[ANN] sph-sgminer: multi-coin multi-algorithm GPU miner | added MaruCoin

[flipme]

Yeah, changed nothing for me too.

Could swapping/buffering be a driver issue then?
Loading/unloading something for the chained algos?
I've looked thru changed files on the system and was astonished how much I/O fglrx actually does.

[1714952613]

1714952613

[1714952613]

1714952613

[1714952613]

1714952613

[flipme]

I just started again with a fresh copy of BAMT 1.6.
It comes with an sgminer for X11, but it crashed.
But it showed temp and fanspeed.
Who made that version? Bit confusing already.

Now I installed yours and the one from badman and it runs but doesn't show temp/fanspeed... and has the 10% drop of course.
Gonna try BAMT 2 later on, with 64bit ubuntu.

[badman74]

I just started again with a fresh copy of BAMT 1.6.
It comes with an sgminer for X11, but it crashed.
But it showed temp and fanspeed.
Who made that version? Bit confusing already.

Now I installed yours and the one from badman and it runs but doesn't show temp/fanspeed... and has the 10% drop of course.
Gonna try BAMT 2 later on, with 64bit ubuntu.

you need to put the adl drivers in /sgminer/ADL_SDK/
there is a readme in the folder telling you where to get them

[flipme]

I just started again with a fresh copy of BAMT 1.6.
It comes with an sgminer for X11, but it crashed.
But it showed temp and fanspeed.
Who made that version? Bit confusing already.

Now I installed yours and the one from badman and it runs but doesn't show temp/fanspeed... and has the 10% drop of course.
Gonna try BAMT 2 later on, with 64bit ubuntu.

you need to put the adl drivers in /sgminer/ADL_SDK/
there is a readme in the folder telling you where to get them

Thanks, it's even included in the BAMT 1.6 distro.
I recompiled and ran, and now it's crashing again after a minute!

A build without those headers runs stable.

[badman74]

I just started again with a fresh copy of BAMT 1.6.
It comes with an sgminer for X11, but it crashed.
But it showed temp and fanspeed.
Who made that version? Bit confusing already.

Now I installed yours and the one from badman and it runs but doesn't show temp/fanspeed... and has the 10% drop of course.
Gonna try BAMT 2 later on, with 64bit ubuntu.

you need to put the adl drivers in /sgminer/ADL_SDK/
there is a readme in the folder telling you where to get them

Thanks, it's even included in the BAMT 1.6 distro.
I recompiled and ran, and now it's crashing again after a minute!

maybe your clocks or intensity is too high

[n00bnoxious]

I have this issue with ALL my 290x BAMT 1.6 rigs too.   They drop 200k+ after a few mins.   So, max I can get it to is 2.41 Mh/s.   Do you have any fixes or workaround for this yet that you have discovered?

Could I have opinion of specialist about my Problem? I suspect a bug of the programm because that'is the only thing I had not check, someone can tell me if it's possible?

I offer on the other topic 30 Darkcoin Bounty to solve that problem, a nobody has a clue, but I get in touch with other that had the same problem.

Here is the facts:
I have 3 different R9 290, other people has 6 290 of other brand, after 10 minutes, the cards looses 10% of Hash.
That probelm happend only on R9 290 cars.
I'm on Bamt 1.6 from Bajawa, He is on windows 7, same problem.
All 290 on the same rig experiment the loss in the exact same second
if you put another card in the rigg that is not a 290, all the 290 had a loss, but not the other card
The cards are cold.
The problem happenedonly with sph-darkcoin miner, if I mine Litecoin, I'm stable.
I split the cards with 2 PSU, does not change a thing, cards crashed at the same time, no matter what psu they are plugged.
I change motherboard, does not change a thing.
I run 4 different sgminer, each one with one card activated: even I did launch them with a delay of 3 minutes, 10 minutes later, all the instance with a 290 had a 10% lossat the same time
I launched sgminer without a .conf: only intensity in the parameters: happened with I:18 and I:21 (0% rejected, cards should worked with high intensity)
Once I got the loss only a cold reboot can recover the missing hash. If I mine restart only the miner, it restart but is capped to 90% of max hashrate.

