EU server is open: eu.clevermining.com
Consider it beta and please setup us.clevermining.com as your backup pool.
Cheers man!
|
|
|
1. Yes, but you should get powered risers if you plan to have lots of cards. 2. It won't reduce any KH/s 3. There's no issue to begin with (see #2) 4. They are plugged via a normal power cable to your PSU (usually a molex connector, although there are sata ones too).
Thanks but if (2) and (3) are not an issue, then why would I want to get powered risers? Because you don't want all the extra power to go through your motherboard. It's the same reason anyone who uses multiple of any type of card (nVidia or AMD) uses powered risers. Most motherboards are designed to handle maximum 2 additional cards boarded on them, which equates to additional 150w (+/-). More cards (75w+- per card) means more power is needed to power them - more current goes through the motherboard, which means increase in likelihood of frying said motherboard.
|
|
|
Hey guys, so I've been looking through the thread and it has been commented several times that I should base the hashrate shown on the site vs my WU on my miner. Is there any way to increase this number? My hashrate is averaging out at 820khash (due to heat reasons) and WU at 750/m. http://imgur.com/AA4LCxtBut on the stats page it is showing that I seldom get above 700 khash accepted. users/1Gz3QtQXTEVTZoSNpzxjrU4fq3mKSAQ5ky WU is normally about 90% of your KH/s, so by increasing your KH/s you also increase your WU, although different settings will yield different ratio between KH/s and WU - but don't expect it to greatly vary from the 90% mark. Am I doing something wrong? How can I make my WU reflect my actual hashrate closer?
Any configuration that you do with your miner should revolve around maximizing your WU and not your KH/s, so in a sense, your question is backwards and should be "How can I make my KH/s reflect my actual WU?".
|
|
|
1. Yes, but you should get powered risers if you plan to have lots of cards. 2. It won't reduce any KH/s 3. There's no issue to begin with (see #2) 4. They are plugged via a normal power cable to your PSU (usually a molex connector, although there are sata ones too).
|
|
|
"only 868 KH/s" - that's not so bad considering your power consumption should be (that was in my case) at around 250w per card, which gives it a 3.47 KH / watt ratio! Granted, you need to undervolt, but yeah, the potential is there.
I get 900KH/s now with 275w (3.27 KH / watt) per card with 1015/1250, 20481 TC 1.07v. Try it out.
FYI, your clock settings will vary to get optimal settings if you got Elpida or HynixMFR memory (I got HynixAFR).
|
|
|
No one mentioned them, and I guess it's because they are rather new to the scene, but https://www.kraken.com/ is also pretty good - been using them for a while and they got a reputable bank behind them. I tried BTC-e but their fees are a deathblow for small-medium transactions (especially the SEPA one).
|
|
|
Not here to badmouth hashcow but I really dunno how people get 0.015 BTC here, I tried it for 12 hours and I got a mere 0.00027175 (confirmed + current) BTC for 2.6MH/s...
A couple of days later and I got 0.01796583 in my balance for 12h mining with 2.4MH/s. Very good!
|
|
|
those are my settings. STOCK! -g 1 -w 512 --lookup-gap 2 --thread-concurrency 32765 -I 19 --gpu-fan 85 --gpu-engine 947 --gpu-memclock 1250 --temp-overheat 92 --temp-cutoff 96 its so weird the rig worked for couple of weeks smooth with powertune+20! it means that the psu should do the job right? now when i'm on zero it crashes! please help.. forgot to mention. when the rig crash, i can't power it on i need to unplug the AC and plug it again. and the motherboard lid is still on after the crash. It's not that weird. If memory serves me right, than at stock values a single R9 290 eats about 330w-350w, 2 of those will make it 660w-700w and when you add cpu, mobo and ram on top of it you exceed the limits of your PSU. A quality PSU can provide more power than it lists as, but it's not meant to exceed its limit for long periods of time (weeks in your case). I can't say for sure if your PSU is toast, but try undervolting your R9 290 with stock clock speeds and see what you get. A really conservative settings I used to have were 911/1250 with voltage down to 1.023v, those gave me like 225w per card but hashed at like 800KH/s. thank you my friend. it seems i need to change my PSU. i tried now your settings and i get around 700KH/s. maybe i can push it to 750? what would you change to make it stable Well, I also got R9 290 with HynixAFR, so it might have account for the difference in speeds you and me get (in case you don't have HynixAFR youself). I suggest for you to set back the voltage to its original value and than play with the core and mem speeds, once find KH/s you like, start undervolting. You could always go the reverse route if you wish - set a certain voltage (never set it above stock!) and than find the optimal core/mem clock speeds that give you the most KH/s. You can start by trying mine: 1015/1250 with 1.070v - those give me 900KH/s with 275w usage. But I wanna emphasize again that if you got different memory type (Elpida or HynixMFR) than my clock speeds would probably net you different results, most likely worse so you'll need to do some trial and error until you find the sweet spot.
|
|
|
those are my settings. STOCK! -g 1 -w 512 --lookup-gap 2 --thread-concurrency 32765 -I 19 --gpu-fan 85 --gpu-engine 947 --gpu-memclock 1250 --temp-overheat 92 --temp-cutoff 96 its so weird the rig worked for couple of weeks smooth with powertune+20! it means that the psu should do the job right? now when i'm on zero it crashes! please help.. forgot to mention. when the rig crash, i can't power it on i need to unplug the AC and plug it again. and the motherboard lid is still on after the crash. It's not that weird. If memory serves me right, than at stock values a single R9 290 eats about 330w-350w, 2 of those will make it 660w-700w and when you add cpu, mobo and ram on top of it you exceed the limits of your PSU. A quality PSU can provide more power than it lists as, but it's not meant to exceed its limit for long periods of time (weeks in your case). I can't say for sure if your PSU is toast, but try undervolting your R9 290 with stock clock speeds and see what you get. A really conservative settings I used to have were 911/1250 with voltage down to 1.023v, those gave me like 225w per card but hashed at like 800KH/s.
|
|
|
Did you undervolt? a 750w can be enough for 2x R9 290 (and of course mobo+ram+cpu) if you set it correctly and go a bit conservative with the settings.
