Jump to content

LAR_Systems

Member
  • Posts

    729
  • Joined

  • Last visited

Everything posted by LAR_Systems

  1. New version selector for the web client has been installed and a "stable?" RC for 8.3.5 is now posted for you and everyone else to break lol You can give my first crack at 8.3.5 RC at the new version selection page https://webclient.lar.systems/ Cheers
  2. Depends on what version of 8 you are running. F@H 8.5 beta requires much re-programming on my part, so if you have upgraded to their latest, I'm not there yet. Mine is only compatible with their original V8.1 beta. I'm working on it though... trying to get something stable up asap.
  3. Yeah, I was away from home for a week so of course that's when a UPS died in my rack because hardware always knows when I'm gone. Back up as of yesterday, almost all services restored.
  4. Which I don't have access to in a web extension, for the remote system etc. Hence the issue. Best that could be done on 7, know the vast majority are not port forwarding, get the info the extension is allowed to get and have to sweep moderate stuff that is whacky... but that's manual... and I only got so many hours in the day. If V8 becomes the main app, problem solved, but been sitting on the same beta version for a long time, so who knows.
  5. In Version 7 the JSON with the CPU slot PPD does not identify it other than it's a CPU and number of threads. The work work around was needing to use the plugin/browser extension framework to get the actual CPU name, make, min/max threads etc. In Version 8 they handle this correctly where all hardware is identified over the socket connection of the remote machine by the folding client and passed in the early connection data. In Version 7, the default web interface and it's underlying code / APIs appeared to be an after thought with the focus on the trey application which reports all the correct full remote information via the logs etc.
  6. The system working on averages smooths out that kind of thing... like unless you run on LNG for weeks you should not taint the record sets, as there would be tons of regular entries and the few strange outliers. Same is on GPU PPD when some change clocks, bios, water cool etc. The big killer is when they attach to something remotely and forward the ports and then log hundreds of data samples over the course of weeks as that will start to move the needle if it's a server CPU with massive number of cores, but the machine they observe on is some low core regular chip... that will mis read the number of threads vs the PPD (lower threads, higher PPD) and juice the scores. The records in question on the reported case where in the thousands... someone left that running some strange way for weeks on end.
  7. This is a known pain in my ass when one of two things happen with the Folding@Home V7 client. 2 things can cause this. The first is unicorn work units, if you go to the bottom of the page you will see some work units / projects might report some massive anomaly of PPD for a project that drags the average up on a device well beyond how it would perform on the majority of projects. The other is some people will port forward the web client to a remote machine so they can monitor that machine... but the V7 Client does not report back the real CPU running on the slot as part of the remote data, it ends up reflecting the local machine the client is viewed on CPUs. So if you have someone with a Xeon / Epic nasty CPU server, but they view it on some old i3 POS laptop... the I3 might get recorded as having awesome stats... there is no way for me to tell that's happening automatically, all I can do is sweep moderate records that don't make sense from time to time. The good news... the new V8 F@H Client Beta reports the real CPU on a local or remote machine removing the bulk of these issues... as I see I more users start using the V8 Beta Client for my system, I will remove CPU reporting completely from the V7 version to have better data overall. Thank you for attending my Ted Talk. Edit: Investigated, this particular CPU seems to be someone remoting in to something bigger, I just searched by the irregularities and they are all grouped. Thanks for the report.
  8. It's a Monthly Active day count, and it's December, 1st The GUI is probably cached on the word November and will reset shortly based on your location / timezone. Edit: Pushed so much new stuff this year, I will be bug squashing most of next year.
  9. The day league is done by raw PPD production, so you can be a daily AllStar but a monthly Minor league etc. The monthly leagues are a combination of PPD production and consistency of daily folding. So if you have the highest PPD production, but only a handful of active days in the Month you will not be a Monthly Allstar, the people that have been in the AllStar daily ranks with the most days will. This is to level the field a little bit and reward regular folding vs someone turning on their works compute server a couple days a month to decimate the hobby folders doing the best they can everyday. In the example you provided in the first screen shot that user in the minors with massive production for the month has only been active 13 of 26 days, so there is no way they are going to be a November All Star. They keep it up, they will crush in December with daily consistency. But the model is kind of like if a pro-athlete wants to be an All Star they can't be amazing a few days a season and off injured not playing the rest of the time, that time away counts against them. P.S. If the user in the example keeps folding the rest of the month, they can move up in the Leagues as well, just not likely to hit AllStar but could get from Minors to Majors etc. but they are not going to see those jumps until closer to the end of the month as the number of Active days gets larger.
  10. Had to go out of town, so all work was finished then machines offline. I'm back now, will be ramping back up over the next 24-48 but yeah, bummer it worked out like this with Folding Month.
