Most people are probably giving OP a hard time because it doesn't really make sense. Deepbit has a large variance even averaged over long periods of time. Here are my last 7 solved blocks on Deepbit:

02 Jun 2011 12:38:30
202 / 86798
3615.630 MH/s
+ 0.11636213 BTC
02 Jun 2011 12:34:32
55 / 31106
2693.103 MH/s
+ 0.08840738 BTC
02 Jun 2011 12:33:05
228 / 109837
3281.351 MH/s
+ 0.10379016 BTC
02 Jun 2011 12:28:09
314 / 149217
3327.463 MH/s
+ 0.10521589 BTC
02 Jun 2011 12:21:26
455 / 236195
3047.642 MH/s
+ 0.09631872 BTC
02 Jun 2011 12:10:49
1383 / 728234
2691.448 MH/s
+ 0.09495574 BTC
02 Jun 2011 11:34:20
944 / 472862
3129.525 MH/s
+ 0.09981771 BTC

My current Hashrate is roughly 3200Mhash/sec give or take as reported by miners. You can see deepbit sees wild variation from between 2600MHash to 3600Mhash/sec. A 1GHash/sec difference.

The best data would be to mine for 24 hours on polcbm, and look at the statistics page and view your rewards. Average it out if you wish. Then mine for 24 hours on phatk, and look at statistics again and view your rewards. If you really see over 24 hours a consistently lower payout block after block then you might be on to something. But just looking at the averaged MHash rate isn't really going to work.

I'm not totally sure where your numbers are coming from, but I'm fine with accepting them, CPU mining is a terribly wasteful application of resources anyway. I don't mean to be one of those eco-freakos but the application of 95-140W of electricity usage to gain a dollar or two x tens of thousands of people = massively wasteful use of limited resources.

When we all get some awesome solar energy generation or something go nuts, but til then I'll balk at super inefficient uses of power.

Sigh no multiquote.

Network hashing power peaked at 4THash/sec not 6? To suggest that there would be 10x as many cards hashing it out would require 10x as much return, not impossible but certainly improbable, and one of the scenarios in which the argument "buying coins is better than buying hardware" applies.

If you mean to contact one of the pcb manufacturers, you are still suggesting a design revision which is actually very expensive. I don't think you guys understand the scale of actual production that allows for these things to exist.

http://www.techradar.com/news/computing-components/graphics-cards/amd-shipped-16-million-dx11-cards-in-9-months-703763

In 9 months from launch alone AMD put out 16 million 5xxx series chips. Again I will note that AMD manufactures the GPU not the smaller companies that utilize the chips. They have no ability to fabricate 40nanometer chips. This is an insanely expensive process that is far beyond the scope of any smaller companies. I can only assume that their last gen fabrication line has been shut down but if they had some excess stock I suppose it would be possible for a smaller company to gobble it up and try to make some cards out of them, but again, 20,000; even 200,000 is such a tiny portion of the market that it is not appealing at all. Especially given the healthy appetite people have for the existing unmodified platform that exists.

Not particularly profitable for anyone to build just for mining.  The most powerful mining card the 5870 is fabricated on a 40nm process. 40 NANO-meters. Think about that scale and what it implies as to the difficulty of designing and creating a working piece of technology. There is a reason there are literally only 2 gaming GPU manufacturers out there, cost is prohibitive. The only current alternative is FPGA or some other form of ASIC, which still represents thousands of dollars of investment into silicon for calculation power far inferior to gaming GPUs, and still has no code base to support it.

There may be 10 or 20 thousand hardcore miners? I'm guessing that just contracting a fab plant alone would cost more money than all of them would be able to scrape together.

Not to mention that it wouldn't really be profitable for miners either, well, perhaps the few larger groups who could afford such a project (which would further squeeze out the small fry who didn't / couldn't get on board). Higher hashing power does not increase block rate generation, it merely makes it harder to generate.


