My question is... with 1.6Million RAC....... WHO PAYS THE ELECTRIC BILL?????
What's that work out to? An easy thousand + $$ a month??? Not really my business but a man thinks these things.

---

I pay the electric bill once a year because of being a solar customer. Coming due in May. Around $600 a month this past year after building two new hosts. And my solar production was way down this past year because of two summer months of forest fire smoke reducing generation. A very wet and cloudy winter this year too. Output is going to be about 1 MWh down in generation this year compared to my normal average yearly production.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

SOLAR! That's cheating! LOL. 1MW/hr loss is a lot. That's weeks of lost peak time on a 20KV system.
Winter here in Colorado Springs has been great. Luckily we didn't get much smoke from the fires to the South of us.

Once a week I maintain a MET Station for NextEra Energy out in Peyton. They are planning on a GW of solar production in addition to the several hundred wind turbines already installed on site.
Ranchers are making more money off the energy production than the cattle. Once the solar goes in, the cows will have to go elsewhere. So I assume my cheeseburger will go up in price. Bullish on beef.

---

My question is... with 1.6Million RAC....... WHO PAYS THE ELECTRIC BILL?????
What's that work out to? An easy thousand + $$ a month??? Not really my business but a man thinks these things.

it's almost not worth asking, because a person's bill will be down to what their rate is, which can vary wildly across the US, and the world.

Keith is in CA with rather high electric rates (even when offset by solar)
and me for example, on the east coast USA, my electric rate is about 11c/kWh.

my bill is in the range of $3-400/month (estimated because I don't really track the cost of the SETI machines exclusively)

and someone in China or other areas with incredibly cheap electric rates can pay next to nothing.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Winter here in Colorado Springs has been great. Luckily we didn't get much smoke from the fires to the South of us.

You all just had that cyclone bomb last week. Used to live there too. North side by South Gate of the Air Force academy. Worked at Memorial. Office on pikes peak by the old Osteopathic Hospital.

---

My Peak hours summer rates are about $0.50/kWh. Winter Peak hour rates are around $0.35/kWh. Off-Peak is around $0.13/kWh. That period is always the most costly because of no solar production. About to change back to summer rates shortly. Who knows what the rates will be this year. Just had another notice of rate increases before the PUC. See one in every bill. PG&E is in bankruptcy because of the fire damage liability and lawsuits. Don't even know if there will be a power company by that name in the future.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Ah, nice man. The Springs says hello!
Osteopathic - I had to look that word up. LOL
Yea we got a bomb all right. At 9AM it was raining, with zero snow on the ground, then the temp dropped about 10AM and started to snow. By 8:30PM, I had 27" in my backyard. Knocked down 40 feet of fencing. 97MPH wind gust at COS airport. There were 1200+ vehicles stranded and abandoned in the Springs. 400+ vehicles on Woodmen Rd alone. The National Guard was helping out.

The Facebook group 4xForce activated over 300 volunteers to go rescue the vehicles. I snatched 7 vehicles in total and called it quits. Stuff makes you hungry. Broke one of my foldable shovels on the ice.

https://gazette.com/news/more-than-drivers-stranded-in-colorado-springs-area-gov-polis/article_9056f158-458a-11e9-9974-53e3ebdd484a.html
https://gazette.com/news/stranded-falcon-woman-says-storm-went-from-virtually-nothing-to/article_02b9ad1c-468a-11e9-b659-3f82b993f932.html
https://denver.cbslocal.com/2019/03/15/colorado-march-blizzard-recap/

Falcon got nailed. The snow was 6' piled over the fences. You could walk up the snow embankment and jump down into people back yards.
If your ever back in the Springs area, hit me up. We can grab a bite to eat at Ted's Montana Grill.

---

They may be forced to declare bankruptcy so they can negotiate. Its definitely a major problem, and the customer is the one ultimately who's going to pay for it all.

---

I understand those variables. In China many areas are subsidized by the Gov., making it attractive for Coin Mining.
Norway is ending its subsidy causing the price to go to $0.019 kWh, up from approximately $0.0056/kWh
https://news.bitcoin.com/sweden-expects-to-attract-hoards-of-norwegian-bitcoin-miners-after-brutal-tax-hike/

I pay about the same as you, 11c/kWh. Industrial accounts in Colorado Springs pay 7.6c/kWh. Commercial is in the 9 ish range alt time I checked.

