Get Trixx for complete voltage and clock control.

https://www.sapphireselectclub.com/ssc/TriXX/

Thanks for the replies. Seems like its an issue then between cgminer and windows/drivers.

They've disabled overvolting in afterburner and you have to jump through alot of hoops to get it working - search engine it and there are guides. In addition, some cards are now voltage locked.

As for the high temps, its possible(but not likely) you got a bad thermal paste job, or have an airflow problem.

80 is really not so horrible. My old 8800GT once hit 130 C and lived to tell about it. My old x1950 used to run regularly at 100C when gaming, 75 at idle. I ran that card 24/7 for 6 years straight - it still works. Modern GPU's can really take some heat. They're usually obsolete long before they burn out. I had a Radeon 9800 pro that I bought back in 2003. That card burned out in 2010 only because the thermal paste dried out and the GPU fried. But now I'm rambling like Grandpa Simpson.

My new 7850 runs at 62 C full load (1050/1300 @1.075v) though it has the XFX double dissipation cooler on it. The XFX Double D cooler models are the only ones now with a lifetime warranty.

When I go back to the desktop after gaming and start cgminer, I get really low hash rates in the single digits (usually get well over 350k for scrypt). If I reboot, all returns to normal. Of course its a pain to have to reboot to switch from gaming to mining, so I'd rather not have to do this. Seems like the video card is not being restored after the games are shut down. I tried the GPU restart in cgminer, but it didn't help. Anyone know what might be causing the issue or have this happen?

Here are system details:
Games this happens with: Stalker Pripyat, Call of Juarez, Battlefield 3, maybe others
Video Card: XFX 7850 Double D 1GB
Drivers:13.1 Catalyst
Windows 7 Ultimate, DX 11
CPU: E8400 @ 3GHz
RAM: 3GB (Max MB will run reliably)
MB: Gigabyte P35-DS3L
cgminer 2.11.4

I've got a 7850. Out of the box it did 250Kh/s. After some tweaking its now averaging 360Kh/s. I think max it would do is 380 headless. So a 7870 should easily hit these values. But alas its not always so easy.

Maybe try a driver cleaner and then install the 13.1 catalyst (App SDK is included)

I'm using afterburner to OC.
No Overvolting
Core Clock: 1050
Mem Clock: 1300
Powertune: +10%
Temperature: 62 C

These are my cgminer 2.11.4 settings:

Your results may vary. Set either shaders or thread concurrency, but not both. I would suggest thread concurrency at 8192. Higher worksize can sometimes increase hash rate. Do the setx commands. Start the intensity at 12 and work up. Make sure there are no other apps running that might steal GPU cycles. Make sure your motherboard FSB and PCIe bus speeds are set properly. Aero can be enabled - some report disabling it can reduce performance. Thats all I can think of right now. Give us more details and maybe we can figure it out. Good Luck.

I have a 7850 in a PCIe 1.1 x16 slot on a socket 775 MB with 3GB RAM and E8400 CPU @ 3GHz. Its pulling 360 Kh/s avg. So it works quite well.

Yeah, maybe he can change it to 4 decimal places, which would give us 1 decimal place when converting to kh/s. Of course this would cause a problem with Mh/s, which would then have to be truncated or rounded so as not to be unsightly . Ideally, cgminer should check the algo being used and return the appropriate precision. Maybe I'll post a request over in the cgminer thread. Thanks!

Not a problem - thank you.

Works now (yay!!!)

http://i.imgur.com/VCUmtSc.png


OK, not to be picky, but I guess the API is only returning 2 decimal places, which isn't an issue when dealing with Mh/s, but with Kh/s we lose a digit - any way to get more? Not really a big deal though, so don't worry about it too much. Thank you again for your work.

Thanks for your reply. It must be something on my end then. I notice the values below are in MHS, instead of KHS. Also to note I am not passing any values from CGWatcher in the arguments box, I'm using a conf. Shouldn't matter though, if the API is picking up the values.

Result from gpu|0:

We're getting closer but still no cigar. Units are now correct but value is still 0.

http://i.imgur.com/9PHGinn.png

Yes, these are from my conf, but your results may vary. You may want to start with intensity at 13 and work up.

