Posts by Tigers Dave

1) Message boards : Projects : BOINC Project Collatz Conjecture is FAKE? (Message 110980) Posted 30 Jan 2023 by Tigers Dave Post: If it were up to me, I would give Slicker until the 1-year anniversary (early June 2023) before removing Collatz from the project list.
2) Message boards : Projects : BOINC Project Collatz Conjecture is FAKE? (Message 109880) Posted 27 Sep 2022 by Tigers Dave Post: Hi Ian&Steve C! Could you please provide us with a gist of the argument that the methodology was flawed? Or, could you please provide me with a link to that argument? I have had a number of exchanges over the years on this topic and I remain unconvinced that the methodology was flawed. poppinfresh99 has posted several times his concern that the algorithm employs a sieve and thus the algorithm fails to analyze all possible candidate exceptions to the Collatz Conjecture. For an example of that argument, see this thread: https://boinc.berkeley.edu/forum_thread.php?id=14159. This particular argument from poppinfresh99 is flawed because any even candidate number N can be reduced to N/2 and N/2 was analyzed approximately 18 months ago. Thus, the project does not even request that any even candidate number N be analyzed. Furthermore, I postulate, but cannot confirm (because I have not seen the source code), that once any candidate number N has been reduced to less than N/2, analysis of that number can cease, as the result can be determined using previously-determined results. My hypothesis is supported by the observation that Collatz work units have not been taking longer to analyze as N increases; the algorithm only has to reduce N to less than N/2, regardless of how large N gets. But, I could be convinced by additional analyses and data. "Points faucet for low-powered systems"? Don't you think that is a little harsh? the project isnt "fake". it exists (or did). whether it comes back or not remains to be seen. but their methodology was shown to be flawed in the referenced old post, and the project admins refused to change anything about it when it was brought to light. they were more than happy to continue on crunching numbers uselessly. people are free to waste their compute power there if they wish. it's just a points faucet for low powered systems after all.
3) Message boards : GPUs : Any projects use AMD Radeon Pro Vega II OpenCL: 1.2 (Message 101580) Posted 15 Nov 2020 by Tigers Dave Post: Your GPU should work on Collatz @ Home, too.
4) Message boards : GPUs : Nvidia/AMD Cuda/OpenCL on Boinc projects - which card to buy? (Message 101579) Posted 15 Nov 2020 by Tigers Dave Post: Here are Collatz @ Home performance data, generated using my Macs and optimized crunching parameters. AMD RX 5700 XT - ~7 M credits/day NVIDIA GTX 1080 Ti - ~7 M credits/day AMD RX Vega 56/64 - ~6.5 M credits/day NVIDIA GTX 1080 - ~5 M credits/day AMD RX 580 - ~4 M credits/day NVIDIA GTX 980 Ti - ~4 M credits/day NVIDIA GTX 980 - ~3 M credits/day Some folks may achieve better numbers than me through undervoltage, overclocking, and better optimization of the crunching parameters. And, of course, the performance of a particular GPU under MacOS may be different than under Windows or Linux. Finally, some projects may be better at utilizing AMD GPUs than NVIDIA GPUs and vice versa. So, I agree with the advice of one of the previous posters that you check the message boards of the projects for which you crunch. FWIW, NVIDIA GPUs appear to be a much better bang for the buck than AMD GPUs when crunching Collatz @ Home tasks. But the most recent versions of the MacOS are not compatible with contemporary NVIDIA GPUs. Good luck!
5) Message boards : Questions and problems : BOINC issue with MacPro Late 2013 GPU AMD FirePro D700 (2) (Message 101578) Posted 15 Nov 2020 by Tigers Dave Post: I agree with Jord that trying another project may help you distinguish between hardware and software problems. Collatz @ Home also supports AMD GPU tasks on Macs. There is at least one user on Collatz @ Home that is crunching GPU tasks using a MacPro 6,1 with dual D500s. You can learn about that system here: https://boinc.thesonntags.com/collatz/show_host_detail.php?hostid=806697 There is a very small community of Mac users at Collatz and they try to help folks with hardware and software issues. Check out the Macintosh message board there. Good luck!
