Bitcoin Forum

--------------------------------------------------------------------------------------------------------
UPDATE: Pool is closed.
--------------------------------------------------------------------------------------------------------

--------------------------------------------------------------------------------------------------------
NEWS: I've decided to go open source: https://bitcointalk.org/index.php?topic=602292.0
--------------------------------------------------------------------------------------------------------

After a few days of testing my new Primecoin pool is now open to the public:
http://54.187.143.168/

I don't have a domain name yet, sorry. The pool works with my new GPU miner which you can download here:

EDIT: this is new version rc1 for i5 i7 fix!
32bit Windows: https://mega.co.nz/#!QBpnDSAR!5_AVpyO2ow1Y5NAAryZdeRcUtOH0rZTmalMfWMoeB_8 (https://mega.co.nz/#!QBpnDSAR!5_AVpyO2ow1Y5NAAryZdeRcUtOH0rZTmalMfWMoeB_8)

EDIT: same fix for linux will follow shortly.
64bit Linux: https://mega.co.nz/#!NEomzA4B!DqKHLTaA6jNPVaRi7BOIERKHbDkicTJDW80vKTWhV4o (https://mega.co.nz/#!NEomzA4B!DqKHLTaA6jNPVaRi7BOIERKHbDkicTJDW80vKTWhV4o)

To start mining:
1.) Download miner and unpack.
2.) Put your payout address in config.txt
3.) Now you can start the miner
4.) After you find your first share you can view your account and workers on the website

Pool System:
- Similar to beeeeer.org
- The automatic payout threshold is 11 XPM
- Exception: Your very first payout will be requested once you hit 1 XPM to confirm your address
- There is no manual payout (yet)
- Payouts will be executed after ~2 days when the coins are mature
- The minimum share is 8-chain
- Each share has a value that is recomputed every 30min based on the last 24h
- The poolfee is 10%

Known Performance:

3.3 CPD on 7950 @ 1000/1250
4.1 CPD on 7970/280X @ 1100/1250

Yes those numbers are real :). They are accurate to +-0.5% when averaged over a long time.
Also, you can see the memory clock does not need to be very high.

Some things you might want to know:
- The Linux version is untested.
- The miner works with Catalyst 14.4, older versions might not work as good or not even at all (EDIT: 14.3 should be ok too)
- The default settings are optimized for 79XX and 280X
- For 290/290X you can try sieveprimorial = 12

The website is still very basic, but get's the job done. I'm still working on it.
The site will automatically reload every minute, so don't use F5 please.

Happy mining! :)

Cool!! Definitely taking a look!

sas that need 14.4 driver - its very bugged

Catalyst 14.3 should be ok too.

Can anyone vouch for this pool/miner?

I don't have the means to test it atm.

i start to test with some of my rigs...wil report later

why so high latency at pool?

It's the average of all connected workers. But ~200ms is still very low, keep in mind this is 2x network ping latency + server cpu time needed to create all work.
On other pools this latency is in the order of multiple seconds and that's why they have so many stale/invalid shares.

got 1 xmp on balance allready & didnt reciew it yet(

What are you valuing 11 and 12 chains? Also is there a block winning reward?

It takes ~2days for the coins to mature. I don't have the funds to create a buffer to be able to pay immediately.
Although in the future this might be possible, once the hashrate has stabilized and is not increasing.

Anything above 10 is valued as 10, because these chains are so rare that you don't want to depend on them. Therefore 10 chains have higher value of course.
And there is no block reward for the same reasons.

Does anyone have a link to 14.3 drivers for windows 8.1? At AMDs page I only find 14.4 now...

error: ProbablePrimeChainTestFast 0/4

in next version of miner please add GPU fan control

It happens when you OC too much. Try lowering memory clock first, this helps a lot in my experience. Even stock memory clock isn't stable on some cards.
EDIT: Once a GPU has crashed you need to reboot for it to work again without errors.

Can be done. Do you mean a static setting of how much % or an automatic control that tries to hold a certain temperature?

60-80%

No ammount of overclock change has seemed to get it to stop crashing on startup i had the same error and now that i have lowered the overclock i don't even seem to get that error but its just straight up crashing as soon as it gets work from the server.

 I'm testing it on a rig of 4 sapphires 290x, windows 8, i7 3820 and 14.3 drivers however on my other rig using windows 7 it all seems to work nice and if you actually pay out then this is one hot smoking miner but would love a fix for windows 8 rig if possible brah

gj man, i see a little problem, i connected 6 rigs and ur pool shows me only 2 of them - can u check is everythink allright?

AMQfWsuHcXK1ZznDDTKdgCJt6S8etJB44F

Hi,

Thank you, finally one for Linux!! I was lazy to try out other's XPM miners as all of my rigs are Linux.
Just downloaded Linux version and testing with a 5xR270. Seems working but all the set core and powertune failed.
Does it even support R270? I will test with 7950/7970 later.
The followings are the corresponding error messages.

Found 5 devices
Using device 0 as GPU 0
Using device 1 as GPU 1
Using device 2 as GPU 2
Using device 3 as GPU 3
Using device 4 as GPU 4
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 20 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 435656388001 (39 bits)
GPU 0: sieve size = 55050240 (26 bits)
GPU 0: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 1: has 20 CUs
GPU 1: hash primorial = 30030 (14 bits)
GPU 1: sieve primorial = 435656388001 (39 bits)
GPU 1: sieve size = 55050240 (26 bits)
GPU 1: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 2: has 20 CUs
GPU 2: hash primorial = 30030 (14 bits)
GPU 2: sieve primorial = 435656388001 (39 bits)
GPU 2: sieve size = 55050240 (26 bits)
GPU 2: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 3: has 20 CUs
GPU 3: hash primorial = 30030 (14 bits)
GPU 3: sieve primorial = 435656388001 (39 bits)
GPU 3: sieve size = 55050240 (26 bits)
GPU 3: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 4: has 20 CUs
GPU 4: hash primorial = 30030 (14 bits)
GPU 4: sieve primorial = 435656388001 (39 bits)
GPU 4: sieve size = 55050240 (26 bits)
GPU 4: max bits used 341/352
ADL Initialisation Error! Error -21!Set powertune not supported
set_powertune(0, 20) failed.
Set powertune not supported
set_powertune(1, 20) failed.
Set engineclock not supported
set_engineclock(2, 1050) failed.
Set memoryclock not supported
set_memoryclock(2, 1500) failed.
Set powertune not supported
set_powertune(2, 20) failed.
Set engineclock not supported
set_engineclock(3, 1050) failed.
Set memoryclock not supported
set_memoryclock(3, 1500) failed.
Set powertune not supported
set_powertune(3, 20) failed.
Set engineclock not supported
set_engineclock(4, 1050) failed.
Set memoryclock not supported
set_memoryclock(4, 1500) failed.
Set powertune not supported
set_powertune(4, 20) failed.

