5830s selling out - no more at Newegg

[jfourmo]

A simple estimate of the bitcoin community puts miners in the roughly 10-20,000 range. New card sales are likely less but even assuming an optimistic estimate, we'll say 20,000.

In the first 9 months after the 5xxx series was released, AMD sold 16million 5xxx chips. 20,000 / 16million = .1% Bitcoin is a drop in the bucket at the moment. I'm glad people are being optimistic about bitcoin, that's what it needs, but really let's keep things in perspective.

Screw bitcoin :p I'm in it for the money

[Adam]

The 20,000 is probably pretty close to accurate, so if the 16M number sold is accurate why has there been such an impact on the video card market?  You can't reliably find 5850s, 5870s, 5970s, or 6990s in stock anywhere, and even 5830s are going.  If we're only .1% of the market, it shouldn't have had that big of an effect.  Does that mean there is like 50k G/hash in cards purchased recently and still waiting to come online?

It's possible, and I know I'm part of it since I bought 27 cards in the past week 

[Freakin]

Yeah it's NEVER been this hard to find such a large number of different current-gen cards.  We are definitely having a measurable effect

[noobboon]

I just bought one from newegg too.

got it today.

waiting for the extender cable.

Hopefully, if these card are sold out, less people are willing to join mining since they will have to buy a much more expensive hardware.

[jfourmo]

The 20,000 is probably pretty close to accurate, so if the 16M number sold is accurate why has there been such an impact on the video card market?  You can't reliably find 5850s, 5870s, 5970s, or 6990s in stock anywhere, and even 5830s are going.  If we're only .1% of the market, it shouldn't have had that big of an effect.  Does that mean there is like 50k G/hash in cards purchased recently and still waiting to come online?

It's possible, and I know I'm part of it since I bought 27 cards in the past week 

Is the 20,000 miners or cards sold to miners? I personally have bought around 30 cards so far, other people who look at this as an investment are buying 5-10 without much hesitation. How many of the 16,000,000 cards are sitting in stores or in prebuilt computers waiting to be sold?

[Jack of Diamonds]

The 20,000 is probably pretty close to accurate, so if the 16M number sold is accurate why has there been such an impact on the video card market?  You can't reliably find 5850s, 5870s, 5970s, or 6990s in stock anywhere, and even 5830s are going.  If we're only .1% of the market, it shouldn't have had that big of an effect.  Does that mean there is like 50k G/hash in cards purchased recently and still waiting to come online?

It's possible, and I know I'm part of it since I bought 27 cards in the past week 

Is the 20,000 miners or cards sold to miners? I personally have bought around 30 cards so far, other people who look at this as an investment are buying 5-10 without much hesitation. How many of the 16,000,000 cards are sitting in stores or in prebuilt computers waiting to be sold?

Most are sitting in gamers' PC's. They purchase the vast majority of all graphics cards.
The cards are also largely optimized for gaming; If they were optimized for OpenCL (besides it just being another feature) instead of gaming we'd be seeing double hash rates or better.

[bsd]

Well keep in mind that the 5xxx gpus are from late 2009: https://secure.wikimedia.org/wikipedia/en/wiki/Evergreen_%28GPU_family%29#Products

I wouldn't be surprised if the 5830s at Tiger Direct sell out soon  

[Adam]

If all those gamers found out about bitcoin, the hash rate could go to 5,000,000 Ghash/sec.

[kcobra]

I seriously doubt that BTC mining has anything to do with the lack of 5000 cards. The line has been replaced with the 6000 line. Manufacturers have probably not been producing the 5000 cards for quite a while (the Sapphire being the exception as it was a revised version just released in April, 2011).

[airdata]

I'm glad all of my hardware is paid off and earning extra cash - at this rate, by the time you people with 30+ cards make your money back, the difficulty will be over 2 mil.

As long as the bitcoin value rises at a fair rate along w\ the difficulty, the difficulty really isn't too big of a deal.

