Topic: [ANN][TAG] TAGMining.Com is now the official pool of Tagcoin! - page 7. (Read 45653 times)

hahahahha. Thats awesome, we really need one of those though!


EDIT: edit double giant batman signal out.

edit- damn, that was big... sorry. lol

Having some trouble getting Ubuntu to recognize both of these 7950's 

Anyone have any tricks?

Thanks.

Thanks for the trust, mate!  Ehalvers, thanks for yours as well!

Yeah, the other pools aren't synced, so you will need to create a username for each.  One day when I become high speed and figure that stuff out, I'll merge the DBs and make possible to use one login on every pool.  I should have done the research and did that from the get-go.  I'll be the first one to admit that lol.

-Fuse

Feb 2nd can't come soon enough.

Fuse, do we need to sign up separately to your other pools or did the user names/pw's sync between pools?

BTW, updated your trust with our two transactions.

FYI...

I'm not going to sugarcoat it... we were on a fork that wasn't the highest, we resynced and the fork bent us over and made us it's play thing.  I should have caught it because we didn't have an orphan all day  .

I'm working on clearing the issues in the DB when you fork and blocks were recorded as good and then end up being bad.  Mining on the pool is down right now.  Payouts should still work.

If you have large outstanding balances on the pool, cash yourself out.

I will be finishing up the DB cleanup and I will probably start the MPOS update that I wanted to do this weekend anyway.  The DB fixes will take some time, so you will have a chance to pay yourself out if you're worried.  However, your coins are safe regardless.

I will keep everyone updated as I progress.

I am sincerely sorry for the downtime.  Hopefully when this BS fork/orphan issue is resolved, we can get back to business as usual.

-Fuse


EDIT-

Also- there will not be a jackpot tomorrow.
The orphans accounted for more than half our blocks this week, and the total liquid funds on the server aren't even enough to cover the server costs for the month.

Shit happens lol

Indeed 31 tag unconfirmed gone.....

And everything what I mined for the day orphaned

Pool down?

EDIT: nevermind, back up!

NP, mate.  I just wanted to make sure you weren't some random spammer.  This forum is full of them.

-Fuse

Alright cheers again for more information! I tried yesterday something but didn't have much time...At least I was able to get 3x GPU to be stable for around 4h and after that cgminer crashed. Seems to be going to right direction since my clocks were quite high running ~640khash. To my understanding the problem seemed to be the first GPU overheating as you mentioned, so it's running lower and only with intensity 18 now, and a weird thing was that it was not able to run thread-concurrency 24000 as other cards. Let's see if I can get some good results today, thanks!

P.S I'll try that undervolting

Sorry, was going through new replays. It was wrong thread.

Here are some GPU PRON pics.

Someone used 6 of those MSI 270s on that new ASRock motherboard.

AB - USE it while you can. Although I'm pretty sure you are limited to a maximum of 4 fully powered rigs or whatever is the amp rating / breaker on the circuit.

Yeah, you should still be able to make those work.  Those cards are pushed hard out of the box at default settings.  You'll have to use TriXX to tweak the voltages to hopefully cool down whatever it is that is overheating (if that is the issue).    I'd suggest start out lower at stock clocks but with the core voltage lowered to 1100mV.  Once you get it stable, then start tweaking.

The stable settings (getting 625kh/s) I have on my Dual-X cards is:

core voltage 1100
clock 1030
mem 1550

Unfortunately, the card lacks VRM temp sensors (according to other posts elsewhere) so you won't see if the VRM's are running hot on the cards.  At stock voltage, it wouldn't surprise me.  Free power or not, lowering core voltage really adds headroom to push clock speeds up safely.  My cards run at 73 degrees with fan at 30%.  It might be a good idea to force the fans to 50% to make sure there is decent air movement until you get things to run.

Another thing I found these cards specifically appear to hate when mining is all of Windows' cutesy display slides, fades, expands, etc and screen savers other than 'blank' screen.  Turned them all off.  These cards were a bit more challenging to tune than other cards I have, but I have learned to like them.  They are still on my watch list.

Sheesh... bottle some of that up and send it my way! 

-Fuse

Alright thanks a lot Frank and Vesper! I'll try to tweak these cards asap when I have some time from work. I referred to Dual-X model (I have also some of that single fan) but all of these new ones are dual-x. I'm not concerned about voltages since I don't have to pay for electricity, student accommodation hah so I could use more voltage if necessary.

@Chofoe

Are you referring to the single fan at the front of the card version?  Those are great cards that I've found to be solid and versatile.  Unless they are new and something has dramatically changed since mid last year in their design?  I ask because it seems the dual-X is referred to the same way in some circles.  Maybe.  Also a great card, but less voltage control.  I have both, but really prefer the single fan version for many reasons.

It sounds like something is overheating somewhere in the card and becoming unstable.

Going on with the single fan version, here is what i have for settings.

AMD SDK 2.7
Cat 12.6
cgminer 3.3.2

Using MSI Afterburner, undervolt core to 1038 and mem to 1550. (these cards can be made stable at voltages as low as 900 and 1500 for obscene efficiency in GPU terms)
That should run stock clocks no problem.  From there, you can really start tweaking the settings.
I used the following line in the .bat file after much trial and error:
cgminer --scrypt -o pool -u user -p pw --gpu-reorder --gpu-threads 1 --thread-concurrency 21712 --intensity 19 --worksize 256 --lookup-gap 2 --auto-fan --temp-target 70 --gpu-engine 1060,1005,1055,1005,1070,1045 --gpu-memclock 1550 --no-submit-stale

I pushed each card (total 6) to the max the engine would go with that voltage without crashing.  That config got me to an average of 630kh/s each.  I've since reworked some of my cooling on the rig and have it running higher clocks and getting close to 665kh/s each and it still is at a comfortable 2.6kh/s per W.

The Dual-X voltage can't be controlled with Afterburner, but core voltage can be controlled with Trixx, but not mem voltage.

I also have some running on Cat 13.1, SDK 2.8, and cgminer 3.3.2. that run stable and with good performance.

Hope that helps.


A side question to everyone else... anyone tried sgminer and compared it to cgminer here?  I've read a bunch about it and am real tempted to try it out and see what happens.

Mate, you've made 2 posts in this thread.  One to ask "is this still working", and this one to tell people to mine another coin.  If you were an active member of this thread or the TAG community, I would be open to your suggestions.  Please post this elsewhere if you're trying to spam infinitecoin info.

-Fuse