6) Message boards : Projects : News on Project Outages (Message 99183) Posted 8 Jun 2020 by Tigers Dave Post: Looks like Collatz is back up.
7) Message boards : GPUs : Projects that use Mac AMD GPUs? (Message 99182) Posted 8 Jun 2020 by Tigers Dave Post: Welcome to the relatively small community of Mac crunchers! The short answer is it depends on the age of your computers and your GPUs. I can use a MacPro 5,1 with a flashed RX Vega 64 to crunch Collatz @ Home WUs under Mojave. I can also use a MacPro 5,1 with a flashed RX 580 to crunch Collatz WUs under Mojave. I also crunch Collatz WUs using a variety of eGPUs (Akitio and Sonnet) that house RX 580s, RX Vega 56s, and RX 5700 XTs. These are connected via TB1, TB2, and TB3 to a wide range of Mac Minis, MacBook Pros, and iMacs under Mojave and Catalina. PM me if you need additional information. With the hibernation of SETI@home, I worry that my GPUs will be underutilized. The list of projects that use GPUs seems to be 5 years old. Is there an update to that list? What projects can I run on my Mac Pros with AMD GPUs? Thanks, Dave
8) Message boards : Projects : News on Project Outages (Message 99181) Posted 8 Jun 2020 by Tigers Dave Post: Looks like Collatz has been down for about an hour.
9) Message boards : GPUs : Trouble getting BOINC to recognize second NVIDIA GTX 1070 in MacPro 5,1 (Message 97039) Posted 24 Mar 2020 by Tigers Dave Post: I have diagnosed the problem. MacOS 10.13.6 High Sierra does not load OpenCL for the second GTX 1070, even though the System Profiler acknowledges the presence of the second GPU and the second GPU can successfully drive a monitor. However, MacOS 10.12.6 Sierra DOES load OpenCL for the second GTX 1070 and BOINC is capable of utilizing both GPUs. It is possible that this "bug" is specific to 10.13.6, as my BOINC recognizes two AMD RX 580 eGPUs connected to an iMac 18,1 running MacOS 10.14.6 Mojave. But, the conspiracy theorist in me believes the "bug" is specific to the combination of MacOS beyond Sierra and NVIDIA GPUs, as it is well evidenced that Apple and NVIDIA are no longer cooperating as they once did. More importantly, eGPUs that utilize the NVIDIA GPUs with Maxwell architecture and beyond are officially incompatible with Macs. It's a pity, as there are some really great NVIDIA GPUs that I would love to run in my MacPro 5,1 boxes or as eGPUs connected to my more contemporary iMacs, Mac Minis, and MacBook Pros.
10) Message boards : GPUs : Trouble getting BOINC to recognize second NVIDIA GTX 1070 in MacPro 5,1 (Message 96515) Posted 8 Mar 2020 by Tigers Dave Post: My MacPro 5,1 has two NVIDIA GTX 1070 GPUs. One is connected to the internal power supply. One is connected to an external power supply. I can connect either GPU to a monitor (hot swappable). See links for photos: https://www.dropbox.com/s/vmpyyuh60l9k9ak/Config%201%20Slot%201.jpeg?dl=0 https://www.dropbox.com/s/2en0cyf2hi4an3y/Config%201%20Slot%202.jpeg?dl=0 However, BOINC only recognizes one of the two GPUs. I don't have any problem getting BOINC to recognize two NVIDIA GTX 750Ti GPUs. Any idea on how I can get BOINC to recognize the second GTX 1070? Thanks in advance for any advice. Log file excerpt is below: Thu Mar 5 20:06:25 2020 \| \| Starting BOINC