Some areas in the US enjoy special arrangements with power providers and miners have flocked to them as well.
https://www.coindesk.com/washington-states-chelan-county-moves-limit-new-bitcoin-mining-firms


>>> A better question would be: What is your power consumption to run Skynet Mr. Myers?

---

We can grab a bite to eat at Ted's Montana Grill.

I remember that place. Bison burgers..Disintegrated after the first bite, lol... No fat to hold the burger together. Ah that brings back memories. Last Bomb was '97 if I remember correctly. Glad you all were able to dig your selves out. Ok, back to the topic...

---

Indeed.

I deleted the computecache, uninstalled BOINC and reinstalled it. Then did lunatics installer. Added the app_config back in. Then added the SETI project to BOINC.

Seems the server gave me a different kind of WU app this time "opencl_nvidia_sah"

So I added this to the app_data file:



<app_name>setiathome_v8</app_name>
<plan_class>opencl_nvidia_sah</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.5</ngpus>
<cmdline>-sbs 1024 -high_perf -hp -high_prec_timer</cmdline>
</app_version>


I'm unsure of the cmdline though. Haven't tried to run a task yet.

---

post the first 30 lines of start up from event log.

sah was the precursor to SoG. SoG is still faster. Cmdline you used is ok.

---

So I began to run the tasks and they all go to 'postponed'. 1CPU + .5GPU (opencl_nvidia_sah)
Seems like that's a no go.


I did some OpenCL benchmarks with Luxmark before installing BOINC to see if there was an OpenCL issue. Mic, Ball, Hotel CPU+GPU and GPU only. Scores were right where they should be.

#13 in top results: http://luxmark.info/top_results/Microphone/OpenCL/GPU/1

---

In file stderr:

19:36:22 (15044): Can't open init data file - running in standalone mode
19:36:22 (15044): Can't open init data file - running in standalone mode
Not using mb_cmdline.txt-file, using commandline options.
Priority of worker thread raised successfully
Priority of process adjusted successfully, below normal priority class used
19:36:22 (15044): Can't open init data file - running in standalone mode
WARNING: init_data.xml missing
Error: Getting Platforms. (clGetPlatformsIDs)
BOINC assigns device 0
WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities
ERROR: OpenCL kernel/call 'clGetDeviceIDs (second call)' call failed (-32) in file ..\..\..\src\GPU_lock.cpp near line 1306.
Waiting 30 sec before restart...


