Posts by Sergei Chernykh

1) Message boards : Bug tracker : Amicable pairs counter defect? (Message 2080)
Posted 24 days ago by Sergei Chernykh
Post:
We're searching through numbers of the form "X*P" where P is a very large prime number. There are not many amicable pairs of this form. In fact, I estimate that there are only 300-400 amicable pairs left to discover in the current search.
2) Message boards : Number crunching : Android support? (Message 2072)
Posted 16 May 2025 by Sergei Chernykh
Post:
Amicable use 32-bit integers

64-bit integers for the most part, because we're dealing with 21-, 22-digit numbers already.
3) Message boards : Bug tracker : Issue with arm64-apple-darwin (Message 2056)
Posted 9 Apr 2025 by Sergei Chernykh
Post:
Your other Macs run x86_64-apple-darwin version, so I guess they use Rosetta. This is not an optimal way, but it's the best for Apple Silicon for now on this project. I need to finally add support for arm64-apple-darwin platform when I have more free time.
4) Message boards : News : WU size and credits are doubled now (Message 2049)
Posted 3 Apr 2025 by Sergei Chernykh
Post:
No, task sizes are the same now. Maybe your GPU or the room it's in got more warm?
5) Message boards : Number crunching : FYI. Some credits didn't post correctly to the consolidators (Message 2039)
Posted 5 Mar 2025 by Sergei Chernykh
Post:
Credits are given out when two users finish the same task, so it depends on who you got paired with for each task.
6) Message boards : News : WU size and credits are doubled now (Message 2032)
Posted 25 Feb 2025 by Sergei Chernykh
Post:
All these tasks have the same amount of work, so the same amount of points. I don't know why your GPU processed them slower - maybe something else was using the GPU?
7) Message boards : Number crunching : Intel Arc B580 (Message 2028)
Posted 20 Feb 2025 by Sergei Chernykh
Post:
I've updated the regex, it's 
Intel\(R\) (Arc|Graphics)
now. Your GPU should start getting tasks now.
8) Message boards : Number crunching : Intel Arc B580 (Message 2026)
Posted 19 Feb 2025 by Sergei Chernykh
Post:
Not "they", it's me. I set 
<host_summary_regex>Arc</host_summary_regex>
in the project's settings. If you can suggest a better regex that covers all Intel Arc GPUs, I'll be thankful.
9) Message boards : Number crunching : Intel Arc B580 (Message 2024)
Posted 19 Feb 2025 by Sergei Chernykh
Post:
The project limits Intel GPUs to only ones that have "Arc" in their host summary text (i.e. Intel Arc GPUs). This is done to avoid integrated Intel GPUs as they're almost useless. If you can fix your host's description to include "Arc" in GPU name, it will start receiving tasks.
10) Message boards : Number crunching : How do I increase number of pending downloadable GPU tasks for BOINC? (Message 2013)
Posted 4 Feb 2025 by Sergei Chernykh
Post:
This is the server-side limitation. I put max 4 tasks per GPU in the config long time ago. There are different BOINC competitions that run on Amicable Numbers from time to time, and this is to prevent people from fetching a lot of tasks before the competition starts to submit them and get credits as soon as it starts.
11) Message boards : Number crunching : Nvidia RTX 50-seris - estimated Amicable Numbers performance? (Message 2005)
Posted 27 Jan 2025 by Sergei Chernykh
Post:
Then I looked at 4060 Ti which has an average of 440.5 sec but mine consistently completes < 287 or almost a 4070 Ti. I wish they had four columns on the GPU page: linux, win 11, win 10 and "other".

It also depends on how many parallel tasks per GPU you run. You should check "tasks per 24 hours" metric instead.
12) Message boards : Bug tracker : Computing - CPU models - Apple Silicon? (Message 1999)
Posted 19 Jan 2025 by Sergei Chernykh
Post:
Apple Silicon CPU/GPU applications are not available officially, but it's possible to build it manually and run it under the BOINC's "anonymous platform". Some people do.

