Yeah I didn't *need* JSON but it sure makes things easier to have it

Am I wrong in thinking it'd be more appropriate to ask what the pornography laws in Finland are like?

Internet != America.

Just saying.

So now that you've got JSON implemented, I think I'll start work on some monitoring software based on clocktweak

I bashed something together for myself but it's very hackish, nothing trapped for or commented, messy code and bashed together in the quickest language I knew (Classic ASP).

It's up at http://codinginmysleep.com/miners/ for critique. Pretty basic interface, color-coded HUD at the top for each miner and the simple clocktweak -r -n output below. Not pretty but it functions. Would anyone be interested in something of this nature?

This looks phenomenal. I'll be giving it a shot tonight

That's my kind of script 

Awesome, updating from git then

Edit: you should also probably modify the readme to indicate password.py as the new location of the account info. Didn't take a substantial amount of brainpower to figure that one out but I've worked tech support enough to know that no matter how obviously wrong the directions are there will always be some idiot who will follow them and then complain when it didn't work 

Not client disconnects, the error I posted gets spammed several times per second, so quickly that it fills the buffer before I'm able to pause and see what occurred immediately prior.

Here's my actual error, just crashed again:
Edit: It's now crashing like this ~once per hour... Help!

Got it to crash again, came back here to post about it and saw the exact error already there. Same error for myself as well

Crashed during the night. I was running it directly so there was no helpful error message waiting for me, the window simply closed itself while I slept. I ran it from within a command prompt this time so that if it crashes again I'll have an error to post. Hopefully it's just a fluke 

For the record, the most common kind of color blindness is red/green, so I think the complaint wasn't so much that you used colors to differentiate, but that you placed red and green RIGHT next to each other. Swap either one for another color so there's buffer between them and the chart should serve a much larger audience. You can't account for every kind of color blindness, but you can at least account for the single most common.

This thing is FANTASTIC. If this is beta I can't wait to see a gold release

I also can't wait for some stats.

Finally got RBE to work. Can't believe how dumb my mistake was - forgot to remove MSI Afterburner from the startup group and it was overwriting my changes thanks to "unofficial overclocking mode"

Pointed a couple of my miners at a test setup last night, and so far they're performing about 20% over calculated estimates from hashrate alone. I think I'll be rolling out a full-scale setup very soon

Well isn't that just lovely.

And we just got a block (maybe) so this is the part that'll tell whether it's a scam or a major statistical freak occurrence - will we KEEP getting them at or around a rate that is expected for our MH/s...

$750 is an amazingly lowball estimate. Play devils advocate for a moment (I know I am) and imagine that they're just redirecting our efforts to some other pool and taking the payouts. Now if I'm a clever scammer then I'm going to operate the pool properly for a bit until I reach some predetermined size that I think is achievable. I'm going to pay people out and run at a tiny loss until I hit my magic number, say 50 GH/s. Then I create a "long block" that simply never ever happens. I can TALK about doubling the minipool rewards when the block hits, but it never will so that talk costs me nothing. The "long tail" will last for several more days, so I make, we'll say two blocks worth of profit before people get suspicious and a couple pool hop. Then another estimated-block-time later people start talking, one more shift down in my hashrate, but it's not until we hit numbers so statistically improbable that people really start to run. Case in point, it's been over 130 hours, your own math says the chances of this happening legitimately are less than 3% and dropping by the minute, yet there are still some 30 GH/s worth of miners contributing efforts. Now if I were a smart scammer I'd pay off a block soon, since I've already made my ~178 btc and stretch out this long tail just a bit more. One more "long block" worth that much and I'd have netted about 306 BTC, worth ~$4,400 at current exchange rates.

NOW is it worth the scammer's time to set up a web site that requires attention perhaps once a day?

A 50 GH/s pool should be finding a block about every 37 hours on average. Now I understand that sometimes blocks go longer, but by your own math the chances of this occurring (for this long anyway) are just under 3%. A brand new pool with only two blocks to its name that has now gone 132 hours without a block is either the least lucky pool on the planet or the most unfortunate statistical anomaly ever. As for there being no benefit to not push out a block, are you actually listening to yourself? That's like saying that as a business owner your employees couldn't possibly benefit from stealing from the register or short-changing someone. Right now this pool should have found 132/37~=3.57 blocks, at 50 BTC each that's a bit over 178 BTC - $2,562.89 at the current exchange rate. I'd call that a pretty good paycheck for a few days efforts and zero actual work. Oh and they've dwindled down to 34.5 GH/s now but that's still a LOT of hashing power that is still making them money which may or may not actually get shared with the rest of us.

The thing that bugs me the most, though, is that ANY pool could probably get away with this - every so often they could just steal a couple blocks in a row and say "wow guys that was a terrible round" and it'd never come out unless someone did long-term analysis of hashrate, expected payouts and actual payouts - something which is impossible with a pool this small and this new. That's also assuming that the API data the pool provides isn't doctored too.

I'm half tempted to just start mining namecoins and take the slight loss to convert to BTC. At least those I can mine solo still, and not rely on pool operators not to screw me.

Trixx only seems to affect the first card in each of my rigs, whether I check the checkbox to sync them all or not. RBE seems to have done the trick in one of my rigs but in some others I flash BIOS and the clocks just go right back to stock settings. I've also got one card that freezes the system if I try to underclock the memory at all. All are the Sapphire 5830s that everyone raves about. Is there some trick I'm missing to make Trixx or RBE work properly?

Any statistical mathematicians want to tell us what the likelihood of a 50 GH/s pool going 128 hours without a block are? (And don't say 100% because it's happening, I hate that smartass answer)

IMO the percent chance that this is a scam is inversely proportional to the chances of going 128 hours without a block legitimately.

Also, Mr Sam hasn't directly responded to the accusations that he's just keeping blocks - this is NOT behavior typical of the innocent, although we're not exactly interviewing him in person to read body language. He could be as outraged as an innocent person usually is at the accusations but is simply keeping the outrage off of the forums.

In any case, I'm back at eligius which is only 4x the size of this pool and yet somehow manages to hit blocks with some regularity.