volatility, people would feel more comfortable with purchasing more equipment even for smaller margins, and we would EVENTUALLY possibly reach a thin, business-like margin.  But I very much doubt the price will stay consistent to allow that to happen, and I don't see where or how you could extrapolate that it would.  Bitcoin is going to either go up or down from here, but it sure as heck isn't going to stay in the same place.

You said that a 5970 purchased today at $350 would never pay for itself, and would be worthless in 130 days.  That is just plain wrong, and is based on the absurd assumption that difficulty will continue to rise at an exponential rate while the price stays the same.  Difficulty will not continue to rise exponentially while the price stays the same, as you even admitted yourself.  In fact, it will continue to rise as an exponentially DECREASING rate if price stays the same.

Now, show me a reasonable model where the price stays the same and the difficulty rises at a exponentially decreasing rate, and show me that you still cannot pay for a 5970 through mining, then reselling the hardware when it becomes unprofitable.  Then, I might believe what you are saying.  But the outright lies that you posted in the OP is just you trying to plant some FUD to keep people away from mining.

Um...another major misunderstanding. I just read my OP, and I don't see this ANYWHERE. Care to point it out to me?

You seem to be contradicting yourself, and there's something you're gravely missing.

You mention 300-400% margins being incredibly attractive, but don't seem to understand that people are going to continue to pile in (albeit at a slower rate than early June), until margins are much more like a business "15% margin".

That's my point: even if price/difficulty stays about where it is, mining capacity will be added THROUGH INERTIA ALONE -- existing miners adding a card here and there, big-time miners adding a few dozen cards here and there, gamers firing up their ATI cards (remember, there are MILLIONS of them) after they find out about Bitcoin mining, and new miners buying a card or two and getting started "because I make $10 -- and only $5 of it is for electricity!" (that would be an example 100% margin -- still a ways in the future).

Right.  My point is, difficulty follows price at a distance.  I do not believe we will ever see a point in time where mining ends up with 15% margins over the cost of electricity.  The market is simply too volatile to warrant people making an investment in hardware for that small of a margin.  And that's why the OP's prediction of mining off a 5970 eventually reaching $0.60 USD is just plain wrong - it will NEVER happen.  Well, at least not until some other graphics card outpaces it by 100x or so, which may as well be never.

Probably not, considering the 28nm manufacturing process only began mass production in late June.

Hell, the 6990 came out barely 4 months ago.

Supposedly difficulty follows price.  And we've been relatively stagnant for a couple weeks now.

1.  130 days to a 100% ROI is a VERY good investment.  Just ask any businessman.
2.  The card won't be worthless when the 7000 series comes out.  Perhaps not worth as much, but far from fully depreciated.
3.  If you're a miner, you want to start mining ASAP.  You don't want to wait for newegg to ship it, and you don't want to pay extra to have it expedited.  Every day not mining is some amount of bitcoins lost.  Craigslist holds the advantage in that you can go pick it up, bring it home, and start mining immediately.
4.  New is better, I'll agree with you there.
5.  Some people don't like rebates, so if $350 is better than the before-rebate price on newegg, I can understand why some people would go for the CL card to save a few bucks.
6.  You can't assume that bitcoin difficulty will continue to increase while price will stay the same.  That's NOT going to happen.  We've already seen a level off of the difficulty, and we're still several multiples above the cost of electricity.  Many companies are happy with a 15% margin on their operations - the typical bitcoin miner is operating on 300-400% margins.  Uncertainty of the future will always keep mining profitable for the vast majority, unless/until the price drops significantly from where it is today.  But as long as the price stays the same, or goes higher, mining will remain profitable.

I just bought a 5870 for $189.99 and already got an offer from someone in my guild for $250   And I haven't even received it yet!

stop babbling dude

a friend of mine sold his 2 month mining 5870 for 40% then he bought: actually he sold for 320$...

we don't give a fuck about you or your efforts in buying gpu cards.
Corner the fucking market who cares.
Stop mining who cares.
Stop the fucking QQ
we are going to die also eventually