P.S. You can download the unofficial Apple Silicon GPU application from https://github.com/SChernykh-team/actions/actions/runs/11844297083 (scroll down for the build artifacts) - I tried to build it to check if it's possible, and it built fine and ran fine.
13) Message boards : Random stuff : website dark mode and donation (Message 1982)
Posted 29 Dec 2024 by Sergei Chernykh
Post:
There is a "Dark Reader" extension for Chrome that can fix it for all websites that you visit.

This project is self-funded, and runs on one server so I don't need to spend a lot of money on it and there is no need for donations.
14) Message boards : Bug tracker : OpenCL_Intel_GPU not giving work, application not found (Message 1981)
Posted 29 Dec 2024 by Sergei Chernykh
Post:
From the project config:
	<plan_class>
		<name>opencl_intel_gpu</name>
		<opencl/>
		<gpu_type>intel</gpu_type>
		<host_summary_regex>Arc</host_summary_regex>
		<min_gpu_ram_mb>2600</min_gpu_ram_mb>
		<gpu_ram_used_mb>2600</gpu_ram_used_mb>
		<cpu_frac>0.0001</cpu_frac>
	</plan_class>

So it should be an Intel Arc GPU - if there is no "Arc" in the host description, it will not work.
15) Message boards : Number crunching : Apple M-series CPU/GPU Applications (Message 1970)
Posted 15 Nov 2024 by Sergei Chernykh
Post:
Technically it's possible to build the binaries for the arm64-apple-macos-11 compiler target. But first I need to figure out how to add arm64 macOS platform to the project because it didn't even exist when this project started.
16) Message boards : News : WU size and credits are doubled now (Message 1967)
Posted 9 Nov 2024 by Sergei Chernykh
Post:
I've increased the deadline for new work units to 1 week. It is enough for all modern CPUs, if you run BOINC on more than 1 CPU core.
17) Message boards : Random stuff : MacOS support? (Message 1964)
Posted 5 Nov 2024 by Sergei Chernykh
Post:
Yes, see the list of supported platforms: https://sech.me/boinc/Amicable/apps.php
18) Message boards : Bug tracker : No GPU work for AMD/ATI GPU since 09:00 UTC (Message 1961)
Posted 29 Oct 2024 by Sergei Chernykh
Post:
You can check Computing -> Server status page: https://sech.me/boinc/Amicable/server_status.php

It says 3235 tasks ready to send right now.

Edit: you already have 4 in-progress tasks, and the project doesn't let you to have more (per GPU).
19) Message boards : News : WU size and credits are doubled now (Message 1955)
Posted 28 Oct 2024 by Sergei Chernykh
Post:
No, GPU and CPU RAM requirements stay the same.
20) Message boards : News : WU size and credits are doubled now (Message 1953)
Posted 28 Oct 2024 by Sergei Chernykh
Post:
First of all, thanks to everyone participating in the project!

I've noticed today that solved work units come in so fast that the server can't keep up. I had to double the size of each work unit (together with the credit for it) to reduce the load on the server.

