You used to see a good amount of blocks getting pushed that didn't have transactions on them.  This wasn't because transaction volumes were low, mind you; these blocks could be ten minutes after the last blocks, with an average number of transactions per normal block above ten, and yet they'd still have zero transactions on them.  My best guess was that someone (or some group) was running custom mining software that didn't bother to include transactions on blocks (as it's harder to code in handling all of the additional transactions coming off the network than to just code a single generate transaction and be done with it).  With the network's huge growth in recent months, however, these "null" blocks have gotten a lot, lot less frequent, so whoever was doing them is becoming a much smaller part of the network percentage-wise.  All of the big pools are using proper software that includes transactions in the mined blocks, obviously.

I wonder what would happen if you used another x1 adapter to connect the graphics card in the x16 slot?  Might that free up more lanes?

In my (working) configuration on a different Gigabyte motherboard, I have two videocards plugged into x16 slots, one using an x1 extender into an x1 slot, and another using an x1 extender into an x1 slot.

As long as you aren't opening any ports in your firewall for it you should mostly be good to go.  However, I would definitely as least recommend changing the password for the "user" account from "live" to something that's actually secure.  Other than that -- well, it is Debian GNU/Linux, which isn't exactly a security lightweight (it's not going to have the same vulnerabilities as, say, running an unpatched version of Windows XP).

I wouldn't bother trying to slim it down to make the 700 MB limit if it's going to slow down the release.  People here are running Linuxcoin on USB sticks, not Live CDs.  And regardless, if they really want to use optical media for some reason, they can just burn it to a DVD instead of a CD.

Nice, what mining program?

Run this to find it:

I've heard of some really hacked up solutions that involve wiring up the servo motor on one server's CD-ROM tray to the momentary reset switch on another server's motherboard.  And then do it vice-versa.  So long as both servers aren't down, you can always log into one to reset the other.

I don't see any possible advantages here?  If you ever received a private key from someone as payment, the very first thing you would do is generate a transaction to send all the coins from that private key to one of your own addresses that you know no one else has access to.  So any time money is transferred between two people you'd have to (a) send the private key and then (b) do a transaction to a non-shared address.  Just doing a normal transaction from the get-go is simpler.

I had to install the package firmware-atheros to get networking working on my WiFi card, but that's not a big deal, and I wonder how many other people are going to be mining on WiFi cards with this similar chipset.  As long as I'm using persistence I'm not worried about something I only have to do once.

The major thing is to just get the latest version out there so I can overclock my cards beyond the limits that the Catalyst 11.5 drivers allow! 

Thanks for this thread!  I looked and it turns out that my state (Maryland) is deregulated.  I believe I've gotten cheaper electricity through another provider by switching, but I'll only believe it once I see that first bill.

How are you so sure that "you WILL get your money back"?  You can do the math really simply -- based on the output of that card, and exponentially rising difficulty levels, it won't pay for itself (or, at the very least, it will output fewer BTC than if you just put the $900 into Bitcoins now).  $900 would buy around 50 BTC at the current market rate, which is more than a 6990 will ever mine over its lifetime.

I'll accept the argument that people might just want a bitching gaming machine, and then want to have it partially pay for itself in what would otherwise be down time, but as a straight money-making proposition, buying a 6990 is a loser.

Well, there's two kinds of people.

#1  Otherwise diligent people who make a colossal mistake and then go to great pains to never, ever let it happen again.

#2  People who make a colossal mistake simply because they're just not careful.  Making one colossal mistake doesn't make them any less likely to have another one in the future.

We don't know what type allinvain is.

I made much the same rig as that recently, and for the same price, except I did it with 4x5830s.  I did run a more expensive motherboard that had 4x PCIe x16 slots, so to shave off that additional money, I ditched the case, ran with only 2 GiB RAM, and ditched the HDD in favor of a flash stick loaded with Linuxcoin.

You mentioned you're planning to run Windows -- well you forgot the cost of that, unless you're planning on being a pirate.

The only guarantee is that it will remain profitable for people with pre-existing hardware and free electricity.  People who pay a lot for electricity may find it no longer profitable in the not too distant future, and people who would have to buy hardware up-front to start mining with now are already finding it unprofitable.

Why would you need his endorsement?  His claim to fame is losing a shedload of money through poor security practices.

If you deleted a data partition on your HDD and then re-formatted it with a different filesystem type you're pretty thoroughly @$^#ed.

See Linuxcoin with Persistence?  http://forum.bitcoin.org/?topic=7374.0

The proper way to express this formula would be:


-F(n-1)+F(n) would average to zero over every two blocks (for obvious reasons), so all you're really proposing to do is just reward 50 BTC per block, forever.

Who is spending over $1 per MHash/s right now on new mining equipment?  That's never going to pay itself back at these difficulty levels.

Folax, that change has already been incorporated into the latest versions of all the big miners.  Grabbing the latest version of the miners is a one minute kind of affair, and I had to do it with Linuxcoin 0.2A anyway because, for whatever reason, the included version of Phoenix wasn't working, but a fresh copy from Subversion was.  I guess the only way this is really going to mess you up is if you aren't using Persistence, in which case you'd have to update the miners each time you rebooted the computer.