So I've been pool mining ltc for a while with multiple pools and was not seeing more than maybe 1-3% rejections, mostly when a new block hit...

Last couple of days, I can start a cgminer session on either of my rigs (one at 2.7.5, one at 2.6.1) and it chugs along fine for a minute or two... then it starts periodically flailing about with a ton of rejects for "time-too-old" and does that every couple minutes before recovering, resulting in about 30-50% reject rate... (on any pool)

Nothing has changed with my network connection or bandwidth (20mbit down) as far as I can tell. No updates or changes to the mining sofrware, drivers, sdk, etc...

These things are hasing at 100mh/s per card minimum, so I can't see how they'd be taking too long to submit shares...

So, not limited to pool, not limited to cgminer version, not limited to machine, and bandwidth seems to check out. Any suggestions as to why I'm rejecting a ton more?

I would appreciate any constructive input. Thank you.
Is my ISP just crap?

-M

I'm about to send in an RMA for a 1000w Coolmax ZU-1000B power supply.
I had it up and running in my new rig for about a week or two before it went pop (no power, components all work fine connected to a different PSU).

Before I slap the replacement back in there, I figured I'd run it by the community to see if anyone thinks I'm overloading the thing. I did use the PSU calculator and it came up with 925w recommended for my config (accounting for 20% capacitor aging and peak utilization)

The rig is running BAMT 0.5 and cgminer with temp targets and cutoffs such that the fans really never went above 60-70%.

So here's what I was powering...

Main board:
MSI-7548 AM2 mobo
AMD Athlon X2 5050e CPU w/stock cooler
4x1gb DD2 sticks of ram

GPUs:
4xATI Reference 5870 (aka the batmobile)
2 are powered by pci-e
2 are powered with molex->pci-e adapters
all are volted at 1.100
all are memory underclocked to 300
engines are overclocked ranging from no OC to 950mhz depending on what the card would stabilize at.

Low power misc:
2gb usb stick
usb mouse
ps/2 kb
1x80mm fan
2x120mm fan

I'm planning on grabbing a kill-a-watt to see what it's really pulling as soon as I get the replacement, but just wondered if I'm missing something. It seems that a 1000w PSU should be able to power this with no issues, but I ended up with pop goes the weasel.

Have I gone too far or is this just a crap PSU or what?

-Merrick

Trade completed successfully elsewhere. Thank you.

I need to convert $760 worth of Dwolla into bitcoins.

Note: if your Dwolla account has been flagged 'high risk' or is part of any bitcoin business, they will not let me complete a trade with you at this time.

Less is ok as I don't mind doing multiple trades to get this converted, but I have someone waiting on a bitcoin payment, so I'd like to get these converted as soon as possible.

Thanks!

-M

I have made arrangements to purchase these from Gnar1ta$

[WTB] 5x Radeon 5870

Please pm with:
  card details
  price shipped
  preferred payment method

(Shipping will be to Az, USA)

-Merrick

So I've got an AM2+ mobo that has a single pcie 16x and three 1x pcie slots. Here is a pic of my board. http://h10025.www1.hp.com/ewfrf/wc/document?docname=c01635688

I currently have a gpu in the 16x slot, but I am hoping to move it to an open-style case, upgrade the power supply, and use extender ribbons to see if I can get 4 GPUs running in there at once (5850s or 5870s).

Trouble is, I'm afraid of overloading the board and causing terrible tragedy. Does anyone have experience doing this kind of setup with an older board or any advice on how to avoid catastrophy?

I was thinking of looking at those powered 1x -> 16x adapters to help relieve some of the strain on the slot power...