Try having a look at the readout in the console i have a feeling you will see your cards saying crashed

Just checked, you are using the same name for multiple machines! All the rigs are there but you can't see them if they all have the same name.

ok thank u i allready changed names:) 

It looks like ADL doesn't work.
I will test this on my linux machine and fix it if possible, until then you can ignore the errors and set your clocks with some other tool.

270/270X should work as they are GCN, but i don't know how the performance will be.
You should def. try lower sieve primorials with these, like 12 or 11 (see config.txt).

Does it show any OpenCL error when crashing?
You need to manually open a CMD window and start the miner from there to see the error. Otherwise the console closes immediately.
EDIT: just to make sure, have you tried to reboot the machine?

Yes i reset the machine many times and i've sent you a screenshot of the console and i can also send you the output if needed but now its not even showing the error its just closing right after it gets work from the server

I'm running a MSI Twin Frozr III HD7970 OC BE @ 1150core/1500mem getting 4.29~4.35 CPD Temp=62C and a Gigabyte HD7850 2GB @ 1150core/1500mem getting 2.08~2.17 CPD Temp=52C.

Nice work on the new pool and nice to have an alternative to ypool.

What will happen to the rewards system once it hits 11 on the Diff?

How does the miner calculates the CPD? ie. 1-ch,2-ch,3-ch,4-ch, etc.....

Found 1 devices
Using device 0 as GPU 0
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 40 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 435656388001 (58 bits)
GPU 0: sieve size = 550556388001 (7 bits)
GPU 0: max bits used 341/352
Failed to ADL_Adapter_ID_Get. Error -1Failed to ADL_Adapter_ID_Get. Error -1Con
ecting to frontend: 54.187.143.168:6666 ...
Connecting to bitcoin: 54.187.143.168:60002 ...
Connecting to signals: 54.187.143.168:60003 ...
Work received: height=529501 diff=10.890152 latency=842ms
error: ProbablePrimeChainTestFast 0/2
error: ProbablePrimeChainTestFast 0/2
error: ProbablePrimeChainTestFast 0/2

W8x64 with 14.4

Any suggestions?

Nice to know it works on 7850, and pretty good actually. How do you keep that 7970 so cool? :)

2.) Don't know yet. Keep in mind diff 11 is still easily ~5 times more difficult than 10.9 so it's gonna be a while.

3.) It depends on the cpuload setting: with cpuload=1 it uses 4-ch and with cpuload=2 it uses 3-ch and so on. But still, i'm using a custom formula that has proven itself to be pretty accurate.

There are 2 problems: ADL not working and "error: ProbablePrimeChainTestFast".
The first one is probably win8 related. The second one is either too much OC or some other driver/win8 problem...

Right now i have no time to test/debug win8.
Maybe try 14.3 driver...

yeah this miner does run hotter

Found 3 devices
Using device 0 as GPU 0
Using device 1 as GPU 1
Using device 2 as GPU 2
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 32 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 435656388001 (58 bits)
GPU 0: sieve size = 550556388001 (7 bits)
GPU 0: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 1: has 32 CUs
GPU 1: hash primorial = 30030 (14 bits)
GPU 1: sieve primorial = 435656388001 (58 bits)
GPU 1: sieve size = 550556388001 (7 bits)
GPU 1: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 2: has 32 CUs
GPU 2: hash primorial = 30030 (14 bits)
GPU 2: sieve primorial = 435656388001 (58 bits)
GPU 2: sieve size = 550556388001 (7 bits)
GPU 2: max bits used 341/352
Connecting to frontend: 54.187.143.168:6666 ...
Connecting to bitcoin: 54.187.143.168:60002 ...
Connecting to signals: 54.187.143.168:60003 ...
Work received: height=529506 diff=10.890167 latency=760ms
error: ProbablePrimeChainTestFast 0/4


I have win7, 13.2 driver and I clocked both mem and core to 1000. 280x. Restart and still same error. I know newer driver should be installed but I got the same software setup on other rigs and it works fine.

Does it crash and stop mining? or shows only one error and keeps going?

It crashes.
Sometimes it has several rows of error: ProbablePrimeChainTestFast 0/4. Like 20 or so. On two other rigs I have pretty much all same except processors are Sempron 145 and this rig has i5. And I use this rig for gaming and stuff also, so maybe there is some kind of other issues.

I will try driver update later.

Cool, great to know the calculation part.

The 7970 has an aftermarket cooler, Arctic Accelero 7970 Extreme and it's in Corsair 400R case also. The 7850 2GB I own runs like a champ and cool. LOL

As of now, these are my stats:
7970: CPD=4.33
7850: CPD=2.24

7850 is using the same sieveprimorial as the 7970. Same temps stills, both cards are in the same ATX case.

Is this miner free or is it a 10% dev fee like others?

Pool fee is 10%, so yup, it's the same. At least this miner is the fastest one currently.

Hey madMAX, I notice the share values change. How is that calculated? What's the value of 11-ch/higher or is it the same as 10-ch on the pool?

Actually there is something odd: GPU 2: sieve size = 550556388001 (7 bits)
This value is wrong and it's no wonder the miner crashes. I'm gonna have to investigate how this can happen...
It should be: GPU 2: sieve size = 55050240 (26 bits)

will work on catalyst 13.12?