Keep crunching! Thanks again!
21) Message boards : Bug tracker : Workunits that are stuck but both computers finished the task 2024/10/26 (Message 1951)
Posted 28 Oct 2024 by Sergei Chernykh
Post:
Update: someone added a lot of computing power to the project, so WUs are piling up right now - but all scripts are working. I'm keeping an eye on the server in the meantime.
22) Message boards : Number crunching : GPU version: kernel size tuning and less UI lags (Message 1950)
Posted 28 Oct 2024 by Sergei Chernykh
Post:
If you're not getting full GPU utilization on RTX 4090, it's better to run several WUs per GPU in parallel than to increase the kernel size above 23.
23) Message boards : Bug tracker : Workunits that are stuck but both computers finished the task 2024/10/26 (Message 1947)
Posted 26 Oct 2024 by Sergei Chernykh
Post:
I've restarted all services to be sure that they're not stuck.
24) Message boards : Bug tracker : Workunits that are stuck but both computers finished the task 2024/10/26 (Message 1946)
Posted 26 Oct 2024 by Sergei Chernykh
Post:
No, it's not stuck. Just every time I'm running some HDD-intensive tasks on the server, validation starts lagging. I stopped these tasks now.
25) Message boards : News : Maintenance on 2024-11-07 (7th of November) (Message 1943)
Posted 24 Oct 2024 by Sergei Chernykh
Post:
Project server will be unavailable on 2024-11-07 (7th of November), from 03:30 AM UTC to 05:30 AM UTC due to the maintenance in a Hetzner datacenter where the server is hosted.
26) Message boards : Random stuff : Russia and Ukraine (Message 1942)
Posted 22 Oct 2024 by Sergei Chernykh
Post:
No politics here. If it puts your mind at ease, I'm the only scientist/developer on Amicable Numbers, and I live in Sweden.
27) Message boards : Bug tracker : Workunits that are stuck but both computers finished the task 2024/10/07 (Message 1937)
Posted 8 Oct 2024 by Sergei Chernykh
Post:
I've added a script to unstuck such WUs. They shouldn't be stuck for more than an hour now.
28) Message boards : Bug tracker : Workunits that are stuck but both computers finished the task 2024/10/07 (Message 1936)
Posted 8 Oct 2024 by Sergei Chernykh
Post:
I've fixed it. Normally it shouldn't happen so there is no need for such script. I was restarting services recently, maybe this is why some WUs got stuck.
29) Message boards : Bug tracker : No GPU work for AMD/ATI GPU since 09:00 UTC (Message 1933)
Posted 6 Oct 2024 by Sergei Chernykh
Post:
I've restarted all services. Let me know if it solved this issue.
30) Message boards : Bug tracker : Workunits that are stuck but both computers finished the task 2024/10/06 (Message 1928)
Posted 6 Oct 2024 by Sergei Chernykh
Post:
Thanks for the report. Fixed.
31) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1921)
Posted 4 Jun 2024 by Sergei Chernykh
Post:
Parameters are "/lpr 127658160073 127658185709 /task_size 200102005550" and they set the range of primes to check, and the total amount of checks to perform (for progress reporting to BOINC client).
32) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1913)
Posted 22 May 2024 by Sergei Chernykh
Post:
Yes, copy the .exe file to a separate folder, and run it from command prompt there.
33) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1909)
Posted 19 May 2024 by Sergei Chernykh
Post:
For that specific WU, just run
Amicable_OpenCL_v_3_09.exe /lpr 127658160073 127658185709 /task_size 200102005550

In a separate folder, then check the output. One of the files will have start/stop timestamps.
34) Message boards : Number crunching : No new amicable pairs (Message 1908)
Posted 19 May 2024 by Sergei Chernykh
Post:
It is expected because part 3 of the search is sifting through numbers with large prime factors in them (such amicable numbers are rare).
35) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1905)
Posted 19 May 2024 by Sergei Chernykh
Post:
Can you try to run the same WU again, manually? And without anything else running on the PC. I think it was a random fluctuation not related to the WU itself.
36) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1894)
Posted 14 Apr 2024 by Sergei Chernykh
Post:
There are 2 pages for preferences, please check both:
https://sech.me/boinc/Amicable/prefs.php?subset=global
https://sech.me/boinc/Amicable/prefs.php?subset=project
37) Message boards : Number crunching : Long running CPU tasks are all timing out. (Message 1891)
Posted 13 Apr 2024 by Sergei Chernykh
Post:
A single-core ARM cloud VM is too weak for this. These WUs were tuned to be not too fast on the fastest GPUs (2 minutes per WU on RTX 4090), and not too slow on Raspberry Pi 4.

