Pictures of your mining rigs!

[Turbor]

I have a personal preference for page 43.  

I hate you

[P_Shep]

Try not to be too jelly of my gargantuan 150MH/s...

[rjk]

Here is a "Mini-miner" that I put together:




(That is one 5870 and one 5850)
And here are some more pics of the 6-card rig, taken with a better camera:




(All 5870s)

[smracer]

[Electricbees]

Monitors on walls have my approval! 

[paraipan]

Monitors on walls have my approval! 

nicee, looks like a halflife 2 lab

[Transisto]

[rjk]

Ahahaha, awesome!

[cuz0882]

First PR-2000 off the line. Generates over 2.0 GH/s using 3x 6990s at stock clocks.

I would remove the hot swap and stick a 140mm fan in its place. The HAF-X has terrible airflow despite its name. But that helped mine a lot. Still 7 degrees hotter then my corair case though. I also replaced the 200 mm front fan with a 140mm fan.

[Electricbees]

2 rigs, 7 cards, 3 PSU's between them. Hazard waiting to happen.

[rjk]

2 rigs, 7 cards, 3 PSU's between them. Hazard waiting to happen.

Lol! I remember (ab)using a bunk bed in that manner, except that it was with Romex, christmas lights, and an old mechanical time switch. Definitely a fire hazard

EDIT: Ah, I see the metal racking everything is mounted on, that is better than my setup where I used wire staples and fastened it directly to the bed. 

[MrGaSp]

First PR-2000 off the line. Generates over 2.0 GH/s using 3x 6990s at stock clocks.

I would remove the hot swap and stick a 140mm fan in its place. The HAF-X has terrible airflow despite its name. But that helped mine a lot. Still 7 degrees hotter then my corair case though. I also replaced the 200 mm front fan with a 140mm fan.

Mind telling me how hot your hard drive gets?

[molecular]

2 rigs, 7 cards, 3 PSU's between them. Hazard waiting to happen.

Do you sleep there?

[Electricbees]

2 rigs, 7 cards, 3 PSU's between them. Hazard waiting to happen.

Do you sleep there?

Hardly, but yes.  It's some 100 square feet of dorm room, which turns to 100 square feet of hellfire if the door and window are shut.
Window's open now, average temp inside is about 48 Fahrenheit. It's a little warmer when I actually SLEEP there, but I don't do that often...
The building has a mouse problem.

[portron]

Couple of these 4x 5870

[yochdog]

Can you please tell me how much do you estimate this setup cost you?
It's quite magnificient and quite a job!

Cost for you to build this system today:

16 X6500s at $550 each: $8800
Power supply (I used this one): $65
4 120 mm fans at $15 each: $60
Various Molex splitters and adapters: $20
16 USB cables: $50
USB hubs: $30

Total: $9025 / 4.2 GH/s or 2.12 $/MH

A lot of those extra things you probably have sitting around already, or you could find better deals.

The next price break above that is at 25 units. You could still run that many boards off of a single supply (400W) so your overhead cost would only be slightly higher.

25 X6500s at $535 each: $13,375
Power supply: $65
Fans: $90
Various Molex splitters and adapters: $30
25 USB cables: $75
USB hubs: $40

Total: $13,675 / 6.6 GH/s or 2.05 $/MH

You will also need a host computer to run the system, but you most likely already have that and the performance requirements on that computer are quite low.

Note that those $/MH are based on 266 MH/s for each X6500. I have every reason to believe that we will reach higher hashrates with improvements to the software and firmware in the future.

This is fascinating to see someone actually do an FPGA farm in scale.  I could not stomach the upfront capital cost needed to replicate my 18,000 MH/s.  My my rough math, it would cost $35,000+ to get to that speed!

Now clearly cost savings would start to offset the capital cost, but that is a loooong payback.  Also, the salvage value is much higher on GPU's, which has a huge impact on the NPV calculation. 

I have invested $12,500 into my operation, and get 18,000 MH/s.  $.69/MH.   Pretty damn good!

[fizzisist]

Note that those $/MH are based on 266 MH/s for each X6500. I have every reason to believe that we will reach higher hashrates with improvements to the software and firmware in the future.

This is fascinating to see someone actually do an FPGA farm in scale.  I could not stomach the upfront capital cost needed to replicate my 18,000 MH/s.  My my rough math, it would cost $35,000+ to get to that speed!

Now clearly cost savings would start to offset the capital cost, but that is a loooong payback.  Also, the salvage value is much higher on GPU's, which has a huge impact on the NPV calculation.  

I have invested $12,500 into my operation, and get 18,000 MH/s.  $.69/MH.   Pretty damn good!

I think it's time to update those calculations with our newer performance figures. As of the last few weeks, I've been running all of these boards at 180 MHz, or 360 MH/s per X6500. In the last couple days, I've been experimenting with 200 MHz, which seems to not work reliably with every board (some handle it fine as is). On one board that was having trouble, I removed the heatsinks and replaced the thermal tape with thermal epoxy. After that it has been running perfectly at 200 MHz. Power usage measured at the wall with a Kill-a-watt (before the 80 Plus Bronze PSU) is roughly 20 W (still need to measure this carefully at the board).

Assuming that switching to thermal epoxy will work well for the rest of them, let's say each X6500 will do 400 MH/s. This gives, for the 25 unit system, 10 GH/s instead of the 6.6 GH/s I calculated before. The price is still the same, so the cost/performance is reduced to only 1.358 $/MH/s.

Not too shabby, right?  

[fb39ca4]

I read through this whole thread 

[paraipan]

I read through this whole thread 

nice, now you have the other 99.99% of the forum left 

[MrGaSp]

I'd take a picture of my single card rig, with a fan rigged onto the side of it, but the fan got bumped and lost quite a few fins.  GPU 1, Fan 0