Isn't it required to use 14.3 or higher driver, he is using 13.2.

It is working on my other rigs with 13.2. And it shows 11.5 CPD for 3 card rig with 1080/1250. With cpuload="0".

Yeah they change based on pool luck. They are calculated based on the blocks vs. shares found in the last 24h. Once the pool has more hashpower the share values should stabilize.
11-ch is the same as 10 because they are too rare to make an exception.

Don't know. You can try, but it might be slower, so compare your results with other miners.

In this case juve4v has the same error and he uses 14.4 ... i'm looking for the bug right now.

13.12 dont work

OpenCL error: -45 at ..\xpmclient.cpp:101
OpenCL error: -61 at ..\xpmclient.cpp:196

That is strange, I've been running 2 hours plus without an issue. Maybe need to know their specs of their setup to see if the issue can be duplicated.

might be the way he compiled it and its not working for sandy bridge

I can confirm 13.12 is working on one of my rig. You probably have unsupported cards mixxed ? If that is the case make sure you use the correct setting  in your conf:

# 0 = don't use device
# 1 = use device
devices =        ["1", "1","0","0","0","0"]; where the first 2 are supported the rest not.In my case I'm using only 1,2 for xpm and the rest on something else. But note that performance on 13.2 is also diminished

It's looking more and more like a GMP cpu problem.

@juve4v and @liifer what kind of CPUs do you have in those rigs that crash immediately?

EDIT: @LongAndShort i see now you have the exact same problem. And you said this rig has an i7 3820 right?

I have i5-2400 @ 3.10 GHz. This rig crashes.

at me it can't be connected

Connecting to frontend: 54.187.143.168:6666 ...

 >:(

Sandy 17 2600k + 290 .I'll try downgrade from 14.4 to 14.3 and report back.

Yup

I updated driver to 14.4 same error.

my 7970's are at 4.6 cpd and 290x's at 5.7

Its beast isn't it :)

Have you tried 14.3? I'm thinking to downgrade from 14.4 to 14.3 and see the result

I'm trying to put a new version together which will hopefully fix the i5 and i7 crashes.

so far yes it does run hotter but i have all them at 73c  oh and im using the extra cpu usage to mine xpm on beeeeer lol no slow down the way i have it set

In my country they call that "Ringing its neck" (getting everything you can from it)

The website is getting a small scale DDOS right now  ;D

But still only 1% CPU load on the sever  :P

lol u know how i do things

well i hope u were ready for that cause its a war out there

when I oc, my GPU usage goes to 62% is that normal? I tried all drivers.

I have 2x sapphire 7950 Boost

Does setting cpu to 2 increase CPD?

Gonna have to try...I won't support ypool :p

1.) My 7950s with Boost have the same problem. When powertune kicks in the GPU usage can display weird numbers and the mining will slow down. On some drivers (like 14.4 for me) disabling powertune does not work.

2.) higher cpuload might increase CPD a little, but will use more power for sure as the GPUs can do the same work for much less energy.

Thank you very much, you use MSI afterburner to disable powertune in 14.3?

I am getting 3.1 CPD per 7950 @ 925/1250 you think it is ok or should it be more?

Getting the following before the program closes:

C:\Users\Ivan\Desktop\madPrimeMiner-v1.0-rc0-T10-w32>xpmclient
madPrimeMiner-v1.0-T10
ClientName = 'IVANLABRIE'  ClientID = 951955625  InstanceID = 2300866414
Address = 'AT4iEgpfXBDqth1k8tQU5Bb57z9mchwgY9'
Found 2 devices
Using device 0 as GPU 0
Using device 1 as GPU 1
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 32 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 435656388001 (58 bits)
GPU 0: sieve size = 5505 (7 bits)
GPU 0: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 1: has 32 CUs
GPU 1: hash primorial = 30030 (14 bits)
GPU 1: sieve primorial = 435656388001 (58 bits)
GPU 1: sieve size = 550556388001 (7 bits)
GPU 1: max bits used 341/352
Connecting to frontend: 54.187.143.168:6666 ...
Connecting to bitcoin: 54.187.143.168:60000 ...
Connecting to signals: 54.187.143.168:60001 ...
Work received: height=529682 diff=10.890427 latency=1071ms
error: ProbablePrimeChainTestFast 0/3
error: ProbablePrimeChainTestFast 0/3
er
C:\Users\Ivan\Desktop\madPrimeMiner-v1.0-rc0-T10-w32>

:(

i7 3820
2x4gb samsung 1600mhz cl11 ram
2 x msi gaming r9 280x cards
win 8 64
cat 14.4

I'd say 95% of the 7950s will do 950-1000mhz core and 1500mhz memory clock. Give it a shot :p (you might be able to run them at 0.962v too)

Hi madMax,

FYI, 270 works, not sure the clock rate it is using now but it gives ~2.2CPD each.

Question: how should I estimate the payout? The 2.2CPD is for 10ch I believe, so how should I expect # of 8 and 9ch per day?

Also is there any formula to give those payout for 8/9/10ch shares? I expected it is something related to the probabilities of getting each (after 10% for pool).

I'll test run it for 1 or 2 days, before I switch more rigs to your pool.

Thanks

Seems like the Intel Core cpu's have a GMP issue.

I'm having no issues like that with AMD or Core2 Quad cpu's.

DDOS happening!!

First they tried the website, but failed.

Now they go after the primecoin server. In any case, if the server crashes your miners will pause and wait to reconnect.

i still dont connect ...

tracert 54.187.143.168

19   202 ms   201 ms   201 ms  205.251.232.153
20   199 ms   235 ms   223 ms  205.251.232.165
21     *        *        *     Пpeвышeн интepвaл oжидaния для зaпpoca.
22     *        *        *     Пpeвышeн интepвaл oжидaния для зaпpoca.
23  ^C

usa block rus traffic :)

I had to reindex the blockchain, because it got corrupt during DDOS fire.

It's still reindexing and will take a while, like 1h, sry.