I have an XFX 7850 Double D 1GB. I'm averaging 360Kh/s with the following:

No Overvolting
Core Clock: 1050
Mem Clock: 1300
Temperature: 62 C


Voltage seems to be locked. Voltage reported by GPU-z is 1.075. Memory won't go any higher without crashing. Thread concurrency won't go any higher without problems. I did the setx stuff, didn't make any difference whatsoever. I get high discard rates (50-200%), but low rejects and stales. Efficiency is around 140%.

Thank you for the new version. I just downloaded it, but its still showing "Current Total Hashrate" as 0 for Litecoin mining. The graph is flatlined.

Edit: The units on the graph are now correct in Kh/s, but the units on the top for current and avg are showing h/s. The value for avg is displaying correctly, but the value for current is 0.

I'm using cgminer 2.11.4, Full access.

http://i.imgur.com/Ri9A0JI.png

OK, I've had this happen quite often. In my case it was related to thread concurrency and intensity. These values can vary depending on which card you have. Post your card details and settings in a new topic, since we're off topic here. Or check the cgminer thread for the settings for your card.

What type of crash is it (black screen, windows popup error etc) ?

Is the fan running at a good RPM? How are the card temperatures? Is it overclocked? Were any of the power connectors disturbed while disconnecting/reconnecting the hard drive? Try reseating them. Also try reinstalling cgminer and the display drivers. Check over in the cgminer and maybe hardware sections and post this question there once you have enough posts.

Thanks for the replies. I think the card is getting plenty of power, although I may check that as a last resort.

I've played with the thread concurrency a bit:
8000 didn't make any difference.
4096 I get HW Errors.
Any higher than 8192 and I get black screen and the card crashes.

I did the setx stuff.

I'm trying yet another pool now to see if that will make any difference. Will report back.

Update: The other pool seems to be having server issues(wemineltc), so not good for testing right now. I'm still at a loss.

Thats great. Glad I could help. Here is the workaround if you want to log to a text file while using CGWatcher:
(Hopefully the author of CGWatcher can fix this in a future update)

Create a cgminer.cmd(or name it whatever) file with the following line in it:
cgminer.exe 2>logfile.txt

Then point CGWatcher to the cmd file instead of to the cgminer.exe in the Path box on the Settings tab.

I tested this by using a .conf file with the rest of the settings in it that I had in the cmd line previously - not sure if it will work if you have any arguments in the CGWatcher box, but they may propogate through.

Hope this helps.

Thanks for your reply. Here is more info:

OS: Windows 7 32bit
CPU: E8400 @ 3GHz
Vid Card: XFX 7850 DD 1GB
PSU:OCZ 600w ModXtreme Pro
Vid Driver: 13.1 catalyst
GPU Temp: 60 C at full load
Core Clock: 1045
Mem Clock: 1310
GPU Usage:99%
No Overvolting
Stock XFX Double D Cooler (2 fans, copper pipes etc)

I'm using CGWatcher 1.1.2 with cgminer 2.11.4 and I noticed that alot of my work is being thrown away. I have about 2x as many "Discarded" as "Accepted". I assume the discards are caused by the GPU not getting the work done fast enough before the server moves to the next block.

Here are some details:

I've tried a couple different pools, but still high discards.

Is this normal? Is there any way to minimize the number of discards? Would worksize make a difference? Is it a network latency issue? Anything suggestions for what I can tweak to improve this would be appreciated.

Edit Update: I switched to another pool and discard rate has dropped from 200% down to 60%, so I think a large part of the problem was the pool.

Edit Update 2: Saw over in the cgminer thread that discards are "meaningless" which is why it was removed from the cgminer screen. So I guess there isn't anything that can be done about it. I think its related to hash rate - seems the lower the hash the more discards. This would make sense because the miner can't keep up with the pool. The pool moves on to a new block while the miner is still working on the old one, so the work gets tossed, if I understand things correctly.

If the cmd window disappears right away, this is usually caused because an error occured. Double check the command line arguments. Make sure cgminer will run fine by itself first with the same command line - run it from a .cmd or .bat file or use the conf instead. Note that the command line argument to output to a log file(2>logfile.txt) will cause this with CGWatcher(I have a workaround for this - let me know if you need it and I will give details).

The permission issue might be related to the api-allow argument.