I still will not mine on P2Pool due to this attempted scare tactics being used sometimes in this forum.
Do you know that mining on p2pool requires no trust whatsoever as to what the pool op might be doing with your hash rate behind your back. You're running your own bitcoind what makes you the pool op as far as share/block generation is concerned. The only part still requiring a degree of trust is splitting the block reward fairly. You do see the benefits, right? I do agree that we're better off with many competing miner-smart pools instead of one ginormous p2pool. The old (miner-dumb) model should eventually die out, just as proportional payout scheme seems to be doing now. Scare tactics - let's just call it aggressive marketing, ok Kano the Unyielding?  |
|
|
|
Do the effects of electromigration progress over time or is it just an instant thing where the card simply gives up after running great immediately before the total cut in the interconnectors?
If not, I believe it may be your own configuration issues with the drivers, OC, and etc.
Does the card still function fine at stock values?
The chip slowly degrades and requires higher voltage and/or lower clocks to be stable. Yeah, this particular card passed hour long Furmark and OCCT tests so it's still in a pretty good shape, except it's now mining at 935 MHz instead of 942. I haven't had a catastrophic card failure yet, like some users here. Lowering GPU voltage does seem to prolong cards' lives wonderfully, this ASUS is the only card I could not undervolt. All GPU fans are ok (running at 60% RPM max), though I wore out two case fans (they started rattling/buzzing). Those case fans were running at 100% |
|
|
|
|
What's up, Doc? We're getting hit with some un-BitMinter-esque stale rates.
While 0.8% stales might blend into the crowd anywhere else, here it stands out like a nude in a convent.
|
|
|
|
If testing goes poorly, would there be a huge disadvantage to running the OCZ unmodified to two of the cards, and running the other two off a Corsair VX-450w I have sitting around? It's group regulated so I'm not sure how well it like not being on loaded on the 3.3 and 5V rails.
There should be no disadvantage whatsoever. Use the VX-450W exclusively for 12V and never care how far out of spec its minor rails jump. As long as 12V doesn't fall below spec minimum you're ok. A power resistor (e.g. 5W) between neutral and 5V might mitigate the crossload situation somewhat. |
|
|
|
i got noscript. problem?
That makes at least two of us... That "fuck you" attitude needed some sort of comment lest we see newbs catching up the idea and getting zapped. A week later "I lost my bitcoins" posts will crop up like a bad rash on one's behind. Also, don't fall into the delusion of invulnerabillity. There have been exploits aplenty targeting various bugs in html parsing, completely orthogonal to JavaScript. NoScript makes you a less desirable/expected target but it can't mitigate all attack vectors. |
|
|
|
I would second (third) the recommendation on not trying to jury rig the PSU. Mining is very hard on PSU, far harder than just about any retail application including enthusiast gaming.
Mining could best be described as a PSU torture test. Low end PSU are going to die and before they die they are going to choke having all kinds of ripple issues, and inconsistent power, bad regulation, and random OS crashes. If spending lots of time trying to debug systems that can't be debugged because they have "random" power as an input sounds like fun then precede ...
'afternoon DAT. This particular PSU was spanked in virtually any semi-competent review for developing excessive noise and ripple - even going out of spec on higher loads. This is a clear example of a manufacturer cutting corners by using an inadequate number/quality of capacitors to even the ripple out. Yeah, for anyone still unable to grasp it: #define gamer-grade "overpriced low-end product 'upgraded' with a shitload of blue LEDs in an attempt to sell it as something more than it actually is"For the same uninformed crowd, world's best PSU platform is currently Seasonic Platinum. So the cards really don't pull much power through the PCIe bus and most of it will come through the 6 pin connectors. I guess that removes the first option, I won't be able to run four 6pin connectors on each 12V rail if they're loaded down with 75W each. I can still add one additional PCI connector on each rail and still be under the 20 amp mark on each rail.
Looking at the actual topology of the supply, it is actually a single rail split four ways for overcurrent protection as opposed to four physically separate rails. I guess I'll have to take it to work and actually load down the rails to be sure, but if the supply does make 75A between all three rails I should be ok, at least stock or undervolted.
Just to be clear, I'm not going to use molex to PCIe adapters. I'm planning on soldering connectors onto the PSU. I might have to save up for a good 1000W supply eventually, but if I can make a $60 supply work for now I'm going to give it a shot. Thanks for the opinions guys.
All contemporary "multi-rail" PSUs use what's best called "virtual-rails". Redundant circuitry for each rail would be prohibitively expensive, therefore a single 12V rail is divided into a number of "virtual rails" by means of a OCP (overcurrent protection) circuit. Since you seem to have made up your mind, I recommend that you solder one additional PEG-6 connector to both rail 2 and rail 4, and two PEG-6 connectors to rail 3. The reason for that is, the motherboard won't draw a lot of 12V current, neither will the fans and drives. Best of luck! |
|
|
|
|
You could cut corners by purchasing used components (fairly safe with the CPU and ram) but you might end up with no warranty and a hardware failure a month from now.