EDIT: I made the mistake of opening the primecoin port 9911, so the attackers were able to flood the "bitcoin" server with invalid data...

By attackers u mean Ozzy?

http://scr.hu/1xcl/53tnn

I think xpm is certainly the worst coin, I ever seen:
* in terms of fee
* fairplay
* totally defeat the purpose behind cryptocurrency
rename xpm to mafiacoin it will certainly fits more its purpose

haha +1

Keep up the good work!

Miner can't connect for 10 minutes now...

captain? fix?

pool is not working

Connectingo to frontend: 54.187.143.168:6666

and it stays there.... ill keep mining with PrimeGPU in the meanwhile

Still reindexing... will take another 15min.

Cool cool can't wait :P

 :)

What about CPU usage?
2or4or6*280X on Athlone x270 dual core

not up yet?

game over? or even earn pool?

We are running again ;)

EDIT: No we are not... same issue again: ERROR: CBlock::WriteToDisk() : OpenBlockFile failed

I will need to make a complete resync which will take another 2.5h ...

game over...

This just the first day, not even that.

I worked on this miner and pool for over 4 months now, do you think i give up on the first day?  ::)

what about open sourcing your miner ?
sorry I am stupid... the magic 10% fee  ::)

Make a backup of blockchain in case ozzy hits again.

 
 :'(

We are all well aware, we're just kidding  ;)

Any word on the i7 / sb - GMP fix?

looking forward to test when back up again!
Seems to be a very good speed improvent.
Now Xpm miner war is on... 8)

I was in the middle of making the fix, but now i need to check that the server can withstand some script kiddie DOS attacks ;)

DDOS is not terrible, because the server does not work  :)

i cant conect at all, this is due the ddos?

I looks like the resync will take another hour. Until then you can't connect, only view the website.

will support your good work forever  :D

and BTW if you got Stab from the back it's means you are on top of the line ;)

Keep the good work up  :)

and #S@* all the DDoser's  :P

@madMAx510510
Congratulations on the fastest Primecoin miner currently available. I would not be surprised if this miner can perform an estimated 6 CPD with an R9 290X + Core i7 4770K with optimal tuning parameters.

is pool working? My miner shows only following, or this is normal when try to run it on catalyst 13.12?
Can please someone give link to catalyst 14.3?

madPrimeMiner-v1.0-T10
ClientName = 'RIG1'  ClientID = 848067752  InstanceID = 2341488632
Address = 'AJDpWZwrezK6ZuBjTZPVnFEtugsqQgoQzt'
Found 4 devices
Using device 0 as GPU 0
Using device 1 as GPU 1
Using device 2 as GPU 2
Using device 3 as GPU 3
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 32 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 435656388001 (39 bits)
GPU 0: sieve size = 55050240 (26 bits)
GPU 0: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 1: has 32 CUs
GPU 1: hash primorial = 30030 (14 bits)
GPU 1: sieve primorial = 435656388001 (39 bits)
GPU 1: sieve size = 55050240 (26 bits)
GPU 1: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 2: has 32 CUs
GPU 2: hash primorial = 30030 (14 bits)
GPU 2: sieve primorial = 435656388001 (39 bits)
GPU 2: sieve size = 55050240 (26 bits)
GPU 2: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 3: has 28 CUs
GPU 3: hash primorial = 30030 (14 bits)
GPU 3: sieve primorial = 435656388001 (39 bits)
GPU 3: sieve size = 55050240 (26 bits)
GPU 3: max bits used 341/352
Connecting to frontend: 54.187.143.168:6666 ...

It is down for now, same on my miners.
We cant conect until op is done resyncing.

Can please someone give link to catalyst 14.3?

14.4 works with this miner, just download at AMD's website.

Yes, but 14.4 sucks when you want to switch back to SGminer. I just experienced first hand :(
I'm now going to give 14.3 a try.

http://www.guru3d.com/files_details/amd_catalyst_14_3_beta_download.html

when we can farm again?

Thanks for the link!

The pool is back up running now ;)

I made a blockchain backup so this doesn't happen again.

EDIT: Also i made new version for you guys having problems with i5 and i7:
https://mega.co.nz/#!QBpnDSAR!5_AVpyO2ow1Y5NAAryZdeRcUtOH0rZTmalMfWMoeB_8 (https://mega.co.nz/#!QBpnDSAR!5_AVpyO2ow1Y5NAAryZdeRcUtOH0rZTmalMfWMoeB_8)

Hopefully this fixes it, i've compiled GMP with generic i586 assembly in this version.

cool! Hopefully you will put in some ddos protection as well

similar problem
Pentium g2030
win8.1 x64
catalist 13.12
x6 asus r9 280x

The number of 9ch should be ~10x more than 10ch and the number of 8ch should be ~100x times more than 10ch.

The formula is pretty easy: (sum of XPM mined last 24h)*(1-fee)*sharepercent/(number of shares found in last 24h)

Where sharepercent is: 0.3 for 8ch, 9ch and 0.4 for 10ch. This gives a sum of 1.0 for these 3 chain lengths.
And the fee is: 0.1

Have you tried the rc1 version?
https://bitcointalk.org/index.php?topic=598542.msg6597096#msg6597096

Thanks! Will try it asap.

I am going to try this tonight, I will compare to a similar rig on primegpu.

With latest version and 14.4 drivers, it just locks up, freezes.
it was working good with 13.12 drivers and getting 3.9 cpd pre gpu.
what am i doing wrong?

Rollback driver and you're good?!

i got 2 rigs with 5x280x with pentium haswell and 14.4 driver.

both do around 18+- CPD. 2 gpus do 4cpd and the other 3 do 3.3+-
i guess its a cpu bottleneck? overall great performance! 2x290 give me 10cpd!

and a request, please add a page for each user on the pool site so we dont have to input our xpm address each time we check our stats.

thank you! great work

Yeah, that didnt help...
G1620 with 3 MSI and 2 gigabyte 280x-s with btc board.

Miner have lost connection to server. The site shows 0 chains/day.

+1

Pool is down

again dead(

Guys, i'm gonna have to shut it down for a while.

