Bitcoin Forum
June 14, 2024, 05:59:30 PM *
News: Voting for pizza day contest
 
   Home   Help Search Login Register More  
Warning: One or more bitcointalk.org users have reported that they strongly believe that the creator of this topic is a scammer. (Login to see the detailed trust ratings.) While the bitcointalk.org administration does not verify such claims, you should proceed with extreme caution.
Pages: [1]
  Print  
Author Topic: WOW THIS FORUM IS ALIVEE!! NOW I MAY FINALLY GET SOME HELP FROM THE PROS!?!  (Read 742 times)
gtraah (OP)
Sr. Member
****
Offline Offline

Activity: 420
Merit: 250



View Profile
December 21, 2013, 07:04:04 AM
Last edit: December 21, 2013, 07:31:09 AM by gtraah
 #1

    People I have been asking in cryptocointalk, and a few alt coin forums, it take 3 days for a reply, and to top it off the reply i get is only answering 15% of my query it is very frustrating when you want to find the problem, and yes I have googled read forums, etc etc.. Beleive me if I found the answers somewhere I would not bother writing this entire post and bothering any of you. And I thought instead of creating 20 smaller posts, why not incorporate all my queries into the one.

    I know its a long post guys but please read it and and try not to skip anything. I would like to
finally cure my problems.

OK here it is:

I have 3 x 280x R9 Toxic OC Tri-X Because I dont have the riser cables I decided not to put 3 cards on the motherboard as there is no room and they will be too close together, So what I have done in the meantime untill I get the riser cable is put 1 Card On the 1st Slot (MOnitor plugged into it) the 2nd card is on the LAST slot at the bottom this give sufficient room in between and better airflow.

1200 w PSU with a MAX of 1405w with 86.5% effiency which is why they advertise it as 1200w Its a gold edition Silverstone

I have the Asrock motherboard OC Forumla Z87 Full size, 4 x 16x 2 x 1x ( My plan is to eventually build it up to a 6 card Beast. qith powered risers for the 1x slots, and maybe cheaper cards for the other 3x like 7950 or something. So it will be 3x R9 280x TOXIC OC + 3x 7950  - Which should get me 4500KHs AVG


My cards average around 720KH and probably should get more after reading what other people get but I just don't have much clues as to what I am doing and what the configurations should be. The Card thats closest to the CPU up the top is always hotter but this is obvious as the monitor is plugged into it and the cpu and everything is closer to it, so there is more heat around it.

77-80c TOP GPU

70c Bottom GPU

Here is an example of around 12 hrs of mining results in a pool-

Accepted = 888512 < Maybe more but I didnt catch the log, as I am not even sure it restarted or anything, but the main thing is to look at the Stats of accepted/rejected/HW ANd at the bottom I will show you how much coins I received.
Rejected = 7392 < Same as above
HW = 1 > MAIN GPU (Monitor plugs into) < Same as above
 
Time = 12 hours

Now I have noticed that after 12 hrs I start to see tiny graphical glitches, like little lines appear and disappear every now and then, not so bad that it effects anything I can still do everything as normal, I will explain more below.

I am using CG watcher because there has been a time, yes 1 time! where the miner just stopped or GPU - DEAD etc.. which I really didnt understand fully but I am trying to  Sad. Instantly I search the internet for more info on this and realised people use CGwatcher to reset the miner auto or some people lower the configuration and accept lower KH/s- But I went for the 1st option Cgwatcher because I beleive the cards I am using are capable and have seen many people state they are getting 740-780Kh Stable so I thought , why cant I get the same, But I decided to be happy with 720KH as this seems pretty stable in the results, so now that I lowered the config to 720KH stable I also decided to use CGwatcher &  have it scheduled to :

  • Restart when Miner fails
  • Restart when Pool goes down
  • restart COmputer every 12 hrs
(reason I did this is because when I restart the glitches dissapear for awhile , So i thought maybe they need a rest everynow and them - CORRECT ME PLEASE IF I AM WRONG  Embarrassed)
.[/list]

I also have it set to restart when GPU is Sick or device count is less than 2 or hashrates go below 1200KHs between 2 cards. For the last one I dont understand how this works as when I first start the miner the hasrates ARE below 1200KH so my thought was HMM why hasnt it retstarted? Which is good but made me think does that thing even work properly or maybe it knows to wait after your cards went beyond 1200KH before it says "Ok now that you are above the 1200KH, I will shut you down if now start to get lower numbers. (Possible, Your input is welcome here)

There are somethings which I do not understand ranging from: -Coin amount Pool stats Per KH-, to -Miner setup to CG watcher-, I ask the experienced people to please bear with me and try to help me out, I will appreciate it so much, I am a quick leaner once I get the hang of it and I wont bother anyone anymore once I have the answers, I will eventually have some expertise where I can start to help the newbies like myself.

