Jump to content

LAR_Systems

Member
 View Profile  See their activity

  • Posts

    729

  • Joined

  • Last visited

 Content Type 

Everything posted by LAR_Systems

  1. LAR_Systems
    How about if there was a toggle button on the bottom that let you flip the display like this?
  2. LAR_Systems
    You set the hours you want to fold between, then it switches to finishing and stop at the end of that folding time window to wrap up work units so you don’t loose return bonuses paused. so in your case you would want folding to start at 18:00 and stop at 15:00 which would be the next day. Then I would recommend based on the time your gear takes to complete a work unit set it a bit earlier like 13-14:00 to give your system time to finish work units. Last, the timer only works if the extension window is open and running, you close the browser window or tab the script can’t run, so that’s important.
  3. LAR_Systems
    Hey, I can give this a look for an update, but it would be helpful to know what screen resolution/size you're running the window at before I make changes that might just create a similar issue differently. Thanks,
  4. LAR_Systems
    Version 2.0 of Folding at Home in the Dark has hit the Chrome Extension Store and is PUBLIC! Someone needs to let Linus / the LTT writers know that it's ready for the next sleeper PC with the new "Monitor" selector in addition to the LTT RGB Lambo theme.
  5. LAR_Systems
    Version 2.0 has just hit the Chrome Extension Store and is PUBLIC! It's got some new stuff for you to use/play with. NEW - Timer (Time of Use) shifting allows you to set a time to start folding and finish folding so you can fold at off peak power usage times or to manage seasonal heat. NEW - View a recent time allocation breakdown of research your system was working to better understand why your PPD varies due to different research projects. NEW - UI Theme skin picker that allows you the flexibility to chose your style regardless of your team affiliation. NEW - Monitor style picker to allow you adapt your favorite UI Theme to a number of retro style display colors. NEW - Code and UI redesign to streamline the use of space and put more of a focus on the research being completed in a modern responsive UI that has been optimized to feel / be more responsive. Give it a look and enjoy!
  6. LAR_Systems
    Yeah, unfortunately CPU detection is not part of FAH and happens in the client browser... so its an issue I can't get around when users are running it off the local device. It's why the CPU system is still in BETA on my end, now that the core rebuild of the client is over and done I can look in to other methods of running it headless like Python/.NET Core apps just have not been able to get there yet. EDIT: Can you describe in detail what you're doing when forwarding the ports for your headless machine to your laptop? Perhaps I can just do something like disable CPU work unit tracking in the client if I can detect it's hitting a non local machine, but I have limited permissions in networking running in the browser.
  7. LAR_Systems
    Thanks for the feedback, made some changes for that Lambo pop. And yeah, the timer / time of use stuff is a request I have seen in a ton of forums from users old and new, I'm surprised they (FAH) don't have it with an easy mode in the default client cause it has to hurt the number of people willing to fold that don't want to rack up a power bill and heat during peak hours. I know testing it, it's been nice going in to a cool office in the morning with my main system already wrapped up for an hour or two.
  8. LAR_Systems
    Team feelings on the LTT skin I might call RGB Lambo? Team info, stats, badge level shown when you mouse over the logo when live, hidden while not interacting, showing neon LTT logo.
  9. LAR_Systems
    You currently have to visit two sites because those sites like EOC have to process the entire F@H user stats files every few hours and calculate/store them to have those metrics. They are not available on an existing or public API like the metrics that I am currently displaying in the client. The client and the hardware DB have been a big bite to chew on as one dev, not sure I want to take on the kind of work EOC is doing as well to provide those points of data that we can already get, that said I will add a direct link from the team panel to EOC team page as I have for the user stats.
  10. LAR_Systems