I'm getting the same error again: ERROR: CBlock::WriteToDisk() : OpenBlockFile failed

Seems to be a bug i need to fix.

Please fix it asap because compared to claymore miner, this is pure gold. No crappy ypool coin stealing shit. I made 1 xpm in 10 minutes, with claymore miner i got 4 coins in 8 hours.

I always felt like ypool was stealing coins when mining there, vs solo...way more than expected when factoring in fees and whatnot.

Well its turned into pool wars and developer "loopholes?" Whatever it is, it is not good for the coin. Guess Miners have to make the call if it is that centralized? is it worth your effort in the long run?

any eta for fix?

I think you simply was lucky to find 9th or 10th chain :-)

And it is strange that claymore didnt collaborate with any other pool than ypool

Good luck fixing the bug. I hope the pool is back up soon.

In the meantime, I'm trying to set up some machines to start on your pool when it goes live again.
Unfortunately I can't test now the pool is down but I think I have a problem on one machine. It reports:

Found 2 devices
Using device 0 as GPU 0
Using device 1 as GPU 1
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 32 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 435656388001 (39 bits)
GPU 0: sieve size = 55050240 (26 bits)
GPU 0: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 1: has 32 CUs
GPU 1: hash primorial = 30030 (14 bits)
GPU 1: sieve primorial = 435656388001 (39 bits)
GPU 1: sieve size = 55050240 (26 bits)
GPU 1: max bits used 341/352
Failed to ADL_Adapter_ID_Get. Error -1Failed to ADL_Adapter_ID_Get. Error -1Conn
ecting to frontend: 54.187.143.168:6666 ...

My main concern is the: Failed to ADL_Adapter_ID_Get. Error -1
Will this rig work when the pool goes live?

truly is pure gold, hoping for a fix! do you even know how to start dealing with it?

I think i found the issue:

2014-05-07 19:29:17 Using at most 200 connections (1024 file descriptors available)

1024 file descriptors is way too little, thats why it's failing to access the block files.

However i need to make a complete resynch again because the blockchain is corrupted again, and the backups don't work...

EDIT: Managed to get the backup working :) now i only need to find a way to increase the file descriptors in linux...

so 2 hours more need to w8?

i have any strange value
[GPU 0] A=98% 7970
[GPU 1] A=98% 7970
[GPU 2] A=97% 7970
[GPU 3] A=64% 7950

processor AMD Athlon 2 x2 2core, but cpu load 30-35%

ulimit -n ? :)

10240 file descriptors available :)

thanks eXtremal.

pool works again.

EDIT: not... getting strange errors now: *** buffer overflow detected ***

I think this is it for today, been up for almost 40h now... need to sleep. It will be dealt with tomorrow.

To protect against buffer overflows, input validation is required. Any field, form entry, or any other method that input is received could be a potential point of vulnerability for a buffer overflow. This not only includes a form field, but HTTP headers that are consumed, XML interfaces, or any other type of input accepted. Broaden your mind and think of all the possible locations that input are accepted in your software application. There are a lot. The more you think about it the more area's that need to be tested for.
Some of the more simply things that a developer should perform are to set limits to the input's allowed size. Even though someone could have a name with 100 characters in it, limit the size of the field. You need to know exactly the size of the potential data that is coming up. This could be as simple as setting the HTML size tag in the input type field, or using some javascript to check the field and validate the input. But be warned this is not enough. A malicious user won't conform to using a browser that will automatically set the input size and use javascript to validate input. A malicious user could and will use something else. A malicious user will write his own code, use netcat, or some other form of a browser to get around these simple checks. The input must be validated within the web application's code itself.
To recap html form checking, try to do the following.
Limit HTML input tags with the html size tag
URL Encode all input, to protect against special characters
Use JavaScript to perform the initial input validation
Most important, the underlying code must verify and sanitize the input provided

I'm doing a complete fresh resync now with which i can work tomorrow.

Until then.

sorry for your loss Game Over!

GPU 1: max bits used 341/352
N=11 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 2: has 32 CUs
GPU 2: hash primorial = 30030 (14 bits)
GPU 2: sieve primorial = 435656388001 (58 bits)
GPU 2: sieve size = 550556388001 (7 bits)
GPU 2: max bits used 341/352
Connecting to frontend: 54.187.143.168:6666 ...

is this pool alive?

its down again, up... and down, and up, and down... and up and so on...

You should implement a solo with the same fee as the pool miner, just as a back up for situations like this.

Are you able to connect to the pool?

Cannot connect to the pool  ;D

That is exactly what I want to suggest too.  ;)

Mad props for your choice of wire protocol and communication framework. :)

slacker still sleeping(

Thanx!! MadMax, add this link to the top!
It works!

Mad i recommend to make the miner multipool with 10% fee. Or add suport for ypool.

+1

NA screw ypool MORE COINS FOR ZNA

pool working - please upload new vorking version

Alright guys, pool works again, this time properly fixed.

All the issues were cause by the ancient ulimit -n of 1024. Basically the server ran out of file descriptors and this caused the blockchain to go corrupt.
This is also why the DOS was so effective and why it crashed again once we hit >100 workers.

I set the ulimit -n now to 102400, more than enough  ;D

UPDATE: DOS hit again! Had to shut down the server to prevent any further damage!

It's down again.

Pool worked a couple of minutes, now rests a couple of hours?

pool down ;D, again

There is an issue where invalid strings from DOS attacks are accepted. I will need to make a complete DOS proof check of the server code.

This will take at least 1h guys. The server will be offline until then.

madMAx510510 tell me please what is mean the following values

T - ?
A- ?
E - ?
primes - ?
fermat- ?

[GPU 0] T=61C A=99% E=0 primes=0.102144 fermat=391718/sec cpd=4.18/day
[GPU 1] T=60C A=99% E=0 primes=0.102022 fermat=390992/sec cpd=4.13/day
[GPU 2] T=62C A=99% E=0 primes=0.102153 fermat=391804/sec cpd=4.19/day
[GPU 3] T=61C A=99% E=0 primes=0.102158 fermat=343804/sec cpd=3.68/day

move service to cloudflare! perfect ddos protection guaranteed! fee would be enough to pay for it. in other case ddos will be permanent.