client version 7.14.2 for x86_64-apple-darwin Thu Mar 5 20:06:25 2020 \| \| log flags: file_xfer, sched_ops, task Thu Mar 5 20:06:25 2020 \| \| Libraries: libcurl/7.58.0 OpenSSL/1.1.0g zlib/1.2.11 c-ares/1.13.0 Thu Mar 5 20:06:25 2020 \| \| Data directory: /Library/Application Support/BOINC Data Thu Mar 5 20:06:25 2020 \| \| OpenCL: NVIDIA GPU 0: GeForce GTX 1070 (driver version 10.33.0 387.10.10.10.40.134, device version OpenCL 1.2, 8192MB, 8192MB available, 1712 GFLOPS peak) Thu Mar 5 20:06:25 2020 \| \| OpenCL CPU: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (OpenCL driver vendor: Apple, driver version 1.1, device version OpenCL 1.2) Thu Mar 5 20:06:25 2020 \| \| Host name: DJR-Home-Office-MacPro-51-1.local Thu Mar 5 20:06:25 2020 \| \| Processor: 16 GenuineIntel Intel(R) Xeon(R) CPU E5520 @ 2.27GHz [x86 Family 6 Model 26 Stepping 5] Thu Mar 5 20:06:25 2020 \| \| Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clfsh ds acpi mmx fxsr sse sse2 ss htt tm pbe pni dtes64 mon dscpl vmx est tm2 ssse3 cx16 tpr pdcm sse4_1 sse4_2 popcnt Thu Mar 5 20:06:25 2020 \| \| OS: Mac OS X 10.13.6 (Darwin 17.7.0) Thu Mar 5 20:06:25 2020 \| \| Memory: 48.00 GB physical, 667.06 GB virtual Thu Mar 5 20:06:25 2020 \| \| Disk: 930.71 GB total, 666.82 GB free Thu Mar 5 20:06:25 2020 \| \| Local time is UTC -6 hours Thu Mar 5 20:06:25 2020 \| Rosetta@home \| Config: excluded GPU. Type: NVIDIA. App: all. Device: all Thu Mar 5 20:06:25 2020 \| collatz \| Config: excluded GPU. Type: ATI. App: all. Device: all Thu Mar 5 20:06:25 2020 \| \| cc_config.xml: bad URL in GPU exclusion: http://einstein.phys.uwm.edu/ Thu Mar 5 20:06:25 2020 \| \| cc_config.xml: bad URL in GPU exclusion: http://setiathome.berkeley.edu/ Thu Mar 5 20:06:25 2020 \| collatz \| cc_config.xml: bad type 'ATI' in GPU exclusion; valid types: NVIDIA Thu Mar 5 20:06:25 2020 \| collatz \| URL https://boinc.thesonntags.com/collatz/; Computer ID 850946; resource share 40 Thu Mar 5 20:06:25 2020 \| Rosetta@home \| URL http://boinc.bakerlab.org/rosetta/; Computer ID 3700872; resource share 20 Thu Mar 5 20:06:25 2020 \| \| General prefs: from http://boinc.thesonntags.com/collatz/ (last modified 09-Oct-2014 08:54:01) Thu Mar 5 20:06:25 2020 \| \| Host location: none Thu Mar 5 20:06:25 2020 \| \| General prefs: using your defaults Thu Mar 5 20:06:25 2020 \| \| Reading preferences override file Thu Mar 5 20:06:25 2020 \| \| Preferences: Thu Mar 5 20:06:25 2020 \| \| max memory usage when active: 24576.00 MB Thu Mar 5 20:06:25 2020 \| \| max memory usage when idle: 44236.80 MB Thu Mar 5 20:06:27 2020 \| \| max disk usage: 674.41 GB Thu Mar 5 20:06:27 2020 \| \| max CPUs used: 13 Thu Mar 5 20:06:27 2020 \| \| suspend work if non-BOINC CPU load exceeds 25% Thu Mar 5 20:06:27 2020 \| \| (to change preferences, visit a project web site or select Preferences in the Manager) Thu Mar 5 20:06:27 2020 \| \| Setting up project and slot directories Thu Mar 5 20:09:42 2020 \| \| Checking active tasks Thu Mar 5 20:09:42 2020 \| \| Setting up GUI RPC socket Thu Mar 5 20:09:42 2020 \| \| Checking presence of 2610 project files
11) Message boards : Projects : News on Project Outages (Message 91324) Posted 1 May 2019 by Tigers Dave Post: Collatz now appears to be functioning properly.