Whatever you do, you really need a good, efficient PSU.
A double 6870 rig does not make economic sense - you might not even make up for electricity costs with such low hash rate, let alone recover the hardware purchase costs. Until the latter is achieved you aren't actually making any money.
With such limited budget your best bet might be speculation. Mining isn't the only way to make some money with Bitcoin, you know.
During the recent price crash (after TradeHill suspended their activity) i purchased ±100 bitcoins I sold back for just over 110% of their price a week later.
Some food for thought...
|
|
|
|
A search on "tq_push in queue_request" turned up nothing useful...
How about doing a search of queue_request inside cgminer source code? static bool queue_request(struct thr_info *thr, bool needed)
{
(...)
applog(LOG_DEBUG, "Queueing getwork request to work thread");
/* send work request to workio thread */
if (unlikely(!tq_push(thr_info[work_thr_id].q, wc))) {
applog(LOG_ERR, "Failed to tq_push in queue_request");
workio_cmd_free(wc);
return false;
}
(...)
}
Some nVidia-specific bug? The hash rate is obviously bogus, it looks as if the kernel failed to perform some check and treated garbage as valid shares. Which kernel did you try? |
|
|
|
I went and gave them this: Username: fuck Password: you  Not prudent - you can't tell what browser exploits the fake site might have been enriched with. While your act of defiance is duly noted, never actually touching the attacking server is the way to go. And *PLEASE* don't tell me that running <your_favourite_antivirus> justifies this attitude. |
|
|
|
|
So the scammer-bitch ran? No surprise here...
I really think bitcointalk's terms of service should sanction IP-address ban against such individuals as A1BITCOINPOOL/Mu50stang if static IP address is being used:
"Circumventing user account ban may result in IP address ban where applicable." or something the like.
I'm fully aware this approach won't fully protect the general public from A1's upcoming scams but it would at least be some form of defense/sanction against him.
|
|
|
|
Yes, I know it's kind of a POS, but the price was right.
Except it probably wasn't right. Crappy rail topology and connector count (1) basically make this PSU a three GPU capable device thus degrading it to a 650W level. Is there any excuse for not fitting a 1kW unit with even one sodding PEG 8-pin connector??? (2)A decent 800W unit would have no problem with powering four 5850 cards which this PSU just can't do without some hardware hacks and thus voiding your warranty. A 5850 is a 150W-ish card. 150W at 12V is 12.5A. A PEG 6-pin connector is rated for 75W by the ATX spec. Without soldering in additional connectors the best you can do is: rail2 -> a single 5850 card
rail4 -> a single 5850 card
rail3 -> a single 5850 card using two double molex -> one PEG-6 connector adapters (be sure to have read this topic) Each of those rails is only loaded in 62% but there are no wires to fit in a fourth card. One option I thought I might do is splice a second PCIe connector onto each PCIe rail...
We're talking of a crappy PSU here - isn't it better to sell it off on e-bay and purchase something decent? Is it worth all the hassle? If you choose to proceed with your wacky plan, your best bet would be to solder in one PEG-6 connector to rail2 and one to rail4 and run the fourth card off these: rail2, rail4 -> a single 5850 cardWhether you do so by (1) clipping off one PEG-6 connector and soldering/using wire nuts to attach two PEG-6 connectors in its place, or (2) opening up the PSU and soldering in some additional wiring fitted with a PEG-6 connector (e.g. recycled from a dead PSU) is of no great importance. The first option is the quick-and-ugly but easy path. The second, while more labour intensive, is the elegant solution. If you go with wire nuts, make sure to twist all those wire strands firmly together - the last thing you need is a short circuit taking out your cards. Remember, however, that I take no responsibility for whatever damage or physical harm you might cause by messing with your PSU. Whether it's you or your mining rig that gets zapped, whether your PSU, your house, or your neighborhood burns down, don't come crying here. I'm not too worried since I expect the draw at the wall of the system to be in the 700W range, but I am wondering what is the best way to configure the system for mining.
Simply put, without voiding your warranty you can't. That's what you get for getting lured with the price. Notes: (1) :chuckles: It might only have four PEG-6 connectors but hey, it's got not one but two floppy connectors, fancy that!
You can hook up two floppy drives and copy floppies in a most convenient fashion! (2) the only explanation I can come up with is that those 12V rails really can't deliver the promised 20A: A PEG-8 connector (rated for 150W) and a PEG-6 connector (75W) combined might require up to 225W (18.75A) from the rail. If the manufacturer didn't feel confident that their PSU was capable of actually delivering such loads, they would naturally go with two PEG-6 connectors. |
|
|
|
Is it possible to turn it on for a specific gpu?