T - temp GPU
A-  Acept
E - Error

Once i fix these invalid data issues the server should be ok with these small scale attacks.

T = Temp
A = GPU activity (ie. load)
E = error count
primes = average prime probablility
fermat = number of fermats tests per second

GPU activity:

[GPU 0] A=98% 7970
[GPU 1] A=98% 7970
[GPU 2] A=98% 7970
[GPU 3] A=64% 7950

CPU load 25-30%

AMD Athlon X2 2 core

It`s not good values ?

I know this up and down again sucks, but in software development stuff never works on first try, it's always a process.
If you don't want to be part of the process then just wait a few days ;D

Some 7950s have issues with powertune which will make the A=% reading meaningless. Your CPU load seems fine, you can also try cpuload="0" in config.txt for lower cpu load.

MadMax, please add timestamps to miner output (may be in the next version)...

Are those coin that have been mined already still eventually reach my wallet?

UPDATE: I will take my time with this fix and make sure i don't miss anyting. So it will be at least another 2h with testing etc.
Basically i need to assume that any data being sent to the server can be invalid or garbage from DOSers and handle this situation accordingly.

Yes they will, but they are not even mature yet  ;)

at me it can't be connected

Connecting to frontend: 54.187.143.168:6666 ...

...and up, and down, and up and down anddddd,,,,, upp and down, here weeee goooo, and uppp and down and up and down, do your exercise here, cmon (personal trainer voice), and up and down (techno music as the background)

你个逗比

笑死了 ;D

when we can farm again?

2015 perhaps

Should just make a miner for ypool with dev fee and not deal with the pool headaches.

Doing some final tests right now.


too late.

ypool suxx - he setealing some money

Well, you probably won't have any time for it, but if you had one with a higher dev fee for times when site goes down, people could still mine for you :)

Ok, i fixed the issues, pool is up again.

Let's see how long  ;D

NOTE: Because of the recent downtime we mined only 18 blocks in the last 24h, which means that the share values will be a little unstable (ie. they are depending more on luck) until we have more blocks in the history.

good job!

when we start reciew first payments?

Pool System:
- The automatic payout threshold is 11 XPM
- Payouts will be executed after ~2 days when the coins are mature

GAMEOVER, pool down///

again dead(

I just have one PC with 4 GPUs  successfully connected to the pool and starts to mine.
But my other PC with 1 GPU crashes after connection.

xpmclient.exe crashes after showing "error: ProbablePrimeChainTestFast 0/4"


GPU 0: has 32 CUs
GPU 0: hash primorial = 30030 (14 bits)
GPU 0: sieve primorial = 10131543907 (63 bits)
GPU 0: sieve size = 55051543907 (7 bits)
GPU 0: max bits used 346/352

Connecting to frontend: 54.187.143.168:6666 ...
Connecting to bitcoin: 54.187.143.168:60000 ...
Connecting to signals: 54.187.143.168:60001 ...
Work received: height=531821 diff=10.895481 latency=702ms
error: ProbablePrimeChainTestFast 0/4
error: ProbablePrimeChainTestFast 0/4
error: ProbablePrimeChainTestFast 0/4
error: ProbablePrimeChainTestFast 0/4
error: ProbablePrimeChainTestFast 0/4
error: ProbablePrimeChainTestFast 0/4
error: ProbablePrimeChainTestFast 0/4



I have modified the config.txt to something like:


# 0 = don't use device
# 1 = use device
devices =        ["1"];

# 5 is the best setting regardless of GPU type
hashprimorial =  ["5"];

# Use 13 for Tahiti (79XX/280X)
# You can try 12 for 290/290X
sieveprimorial = ["12"];

# -1 means don't change
corefreq =       ["-1"];
memfreq =        ["-1"];
powertune =      ["20"];

Is my modification correct? Or it is a bug?  ???

Started mining again! thx for your hardwork! the performance diffrence between your and other miners is disturbing :O

This is a capture from one of my rigs with 5x280x 1100/1500 and pentium 4gb ram

http://s29.postimg.org/ogjeoe0rr/Capture.jpg

you see the last card is slower and generates less heat, can this be caused cuz of cpu bottleneck?

this is a capture from the other rig with the same hardware and better cooling. 2gb ram

http://s14.postimg.org/4xlaxpvc1/Capture2.jpg\

not exactly ideal. any help appreciated

EDIT: after submiting comment, pool dead -_-

Getting DOSed again, this time linux is failing...

Attacker IP address is: 46.16.33.40

Banning by iptables doesn't help?

You can not put a swimming pool on a normal dedicated server?

IP Address     Country           Region   City           ISP
46.16.33.40     United Kingdom   England   London   Anchorfree Ltd Network

http://rbls.org/46.16.33.40
Seems a compromised node used by spammer...

The site and miner are on.

I've been getting multiple requests to open a private pool which is a great idea.

Yes i'm reading your PMs ;) Just don't have time to respond right now.

But this is a good idea, however it means only people with static ip or dynamic DNS might be able to connect.

Yes, but it's a TCP battle right now  ;D

U can add only who farmed last days till buy hosting on cloudfare

I think this is a very bad idea, what about us that do not have a static IP?

Yeah, best way to stay running is using cloudflare protection.

Down again?

Yep, they are flooding the server with TCP connections. It's getting to the point where someting like cloudflare is the only option.

The private pool i mentioned would be running parallel to the public one, but i would be free from DOSers.

so when we weill see new pool or something?

Interested in the private pool. Any ETA estimate?

there are tons of public pools that get ddosed , it isn't end of world , there are just far more suitable providers than amazon aws for this purpose

What about a pool with invitation signup???

then we would not need a static IP to participate, and madmax would provide invitations via pm/email...

Any hopes to get something running in couple hours?

IP changes when you restart router without static? For me it usually runs for months straight.