Anyway, post your settings here.
|
|
|
It shows accepted #rate. I've got 2.4 too in theory 780/810/790 on my 3x280 rig, but #rate on stat page shows usually 2.1-2.5
If u got below 2.0 i bet it's ur settings.
You need to compare it to your WU not KH/s...
|
|
|
Usually a display driver crash is due to misconfiguration of the clocks speeds and/or misconfiguration of the voltage and/or bad drivers.
So post your config file here and and driver version you're using.
|
|
|
I have come to the conclusion that the amount of system RAM required heavily depends on - Thread concurrency
- Number of GPUs
I have also come to the conclusion that the system RAM is only required to create the kernel code that cgminer then transfer to the GPUs. In other words, once the kernels are built, you no longer need the system RAM *until* you change whatever parameter in cgminer that affects the kernel, then cgminer will need the RAM again to re-build the kernel. I may be totally wrong though, asked and still looking for certainty about cgminer and system RAM requirements. Given that CK has now totally removed GPU support from cgminer, I didn't ask him directly (as doing so might be perceived uninformed/trollish) You are correct, to an extent. It was my impression that the system memory was only used to build the initial kernel, which does get larger the higher your TC goes. My rule of thumb is to match the video ram of your largest card, and then +1GB for the OS. For example, a 3GB 7970 or 280x works fine on 4GB (3GB for card, and +1GB for OS). A 290 has 4GB RAM, so you would need at least 5GB of system RAM (we use 8GB, and have never had any issues). I got R9 290 and a total of 4GB RAM in my system and everything works fine.
|
|
|
Thanks for a reply. But what I was wondering is how the pool calculates pool difficulty for a particular worker. Which will be more optimal and preferable from server side, having all rigs under one login or split them individually.
VARDIFF makes no difference to your profits. Small rig or large rig. Doesn't matter. It's been explained on this forum and every other pool forum. You can read up on it.
It doesn't matter, AS LONG AS THE REJECTS ARE CLOSE TO ZERO. No. It doesn't matter. Ever. Period. +1 I think it does. Otherwise tuning vardiff would not exist. On wemineltc I am receiving various difficulties ranging from 64 to 600 depending on the particular rig's hashing power. Taking into account that there is only one server in US and reject ratio is high due to latency I suspect that that it matters even more. Currently i see across all rigs that difficulty is set to 512. Might be hard coded value. Since clevermining is still in development I wonder what is Terk's opinion. Should we use separate btc addresses for each mining rig?! Is he working on sub-workers. Like for example "btcaddress_worker1"?! Or should we just use one address for all of them?! What is Terk's preference on this? Sure vardiff matters, the guy just doesn't understand. You're the one who doesn't understand. I'm tired of explaining it, it has been done about a billion times before, just use the search function. The only way vardiff affects your profits is that it increases the CPU load on the server, therefore increases the costs for the pool operator, therefore makes a higher fee necessary and thus slightly reduces your payouts in the end. Let them think it matters. Ignorance is bliss.
|
|
|
@aulonocara does this look good --gpu-engine 920 --gpu-memclock 1300 --lookup-gap 2 --thread-concurrency 8192 -g 1 -I 17 --temp-target 75 --gpu-powertune 20 --gpu-vddc 1.2 --shaders 1536 --auto-fan
anything to add
DONT overvolt... stock is 1.1v so that's the highest value you should ever set. Remove 'shaders' as it has no affect when you set thread-concurrency.
|
|
|
I have 600 watts cooler master silent pro supply...i have running now at 297 kh/s for 1 hour now without any restart on 14.2 beta with below settings:
--gpu-engine 600 --gpu-memclock 800 --lookup-gap 2 --thread-concurrency 7000 -g 1 -w 256 --intensity 18 --temp-target 80 --auto-fan --gpu-vddc 1.030
it has flashed to 6970 but i switched back to the old one....both have same problem.
Remove that gpu-vddc from the config and set your core to 800 and memclock to 1250. If your PC still crashes than download and install MSI Afterburner (ver 3.xx - beta) and let it detect your clock speeds and voltage (you might need to restart after installing). Stock voltage for the HD 6950 is 1.1v, so if MSI Afterburner reads anything else you might have a problem there (I assume you didn't flash any bios) and you would probably need to set it manually back to 1.1v and than change the clock speeds back to their stock settings (if you haven't already - 800/1250) and work from there.
|
|
|
Can someone confirm 0,015 BTC per day and MHs or any payout that occured within last 48 hours? Coz it seems hashcow sucks.
Not here to badmouth hashcow but I really dunno how people get 0.015 BTC here, I tried it for 12 hours and I got a mere 0.00027175 (confirmed + current) BTC for 2.6MH/s...
|
|
|
Sorry, but why are you saying those coins are undervalued? Severely even. What's their current market-size? And I mean their real market-size - which retailers actually accept any of those as payment?
|
|
|
Thanks for a reply. But what I was wondering is how the pool calculates pool difficulty for a particular worker. Which will be more optimal and preferable from server side, having all rigs under one login or split them individually.
VARDIFF makes no difference to your profits. Small rig or large rig. Doesn't matter. It's been explained on this forum and every other pool forum. You can read up on it.
It doesn't matter, AS LONG AS THE REJECTS ARE CLOSE TO ZERO. No. It doesn't matter. Ever. Period. +1
|
|
|
|