Hrm. I'm curious about this premise. The current network hash power is 4Thash / sec no? 4THash / sec = 4000 Ghash /sec = 4000000 MHash/sec no? A 5870 can do ~400MHash/sec, so 4THash = 10,000 5870s. We know for a fact that the network has tons of 5850s, 5830s, 5770s, 4870s, etc., but even assuming just 5870s that number is quite low.

Are we suggesting that globally there were less than 20,000 5870s created for sale?

Any pictures?

Solo mining is a lot like pooled mining except that you will have a much larger variance in block generation. On average you can expect to find 1 block in 15.5 days. Over a long enough timeline you will average out to make a little more than you would donating to a pool, but not by that much, and it's a bit nerve wracking waiting 15-30 days to see any result. It's of course a personal choice.

EDIT: One thing I'm unsure of, maybe someone else can answer. If you are working on a block of Difficulty A, and you do not solve it before the network implements a raised difficulty, do you still work on the block at its original difficulty or the new difficulty? That could be a bummer for a solo miner.

If you want to adjust multiple cards using trixx just select the option to "apply overclocking profile to all similar cards" or whatever the option is called (I don't use trixx but I remember that option when I tried it).

EDIT: Found it, Under settings "Synchronize cards in Multi-GPU config"

You could do 6 cards by using PCI-E extenders. Just a warning though you run the risk of exploding the motherboard if you run 6 5870s, as the PCI-E bus still delivers power to cards despite having external power connectors. I don't know the exact numbers, the spec calls for 75W max, so say 40 * 6 = 240W running through the PCIE Bus.

Of note, a $200 motherboard that requires a $200+ CPU isn't necessarily that money saving. You can find cheap AMD AM3 boards with 5 PCI-slots (x1 works fine) for ~$120 with a $40 sempron that saves you > $250.

In fact you could get 2x $120 mobos + 2x $40 CPUs, have 10 slots for cards and still have spent less than you would have on LGA1366.

Crossfire can cause some problems with mining, but it's usually not detrimental. It has the upside of not forcing you to create dummy plugs for slave cards though which is a plus.

Non-reference cooler performance varies as much as the coolers themselves. I have a dual fan cooler that essentially blocks the next slot down (the cards are literally touching), but it still cools well if I jam a card in with it (I'm a little worried about physical damage but c'est la vie). I have another non-reference cooler that has no clearance issues but if the fan is not given enough space / fresh air it literally chokes. Jumps to 100% fan speed >90C temps and starts to thermally throttle. I have a lot of reference coolers that aren't spectacular either, basically multi-card setups are a huge hassle.

I'd go with a 5870 personally, cooling does not ensure overclocking, there are very few (I don't know of any) 5850s that can hit 400MHash/sec. Most 5870s can hit at least 420Mhash/sec without volt mod on reference cooling. It's hard for me to think in GBP though, so YMMV in terms of returns.

saw a few of these on ebay, haven't seen them for sale in stores (as they are usually bundled with laptops). ~$400. Doesn't seem a great buy to me.

The 6 series actually does not use "significantly more electricity".  In fact a 6870 uses less electricity than a 5850, so... yeah. The 69xx series is a bit more power hungry due to additional tesselation units, rasterizers, higher clocked mem profiles etc, all things that don't help mining at all, and doesn't match up against the 5870 due to the 5870s ridiculous number of stream processors. The 5870 is just a beastly card though, which is why it is the golden miners card, and is not really a fair comparison.

The 6 series has a reduced number of stream processors that are clocked higher and a slightly improved engine. This makes it more power and temp efficient, it scales better in crossfire and does better at tesselation, all great qualities for gamers. For miners the improvements provide essentially nothing, and so you get a slightly more power efficient card that typically overclocks much more poorly for mining.

It has the advantage of being actually available for purchase however.