I'm getting out of hints. If the problem is in the programm, I shoud stop wasting my time. If it is not, I will keep investigate, even if I think I already get pretty far.
PS: as usual, english is not my first languagae, so if I offensed any-one, it was not intended.

I've not been following your problem since the start so can you confirm what intensity you're on? I noticed on my R9 280X cards that if I put I up too high, first of all I'd get an extra boost of power right at the start, but my stale shares went up hugely and then the speed dropped down to its previous level.

My suggestion is to run it on I13, and make sure that's stable for the entire time, then move up 1 at a time until you find the magical balance between speed and stability.

Let me know if this helps :-)

[CryptoGretzky]

So... this is on my Asus R9 290x Elphida flashed with the Stilt Bios running on PiMP 1.1 beta (64 bit BAMT)...  Motherboard is a Gigabyte GA-EP45-UD3P.

Don't know if it's coincidence or what yet, I will test it on another rig, but using PiMP, but so far, it hasn't dropped after many hours running on it....   getting around 2.78 Mh/s



Oh... forgot to mention that I already tested this version of sgminer 4.1.0-96-g983e on BAMT 1.6 running on Asrock H81 Pro BTC with, but that dropped after a few mins still.   Will post again after I test PiMP on another rig.

[maniacmusic]

I tried to use sph sgminer to mine Offerings (Cthulhu).... what value should I give to the -k option?? I tried -k quark, -k  quarkcoin and even --kernel-path kernel --kernel quarkcoin but sgminer won't start
Thanks..

[EnTiTy77]

I have this issue with ALL my 290x BAMT 1.6 rigs too.   They drop 200k+ after a few mins.   So, max I can get it to is 2.41 Mh/s.   Do you have any fixes or workaround for this yet that you have discovered?

Could I have opinion of specialist about my Problem? I suspect a bug of the programm because that'is the only thing I had not check, someone can tell me if it's possible?

I offer on the other topic 30 Darkcoin Bounty to solve that problem, a nobody has a clue, but I get in touch with other that had the same problem.

Here is the facts:
I have 3 different R9 290, other people has 6 290 of other brand, after 10 minutes, the cards looses 10% of Hash.
That probelm happend only on R9 290 cars.
I'm on Bamt 1.6 from Bajawa, He is on windows 7, same problem.
All 290 on the same rig experiment the loss in the exact same second
if you put another card in the rigg that is not a 290, all the 290 had a loss, but not the other card
The cards are cold.
The problem happenedonly with sph-darkcoin miner, if I mine Litecoin, I'm stable.
I split the cards with 2 PSU, does not change a thing, cards crashed at the same time, no matter what psu they are plugged.
I change motherboard, does not change a thing.
I run 4 different sgminer, each one with one card activated: even I did launch them with a delay of 3 minutes, 10 minutes later, all the instance with a 290 had a 10% lossat the same time
I launched sgminer without a .conf: only intensity in the parameters: happened with I:18 and I:21 (0% rejected, cards should worked with high intensity)
Once I got the loss only a cold reboot can recover the missing hash. If I mine restart only the miner, it restart but is capped to 90% of max hashrate.

I'm getting out of hints. If the problem is in the programm, I shoud stop wasting my time. If it is not, I will keep investigate, even if I think I already get pretty far.
PS: as usual, english is not my first languagae, so if I offensed any-one, it was not intended.

I've not been following your problem since the start so can you confirm what intensity you're on? I noticed on my R9 280X cards that if I put I up too high, first of all I'd get an extra boost of power right at the start, but my stale shares went up hugely and then the speed dropped down to its previous level.

My suggestion is to run it on I13, and make sure that's stable for the entire time, then move up 1 at a time until you find the magical balance between speed and stability.

Let me know if this helps :-)

I did exactly what n00bnoxious did but with xintensity just to see what would happen. Using SPH-SGminer on Darkcoin with 3 R9 290.

Went all the way to 620 and they hash was good there although I was getting 400k/h swings up and down which isn't real stable and the computer would stop responding after 10-20 min which I have never seen before.Ended up on 410, swings were tighter but still there and for some reason the 3rd gpu would sit 200k/h below the other cards.


Decided to play with the GPU threads as on scrypt running 2 instead of 1 gets me an extra 20-30k/h per card. Started at 4 per card , went to 8 - no real change, went to 12 - WOW! Swinging gone! Cards stabilized on hashrate and no drop.