Wow you guys really are suckers for punishment right..its like gox you all knew it was insolvent and gonna blow yet you just wanted to trade there even more because it had a higher looking volume and trade ammount. Its this type of mentality that keeps us sick and  in bonadage! and is how the media landscape has groomed us even though you know this guy dosn't have the first clue what he is doing you are willing to sit here waiting begging for his miner and pool thats dosn't work properly even though you know something bads surely, possibly about to happen to your coins that you mine with him!.. Read the writing on the wall!

He messed up his chance and should just give up now cause he really isn't capable of securing peoples coins! (look at all the devs that just keep coming back trying to fix their problems...they keep losing peoples coins and getting hacked!) So facilitating these kinds of actions by allowing someone a second third fourth chance like this you are just insane he really needs to just focus on the miner and give up the rest of his ordeal cause i'm concerned, good miner or not that he hasn't the faintest clue what could happen at any given time with regards to the server and you should too!

No offense Mad and like i said if you actually do payout then you have one hot smoking miner but until then i figure you should just step back into the shadows mate and focus on what you are good at. The fact is ypool has a great platform for this kind of work but you don't have to go with ypool there are a few others mate that allow a dev fee to be taken for their miner i just cant remember the pools names

with 7950 card your miner have some problem with GPU activity !!!

in claymore miner activity 96% and more
in cgminer too

in your miner GPU load 64% up to 100% to 1-2 second down to 64% - 10-20 second and up down up down
and CPU clock i set 1100, in you miner this value jump 850 to 1100.
and CPU memory i set 1500, in you miner this value jump 1250 to 1500.

it is very interesting problem !!!

@Tall and Tiny.

Stop bringing your bible to others church.

First they ignore you, then they laugh at you, then they fight you, then you win. -Gandhi.

Keed up good fight Mad.

im pretty sure beeeeer allows dev fees

this is the official announcement for our Primecoin mining pool www.beeeeer.org
As of writing this, ypool is not the only Primecoin pool any longer. ;)

we have our own board now: board.beeeeer.org

Some stats/features:

    adjustable pool fee (per miner, default: 2%)
    integrated developer fee, depends on the used miner
    no registration needed, username is your primecoin-payout-address*
    custom protocol & miner (based on mikaelh's high-performance client)
    your payouts are executed, when you reach at least 3.01 XPMs
    CPPSRB-based payout system, you're paid per share, current share values are shown on the main page --> *click* READ ABOUT CPPSRB *click* (also: donate!)
    high end visual stunning plain-text web interface
    beta testing phase (i'll work on every little bug/problem ASAP)
    third party stats: e.g. http://xpm.syware.de/

Yeah beer is a great pool there is one other that im thinking off too kev. I swear i saw him advertise in the pgpu or claymore thread a little while ago..know who i'm talking about?

The problem with the other pools is, they have such high latency that you loose 5-10% to stale/invalid shares. That's why i made my own.

While i'm working on a more permanent solution i will now accept ip-addresses that will be able to connect to the sever.
Please PM me, your ip cannot be known publicly of course.

Yeah i hear ya mate but there seems to be a reason they have high latency right cause they need to be protected as do the mining ports anyhow here is the pool i was referring to
http://www.minin.gs
and a cloudfare signup trial is free btw

i'm getting a ping of 200ms to that server which is the same as yours

Ping is not the issue, it's the time it takes to create work for all connected miners.

The first payments have just been executed.

They are even included in our own block ;D height 532078
What a coincidence.

I kept getting this error;

warning: ran out of hashes. pipeline stalled

Anyone have this issue as well?

It's only a warning and should happen rarely. It was only used for debugging, i just forgot to comment it out.

This one is funny  ::)

Sent you pm

is this a bug or what?
i was trying to run the miner a lot of times but result is the same

Connecting to frontend: 54.187.143.168:6666 ...

Guys? the pool site is up and reporting that the pool is up but my miners cant connect?

Right now your ip needs to be on the list, because of recent DOS attacks.
So send me a PM with your ip, then you can connect ;)

I have sent it to you

I think I have found a BLOCK for height=532230 but not shown in the "Last Blocks found by Pool" on your server.
(Or it is shown as height 532231 instead?)

Work received: height=532230 diff=10.896351 latency=169ms
[GPU 0] T=-1C A=-1% E=0 primes=0.105853 fermat=151145/sec cpd=2.31/day
[GPU 1] T=-1C A=-1% E=0 primes=0.105098 fermat=150501/sec cpd=2.14/day
[GPU 2] T=-1C A=-1% E=0 primes=0.105892 fermat=150452/sec cpd=2.30/day
[GPU 3] T=-1C A=-1% E=0 primes=0.105486 fermat=150831/sec cpd=2.22/day
[GPU 4] T=-1C A=-1% E=0 primes=0.105545 fermat=150883/sec cpd=2.24/day
(ST/INV/DUP): 9x 8ch(1/0/0) 3x 9ch(0/0/0)
GPU 4 found share: 10-ch type 2
GPU 4 found BLOCK!
Share accepted.
Work received: height=532231 diff=10.896364 latency=169ms
[GPU 0] T=-1C A=-1% E=0 primes=0.105805 fermat=150336/sec cpd=2.28/day
[GPU 1] T=-1C A=-1% E=0 primes=0.105113 fermat=150127/sec cpd=2.14/day
[GPU 2] T=-1C A=-1% E=0 primes=0.105878 fermat=150168/sec cpd=2.30/day
[GPU 3] T=-1C A=-1% E=0 primes=0.105465 fermat=149805/sec cpd=2.20/day
[GPU 4] T=-1C A=-1% E=0 primes=0.105644 fermat=149644/sec cpd=2.24/day
(ST/INV/DUP): 9x 8ch(1/0/0) 3x 9ch(0/0/0) 1x 10ch(0/0/0)

Now works great.

Can anybody say which sieve 12 or 13 is better for 290\290x in real tests?

Yep  ;)

PM Sent

PM sent!

PM sent

PM sent..

PM sent

will it run on 58xx?

Doing a server restart, something aint right...

I think it is about stale shares?
I see a lot of them...

Server is now crashing consistenly. Some bug in bitcoin function ConnectSocketDirectly is causing the issue.