  11. https://webclient.lar.systems/ has been updated today to include your F@H points, work units etc. that have been uploaded and processed by F@H officially across all your folding machines, and your rankings in folding.lar.system leaderboards / leagues.
  12. Lol, why oh why would you ever not want to go full BETA during folding month!
  13. @Gorgon @leadeater The latest update to the Version 8 Beta of the web client has been published. It now has built in support for connecting to remote system configured to allow it. But this only works with F@H Version 8 Beta. https://webclient.lar.systems/ Full details on this comment here, but I do have it remoting to machines on my test network, while applying the IP/Port in the URL so you can have multiple / bookmark systems if wanted.
  14. Connecting to Remote Folding Machines with Version 8 BETA This is now possible. The client now has an area at the bottom of the page to enter an IP / PORT which will be appended to the URL for the web client enabling you to change the IP and/or port that it's connecting to. This is added to the query string so you can bookmark the pages / have multiple open etc. as you need to. Example: https://webclient.lar.systems/?remote_ip=127.0.0.1&remote_port=7396 This like other remote access to F@H requires that you have opened firewall ports and enabled remoting access in the F@H config file of the folding machine in addition to needing to allow a change to your browsers security settings for the webclient.lar.systems domain. To get an idea what is involved just put any IP or Port that will not load a client in the boxes on the client and the summary of instructions will appear. Example: https://webclient.lar.systems/?remote_ip=127.0.0.99&remote_port=7396 This is about as alpha build as it gets so keep that in mind, as it make take a bit of playing around to get everything open and configured for remote access. Cheers,
  15. You seem to be right on track for a 3070 2-3 Mill PPD per day. Mobile https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/ga104m_geforce_rtx_3070_mobile_max-q_8gb16gb You will see fluctuation in PPD rewards depending on what work units you get assigned randomly, so some days you will get more PPD than others. That also said on the mobile, power constraint is likely a secondary factor to thermal throttling on a Laptop so do what you can to keep it cool, fans / vents clear of obstructions etc. and you might find that helps squeeze a bit more PPD. Related to this, turning off CPU folding is a good plan as doing both on a laptop almost always ends up with massive thermal throttling as many designs share the same heat pipe and fan for both CPU and GPU.
  16. Yep, possibly on a power supply / cables you already own as well. Way lower total cost of ownership. Some of my gear is getting pretty old / inefficient but I'm rolling with it in part as the mods / upgrades I would have to make to go to the latest gen cards in terms of physical size to house, power supply, heat dissipation etc. are also cost factors. When you go ALL in on a parts list... suddenly it gets expensive if it means not replacing everything you plug cards in to 4060tis from a size/power connector and wattage perspective almost start making sense. That said... I'm more likely to just wait and start building a new custom GPU housing and stack PSU for my next build as I'm not likely to fit 4x of any card on a single PSU in an enclosed case much longer at a price and noise I'm willing to run in an area where I also work.
  17. 4080 is currently best for power usage per PPD. 4070ti is the sweet spot for cost of card & power usage per PPD as @GOTSpectrum mentioned. 4090 is a beast for gaming / machine learning etc. but for folding you can buy two 4070tis for the same price as a 4090, get more PPD overall while completing 2x the work units. https://folding.lar.systems/gpu_ppd/overall_ranks_power_to_ppd
  18. The hardware classification is "more efficient cards I can not afford" lol.
  19. You are there, you are just new to this months ranks by the looks, takes some time for my systems to parse and post combined with the delay of F@H reporting WUs https://folding.lar.systems/league/team_ranks?id=223518
  20. Day 3 of the folding event.... (Scrolling hardware sites looking at GPUs...)
  21. Really happy to hear that! Made some changes to account for slight differences in how GPU drivers are loaded for AMD devices and that seems to have fixed the data load. Think I’m going to enjoy operating this version of the client as a web app instead of a browser extension as I can push updates in minutes, not days.
  22. This is the truth, and frankly if you're going to cave it's best to do it early in the month, than mid month like seems to be the regular pattern. Easier to get out in front early than to catch up!
  23. For anyone interested in keeping it competitive/seeing who is creeping up in production on you throughout the day I have put the teams 24 Hour Rolling PPD ranks and Monthly Team Ranks pages in to event mode highlighting the users signed up as listed in the Google Doc. These are ROLLING PPD ranks so they change during the day as the system finds updated data from F@H. THESE ARE NOT OFFICIAL EVENT RANKS AND WILL NOT LIKELY MATCH OFFICIAL EVENT RANKS. THIS IS DUE TO THE ROLLING UPDATES VS THE OFFICIAL VALIDATED ONES PUT IN THE EVENT BLOG POSTS DAILY. PLEASE DO NOT BUG THOSE RUNNING THE EVENT ABOUT THESE RANKS, THEY ARE JUST FOR FUN TO SEE WHO MIGHT BE SNEAKING UP ON THE REST OF US WITH MERC ARMIES "BORROWED" 4090s ETC. DURING THE DAY. LINUSTECHTIPS_TEAM ROLLING 24HR PPD RANKS FOLDERS PARTICIPATING IN THE NOV 2023 EVENT ARE HIGHLIGHTED https://folding.lar.systems/league/team?id=223518 LINUSTECHTIPS_TEAM OVERALL RANKS: NOVEMBER 2023 FOLDERS PARTICIPATING IN THE NOV 2023 EVENT ARE HIGHLIGHTED https://folding.lar.systems/league/team_ranks?id=223518
×