Give it a go.

boubou you can send the 30 darkcoin bounty to Xpw1MPT1Fuv2xcod9F7DcNo6jhnBHEQht9  

[boubou]

I have this issue with ALL my 290x BAMT 1.6 rigs too.   They drop 200k+ after a few mins.   So, max I can get it to is 2.41 Mh/s.   Do you have any fixes or workaround for this yet that you have discovered?

Could I have opinion of specialist about my Problem? I suspect a bug of the programm because that'is the only thing I had not check, someone can tell me if it's possible?

I offer on the other topic 30 Darkcoin Bounty to solve that problem, a nobody has a clue, but I get in touch with other that had the same problem.

Here is the facts:
I have 3 different R9 290, other people has 6 290 of other brand, after 10 minutes, the cards looses 10% of Hash.
That probelm happend only on R9 290 cars.
I'm on Bamt 1.6 from Bajawa, He is on windows 7, same problem.
All 290 on the same rig experiment the loss in the exact same second
if you put another card in the rigg that is not a 290, all the 290 had a loss, but not the other card
The cards are cold.
The problem happenedonly with sph-darkcoin miner, if I mine Litecoin, I'm stable.
I split the cards with 2 PSU, does not change a thing, cards crashed at the same time, no matter what psu they are plugged.
I change motherboard, does not change a thing.
I run 4 different sgminer, each one with one card activated: even I did launch them with a delay of 3 minutes, 10 minutes later, all the instance with a 290 had a 10% lossat the same time
I launched sgminer without a .conf: only intensity in the parameters: happened with I:18 and I:21 (0% rejected, cards should worked with high intensity)
Once I got the loss only a cold reboot can recover the missing hash. If I mine restart only the miner, it restart but is capped to 90% of max hashrate.

I'm getting out of hints. If the problem is in the programm, I shoud stop wasting my time. If it is not, I will keep investigate, even if I think I already get pretty far.
PS: as usual, english is not my first languagae, so if I offensed any-one, it was not intended.

I've not been following your problem since the start so can you confirm what intensity you're on? I noticed on my R9 280X cards that if I put I up too high, first of all I'd get an extra boost of power right at the start, but my stale shares went up hugely and then the speed dropped down to its previous level.

My suggestion is to run it on I13, and make sure that's stable for the entire time, then move up 1 at a time until you find the magical balance between speed and stability.

Let me know if this helps :-)

I did exactly what n00bnoxious did but with xintensity just to see what would happen. Using SPH-SGminer on Darkcoin with 3 R9 290.

Went all the way to 620 and they hash was good there although I was getting 400k/h swings up and down which isn't real stable and the computer would stop responding after 10-20 min which I have never seen before.Ended up on 410, swings were tighter but still there and for some reason the 3rd gpu would sit 200k/h below the other cards.


Decided to play with the GPU threads as on scrypt running 2 instead of 1 gets me an extra 20-30k/h per card. Started at 4 per card , went to 8 - no real change, went to 12 - WOW! Swinging gone! Cards stabilized on hashrate and no drop.

Give it a go.

boubou you can send the 30 darkcoin bounty to Xpw1MPT1Fuv2xcod9F7DcNo6jhnBHEQht9  

I just went back home...

Are you serious?

I planned to go to bed but I have to try that before.
So gpu-thread 12 same intensity (21)?
I check that right now.

EDIT: the line should be "gpu-thread": "12",
It does not seems to work, i don't seemore threads.

[boubou]

So... this is on my Asus R9 290x Elphida flashed with the Stilt Bios running on PiMP 1.1 beta (64 bit BAMT)...  Motherboard is a Gigabyte GA-EP45-UD3P.

Don't know if it's coincidence or what yet, I will test it on another rig, but using PiMP, but so far, it hasn't dropped after many hours running on it....   getting around 2.78 Mh/s



Oh... forgot to mention that I already tested this version of sgminer 4.1.0-96-g983e on BAMT 1.6 running on Asrock H81 Pro BTC with, but that dropped after a few mins still.   Will post again after I test PiMP on another rig.

Have you got a link to your bamt image that you are using?

[CryptoGretzky]

So... this is on my Asus R9 290x Elphida flashed with the Stilt Bios running on PiMP 1.1 beta (64 bit BAMT)...  Motherboard is a Gigabyte GA-EP45-UD3P.

Don't know if it's coincidence or what yet, I will test it on another rig, but using PiMP, but so far, it hasn't dropped after many hours running on it....   getting around 2.78 Mh/s



Oh... forgot to mention that I already tested this version of sgminer 4.1.0-96-g983e on BAMT 1.6 running on Asrock H81 Pro BTC with, but that dropped after a few mins still.   Will post again after I test PiMP on another rig.

Have you got a link to your bamt image that you are using?

http://getpimp.org/   

It's by the same guys (Bee) that made the older 1.3-1.5 Bamt https://litecointalk.org/index.php?topic=2924.0

[EnTiTy77]

I just went back home...

Are you serious?

I planned to go to bed but I have to try that before.
So gpu-thread 12 same intensity (21)?
I check that right now.

Intensity doesn't seem to make any difference after 20 when you have high gpu threads for some reason.

But yeah same intensity, higher threads.

[CryptoGretzky]

I just went back home...

Are you serious?

I planned to go to bed but I have to try that before.
So gpu-thread 12 same intensity (21)?
I check that right now.

Intensity doesn't seem to make any difference after 20 when you have high gpu threads for some reason.

But yeah same intensity, higher threads.

Can you please post your full conf file?

[EnTiTy77]

]
,
"intensity" : "0,0,0",
"xintensity" : "410,410,410",
"rawintensity" : "0,0,0",
"worksize" : "128,128,128",
"kernel" : "darkcoin,darkcoin,darkcoin",
"lookup-gap" : "2,2,2",
"thread-concurrency" : "25601,25601,25601",
"shaders" : "0,0,0",
"gpu-threads" : "12,12,12",
"gpu-engine" : "900-1000,900-1040,900-1000",
"gpu-fan" : "0-85,0-85,0-85",
"gpu-memclock" : "1375,1500,1375",
"gpu-memdiff" : "0,0,0",
"gpu-powertune" : "20,20,20",
"gpu-vddc" : "0.000,0.000,0.000",
"temp-cutoff" : "95,95,95",
"temp-overheat" : "85,85,85",
"temp-target" : "75,75,75",
"api-mcast-port" : "4028",
"api-port" : "4028",
"auto-fan" : true,
"auto-gpu" : true,
"expiry" : "28",
"failover-switch-delay" : "60",
"gpu-dyninterval" : "7",
"gpu-platform" : "1",
"cl-filename" : "scryptR9290,scryptR9290,scryptR9290",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "7",
"tcp-keepalive" : "30",
"temp-hysteresis" : "3",
"shares" : "0",
"kernel-path" : "/usr/local/bin",
"no-submit-stale" : true
}

[HooDude]

I have this issue with ALL my 290x BAMT 1.6 rigs too.   They drop 200k+ after a few mins.   So, max I can get it to is 2.41 Mh/s.   Do you have any fixes or workaround for this yet that you have discovered?

Could I have opinion of specialist about my Problem? I suspect a bug of the programm because that'is the only thing I had not check, someone can tell me if it's possible?

I offer on the other topic 30 Darkcoin Bounty to solve that problem, a nobody has a clue, but I get in touch with other that had the same problem.

Here is the facts:
I have 3 different R9 290, other people has 6 290 of other brand, after 10 minutes, the cards looses 10% of Hash.
That probelm happend only on R9 290 cars.
I'm on Bamt 1.6 from Bajawa, He is on windows 7, same problem.
All 290 on the same rig experiment the loss in the exact same second
if you put another card in the rigg that is not a 290, all the 290 had a loss, but not the other card
The cards are cold.
The problem happenedonly with sph-darkcoin miner, if I mine Litecoin, I'm stable.
I split the cards with 2 PSU, does not change a thing, cards crashed at the same time, no matter what psu they are plugged.
I change motherboard, does not change a thing.
I run 4 different sgminer, each one with one card activated: even I did launch them with a delay of 3 minutes, 10 minutes later, all the instance with a 290 had a 10% lossat the same time
I launched sgminer without a .conf: only intensity in the parameters: happened with I:18 and I:21 (0% rejected, cards should worked with high intensity)
Once I got the loss only a cold reboot can recover the missing hash. If I mine restart only the miner, it restart but is capped to 90% of max hashrate.

I'm getting out of hints. If the problem is in the programm, I shoud stop wasting my time. If it is not, I will keep investigate, even if I think I already get pretty far.
PS: as usual, english is not my first languagae, so if I offensed any-one, it was not intended.

I've not been following your problem since the start so can you confirm what intensity you're on? I noticed on my R9 280X cards that if I put I up too high, first of all I'd get an extra boost of power right at the start, but my stale shares went up hugely and then the speed dropped down to its previous level.

My suggestion is to run it on I13, and make sure that's stable for the entire time, then move up 1 at a time until you find the magical balance between speed and stability.

Let me know if this helps :-)

I did exactly what n00bnoxious did but with xintensity just to see what would happen. Using SPH-SGminer on Darkcoin with 3 R9 290.

Went all the way to 620 and they hash was good there although I was getting 400k/h swings up and down which isn't real stable and the computer would stop responding after 10-20 min which I have never seen before.Ended up on 410, swings were tighter but still there and for some reason the 3rd gpu would sit 200k/h below the other cards.


Decided to play with the GPU threads as on scrypt running 2 instead of 1 gets me an extra 20-30k/h per card. Started at 4 per card , went to 8 - no real change, went to 12 - WOW! Swinging gone! Cards stabilized on hashrate and no drop.

Give it a go.

boubou you can send the 30 darkcoin bounty to Xpw1MPT1Fuv2xcod9F7DcNo6jhnBHEQht9  

This worked wonders for me. I can't get --gpu-threads 12,12 in my .bat to work (SGminer says its an invalid argument) but when I used --gpu-threads 8,8 it brought my hashrate up significantly on my R9 290x Tri-X's. The hashrate still fluctuates a fair bit, but on average it is much higher.

[badman74]

I have this issue with ALL my 290x BAMT 1.6 rigs too.   They drop 200k+ after a few mins.   So, max I can get it to is 2.41 Mh/s.   Do you have any fixes or workaround for this yet that you have discovered?

Could I have opinion of specialist about my Problem? I suspect a bug of the programm because that'is the only thing I had not check, someone can tell me if it's possible?

I offer on the other topic 30 Darkcoin Bounty to solve that problem, a nobody has a clue, but I get in touch with other that had the same problem.

Here is the facts:
I have 3 different R9 290, other people has 6 290 of other brand, after 10 minutes, the cards looses 10% of Hash.
That probelm happend only on R9 290 cars.
I'm on Bamt 1.6 from Bajawa, He is on windows 7, same problem.
All 290 on the same rig experiment the loss in the exact same second
if you put another card in the rigg that is not a 290, all the 290 had a loss, but not the other card
The cards are cold.
The problem happenedonly with sph-darkcoin miner, if I mine Litecoin, I'm stable.
I split the cards with 2 PSU, does not change a thing, cards crashed at the same time, no matter what psu they are plugged.
I change motherboard, does not change a thing.
I run 4 different sgminer, each one with one card activated: even I did launch them with a delay of 3 minutes, 10 minutes later, all the instance with a 290 had a 10% lossat the same time
I launched sgminer without a .conf: only intensity in the parameters: happened with I:18 and I:21 (0% rejected, cards should worked with high intensity)
Once I got the loss only a cold reboot can recover the missing hash. If I mine restart only the miner, it restart but is capped to 90% of max hashrate.

I'm getting out of hints. If the problem is in the programm, I shoud stop wasting my time. If it is not, I will keep investigate, even if I think I already get pretty far.
PS: as usual, english is not my first languagae, so if I offensed any-one, it was not intended.

I've not been following your problem since the start so can you confirm what intensity you're on? I noticed on my R9 280X cards that if I put I up too high, first of all I'd get an extra boost of power right at the start, but my stale shares went up hugely and then the speed dropped down to its previous level.

My suggestion is to run it on I13, and make sure that's stable for the entire time, then move up 1 at a time until you find the magical balance between speed and stability.

Let me know if this helps :-)

I did exactly what n00bnoxious did but with xintensity just to see what would happen. Using SPH-SGminer on Darkcoin with 3 R9 290.

Went all the way to 620 and they hash was good there although I was getting 400k/h swings up and down which isn't real stable and the computer would stop responding after 10-20 min which I have never seen before.Ended up on 410, swings were tighter but still there and for some reason the 3rd gpu would sit 200k/h below the other cards.


Decided to play with the GPU threads as on scrypt running 2 instead of 1 gets me an extra 20-30k/h per card. Started at 4 per card , went to 8 - no real change, went to 12 - WOW! Swinging gone! Cards stabilized on hashrate and no drop.

Give it a go.

boubou you can send the 30 darkcoin bounty to Xpw1MPT1Fuv2xcod9F7DcNo6jhnBHEQht9  

This worked wonders for me. I can't get --gpu-threads 12,12 in my .bat to work (SGminer says its an invalid argument) but when I used --gpu-threads 8,8 it brought my hashrate up significantly on my R9 290x Tri-X's. The hashrate still fluctuates a fair bit, but on average it is much higher.

max threads is set at 10

Code:

char *set_gpu_threads(char *arg)
{
	int i, val = 1, device = 0;
	char *nextptr;

	nextptr = strtok(arg, ",");
	if (nextptr == NULL)
		return "Invalid parameters for set_gpu_threads";
	val = atoi(nextptr);
	if (val < 1 || val > 10)
		return "Invalid value passed to set_gpu_threads";

	gpus[device++].threads = val;

	while ((nextptr = strtok(NULL, ",")) != NULL) {
		val = atoi(nextptr);
		if (val < 1 || val > 10)
			return "Invalid value passed to set_gpu_threads";

		gpus[device++].threads = val;
	}
	if (device == 1) {
		for (i = device; i < MAX_GPUDEVICES; i++)
			gpus[i].threads = gpus[0].threads;
	}
	
	return NULL;
}

in driver-opencl.c

[CryptoGretzky]

Ok, on my other rig, I got PiMP running and this is the result after a few mins (all are Sapphire Tri-X 290x on  Asrock H81 Pro BTC):



GPU 0, 1, 2 are from a newer batch.   GPU 3 is an older one where I couldn't get it stable until I flashed it with the bios that I ripped from the newer GPU.

Noticed the first 2 that the temp/fan speed can be read dropped after a few mins back to 2.4 Mh?

GPU 2,3 temp couldn't be read for some reason and they stayed up at 2.78-2.8 Mh.

So... I am thinking that auto-gpu is throttling it back.   Temperature is quite low though...  my temp-target is 75.  Let me test temp-target with 80.

[cinnamon_carter]

First I want to say I think this project has been well done adding support for many coins.

I noticed you have support for Quark.

Is it possible future development could add support for Simple Inflation Coin which is the father of Quark, secure and all similar clones

Simple Inflation Coin is slightly different in that it has just the 6 algo's rounds thus the quark miner will not just work out of the box for obvious reasons.

Here is the original launch thread of SIF  https://bitcointalk.org/index.php?PHPSESSID=fhfogsq68mludm4ip82ar2c8c0&topic=240894.0;all

ocminer has another thread here he put up for his pool and cloud mining https://bitcointalk.org/index.php?topic=404215.0


A little harder to follow since I don't speak/read Russian is the original launch thread of the coin here

https://bitcointalk.org/index.php?topic=240884.0


The english thread was not well 'organized' and originally did not contain the source code which is here

http://sourceforge.net/p/sifcoin/code/ci/master/tree/

and on github here https://github.com/sifcoin/sifcoin

Finally here is the source code to the most efficient cpu miner currently available to my knowledge

https://github.com/ig0tik3d/sifcoin-miner


SIC is not traded on any exchanges yet it has been running since last summer and everytime I check it there are always lots of nodes availalbe.

Difficulty goes up and down a little peaking most last October when there was a rumor it was to be added to BTCE for trading

and an even larger peak in difficulty a few months back when the optimized miner became available.

I hope you will consider it.

I hope others in the 'small' community of supporters of the coin would join me in a donation to you for your efforts if you ever have time.

Regards.....

[EnTiTy77]

I know 12 threads aren't being engaged.

But it is stabilizing the hashrate