Make 1 btc per day and it's worth $8 or make .5 bitcoins a day but they're worth $16...

[enmaku]

As long as the bitcoin value rises at a fair rate along w\ the difficulty, the difficulty really isn't too big of a deal.

Make 1 btc per day and it's worth $8 or make .5 bitcoins a day but they're worth $16...

That's exactly what I keep saying. It's hard to really argue whether mining difficulty and price are *actually* linked by the market or if they just *seem* like they are, but they tend to fluctuate within a certain range, sort of a homeostatic curve if you look at price/difficulty and difficulty/price ratios. To an extent I'd be willing to say that it's a simple supply/demand thing considering that we the miners are, in fact, the supply.

[bcpokey]

The 20,000 is probably pretty close to accurate, so if the 16M number sold is accurate why has there been such an impact on the video card market?  You can't reliably find 5850s, 5870s, 5970s, or 6990s in stock anywhere, and even 5830s are going.  If we're only .1% of the market, it shouldn't have had that big of an effect.  Does that mean there is like 50k G/hash in cards purchased recently and still waiting to come online?

It's possible, and I know I'm part of it since I bought 27 cards in the past week  

Pretty simple really, 5xxx series cards are all last gen. They stopped being produced/bought/stocked half a year (or more) ago. Look online for 6xxx gen cards, they are a little more expensive, don't hash quite as well, but they're still very good hashing cards, they are not out of stock anywhere. The reason is that this is what is being bought up and stocked by retailers, not older generation technology. The market is pushed by gamers, and gamers don't want old stuff, they want new stuff. We managed to scrape up the dregs of what was left over and unwanted.

There is a fairly large used card market, and thats where you'll find some older gen cards, though people are snapping those up now admittedly. But check ebay as an example, there are still > 100 5870s for sale on any given day. At bloated prices, but they're there.

[Adam]

As long as the bitcoin value rises at a fair rate along w\ the difficulty, the difficulty really isn't too big of a deal.

Make 1 btc per day and it's worth $8 or make .5 bitcoins a day but they're worth $16...

That's exactly what I keep saying. It's hard to really argue whether mining difficulty and price are *actually* linked by the market or if they just *seem* like they are, but they tend to fluctuate within a certain range, sort of a homeostatic curve if you look at price/difficulty and difficulty/price ratios. To an extent I'd be willing to say that it's a simple supply/demand thing considering that we the miners are, in fact, the supply.

But remember if that's true then just buying BTCs is way better than investing in hardware.  Difficulty is growing at like 30% a week, if BTCs keep pace with that than it would be the best investment ever.

[enmaku]

But remember if that's true then just buying BTCs is way better than investing in hardware.  Difficulty is growing at like 30% a week, if BTCs keep pace with that than it would be the best investment ever.

I'm also betting that mining is not exponential but homeostatic as well. If we get to the point where the difficulty is too high, there will be less incentive for new miners to join and eventually incentives for old miners to quit, at which point the difficulty:price ratio will decrease again. Simply put, mining is a long-term investment; if you buy at $10 and sell at $20 who cares if the market crashes after, you made your money and got out. Buying is for those who want to make bank and don't care about the movement. Mining is for true believers.

[Freakin]

As long as the bitcoin value rises at a fair rate along w\ the difficulty, the difficulty really isn't too big of a deal.

Make 1 btc per day and it's worth $8 or make .5 bitcoins a day but they're worth $16...

That's exactly what I keep saying. It's hard to really argue whether mining difficulty and price are *actually* linked by the market or if they just *seem* like they are, but they tend to fluctuate within a certain range, sort of a homeostatic curve if you look at price/difficulty and difficulty/price ratios. To an extent I'd be willing to say that it's a simple supply/demand thing considering that we the miners are, in fact, the supply.

[enmaku]

That's beautiful Freakin. 

