Topic: DO NOT BUY GPUs to mine! You will not get your money back! - page 6. (Read 21049 times)
If I had a dollar for every time I have been told this same thing...
Wait. I'm confused. You are saying you will earn $293 on the card but you won't get your money back? When you are done with the GPU resell it. I imagine in a year 7950s will fetch $150 so your $275 investment returned $443 when shutting down the operation.
Unlike ASICs, there is an aftermarket for GPUs.
Unlike ASICs, there is an aftermarket for GPUs.
Just don't tell the buyer you mined with it for a year ;-)
Also, as every GPU out there will be uneconomical in a year, there will be a glut of used cards, and I think $150 is optimistic, given 8xxx cards coming out, etc. Also, my example was for a 7970, not 7950, which is a bit less powerful (e.g 500 MH @ 200 W).
But point taken re: resale value. However, if we even see 15% hash rate growth vs. 10% in my example, that 500/200 card returns a mere $137 over its life, barely over 1 BTC. With resale of $150, you're basically at break-even. You have to be pretty certain that hash rate growth stays under 15% at this point.
My point is that it's a very bad time to be thinking about a GPU farm, and even small underestimations in hash rate increases result in disastrous ROI.
I do agree that for BTC mining that now is not the time to buy a GPU. Because there are more efficient means to mining BTC, I think we'll see the difficulty increase but the price won't fluctuate much due to miner pressure. Or rather, I suppose I should say that more hashing capacity will come online as the price stays stable.
However, I do believe there is room for more than one crypto-currency and I believe LTC has sufficient presence to survive and thrive. I do think though that LTC is undervalued slightly right now and we'll see LTC profitability compared to BTC at 'equal'(1000:1 hash ratio) hash rates increase slightly.
Wait. I'm confused. You are saying you will earn $293 on the card but you won't get your money back? When you are done with the GPU resell it. I imagine in a year 7950s will fetch $150 so your $275 investment returned $443 when shutting down the operation.
Unlike ASICs, there is an aftermarket for GPUs.
Unlike ASICs, there is an aftermarket for GPUs.
I am not 100% convinced there won't be a market for ASICs after the difficulty makes mining BTC unprofitable. I saw a guy on reddit that said he is going to try and use the Erupter USBs to develop an ASIC for Litecoin. If it can be done on the USB version, I am sure someone will do so for the larger ASICs as well. Alt coins are a big question mark, but if there is still interest down the line, there will be a resale potential in that market. Litecoins were supposed to be GPU-proof. They weren't. So they are most certainly not ASIC-proof either.
I have not read much about Scrypt but my understanding is that the algorithm used is so memory intensive that there may not be the performance gain that we saw going from GPU/FPGA to ASIC with BTC. It sounds like you'd basically be building a compute core with access to massive amounts of bandwidth, also known as a GPGPU.
Wait. I'm confused. You are saying you will earn $293 on the card but you won't get your money back? When you are done with the GPU resell it. I imagine in a year 7950s will fetch $150 so your $275 investment returned $443 when shutting down the operation.
Unlike ASICs, there is an aftermarket for GPUs.
Unlike ASICs, there is an aftermarket for GPUs.
I am not 100% convinced there won't be a market for ASICs after the difficulty makes mining BTC unprofitable. I saw a guy on reddit that said he is going to try and use the Erupter USBs to develop an ASIC for Litecoin. If it can be done on the USB version, I am sure someone will do so for the larger ASICs as well. Alt coins are a big question mark, but if there is still interest down the line, there will be a resale potential in that market. Litecoins were supposed to be GPU-proof. They weren't. So they are most certainly not ASIC-proof either.
I am not sure what the ASIC market would be--current ASICs are SHA-256 and are good for absolutely nothing else other than SHA-256 mining. It's in the hardware.
There is nothing preventing scrypt ASICs of course, except the large initial investment, but again, they would be useless for anything other than scrypt mining.
Wait. I'm confused. You are saying you will earn $293 on the card but you won't get your money back? When you are done with the GPU resell it. I imagine in a year 7950s will fetch $150 so your $275 investment returned $443 when shutting down the operation.
Unlike ASICs, there is an aftermarket for GPUs.
Unlike ASICs, there is an aftermarket for GPUs.
I am not 100% convinced there won't be a market for ASICs after the difficulty makes mining BTC unprofitable. I saw a guy on reddit that said he is going to try and use the Erupter USBs to develop an ASIC for Litecoin. If it can be done on the USB version, I am sure someone will do so for the larger ASICs as well. Alt coins are a big question mark, but if there is still interest down the line, there will be a resale potential in that market. Litecoins were supposed to be GPU-proof. They weren't. So they are most certainly not ASIC-proof either.
I'm still able to pay for 2x cards a month but 6 of 8 of the cards will be off free electricity. Then when the cards are no use most likely still worth 60% of the cost on ebay. I don't see the issue. If your plan selling the cards back at the end your going to make the $$ faster.
Obviously free electricity changes the economics entirely, and is:
1. in violation of my assumptions (let's try to stay within them or argue that they are unrealistic)
2. extremely unusual--most of us pay for electricity.
Quote
Also usually the price of BTC goes up with difficulty.
I totally disagree with this. Difficulty (up until now) has changed as a result of price changes. Miners respond to incentive.
I say "up until now" because ASIC miners are basically profitable at any price, and so there will be very little relation between price and difficulty for some time. Price will fluctuate, difficulty will rise.