Definitely do not buy at that price. 400Mhash/sec (high overclock) generates 6 bitcoins per WEEK at current difficulty, currently valued at $8.7 / BTC, that's $52. It would take 6 weeks at the current difficulty (which increases ~every 10 days) to pay off just that one card, not including electricity. Mining is all about buying efficiently.  I wouldn't buy a 5870 for > $200 (and the lower the better).

I've been meaning to write something like that out for a while, as I keep seeing this argument pop up over and over, finally had some free time to sit down and do it heh. I will mention that I may have written that out a bit more grumpily than I should have as I'm sitting in my room that's like 80F from all these stupid mining machines 

Define the parameters of your purely hypothetical situation. When I buy mining stuff in this scenario what is my outlook? Am I forced to play by the rules you coin buyers typically set for someone buying mining stuff such that they must sell coins the instant they are mined?

Actually I will set the parameters myself, if you don't like them, too bad for you.

Theoretical mining rig(s): Caseless -- Used 5870 $175 each // 400MHash/sec; Random mobo choice -- MSI 790X-G45 AM3 $90 each new (cheaper used) // 5PCI-E slots; Cheapo 1x PCIE extenders $5 each; Corsair tx950 $150; some crap ram $20; Sempron 140 $40. There is a thread floating around somewhere about how to build a custom wood enclosure for like $30 of wood and nails that houses 5 x 2 cards very nicely.

2x above (5 5870s) = $2400. Use the last $150-$200 for some fans or something, I don't care. Hashing power of system 4GHash/sec, electricity cost estimated @ 1296kWh / month + fans, let's round it out to 1400kWh/month, locale dependent, 11cents / kWh = $154 / month.

So, I can choose any point in time? Let's go with 2/18 - 4/18. Nice 2 month period of time. This is winter time so the cost of cooling will be opening a window and pointing a fan out of it to vent the hot air.

Miner Start: 0 BTC $2600 Debt $2000 Assets; Buyer Start: 2600BTC $2600 Debt $0 Assets
Difficulty 2/16 ~45000, Miner Generates 1300 BTC in 2 weeks.
Difficulty 3/01 ~65000, Miner Generates 900 BTC in 2 weeks.
Difficulty 3/14 ~80000, Miner Generates 700 BTC in 2 weeks.
Difficulty 4/01 ~80000, Miner generates 700 BTC in 2 weeks.

End of 2 month period:

Miner End: 3600 BTC $2900 Debt ~$2000 Assets; Buyer End: 2600BTC $2600 Debt $0 Assets

Bitcoin Value Start $1 / BTC; Bitcoin Value End: $1 / BTC

If they both sold at 4/18:
Miner Net profit $700 (excl assets), $2700 (incl assets). Buyer Net profit $0.

Source: MtGox all time value list. Bitcoin.sipa.be network computation graph 8/10 - 6/11, bitcoin mining calculator.

If you continue the trend then the miner continues making more and more coins and as bitcoin value explodes pulls even farther ahead. Because I'm lazy I'm going to extrapolate from 4/16 to 6/1 using the 6/1 difficulty (incredibly unfair to the miner) of 434000 for for the remaining 6 weeks. @ 434000 difficulty miner generates 67.5BTC / wk, or 405 more BTC.

on 6/1
Miner: 4005 BTC $3125 Debt ~2000 Assets; Buyer End: 2600BTC $2600 Debt $0 Assets
Bitcoin Value Start $1 / BTC; Bitcoin Value End: $8.7 / BTC

Miner Net: $31718.5 (excl assets); Buyer Net: $20020

Any time when bitcoins are profitable but not exploding in value mining is a superior cost-benefit.
Again, I'm not arguing that either one is superior, just that I'm sick of fanboys touting one method or the other. No one knows the future. Stop pretending you do. If you have chosen one or the other that's fine.

God do we really need to have another one of these discussions?

They're both fine options, and every argument either way presumes some sort of knowledge of the future that the arguer does not actually have.