I like the look of that straight line, which clearly states "bitcoin price is directly proportional to difficulty"

[bcpokey]

As long as the bitcoin value rises at a fair rate along w\ the difficulty, the difficulty really isn't too big of a deal.

Make 1 btc per day and it's worth $8 or make .5 bitcoins a day but they're worth $16...

That's exactly what I keep saying. It's hard to really argue whether mining difficulty and price are *actually* linked by the market or if they just *seem* like they are, but they tend to fluctuate within a certain range, sort of a homeostatic curve if you look at price/difficulty and difficulty/price ratios. To an extent I'd be willing to say that it's a simple supply/demand thing considering that we the miners are, in fact, the supply.

But remember if that's true then just buying BTCs is way better than investing in hardware.  Difficulty is growing at like 30% a week, if BTCs keep pace with that than it would be the best investment ever.

That's not true. That's only true if the number of coins you can mine are less than the number of coins you can buy for the same amount of money. At 30% difficulty increases mining rewards you much more greatly than buying. At 60% difficulty increases however buying rewards you more greatly. Assuming static increases of course.

If for example however difficulty precedes price you might lose out (that is to say if you buy mining hardware right before a large difficulty jump that precedes a corresponding price jump it could be different profitability)

[enmaku]

That's not true. That's only true if the number of coins you can mine are less than the number of coins you can buy for the same amount of money. At 30% difficulty increases mining rewards you much more greatly than buying. At 60% difficulty increases however buying rewards you more greatly.

And since there are both mathematically-minded people who will see this and choose to buy rather than mine as well as a sort of "evolutionary cost" associated with making the wrong decision to mine vs buy at the appropriate time, the number of miners will naturally decrease once this threshold is crossed. Mining will continue on a downward trend until it reaches another critical threshold at which point market evolution selects against buyers and mining increases once again.

A homeostatic curve created by market evolutionary selection between two opposing populations 

[bcpokey]

That's not true. That's only true if the number of coins you can mine are less than the number of coins you can buy for the same amount of money. At 30% difficulty increases mining rewards you much more greatly than buying. At 60% difficulty increases however buying rewards you more greatly.

And since there are both mathematically-minded people who will see this and choose to buy rather than mine as well as a sort of "evolutionary cost" associated with making the wrong decision to mine vs buy at the appropriate time, the number of miners will naturally decrease once this threshold is crossed. Mining will continue on a downward trend until it reaches another critical threshold at which point market evolution selects against buyers and mining increases once again.

A homeostatic curve created by market evolutionary selection between two opposing populations 

Well, I've been arguing that the simple growth model is wrong for some time now. I was merely pointing out that the advice "its more profitable to buy than mine because..." is often touted but is only correct under certain circumstances. Freakin's graph is awesome and helpful.

I don't know that it will necessarily hold true though in the future, so buyer beware.

[enmaku]

That's not true. That's only true if the number of coins you can mine are less than the number of coins you can buy for the same amount of money. At 30% difficulty increases mining rewards you much more greatly than buying. At 60% difficulty increases however buying rewards you more greatly.

And since there are both mathematically-minded people who will see this and choose to buy rather than mine as well as a sort of "evolutionary cost" associated with making the wrong decision to mine vs buy at the appropriate time, the number of miners will naturally decrease once this threshold is crossed. Mining will continue on a downward trend until it reaches another critical threshold at which point market evolution selects against buyers and mining increases once again.

A homeostatic curve created by market evolutionary selection between two opposing populations 

Well, I've been arguing that the simple growth model is wrong for some time now. I was merely pointing out that the advice "its more profitable to buy than mine because..." is often touted but is only correct under certain circumstances. Freakin's graph is awesome and helpful.

I don't know that it will necessarily hold true though in the future, so buyer beware.

Quite true, all markets have some element of inherent risk so definitely "caveat emptor"
That said, I'll keep on mining so long as it's at least paying the electric bill. I may not keep buying additional hardware but the rigs will keep doing their thing; not just because it's profitable, but because I believe in the project and want to see it succeed.