Topic: [ANN] profit switching auto-exchanging pool - www.middlecoin.com - page 440. (Read 829908 times)

About queue, expiry, scantime: Keep them default.  I don't touch them and i have 0 issues with connectivity.

About worksize: Keep this @ 256.  It has nothing to do with the pool difficulty, but rather GPU work unit.  HD 7xxxx series cards appear optimized for -w 256.

About gpu threads: 7950s love gpu thread @ 1, with a high(er) Thread Concurrency number.   7950s appear optimized to run with 1 gpu thread and a high concurrency.  Ensure that your thread_concurrency number is a multiple of 2 (2^n). 

About vector: Unused, good for delete as you already have.


Your primary GPU related settings are: intensity, thread_concurrency, worksize, and gpu_threads.  When you're using the config file format, tune those and leave everything else at default (unless a pressing need presents itself.)

I would also recommend setting your GPU fan to a constant 80-90%.

Naw, .0092/.0077 return is not the norm.  When i see a day like that, it's almost always because coins are maturing etc.

You should be doing .02-.05/.06 as a range.  

I run ~4.1-4.2mhs with 5 7950s / 1 7970 / 1 5850 and i make (with BTC at current prices and .10 kwh) almost $300 profit a month after electricity.

Just make sure you are undervolting your 7950s and keeping them relatively stockish.

Past stock clocks / undervolt @ .975/.1 you are paying roughly 1 watt / khs.  

Hmm, just doing math, with Bitcoin at $440/coin (which it really isnt at yet) and based on last nights payouts (.0092 .0077). My 6 7950s (in 2 rigs, 3 7950s per rig) are basically breaking even with power costs. (.105/kWh).

Crappy.  So if bitcoin continues this crazy rise then it is worth it, but now I am speculating my power costs plus my rigs have not yet paid for themselves.   Time to unplug?

Hi guys,  I'm a noob.   I'm just starting out mining and hooked up to Middlecoin a couple of days ago since it seems pretty straightforward for a noob,  and had a few questions pertaining to optimizing my config file for pool mining.  I have done a lot of googling and read this whole thread also,  so I think that I'm on the right track with my tune,  but I want to ask some questions anyway.

Some setup info: I've got six Sapphire 7950 dual-x cards in one rig that I just put together. These are the 11196-19-20G models with the bios that allows adjustable voltage.  The rig's on a 15 amp circuit for now, which is why I have the intensity turned down, until I can move it down into the basement, where I have a dedicated 20 amp circuit.  So right now they are hitting ~325Mh each,  and ~300 WU on average,  with temps hovering among the low 70's,  and no hardware errors at all.  

I have what I would consider to be a "stable" and "fast enough" internet connection.  It rarely hangs and it downloads at around 6 Mb/s,  and uploads at around 3Mb/s,  as long as the connection is wired. I am going to ping the middlecoin server from the rig but forgot to do it before I started posting.

First, here's what I'm running for my config in cgminer.

{
"pools" : [
   {
      "url" : "--scrypt stratum+tcp://middlecoin.com:3333",
      "user" : "...",
      "pass" : "..."
   }
]
,
"intensity" : "14",
"worksize" : "256",
"kernel" : "scrypt",
"lookup-gap" : "2",
"thread-concurrency" : "22400",
"shaders" : "0",
"gpu-engine" : "1100",
"gpu-fan" : "0-85",
"gpu-memclock" : "1575",
"gpu-memdiff" : "0",
"gpu-powertune" : "20",
"gpu-vddc" : ".900,.900,.900,.900,.900,.900",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "75",
"api-mcast-port" : "4028",
"api-port" : "4028",
"auto-fan" : true,
"expiry" : "1",
"gpu-dyninterval" : "10",
"gpu-platform" : "1",
"gpu-threads" : "1",
"scrypt" : true,
"hotplug" : "5",
"log" : "5",
"no-pool-disable" : true,
"queue" : "0",
"scan-time" : "1",
"temp-hysteresis" : "2",
"shares" : "0",
"kernel-path" : "/usr/local/bin"
}
My main questions have to do with the queue, scantime, and expiry settings,  and the worksize setting.

I'm pretty sure that I want my queue to be "0", or is that wrong?  I have my scantime and expiry both at "1".  Is that too short of a time frame in either case? I read one guy suggesting "5",  and some other guy went with "1" and said that he didn't notice a difference. Does that depend on what is being worked on at any given time and/or what my rig is hashing?  What exactly should I compare my results against if I test out some different settings for those? I ran all day yesterday with the default settings for the  queue, expiry, and scantime,  and then changed the config to the above this morning.  I think I got .02 bitcoin after one day with the default queue, expiry, and scantime. (I also had the voltage at 1.0 for the first day and went down to .9 this morning,  and didn't really notice an immediate difference, but I figured that I would mention that also at this time.)

My worksize is set at 256,  should that be at 512 since the difficulty of the pool is 512?  It seems that from what I have read,  my cards like a worksize of 256 for a reason that I forget right at the moment,  but I don't know if I'm supposed to match the pool difficulty with my worksize or not.

A couple more things that I just thought about is that my unit shuts off and restarts if I set gpu threads above "1".  Does that matter that I can't run it at "2"?  I'm not really sure what that setting affects exactly.  I think that that setting kind of depends on the specific type of card that is being used, and I read that the 7950's seem to like gpu threads set at "1",  maybe someone can confirm that, or tell me what it means and what the difference is.

Also one other thing, I deleted the vectors line,  since I read that they aren't used for scrypt mining,  but then I see a lot of scrypt config files posted that has that line in there. Am I correct in thinking that it doesn't do anything for scrypt mining and that I can leave it out?

Let me know if I forgot any info that would be needed to answer my questions, or if you see anything else that is amiss. I want to optimize my rig for Middlecoin at this time.

You can calculate that from the numbers on http://middlecoin.com/profitgraph.html

Any stats like, one 7950 hasing X amount would have earned X amount of BTC over 24hrs, 48, 7 days, 30 days, etc...?

What version of cgminer works the best?  I've used from 3.1.1 up to 3.6 and they all do the "pool interrupted" and never recovers.  I finally installed CgWatcher and it seems to keep things going for me when cgminer locks up. 

What software are you using?  If it is one of the last few versions of cgminer that might be your problem.  It depends on how you have it configured.

Chad

You will be disconnected for a few seconds every time we switch to a new coin. This happens every 10 minutes or so on average.

This is a great pool, but I'm getting frequent disconnections from multiple miners, and it only happens on this pool. Anyone else having this trouble ?

NVC takes about 3 days, I think that's the longest.

What coin takes the longest to mature?  I noticed that some looks like they mature fast and others don't.

Very good.   

Yes.

H20, do you have LTC in the coin rotation?

Cool  Thanks for the quick work.  I'm really loving your pool.  You seem to be on top of things when they break.

Oops. It's fixed.

It may happen again in the next hour or so, but if it does it should last less than a minute, and only once.

Same here.  At least I know it's not my internet messing up again  

edit:  looks like it started working again.  wooho

Houston we have a problem/  

http://i44.tinypic.com/300czte.jpg

mySQL errors..looks like its attempting to log into a WDC pool that mighta been taken offline due to the recent fork.

Is anyone else getting MYSQL server errors now?  Mine started giving me this error just a few minutes ago on both my miners.