Wait. I'm confused. You are saying you will earn $293 on the card but you won't get your money back? When you are done with the GPU resell it. I imagine in a year 7950s will fetch $150 so your $275 investment returned $443 when shutting down the operation.
Unlike ASICs, there is an aftermarket for GPUs.
Unlike ASICs, there is an aftermarket for GPUs.
Just don't tell the buyer you mined with it for a year ;-)
Also, as every GPU out there will be uneconomical in a year, there will be a glut of used cards, and I think $150 is optimistic, given 8xxx cards coming out, etc. Also, my example was for a 7970, not 7950, which is a bit less powerful (e.g 500 MH @ 200 W).
But point taken re: resale value. However, if we even see 15% hash rate growth vs. 10% in my example, that 500/200 card returns a mere $137 over its life, barely over 1 BTC. With resale of $150, you're basically at break-even. You have to be pretty certain that hash rate growth stays under 15% at this point.
My point is that it's a very bad time to be thinking about a GPU farm, and even small underestimations in hash rate increases result in disastrous ROI.
I'm still able to pay for 2x cards a month but 6 of 8 of the cards will be off free electricity. Then when the cards are no use most likely still worth 60% of the cost on ebay. I don't see the issue. If your plan selling the cards back at the end your going to make the $$ faster. Also usually the price of BTC goes up with difficulty.
You are not doing it right if you aren't making good on your investment with mining.
In my experience if you pay attention and assess which is the most profitable coin to be mining on a daily or weekly basis you will adapt to the shifts in popularity and keep yourself in the green. If you need any tips, feel free to PM me with your issues.
My 2 cents,
Steelburn@Coincepts
In my experience if you pay attention and assess which is the most profitable coin to be mining on a daily or weekly basis you will adapt to the shifts in popularity and keep yourself in the green. If you need any tips, feel free to PM me with your issues.
My 2 cents,
Steelburn@Coincepts
You're basically arguing with my first assumption. While I agree that alt coins can be profitable early on (and I have made some quick bucks on FTC and CNC), I believe this to be a fad, with less interest in each new alt coin. Also, as more and more miners get clued in on this strategy, the difficulty rises so quickly that alt coins quickly decrease to 100% of BTC profitability. As I said, there are quick bursts of profit to be made with new alt coins, but things quickly revert to the mean. Even LTC has effectively been less profitable than BTC for awhile now, given additional power consumption and NMC bonus.
Currently, there is no alt coin consistently more profitable than BTC.
Wait. I'm confused. You are saying you will earn $293 on the card but you won't get your money back? When you are done with the GPU resell it. I imagine in a year 7950s will fetch $150 so your $275 investment returned $443 when shutting down the operation.
Unlike ASICs, there is an aftermarket for GPUs.
Unlike ASICs, there is an aftermarket for GPUs.
This is probably true at the same time, people without electricity costs it's still somewhat worth while even if you can resell the card or you genuinely want a new card, depends if you buy first hand kit for a GPU mining rig then your not to smart as second hand gives you a lot better chances.
ASIC is going to be weird as sure they're the next step but once they become obsolete it's going to be pretty useless, at least graphics and FPGA had some use afterwards and therefore resell value.
ASIC is going to be weird as sure they're the next step but once they become obsolete it's going to be pretty useless, at least graphics and FPGA had some use afterwards and therefore resell value.
You are not doing it right if you aren't making good on your investment with mining.
In my experience if you pay attention and assess which is the most profitable coin to be mining on a daily or weekly basis you will adapt to the shifts in popularity and keep yourself in the green. If you need any tips, feel free to PM me with your issues.
My 2 cents,
Steelburn@Coincepts
In my experience if you pay attention and assess which is the most profitable coin to be mining on a daily or weekly basis you will adapt to the shifts in popularity and keep yourself in the green. If you need any tips, feel free to PM me with your issues.
My 2 cents,
Steelburn@Coincepts
Here are my assumptions to arrive at this conclusion:
If these assumptions are correct (and I think they're essentially best-case), you can expect to make 3.4 BTC per GH of hashing power over the economic life of the card (which is about 237 days at the moment). To my knowledge, this means that there are ZERO economic GPUs out there right now.
Let's take a 7970 at 200 watts and 650 MH/s (which is a generous 3+ MH/J):
Total BTC earned after electricity: 2.53 ($293.77)
Time to 0 BTC/day: 280 days
In short, DO NOT buy a card just to mine.
- Alt coins are basically "dead," or at best, give a quick bonus for mining early, then quickly go down to 100% or less of BTC.
- Difficulty will continue to increase by at least 10% every two weeks for at least 7 months. We've been averaging 17% every two weeks, and we have not even seen the biggest ASIC surges yet.
- Bitcoin price may rise, but it may not. Assuming a miner cashes BTC immediately, later rises in BTC mean very little additional income. If you expect BTC to rise much later, just buy BTC now.
- Assume 2.5 "real-world" MH/J efficiency--that is about the best I've seen.
- Assume a base of $116/BTC
- Assume $0.11/KWh.
If these assumptions are correct (and I think they're essentially best-case), you can expect to make 3.4 BTC per GH of hashing power over the economic life of the card (which is about 237 days at the moment). To my knowledge, this means that there are ZERO economic GPUs out there right now.
Let's take a 7970 at 200 watts and 650 MH/s (which is a generous 3+ MH/J):
Total BTC earned after electricity: 2.53 ($293.77)
Time to 0 BTC/day: 280 days
In short, DO NOT buy a card just to mine.
Jump to: