[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 704.

Legionnairez

full member

Activity: 163

Merit: 100

Quote from: ivanlabrie on March 12, 2014, 04:02:41 PM

Quote from: Legionnairez on March 12, 2014, 03:25:10 PM

I'm having trouble with the hashrate of my ATI if I try to mine with my nvidia at the same time. Could it be cudaminer setting related?
The setup is as follows
8gb ram
Gigabyte z68ap-d3 MB
Intel i5 3570k
1000w seasonic platinum
gigabyte 280x
Zotac gtx 680 4GB

Cudaminer settings for scrypt:
-H 1 -C 1 -d 0 -i 0 -m 1 -l K8x32
results in 330 kh/s, not affected if the 280 is mining. On its own the 280 gives 720 kh/s, with cudaminer on around 620-650. I think I should be able to run it with full hashrate somehow?
I can provide some hardware readings too if necessary. Only thing unusual I notice with both running is the pagefile usage goes up to 8000+ compared to 6500 with just one miner on.

H and C flags are cpu hogs if you set them to 1.

AFter some testing, combination of H2 C1 seems to be best - full hashrate on 280x and just -10 for GTX 680.
Cheers peeps!

ManIkWeet

full member

Activity: 182

Merit: 100

Quote from: 69charger on March 13, 2014, 01:03:25 PM

I had to revert back to the older drivers due to this system freeze issue. Not having problems since I went back. New drivers have an issue.

Yeah I am still on old drivers too, not an issue at all with them, though I haven't really been mining anymore... (Electricity in NL is high cost)

69charger

full member

Activity: 173

Merit: 100

I had to revert back to the older drivers due to this system freeze issue. Not having problems since I went back. New drivers have an issue.

bigjme

sr. member

Activity: 350

Merit: 250

I will give that a try if it happens again. I've moved almost my entire system to certain virtual machines, so the only thing running on my system is cudaminer. I've updated Mt bios to the newest incase it's a bios lockup.

We will soon see how it goes

Morgahl

member

Activity: 70

Merit: 10

Quote from: bigjme on March 13, 2014, 04:32:33 AM

Internet doesn't cut off. I have something checking that for me. And it's running straight off an SSD, I'm going to see what it's like with the wallets on a virtual for a few days. At least if the virtual dies it's not a big issue

The bit bugging me is that it only happens at night while I'm asleep. It will happily mine all day, system temps of my god and cup never hit the 50°C mark so I know it's not temps

When I've had this happen to me it is just the miner not allowing an interrupt to happen to trigger the monitor to open. Basically all I've had to do is unplug the network cable and wait for the miner to stale and lose connection to the pool/chain and stop mining. This allows the interrupt and i can just plug the cable back in and it will mine once connected.

myagui

legendary

Activity: 1154

Merit: 1001

Quote from: Jayjay04 on March 13, 2014, 10:56:12 AM

Anybody has an idea why when I tried overclocking my gtx 465 with the nvidia configuration panel or any other overclocking program, nothing really happen, even worst reaching a certain point, the hashrate on cudaminer starts decreasing to 90 or 50 kh/s....

On another computer I was getting 180kh/s at gpu clock +175 without any problems. (PSU was only 380w)
But now can't reach more than 120kh/s and overclocking do absolutely nothing.... ?!?! (with a corsair 750w and a better motherboard...)

Any clue why ?!?!

That nvidia model strikes me as an oldie. Me thinks you might need to test it again with earlier cudaminer releases, as latest versions might have dropped the love for the oldies.
Other than that, hard to say, perhaps drivers, mb bios, etc. Note also that many times, overclocking produces the opposite to the desired effect, particularly when heat and/or hardware errors start to come up.

Best of luck,
- Myagui

Jayjay04

legendary

Activity: 1364

Merit: 1000

Anybody has an idea why when I tried overclocking my gtx 465 with the nvidia configuration panel or any other overclocking program, nothing really happen, even worst reaching a certain point, the hashrate on cudaminer starts decreasing to 90 or 50 kh/s....

On another computer I was getting 180kh/s at gpu clock +175 without any problems. (PSU was only 380w)
But now can't reach more than 120kh/s and overclocking do absolutely nothing.... ?!?! (with a corsair 750w and a better motherboard...)

Any clue why ?!?!

ivanlabrie

hero member

Activity: 812

Merit: 1000

Yup, good ole k boost works...Or a properly modded Skynet bios.
Sadly he never got back to me on modding the GTX 750 Ti bios.

myagui

legendary

Activity: 1154

Merit: 1001

Quote from: see360 on March 13, 2014, 08:04:44 AM

Quote from: sin242 on March 13, 2014, 07:46:18 AM

For anyone relying on the chrome trick for maximum speed: Have you tried the K-boost option on EVGA's precisionX tool?

I got a system crash the moment I hit the K-boost switch and haven't tried it since. Would love to hear more about it.

This would happen to me too, until I tried resetting all overclock settings and have mining stopped *before* hitting the K-boost feature. With K-boost ON, I haven't needed the Chrome trick, but on the other hand, these are dedicated mining cards, 'cause if I had any idle time on them, I'd avoid K-boost since it keeps the cards in full power mode at all times.

Cheers,
Myagui

see360

member

Activity: 91

Merit: 10

Quote from: sin242 on March 13, 2014, 07:46:18 AM

For anyone relying on the chrome trick for maximum speed: Have you tried the K-boost option on EVGA's precisionX tool?

I got a system crash the moment I hit the K-boost switch and haven't tried it since. Would love to hear more about it.

sin242

sr. member

Activity: 280

Merit: 250

For anyone relying on the chrome trick for maximum speed: Have you tried the K-boost option on EVGA's precisionX tool?

bigjme

sr. member

Activity: 350

Merit: 250

yeh it does seem a little weird, aslong as im getting full speed i dont care. but im wondering if its the motherboard seeing the wrong speed for the bandwidth and locking up every so often, as i only get lockups when mining

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: bigjme on March 13, 2014, 07:41:41 AM

ok HWinfo reports this when mining
PCIe v3.0 x16 (8.0 Gb/s) @ x2 (8.0 Gb/s)

so speed wise its correct, and when im not mining it says
PCIe v3.0 x16 (8.0 Gb/s) @ x2 (2.5 Gb/s)

so its just the speed its running at is messing up, the actual throughput is right

It is also doing this for my cards.

djm34

legendary

Activity: 1400

Merit: 1050

I just have a look into gpu-z
the 780ti on a first pci slot x8
the amd on the second pcie slot (on riser) x1 (strange)

according hwinfo64
pcie link speed 8 Gbps (whatever it is...) for both cards...

bigjme

sr. member

Activity: 350

Merit: 250

ok HWinfo reports this when mining
PCIe v3.0 x16 (8.0 Gb/s) @ x2 (8.0 Gb/s)

so speed wise its correct, and when im not mining it says
PCIe v3.0 x16 (8.0 Gb/s) @ x2 (2.5 Gb/s)

so its just the speed its running at is messing up, the actual throughput is right

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: bigjme on March 13, 2014, 07:17:20 AM

1 GPU
havent checked in the bios but even dropping from PCI-E 3 to 2 the x2 should go to x4
guessing GPU-Z dislikes my MB

Try to install hwinfo, it gives more detailed information regarding the system

bigjme

sr. member

Activity: 350

Merit: 250

1 GPU
havent checked in the bios but even dropping from PCI-E 3 to 2 the x2 should go to x4
guessing GPU-Z dislikes my MB

djm34

legendary

Activity: 1400

Merit: 1050

how many card is there ?
It can also come from the card (unlikely though, I had no problem with my old gtx660)
Did you check in the bios ?

bigjme

sr. member

Activity: 350

Merit: 250

auto updates are disabled, and power settings are on high performance, ive even disabled turning off the monitor

see360

member

Activity: 91

Merit: 10

Quote from: bigjme on March 13, 2014, 04:00:34 AM

...I'm running windows 7 x64, and leaving it running it will run for months. If I leave cudaminer running it will run all day fine. But I leave it all night and my system hangs at random times...

The timing of the crashes leads me to suspect the Windows auto update feature which occasionally happens overnight by default. I am also running on Win 7 x64 (three GPUs) for weeks at a time, but I don't think I have ever had a hang like this. I see that you have tried disabling sleep mode, but be sure to also disable auto updates.

Try this:
For Windows 7 the two most important settings are to disable automatic updates and to disable sleep mode. Windows decides by default after downloading an important update to reboot your system in the middle of the night, which occurs only often enough to make it hard to figure out what happened. Going to sleep is different from turning off the display, and you can adjust them independently as shown below.

I. Click on the Start Button and find
"Control Panel->System and Security->Windows Update->Change Settings"
Change "Important Updates" to something like "Download updates but let me choose whether to install them"

II. Control Panel->System and Security->Power Options
Whatever plan you use click on "Change plan settings" and make sure "Put the computer to sleep" is set to "Never"

Just remember to update manually from time to time, because running Windows without updates leaves a system vulnerable to exploits.

Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 704. (Read 3426976 times)