stdoutdae:
21-Mar-2019 19:22:45 [---] cc_config.xml not found - using defaults
21-Mar-2019 19:22:45 [---] Starting BOINC client version 7.14.2 for windows_x86_64
21-Mar-2019 19:22:45 [---] log flags: file_xfer, sched_ops, task
21-Mar-2019 19:22:45 [---] Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
21-Mar-2019 19:22:45 [---] Data directory: D:\BOINC DATA
21-Mar-2019 19:22:45 [---] Running under account Userper
21-Mar-2019 19:22:45 [---] CUDA: NVIDIA GPU 0: GeForce RTX 2080 Ti (driver version 419.35, CUDA version 10.1, compute capability 7.5, 4096MB, 3541MB available, 15276 GFLOPS peak)
21-Mar-2019 19:22:45 [---] OpenCL: NVIDIA GPU 0: GeForce RTX 2080 Ti (driver version 419.35, device version OpenCL 1.2 CUDA, 11264MB, 3541MB available, 15276 GFLOPS peak)
21-Mar-2019 19:22:45 [---] Creating new client state file
21-Mar-2019 19:22:45 [---] Host name: DESKTOP-H9KJF5Q
21-Mar-2019 19:22:45 [---] Processor: 12 GenuineIntel Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz [Family 6 Model 63 Stepping 2]
21-Mar-2019 19:22:45 [---] Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 dca pbe fsgsbase bmi1 smep bmi2
21-Mar-2019 19:22:45 [---] OS: Microsoft Windows 10: Core x64 Edition, (10.00.17763.00)
21-Mar-2019 19:22:45 [---] Memory: 15.91 GB physical, 22.16 GB virtual
21-Mar-2019 19:22:45 [---] Disk: 838.34 GB total, 549.42 GB free
21-Mar-2019 19:22:45 [---] Local time is UTC -6 hours
21-Mar-2019 19:22:45 [---] No WSL found.
21-Mar-2019 19:22:45 [---] Last benchmark was 17977 days 01:22:45 ago
21-Mar-2019 19:22:50 [---] No general preferences found - using defaults
21-Mar-2019 19:22:50 [---] Preferences:
21-Mar-2019 19:22:50 [---] max memory usage when active: 8147.35 MB
21-Mar-2019 19:22:50 [---] max memory usage when idle: 14665.23 MB
21-Mar-2019 19:22:50 [---] max disk usage: 549.32 GB
21-Mar-2019 19:22:50 [---] don't use GPU while active
21-Mar-2019 19:22:50 [---] suspend work if non-BOINC CPU load exceeds 25%
21-Mar-2019 19:22:50 [---] (to change preferences, visit a project web site or select Preferences in the Manager)
21-Mar-2019 19:22:50 [---] Setting up project and slot directories
21-Mar-2019 19:22:50 [---] Checking active tasks
21-Mar-2019 19:22:50 [---] Setting up GUI RPC socket
21-Mar-2019 19:22:50 [---] Checking presence of 0 project files
21-Mar-2019 19:22:50 [---] This computer is not attached to any projects
21-Mar-2019 19:22:50 Initialization completed
21-Mar-2019 19:22:50 [---] Suspending GPU computation - computer is in use
21-Mar-2019 19:23:00 [---] Suspending computation - user request
21-Mar-2019 19:23:21 [---] Running CPU benchmarks
21-Mar-2019 19:23:52 [---] Benchmark results:
21-Mar-2019 19:23:52 [---] Number of CPUs: 12
21-Mar-2019 19:23:52 [---] 4564 floating point MIPS (Whetstone) per CPU
21-Mar-2019 19:23:52 [---] 18161 integer MIPS (Dhrystone) per CPU
21-Mar-2019 19:24:03 [---] Fetching configuration file from http://setiathome.berkeley.edu/get_project_config.php
21-Mar-2019 19:24:23 [---] Fetching configuration file from http://setiathome.berkeley.edu/get_project_config.php
21-Mar-2019 19:24:41 [SETI@home] work fetch suspended by user
21-Mar-2019 19:24:41 [SETI@home] Master file download succeeded




Restarted BOINC:

21-Mar-2019 19:49:06 [---] cc_config.xml not found - using defaults
21-Mar-2019 19:49:06 [---] Starting BOINC client version 7.14.2 for windows_x86_64
21-Mar-2019 19:49:06 [---] log flags: file_xfer, sched_ops, task
21-Mar-2019 19:49:06 [---] Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
21-Mar-2019 19:49:06 [---] Data directory: D:\BOINC DATA
21-Mar-2019 19:49:06 [---] Running under account Userper
21-Mar-2019 19:49:07 [---] CUDA: NVIDIA GPU 0: GeForce RTX 2080 Ti (driver version 419.35, CUDA version 10.1, compute capability 7.5, 4096MB, 3541MB available, 15276 GFLOPS peak)
21-Mar-2019 19:49:07 [---] OpenCL: NVIDIA GPU 0: GeForce RTX 2080 Ti (driver version 419.35, device version OpenCL 1.2 CUDA, 11264MB, 3541MB available, 15276 GFLOPS peak)
21-Mar-2019 19:49:07 [SETI@home] Found app_info.xml; using anonymous platform
21-Mar-2019 19:49:07 [---] Host name: DESKTOP-H9KJF5Q
21-Mar-2019 19:49:07 [---] Processor: 12 GenuineIntel Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz [Family 6 Model 63 Stepping 2]
21-Mar-2019 19:49:07 [---] Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 dca pbe fsgsbase bmi1 smep bmi2
21-Mar-2019 19:49:07 [---] OS: Microsoft Windows 10: Core x64 Edition, (10.00.17763.00)
21-Mar-2019 19:49:07 [---] Memory: 15.91 GB physical, 22.16 GB virtual
21-Mar-2019 19:49:07 [---] Disk: 838.34 GB total, 549.37 GB free
21-Mar-2019 19:49:07 [---] Local time is UTC -6 hours
21-Mar-2019 19:49:07 [---] No WSL found.
21-Mar-2019 19:49:07 [SETI@home] Found app_config.xml
21-Mar-2019 19:49:07 [SETI@home] URL http://setiathome.berkeley.edu/; Computer ID 8683082; resource share 100
21-Mar-2019 19:49:07 [SETI@home] General prefs: from SETI@home (last modified 20-Mar-2019 16:35:01)
21-Mar-2019 19:49:07 [SETI@home] Computer location: home
21-Mar-2019 19:49:07 [---] General prefs: using separate prefs for home
21-Mar-2019 19:49:07 [---] Reading preferences override file
21-Mar-2019 19:49:07 [---] Preferences:
21-Mar-2019 19:49:07 [---] max memory usage when active: 14665.23 MB
21-Mar-2019 19:49:07 [---] max memory usage when idle: 15479.97 MB
21-Mar-2019 19:49:07 [---] max disk usage: 20.00 GB
21-Mar-2019 19:49:07 [---] don't use GPU while active
21-Mar-2019 19:49:07 [---] (to change preferences, visit a project web site or select Preferences in the Manager)
21-Mar-2019 19:49:07 [---] Setting up project and slot directories
21-Mar-2019 19:49:07 [---] Checking active tasks
21-Mar-2019 19:49:07 [---] Setting up GUI RPC socket
21-Mar-2019 19:49:07 [---] Checking presence of 140 project files
21-Mar-2019 19:49:07 Initialization completed
21-Mar-2019 19:49:07 [---] Suspending computation - user request