Quad-core cloud VM can finish WUs in less than 5 days: https://sech.me/boinc/Amicable/result.php?resultid=76421032
38) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1889)
Posted 12 Apr 2024 by Sergei Chernykh
Post:
macOS OpenCL application works, see for example this task.
39) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1886)
Posted 11 Apr 2024 by Sergei Chernykh
Post:
No, it should've been much faster. Your GPU can complete WUs in ~3 minutes, as you can see from the list of completed WUs, and all part 3 WUs are the same for GPUs. Double check that nothing else is using GPU and that you run 1 WU at a time.
40) Message boards : News : Part 3 of the 10^21 search has started (Message 1884)
Posted 9 Apr 2024 by Sergei Chernykh
Post:
All old v3.08 WUs need to be re-run (each WU missed ~12.5% of amicable pairs), but earned credit will remain.
41) Message boards : News : Part 3 of the 10^21 search has started (Message 1882)
Posted 9 Apr 2024 by Sergei Chernykh
Post:
I found a bug in OpenCL applications, which resulted in many false negatives (around 12.5%). The bug was in the new code that I added for the part 3, so the previous search is unaffected. Version 3.09 has this bug fixed, but I had to restart the part 3. Luckily, we were only a couple of weeks in.
42) Message boards : Number crunching : RAC dropping (Message 1879)
Posted 3 Apr 2024 by Sergei Chernykh
Post:
Part 3 WUs are much faster on GPUs compared to CPUs. I might launch the CPU-only application to search for the remaining amicable pairs (m=m1*p where p > 1014) before the part 3 finishes to give some meaningful work to CPUs. But it's not ready yet because I still need to find a good way to parallelize it.
43) Message boards : Bug tracker : v3.08 (opencl_nvidia) x86_64-pc-linux-gnu Error while computing (Message 1868)
Posted 28 Mar 2024 by Sergei Chernykh
Post:
Error code -4 means not enough GPU memory. Try to run only 1 WU at a time. You can also try to reduce kernel size to 21, it will reduce memory usage.
44) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1865)
Posted 27 Mar 2024 by Sergei Chernykh
Post:
Edit: I re-read your post and you wrote "4 threads per WU". This is a multi-threaded application, it can use all CPU cores - you don't need to run many WUs in parallel. I tuned part 3 WUs to take ~2 hours on a 16-core Ryzen 7950X (using all 32 threads). Please note that GPUs are much faster with part 3 WUs (less than 2 minutes on RTX 4090), so CPUs are expected to be slow.
45) Message boards : News : Part 3 of the 10^21 search has started (Message 1862)
Posted 27 Mar 2024 by Sergei Chernykh
Post:
Part 2 of the 1021 search will be finished, for the most part, today. There are still ~2000 unfinished WUs which can take up to a couple of weeks to complete, depending on individual participants processing them.

Part 3 of the search will look for amicable pairs where the smaller member of the pair has its largest prime factor between 1011 and 1014:

m=m1*p, 1011 < p < 1014

There are too many primes larger than 1014 (all primes up to 5*1019 need to be checked), so they will require a different approach and a new application when the time comes.

Most of the participants have already started getting part 3 WUs. The new work units will require ~2.5-2.7 GB RAM on CPU, and the same amount of RAM on GPU (for GPU applications), so the minimum requirement for GPUs will be 3 GB RAM.

Update April 9th, 2024: I found a bug in OpenCL applications, which resulted in many false negatives (around 12.5%). The bug was in the new code that I added for the part 3, so the previous search is unaffected. Version 3.09 has this bug fixed, but I had to restart the part 3. Luckily, we were only a couple of weeks in.
46) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1857)
Posted 22 Mar 2024 by Sergei Chernykh
Post:
Which WUs, can you give a link? Some WUs are longer than others, sometimes many times longer. Thankfully, part 3 WUs are much more uniform, and it will start in about a week from now.
47) Message boards : News : Beta test of part 3 of the 10^21 search (Message 1855)
Posted 22 Mar 2024 by Sergei Chernykh
Post:
Credit per task will remain the same but tasks will be smaller than we have right now.
48) Message boards : Number crunching : Intel ARC support ? (Message 1843)
Posted 3 Mar 2024 by Sergei Chernykh
Post:
I've updated the GPU application to v3.08 which should have this error -30 bug fixed. But I don't guarantee that there are no other bugs left.
49) Message boards : Number crunching : Intel ARC support ? (Message 1841)
Posted 3 Mar 2024 by Sergei Chernykh
Post:
System restarted and it processed a couple of tasks, but in error


This is a bug in the code, I'll fix it soon. If it doesn't work after this, I'll get back to it tomorrow.
50) Message boards : Number crunching : Intel ARC support ? (Message 1839)
Posted 3 Mar 2024 by Sergei Chernykh
Post:
I've checked in the BOINC's code how "GPUs not usable" error can happen, and it happens when you use remote desktop to login to a computer, or if you switch between different logged in users on the same computer.


Next 50


©2025 Sergei Chernykh