If you expect bitcoin to blow up exponentially in the next few days/weeks then obviously buy coins, if you expect bitcoin to remain stable(ish) then buy mining stuff. If you expect bitcoin to die a horrendous death, maybe you shouldn't be here.

Buying coins is easier, building mining rigs is more interesting for enthusiast/hobbyists. Find your passion.


EDIT: Thanks JJG for demonstrating my point. For the record you've been saying the same thing for a couple months now, and while you might be right in the future, you've been wrong up til now.

One interesting thing to note, last difficulty spike was painful, very very painful. And as a natural result, it has discouraged a lot of people. Last difficulty bump ~70%, next difficulty bump, ~17%. Things are not arbitrary.

Well, it depends on your reasons for mining in the first place. If you just like the idea of bitcoin then that's one thing, but if you are trying to make a profit, you want to maximize your cost / generation. This means either as many cards as possible on one rig or as cheap as possible multiple rigs.

For cases there is what I suggested, the Rosewill THOR, it's fairly reasonably priced (if you can find it) and has 10 expansion slots, enough for 5 cards if you space them properly (they will be cramped and hot though, and it has only ok airflow so you will need to set up some fans). Another option is a tech bench, mostly open air. You can also set up a rig like the one in the picture you list, except that the side card will block airflow from the sides getting to the crammed together cards on the motherboard. You can also get an HAF 912 (again, poor air flow, smallish case, but has 8 slots for some reason) for very cheap, an HAFX (huge case, lots of airflow, kinda expensive) which has 9 slots to uncramp things a little bit. Only the antec nine hundred TWO (v3) will be able to accomodate 4 cards, and  it would do an okay job of it I suppose.

3 cards will fit in almost any case though, 300, 900, 1200, Lian Li k57, etc. But again, when you are building a rig you want to maximize the number of cards you can fit into a single rig, or else cheap out as much as possible on multiple rigs.

Another nice option that is in between cased and caseless are what are known as "Tech benches". They are basically open-air but fully rigged up cases. One example I've been looking at would be:

http://www.amazon.com/MYOPENPC-Transparent-Acrylic-Station-motherboards/dp/B004TZ56GU/ref=sr_1_13?ie=UTF8&m=A3Q4NSDXS8U7OP&s=generic&qid=1306788458&sr=1-13

I'm still trying to work out which ones would best fit the most cards though.

Just a note, you can't fit 4 double slot video cards into an antec 1200, let alone 5. Nor would you be able to fit 5 cards on the motherboard anyway, you'd need an extender for the 5th pci-E slot. That motherboard is kind of pricey for what you plan on doing, I'm not totally sure I understand whether you are building a second computer for gaming or plan on using this motherboard/cpu, but an AMD sempron will bottleneck the crap out of 6970s, so you won't be able to game on it anyway.

The only cases capable of doing 4 cards are ones with 8+ expansion slots. The only case I know capable of doing 5 cards for under $300 is the Rosewill THOR.

EDIT: Also worth noting that you will be hard pressed to find any 5850s these days. Amazon has been out of stock for about a month and there's no real reason to believe they will get any more anytime soon. Your best bet is to buy used cards or else you will have to go skinny with 5830s, another reason why it might be better to go with 2 cheaper motherboards that can utilize PCI-E x1 slots with extenders.

EDIT #2: The antec quattro is an incredible waste of money. You will likely be able to run 4x 5850s on a quality 750Watt power supply (maybe even less but let's play it safe). You can pick up a quality one of those for about half the price.

Trying out the client now, so far I like it, the affinity setting is something I really like, and the backup option was just what I was looking for. One thing though is the OSD is always click-through for me no matter if I hit the button or try to alt-tab to the window or anything. I saw someone else had this problem but I didn't see any resolution, has anyone come up with anything for this?

 

I'm not really sure why these threads keep popping up.

Mining can be fun and profitable. Buying coins can be fun and profitable.

Do whatever you like, your reasoning is not necessary.