Posts by Ian&Steve C.
|
1)
(Message 8300)
Posted 13 Mar 2024 by Ian&Steve C. Post: resource share is handled by BOINC client. it is only set at the project level. once set, the client does all the calculation for which projects to run and when and overall how that resource share setting is actually applied. that said, BOINC doesnt always do a great job, and there are many variables. resource share of 10% doesnt mean that it will use 10% of your cpu utilization. is means 10% of "work done without missing deadlines" or something along those lines. with wildly varying deadlines, computational size, and other knobs to turn on BOINC, it often acts strangely or not as desired. over a long term it should "even out" but it could take a while. if you havent run one project for a while, it will prioritize that project for a bit to bring it's REC value back to where it thinks it should be. you can force BOINC to not run more than X amount of tasks for any one project by setting an app_config.xml file and adding the <project_max_concurrent>X</project_max_concurrent> statement to it. see here for specific usage/configuration: https://boinc.berkeley.edu/wiki/Client_configuration |
|
2)
(Message 8277)
Posted 7 Mar 2024 by Ian&Steve C. Post: It’s how much of a single CPU core that the GPU application is projected to use while the GPU app runs. So 0.01 means essentially that 1% of a single CPU core is needed to run the GPU task. Theres always some element of CPU computation needed for any GPU task. Some applications require more CPU than others. This is also not in any way determining the CPU use. Changing this value does not actually change what the GPU applications will do. This value is only there for the purposes of BOINC to figure out how much systems resources are used in total. It needs to know this so it can best determine how many other tasks to run on the CPU. For example if you had four GPU tasks running, and each one needed 0.5 CPU, that tells BOINC that running these four tasks will need two CPU threads for support. If you have an 8-thread CPU, and you are allowing the CPU to work 100%, then BOINC would run 6 more CPU tasks. |
|
3)
(Message 8230)
Posted 21 Jan 2024 by Ian&Steve C. Post: Running the CPU aps is burning up my CPUs. The sse2 and sse3 aps are the most energy efficient and run cooler. There's no way to allow one to control which of these dangerous aps gets DLed. Best way is to mitigate the problem is eliminate the cause. Run this command after starting BOINC: just run anonymous platform like you did on TN-Grid. add only the sse3 app if thats what you want to run. calling the app "dangerous" is quite hyperbolic. if you have temp problems, look into your cooling solution. |
|
4)
(Message 8221)
Posted 18 Jan 2024 by Ian&Steve C. Post: when the task is running. what does BOINC say about the resource allocation? usually it's something like "0.993 CPUs + 1 NVIDIA GPU" or "1 CPUs + 0.5 NVIDIA GPU" or something like that. can you report back the specific values listed here for this system? |
|
5)
(Message 8211)
Posted 16 Jan 2024 by Ian&Steve C. Post: post the entire cc_config file so it can be inspected for typos or other mistakes. i assume these entries are in the <options> section. did you restart BOINC? changes to cc_config often require a full BOINC recycle. not just a re-read config files. edit, barring any other typos or mistakes, the likely problem is that you've got the master url for the project wrong. the proper URL is: https://asteroidsathome.net/boinc/ |
|
6)
(Message 8208)
Posted 15 Jan 2024 by Ian&Steve C. Post: it's some problem with your boinc configuration. the app itself is not possible to behave the way you are describing. it only runs once. if you have multiple copies running it's because BOINC told it to. remove the asteroids project completely from BOINC, and re-add it. |
|
7)
(Message 8205)
Posted 14 Jan 2024 by Ian&Steve C. Post: it happens about once a month or so. the tasks run down to 0, and it takes a day or two for the Admins to load up more work. then they load a large amount of work again that lasts another month or so. |
|
8)
(Message 8201)
Posted 13 Jan 2024 by Ian&Steve C. Post: still sounds like something wrong with the BOINC configuration. since it's BOINC that controls how many tasks run at a time, not the application itself. you might want to actually inspect the project folder to make sure there isnt some app_config file that you're not aware of. if there is indeed no app_config, then I would uninstall BOINC and reinstall the latest version you can. |
|
9)
(Message 8198)