I've been running one with autogpu because it has a non reference fan. The rest I just use fan auto
This parameter is global and enables ADL-related routines in cgminer, it is not enabled on a by-card basis as you'd surely know had you RTFM'd first. Normally, since auto-gpu is a prerequisite for card speed throttling in case of fan failure (1) you want it enabled anyway. Notes: (1) or the house catching fire - "Whew! Suddenly getting hot in here, better drop those clocks" (I really couldn't abstain from making a bad joke, sorry) |
|
|
|
|
I took a good look at the graphs.
Other than the wacky +3.3V rail which never actually achieved 3.3V but still was within spec, I saw nothing unusual about them.
Tests ran and finished without issues. Voltage and thermals were ok.
|
|
|
|
Can any device run at 100% indefinitely? Any weaknesses the PSU might have will make themselves known. If it's flawlessly built doesn't suffer from inadequate ventilation, I don't see why it would not. The FSP-manufacured Be-Quiet! e9 700W PSU is a four rail device, meaning you will have to figure out how best to wire your cards. One of those rails will feed the CPU, the manual or manufacturer's specifications should tell you which one. Use the remaining rails for feeding GPUs. The the OCP protection chip overseeing those rails is configured to trigger at 18A (216W), 18A, 22A and 22A (264W) but of course you can't apply such load to all rails without seriously overloading the PSU. Remember that it's not prudent to run close to 100% load - should AMD screw up the next driver version and reintroduce some flavor of the infamous CPU bug, the CPU will require additional power possibly overloading the PSU. The same goes for underclocked memory/undervolted core - should you lose your configuration (e.g. while testing another miner) the power usage of the whole rig might rise. I use a kill-a-watt clone to make sure I don't run any rig at over 80-90% load capacity for this very reason. |
|
|
|
I don't know coz I can't stand MySQL, it's a cheap-ass, crappy, LQ database...
sed 's/database/debacle/' There, I fixed it for youUm, you did read what others had to say about A1 and it's predecessor before joining right? Obviously not  You mean in this topic? DAT called A1 a thief and a bunch of miners have not been paid. Since there are no bitcoin laws yet our only weapon is whiplash. I'm being generous with mine. The sob story about blocks not registering with the pool db and hiring not one, not two but three programmers... come on, get serious. Those must have been the coders responsible for Vista, laid off by Microsoft and with no hopes of finding work. |
|
|
|
I'm waiting for the admin to let me know if the database is still on the server. If it is then I will try the method above to retrieve the database. If its not the server then there is nothing I can do.
Bullshit. Frankly, I'm horrified at the fact that an individual without any technical expertise is reckless enough to offer pool services. How can you operate any service when it's clear you don't even know what you're doing?? How dare you offer such disaster-waiting-to-happen to other users? You can't even argue this was the first time you tried to run a pool... Occam's razor seems to support the hypothesis that this was a fraud from day one: an incompetent can't expect to operate a pool successfully but (with outside help) they might be able to run it just long enough to steal some bitcoins and raise the false flag O noes, I've been hacked!The fact you even copypasted the A1 sauce logo doesn't support your case either... |
|
|
|
Taken the technical "knowledge" he has shown so far there's no way he encrypted the password. And finding the above link shouldn't take more then 10 seconds if you actually search for it. So, yet another lie. Definitely looks like it. |
|
|
|
For instance, when I install Ubuntu it always runs bluetooth as it boots up. Since I don't have anything bluetooth I resort to 'apt-get remove 'package' ' but some things are tied to network manager and audio so must be selective when on a usable desktop.
Forget apt-get, use aptitude. It has an awesomely useful parameter why to help you sort out the dependencies: # aptitude why python
i python-pyopencl Depends python (>= 2.5)# aptitude why openssl
i cups Depends ssl-cert (>= 1.0.11)
i A ssl-cert Depends openssl (>= 0.9.8g-9)You can also put any package (it does not have to be installed) on hold which will prevent installation, removal, or upgrade: aptitude remove gnome-bluetooth && aptitude gnome-bluetooth= should do the job |
|
|
|
Only an idiot would set the same password as for his email/facebook whatever. But then that idiot might be using that password for his workers already, so nothing changes there.
Ah, the old 1d10t error code Did you per chance hear about protests against the ACTA treaty being signed in Europe? Many government websites were brought down as an act of defiance. Apparently, Polish Prime Minister needed admin-rights access to his official site... the password was revealed to have been admin1 |
|
|
|
Why the sad face? Do you think I didn't read the man page? I was just faster with man screen | grep Reattach |
|
|
|
|
Show Posts