This is just too much, i've been baby sitting this thing for over 48h straight now.

FINAL UPDATE: The pool will close. All balances will be PAID once they are mature. I WILL release a solo miner.

Need to sleep now...

same here

great for the SOLO MINING!!! Good work

Shame for the pool, but a solo miner with this speed would be best anyway...and a dev fee is fair I believe.

Can we use this miner to mine primecoins? http://www.coinion.com/2013/12/11/child-labor-bitcoin-mines-exposed/ (http://www.coinion.com/2013/12/11/child-labor-bitcoin-mines-exposed/)

good rather it be solo

You should just use beeeer or another one like it.

now thats funny :D

excellent idea! solo mining  ftw.

That's cool! I am trying to convert the opensource solo xpm miner to pool... may i can start with this..

You know other people might have better luck with the pool. Why not see about working with some one else on the forum? I for one have the $$ resources to help secure it just not the coding knowledge.

Dont kill a pool just because of a few bumps.

Instead of releasing a new solo mining fee software, may-be you could release an open source miner and end this shameful madness right now.
You will do a favor to crypto community and you will certainly receive a lot of donation too.
(not mentioning the fact ddosing your pool wasn't really nice and they deserve a lesson)

If you change your mind, I'm happy to help a bit.  I think you've taken some really nice steps forward with your architecture thus far and it'd be cool to see a new pool design succeed.  (Particularly if you were, in the long run, interested in open sourcing parts of the protocol so that others could build upon it).  Getting rid of the hacked up mix of binary gloop and stuff-over-http would be great.

A good place to start is here:

http://www.lognormal.com/blog/2012/09/27/linux-tcpip-tuning/
http://pic.dhe.ibm.com/infocenter/wasinfo/v6r1/index.jsp?topic=%2Fcom.ibm.websphere.edge.doc%2Fedge%2Fcp%2Fadmingd45.htm

A few are important - tuning time_wait will reduce your susceptibility to grade-school-level TCP attacks such as those that were happening the last few days.  But don't set it to 1 second like those guys did - 10 seconds should be adequate.

echo "1024 61000" > /proc/sys/net/ipv4/ip_local_port_range
echo "10" > /proc/sys/net/ipv4/tcp_fin_timeout
echo 32768 > /proc/sys/fs/file-max

syncookies should already be enabled, but if they're not:

echo 1 > /proc/sys/net/ipv4/tcp_syncookies

and check this page's suggestion for using iptables to rate-limit inbound synfloods:
http://www.liquidcomm.net/news/tech_tips/linux_os/how-to-manage-a-ddos-or-dos-attempt-directed-at-your-linux-server.html

It's very possible that the crash you're seeing from ConnectSocketDirectly is related still to running out of file descriptors or something very similar.  General server scalability tuning might make it disappear. 

http://www.nateware.com/linux-network-tuning-for-2013.html

has a few more - particularly the per-user open file limits, etc.  Don't bother with the congestion window and rmem/etc. stuff in there - your server isn't aiming for high tcp throughput on a single connection.

Doing it all via sysctl config in /etc/sysctl.conf is the most straightforward way to have your changes persist after a reboot.

Please don't stop the pool. Ypool is boring...

Dont give up on the pool! there must be a solution and it seems a lot of guys that can help you are here to help!

Don't give up

Same, dont give up!
Competition is great

yea dont give up! i am bored already

u guys dont understand pools these days will always get hit bad with ddos thats a fact i would concentrate on making faster miners not a pool and he seems to be doing this on his own not a team of people the only way a pool would be ok is if there were many many pools and a way to setup backups in case 1 went down which we dont have this isnt like cgminer where u can set that up in case of those situations

NEWS: I've decided to go open source.

See the official thread:
https://bitcointalk.org/index.php?topic=602292.0

EDIT: The pool server will be running and processing payouts until everything has been paid. All balances >= 0.1 XPM will be paid.
Note that you can't mine. It's running only for the payouts.

So is anybody setting up a new pool with this code?

You were right. This is the issue:
/usr/include/linux/posix_types.h:#define __FD_SETSIZE   1024

Basically when a process opens more than 1024 sockets any select() call will cause a buffer overflow.

And this is exactly what ConnectSocketDirectly does in netbase.cpp on line 359:
int nRet = select(hSocket + 1, NULL, &fdset, NULL, &timeout);

The issue boils down to the bitcoin code using the default value of __FD_SETSIZE as defined in posix_types.h
But you can't blame the bitcoin devs, they never imagined to open more than 1024 sockets.

Tomorrow i might implement a fix for this and start the pool again. I still have all your ips in the firewall settings ;)

Cheers man! :)

excellent idea! solo mining  ???

But again, it will be only temporary so there is at least one pool running at all.

In the long run i don't want to be bothered with DOS attacks or manage firewall settings all day long.
I think there should be multiple pools running and then the miner could be auto-switching in case one goes down.
And yeah solo mining of course.

That would be a nice feature of auto switching for pools, that would be a first for a XPM GPU miner. DDoS protection isn't cheap also.

What will the tentative time table when you will release a new version of the miner?

If your interested, i would be willing to host and manage the pool.

PM Me

Will it be possible to use p2pool mode?

Please be sure your solo miner will work with Catalyst 13.12 as its the only driver that work on my 6 x GPU rig (only driver that recognize the 6 x GPUs on same board)

PS: 13.12 worked 100% fine with your pool miner at 4.30 CPD per GPU (R9 280x core 1150 / mem 1500)

For me catalyst 13.12 works too on rig 3x7970+7950

Similar experience here.
13.12 works fine for 79XX / 280X but not with my r9 290s.  Gives an error and closes unless I use 14.x with them

When are u releasing solo miner?

did you tried using 13,12 modded drivers?

yes 13.12 = 13.12 modded

i uses 13.12 modded as the standard driver on all my 6 GPU rigs but with this particular miner for my R9 290s, it only works with 14.x drivers so for now i run it with 5 GPU.  13.12 modded works fine with my 6 GPU 79XX / 280X rigs though