Posted 13 Jan 2024 by Ian&Steve C. Post: sounds like something wrong with your boinc configuration. did you put 0.01 in the gpu_usage section of your app_config by mistake? |
|
10)
(Message 8191)
Posted 1 Jan 2024 by Ian&Steve C. Post: I’m not sure the value in reposting information that I’ve already linked to and mentioned. Intel GPUs at this time largely don’t support FP64. And the workarounds that do exist are very slow and only work in select few environments. This project relies heavily on FP64 calculations. Intel GPU is not viable at this time. That could change if future generations of GPUs have hardware FP64 support, but for right now it’s basically a non-starter. |
|
11)
(Message 8188)
Posted 1 Jan 2024 by Ian&Steve C. Post: Lack of hardware FP64 is the biggest issue with Asteroids. FP64 is required here. |
|
12)
(Message 8169)
Posted 11 Dec 2023 by Ian&Steve C. Post: Please provide detailed 'How To" instructions for making this switch to the Anonymous platform. your i7-4770k does not support this application. it does not have the AVX512 hardware. only AMD zen4 (Ryzen 7000, EPYC 9000/8000), Intel Core-X i9-79xx/99xx/109xx, and Intel Xeon Scalable processors have the required AVX512 extensions to run this app. |
|
13)
(Message 8163)
Posted 2 Dec 2023 by Ian&Steve C. Post: Slight addendum. Intel iGPUs only support FP64 on 6th/7th/8th/9th gen. Devices before or after this won’t work. And not until they get hardware FP64 in future generations of their dedicated GPUs (if they do), current gen Arc cards still lack FP64 at all. https://community.intel.com/t5/Graphics/What-processors-have-an-internal-Gen11-GPU-with-FP64-double/m-p/691214 There also exists a driver workaround for Linux systems to emulate FP64 on their newer Linux driver for the newer devices that don’t natively support FP64, but the same thing is not supported in Windows, and it’s painfully slow anyway. |
|
14)
(Message 8162)
Posted 2 Dec 2023 by Ian&Steve C. Post: the Asteroids code makes heavy use of FP64 double precision and Intel GPUs lack hardware FP64 support. So it likely won’t happen until Intel starts making GPUs with FP64. |
|
15)
(Message 8158)
Posted 29 Nov 2023 by Ian&Steve C. Post: 348K RTS now Add another 0 ;) |
|
16)
(Message 8128)
Posted 8 Nov 2023 by Ian&Steve C. Post: it's probably fairly normal. the GPU app isnt very optimized here. also note that CPUs are more efficient for the tasks here. for the power used, and CPU will get more work done overall. |
|
17)
(Message 8114)
Posted 17 Oct 2023 by Ian&Steve C. Post: Hello, I encountered technical problems. The tasks work up to 100%, then an error message appears at 100%, then no points are awarded and the tasks disappear. New ones are loaded and everything goes in circles - they execute 100% with an error and disappear. I completed a lot of tasks, but never received any points. Correct it please. Here is screen shot is. https://www.facebook.com/photo?fbid=122149061618002019&set=pcb.122149061648002019[/img] your image doesnt work. try loading it to somewhere else, like imgur this is also not really an appropriate thread or even subforum for your question. not sure how you ended up here. this subforum is intended for technical news from the project. tasks going to 100% when they fail is normal. tasks "disappearing" after they are reported is normal (this is how BOINC works). if you look at your task results on the website (same for every BOINC project) you can easily see your errors. they all have the same reason for error. https://asteroidsathome.net/boinc/result.php?resultid=408203001 couldn't start app: CreateProcess() failed - Your organization used Device Guard to block this app. Contact your support person for more info. seems you might be using a work computer to crunch or something and your work is blocking the tasks from running. talk to whoever your sys admin is. |
|
18)
(Message 8060)
Posted 21 Sep 2023 by Ian&Steve C. Post: Radeon VII: 3 validation pending i see your RVII system. what's the power draw when running asteroids? and are you running 2x? or just one at a time? |
|
19)
(Message 8051)
Posted 20 Sep 2023 by Ian&Steve C. Post: Any plans for NVIDIA? must be thinking of Milkyway. Asteroids has always had CUDA apps for Nvidia. but the OpenCL apps for AMD are what's new here. |
|
20)
(Message 8017)
Posted 18 Sep 2023 by Ian&Steve C. Post: I'm not sure how much VRAM is being used by the AMD app here, but very possible 2GB isn't enough. and yes it will be VERY slow.So is my phone but I run Asteroids on it because it's there. If old stuff works, I use it. Why throw it away? Run it into the ground. the RAM use by an Arm app has nothing to do with the VRAM use by a GPU app. apples to oranges. |
Next 20