12) Message boards : Projects : News on Project Outages (Message 91303) Posted 30 Apr 2019 by Tigers Dave Post: Looks like Collatz is down. I was having trouble downloading and uploading tasks for the last few hours prior to the shutdown.
13) Message boards : GPUs : [Solved] 2 GPUs in a MacPro 3,1 running different projects (Message 84887) Posted 23 Feb 2018 by Tigers Dave Post: Now I am having a different problem. When running only Collatz, on an 8-core MacPro with 87.5% of the cores (7 cores) allocated to BOINC, frequently BOINC runs 7 CPU tasks and 1 GPU task. How do I force it to run on both GPUs (NVIDIA GTC 750Ti).
14) Message boards : GPUs : [Solved] 2 GPUs in a MacPro 3,1 running different projects (Message 83639) Posted 6 Dec 2017 by Tigers Dave Post: I apologize in advance for starting a new thread, but I was unable to post this to the existing "Different projects for 2 opus" thread. I have two MacPro 3,1 (Early 2008) computers with two GPUs each (from NVIDIA and ATI). They are running OS X 10.11 and Boinc 7.6 and 7.8. I wanted to crunch Collatz tasks using the NVIDIA GPU and Rosetta, Seti, and Einstein tasks using the ATI GPU. The cc_config.xml file below allows me to do so. This configuration allows tasks to be queued independently for each GPU; in other words, when the first GPU finishes a task, it does not need to wait for the second GPU to finish its task before it (the first GPU) starts a new task. Happy crunching! <cc_config> <options> <exclude_gpu> <url>http://einstein.phys.uwm.edu/</url> <type>NVIDIA</type> </exclude_gpu> <exclude_gpu> <url>http://boinc.bakerlab.org/rosetta/</url> <type>NVIDIA</type> </exclude_gpu> <exclude_gpu> <url>http://setiathome.berkeley.edu</url> <type>NVIDIA</type> </exclude_gpu> <exclude_gpu> <url>https://boinc.thesonntags.com/collatz/</url> <type>ATI</type> </exclude_gpu> </options> </cc_config>
15) Message boards : GPUs : Two GPUs in a MacPro 3,1 crunching tasks from different projects? (Message 80293) Posted 18 Aug 2017 by Tigers Dave Post: I apologize for not exhaustively searching the forum before posting this question. Found the answer in the following thread: "Different projects for 2 gpus". Will see how it works, although I am a bit nervous about the fact that this thread indicates that the tasks are loaded in pairs, meaning that after one GPU finishes its task it will wait for the other GPU to be finished before tasks are loaded for each of them. Hmmmm ....
16) Message boards : GPUs : Two GPUs in a MacPro 3,1 crunching tasks from different projects? (Message 80292) Posted 18 Aug 2017 by Tigers Dave Post: I have four AMD Radeon HD 5770 and 5870 GPUs (Mac Edition) that I pulled out of 2008 MacPro 3,1 units when I installed NVDIA GeForce GTX (680, 770, 960...) GPUs. I use NVIDIA GPUs to crunch Collatz tasks. However, I cannot crunch Collatz tasks using the AMD GPUs. Is there any way to simultaneously crunch Collatz tasks using an NVIDIA GPU and Einstein tasks using the AMD GPU in the SAME MacPro 3,1? TIA!
17) Message boards : Projects : News on Project Outages (Message 75526) Posted 28 Jan 2017 by Tigers Dave Post: It appears that Collatz went down at approximately 06:00 UTC today. Still not back up.
18) Message boards : Projects : News on Project Outages (Message 74220) Posted 19 Nov 2016 by Tigers Dave Post: Security issue has been resolved. Project is down again due to database issues.
19) Message boards : Projects : News on Project Outages (Message 67631) Posted 9 Feb 2016 by Tigers Dave Post: Looks like Collatz has been down for approximately 12 hours.
20) Message boards : Projects : News on Project Outages (Message 63792) Posted 23 Aug 2015 by Tigers Dave Post: Looks like Collatz is down.

Next 20

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.