Anyway Aswell as the above confusions here are some more that I have put down in 4 points -

1. Cgwatcher - Is there a setting which restarts CGminer if the miner hasnt started hashing within 15 seconds of opening? (There are times when I run the Batch file and it stays on the screen - CGminer 3.5.1 has started for minutes upon minutes only to then go to a blank Dos screen doing nothing. I myself have to manually retart it if hasing doesnt begin within 15-20seconds- Yes THe CGwatcher restarts the miner but this is after around 2 minutes which is a waste of time if you ask me. - So If I am missing an option somewhere in Cgwatcher please direct me Smiley

2. The CGminer BATCH settings I have are - -I 13 -g 2 -w 256 --lookup-gap 2 --thread-concurrency 8192 --gpu-engine 1100 --gpu-memclock 1700 --temp-target 70 --auto-fan - Are these good, I have played alot with the GPU engine and Memclock, I mean I do have the fastest 280x R9 around so I thought I should be pushing atleast 760-780KH Stable - But I am only pushing around 720KH Avg - YEs when I increase the Memclock it does get to almost 800KH but I stopped doing this because I thought I maybe killing the card, am I wrong? Can these cards handle it?

3. Coin amount after 24hrs of Pool test session - Goal to see how long I can sit in a pool for and how many coins I get in 24 hrs which will allow me to create a financial forecast. I am EXTREMELY angry/frustrated I have not been able to work it out and I MEAN EX F***KN TREMELEY Excuse the language but this is how frustrated I am for not being able to get an accurate forecast from the miner I spent money on building, I decided to mine Dogecoin as the difficulty and $ are most profitable ATM & I want to see what I can make out of it, so prior to joining the pool I look at the pool stats which says my average KH is around 1375KW meaning 45-50K coins a day. This is what I am confused about, in 12 hrs of hashing I have only seen 11k coins meaning in another 12 hrs I may only accumulate 20-24k worth of coins TOTAL, the Pool stats show I should be getting DOUBLE this with the KH I am mining at.  I do not know whats going on because there as shown above it seems not many HW errors A good percentage of rejections compare to acceptions. Now the longer it takes me to setup my miner and get the hang of it the more money I am loosing as difficulty risers and Electricity is being thrown in the bin as the Machine is still on. My apologies releasing my frustrations in my typing but I am getting sick and tired of not understanding what to do next, it not in my nature to not know what to do, so as you can tell , Frustrations arise.

4. Now to top it off also after 12 hrs of mining I am seeing TINY minor graphical glitches, little Lines, tiny texture\pixels missing scrambled from various random positions of the screen, only VERY Minor and hardly noticable but I can see them, I mean you can still do what ever you want on the computer , watch movies, listen to music. browse internet , view pictures. But you will see little glimpses of glitches every now and then.

Please if someone can help me even PM me I will be very happy EXTREMELY happy  Grin Grin Grin Grin Grin
DrG
Legendary
*
Offline Offline

Activity: 2086
Merit: 1035


View Profile
December 21, 2013, 07:29:52 AM
 #2

Well you should have probably posted this in the mining support subforum.

For the graphical glitches check and make sure your VRMs aren't cooking.  GPUs temps are not the thing to work about when Litecoin mining (they were with Bitcoin since most people downclocked the mem).  With scrypt the whole card will cook.  The VRMs may be running near 100C+ which is not good for longevity/stability.  You may need to undervolt and or clock down a bit.

Just because somebody in Alaska can get 730KH/s doesn't mean you in the desert can do the same.
gtraah (OP)
Sr. Member
****
Offline Offline

Activity: 420
Merit: 250



View Profile
December 21, 2013, 07:34:22 AM
 #3

Thanks for that DRG, So can you explain the quickest way to check if my VRMS are cooking, and do you have any input for my other queries? thanks Yes I understand it happens to be summer here in Sydney and the last few days were between 30c-40c Although I am in aircon room.

So what do you think about my other concerns ? hmm
DrG
Legendary
*
Offline Offline

Activity: 2086
Merit: 1035


View Profile
December 21, 2013, 07:59:24 AM
 #4

The VRMs are usually easy to check with GPUz (google it).  Just make sure to launch 1 instance for each card before mining.  Pick each card on the dropdown at the bottom.   Most higher end cards have the chip that can be queried to supply VRM temps.  Or you can try this nonsense: https://bitcointalk.org/index.php?topic=123235.0

As far as cards being SICK - we've all been there.  It's very frustrating.  But once I started checking the VRMs I often found those to the culprit.  The device driver won't crash.  It will keep mining, then go SICK.  Usually when the VRMs hit 100C+.  You might have to add some extra cooling to them or do some undervolting.