---

Change your settings so that it is always crunching and see if that resolves it. You didn't install this as a service did you? Boinc only allows GPU to run in Windows as a single sign on. I'm throwing out ideas....

---

If you used the Lunatics installer, did you select the SoG app in the Nvidia gpu app panel? If you did and you still are getting nvidia_sah tasks, BOINC is still not believing you have a OpenCL device to the contrary of what your Event Log BOINC startup shows.

21-Mar-2019 19:22:45 [---] CUDA: NVIDIA GPU 0: GeForce RTX 2080 Ti (driver version 419.35, CUDA version 10.1, compute capability 7.5, 4096MB, 3541MB available, 15276 GFLOPS peak)
21-Mar-2019 19:22:45 [---] OpenCL: NVIDIA GPU 0: GeForce RTX 2080 Ti (driver version 419.35, device version OpenCL 1.2 CUDA, 11264MB, 3541MB available, 15276 GFLOPS peak)

Error: Getting Platforms. (clGetPlatformsIDs)
BOINC assigns device 0
WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities
ERROR: OpenCL kernel/call 'clGetDeviceIDs (second call)' call failed (-32) in file ..\..\..\src\GPU_lock.cpp near line 1306

Instead of relying on graphics benchmark products to identify OpenCL drivers, why don't you download the clinfo utility from Seti. Unzip it and run it in a Terminal. It will print out the detected graphics drivers in your system. Both the Nvidia driver and the Nvidia OpenCL components.

https://boinc.berkeley.edu/dl/clinfo.zip

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I changed it on the site. Usually I toggle "run always" and "suspend" when I want BOINC to stop. This is a work PC, so I can't let it run all the time like a dedicated crunching PC.

I installed the BOINC software from the installer without virtualbox. I'm not sure about the 'as a service' you speak of. The PC ran CUDA 50/42 (just not well) and my old AMD card ran openCL for months.

Any news about a possible update to CUDA stuff? Think the last one was Jan 2016 if I read the Lunatics readme file correctly? Mentions CUDA 5.0 - Now its CUDA 7.5 with the RTX cards.

---

How can I convince it otherwise? Slip a $20 bill in the CD ROM tray? LOL?

Yes I did select SoG (last option) during installation.

---

Your installation of the Nvidia drivers is still not putting the OpenCL components of the driver into your system correctly for some reason making them inaccessible to BOINC.

No, the CUDA50 application is the last and latest official Windows CUDA application the project provides. The SoG OpenCL is the fastest official Windows Nvidia application the project provides.

The CUDA 6 and CUDA 9 and CUDA 10 applications are Linux applications only. The Linux CUDA 6 application on the Seti website is the last and latest official CUDA application provided.

The CUDA 9 and CUDA 10 applications are only available from third party sites and developers and are not official applications provided by Seti. They are the fastest current applications for Nvidia now. Only available for Linux hosts.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

https://boinc.mundayweb.com/wiki/index.php?title=Determine_OpenCL_capability_of_GPU_and_CPU

trying this

---