GPUz also helps find other power issues like unstable power from bad PSUs.  The voltage may drop as the PSU gets loaded and runs hotter.  If the voltage drops to much I'm guessing the card will shut down (I'm not enough of a engineering nerd on this).
railzand
Sr. Member
****
Offline Offline

Activity: 462
Merit: 250

Lux e tenebris


View Profile
December 21, 2013, 08:27:08 AM
 #5

on the money side - doge is so last week. and you cant estimate anything from pool promises, history, other miners' claims or just 12 hours minig on some pool. earnings relate to difficulty, network hasrate versus your own, pool and own uptime

gtraah (OP)
Sr. Member
****
Offline Offline

Activity: 420
Merit: 250



View Profile
December 21, 2013, 08:37:19 AM
 #6

Ok in regards to pool promises its not really a promise its just an estimate on what you would get from the current difficulty , Yes in 12 hrs difficulty may change but the interesting thing is it still states that on estimate with these KH I should be getting 45-50k an yes i am not planning keep that ridiculous coin of doge I am just going to off load it for something more future proof. You see if I received like 20k or something then I would not worry but HALF the amount of the estimate is not right

But before I do this I need to setup my miner, its just really pissing me off that I have not got it right yet 4 days after I built it, I am not used to taking this long to do things. And I dont know why I cant work this one out. SOOO eRrRRARGGHHhh frustrating@!!
railzand
Sr. Member
****
Offline Offline

Activity: 462
Merit: 250

Lux e tenebris


View Profile
December 21, 2013, 08:45:04 AM
 #7

hey chillax, you perfectionist you, jk, look 720 is great, better use your time to research better pools, read this entire forum etc

DrG
Legendary
*
Offline Offline

Activity: 2086
Merit: 1035


View Profile
December 21, 2013, 08:45:22 AM
 #8

Ok in regards to pool promises its not really a promise its just an estimate on what you would get from the current difficulty , Yes in 12 hrs difficulty may change but the interesting thing is it still states that on estimate with these KH I should be getting 45-50k an yes i am not planning keep that ridiculous coin of doge I am just going to off load it for something more future proof. You see if I received like 20k or something then I would not worry but HALF the amount of the estimate is not right

But before I do this I need to setup my miner, its just really pissing me off that I have not got it right yet 4 days after I built it, I am not used to taking this long to do things. And I dont know why I cant work this one out. SOOO eRrRRARGGHHhh frustrating@!!


Kids.... I have a couple of Diamond 7970 ref card that in 2 months I can't get mining above 550KH/s despite over 20 reinstalls, mobo/cpu/ram swaps, multiple CGMiner configs, GUIMiner, BIOS flashes, OS changes, etc.
gtraah (OP)
Sr. Member
****
Offline Offline

Activity: 420
Merit: 250



View Profile
December 21, 2013, 08:49:27 AM
 #9

Ok So check this one out, I got the latest version of CPUZ and and i start it up and guess what?

VRM temp 1  0c - ON CURRENT READING

VRM temp 2 0c - ON CURRENT READING

BOTH GPUs

When I turn it to AVG reading they go to 1c


Wow nothing is turning out for me is it

gtraah (OP)
Sr. Member
****
Offline Offline

Activity: 420
Merit: 250



View Profile
December 21, 2013, 09:27:17 AM
 #10

OK GOT IT WITH HWINFO64

My VRM TEMP is 85c with the MAIN CARD and 75c with the 2nd card, aroun d 4c hotter than the normal temp.

SO I BELEIVE THIS IS STILL OK, do then now my question is wHY I am I getting tiny texture glitches around thew screen. randomly. here n there. Happens after around 4-5hrs of mining, computer still perfectly functional though

So now no one here can explain the my Batch CGminer configuration? Or CGwatcher? Beleive me I have tried googling this is what I do before posting on a forum, forum is my final try. as I HATE having to wait for answers, I am literally sitting here waiting for someone to reply. LOL
vm1990
Legendary
*
Offline Offline

Activity: 1540
Merit: 1002



View Profile
December 21, 2013, 08:46:01 PM
 #11

OK GOT IT WITH HWINFO64

My VRM TEMP is 85c with the MAIN CARD and 75c with the 2nd card, aroun d 4c hotter than the normal temp.

SO I BELEIVE THIS IS STILL OK, do then now my question is wHY I am I getting tiny texture glitches around thew screen. randomly. here n there. Happens after around 4-5hrs of mining, computer still perfectly functional though

So now no one here can explain the my Batch CGminer configuration? Or CGwatcher? Beleive me I have tried googling this is what I do before posting on a forum, forum is my final try. as I HATE having to wait for answers, I am literally sitting here waiting for someone to reply. LOL

display glitches tend to be the GPU RAM running to hot and creating tiny errors

Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!