    Can't do it, I have tried. Firefox sandboxes the Extensions in such a way the browser can not communicate with the local folding@home client service (in party because it runs insecure on HTTP) URL required to get the folding clients API stream. The best I could do in their sandbox is drop CSS/JS on top of their HTML (running 2X the resources) and read the values they write to the HTML and try to parse them to send the DB data. Kind of pointless to the user, as it's just the same default broken UI with dark tones. On Chrome / Chromium / Edge I can replace their HTML completely and in the latest update, replacing their code with an LAR version that provides more features than the default client and far more than what I can do in Firefox to make it worth doing that. I'm more about the user experience and fixing/improving the folding client than I am just about feeding the DB, I want something useful that looks good in my own browser and yours with information and utility. Showing a responsive modern UI, (trying to add accessibility) skins, system and slot averages for more accurate folding information for users / DB and what I'm wrapping up now... Time of Use fold scheduling to enable folders that want to not go broke on power to use off peak times to fold only, then auto finish otherwise is my highest priorities and hopefully make using Edge or Chrome worth it. Time Shift / Time of Use Alpha (Testing GUI) That all said, I'm exploring what can be done with Python or .NET core to run a local package for users that are headless or want to load things up in any browser they like... but that has been back burnered due to the need for the Chrome Extension needing to be updated to the latest Google publishing standards... and I only have so many hours to maintain the extension and database site.
  11. LAR_Systems
    New Beta GUI felt cute, don't know, might delete later.
  12. LAR_Systems
    Just a heads up for anyone following the project. There will be GUI changing coming that include a complete re-write/design to accommodate changes to F@H stats sources and links and Chrome Extensions manifest guidelines. This rebuild has allowed me to leverage some features (PPD averages, web client logging, optimizations, more focus on research, modern UI bordering on sci-fi movie feel) I have wanted to add for awhile but was limited working inside the F@H client HTML framework. This is by no means done and requires plenty of testing, controls to show/hide walls of information, and most importantly creating the LTT specific skin / color scheme once the primary interface is completed. But it's shaping up enough for a sneak peak of progress made so here we are. Release TBD based on my time to test and wrap this up. Fully Expanded (open everything) in default non team skin below.
  13. LAR_Systems
    I might be testing some new things out
  14. LAR_Systems
    The difference between 1660 super and TI on average is around 200ppd give or take the week / work units. https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/tu116_geforce_gtx_1660_super https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/tu116_geforce_gtx_1660_ti
  15. LAR_Systems
    I have been busy working and meanwhile this dropped when I was not looking. My cousin just flipped it over to me with an LOL. So the A100 https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/ga100_a100_pcie_40gb work unit that they got was for a Cancer project, so that's going to be lower on the PPD and they only ran the test for one WU so that's all we have a sample. But for comparing the A100 to the 3090 that Cancer Wu as in the DB the site has the data https://folding.lar.systems/projects/folding_profile/18021 based on others submissions on the same project. Average PPD 3090 = 6,333,449 Average PPD A100 = 5,886,775 Given the power usage difference pointed out in the LTT video that is a massive PPD performance for the power used on F@H. Edit: In a form factor you can jam in to a rack case! The A100 if left to run a full spectrum of different work unit types over the course of a week would crush the 3090/3080 in terms of PPD per watt I'm guessing. Anyone know how to get @jakkuh_t to leave it on for us
  16. LAR_Systems
    It was an issue of F@H having a bad character in the GPU.txt string. I have manually accounted for it which seems to fix up the listing and consolidate the data. https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/gp107_geforce_gtx_1050_ti_2138
  17. LAR_Systems
    This happens if F@H changes the entry in the GPU.txt for the same device. It's rare, but it happens, I will give it a look.
  18. LAR_Systems
    Please don't burn your place down, monitor your PSUs at all times.
  19. LAR_Systems
    I would need to see a screen shot of the extension window to get the card WU and project information needed to see what's up. It's possible that there is something going on, or that you're getting new project WUs that don't have enough data for the cards yet.
  20. LAR_Systems
    Hey there, Glad you're getting some alternative uses out of the DB and client. I hade a look at the W4100 and it's one of the cases where the AMD identifier for the card just is not set/picked up correctly for non gaming cards. I have manually set the parameter which should solve the issue ongoing and the profile link once set is this https://folding.lar.systems/gpu_ppd/brands/amd/folding_profile/cape_verde_gl_firepro_w4100 Cheers
  21. LAR_Systems
    That is correct, all the CPU detection is on the local browser, no way around that as F@H APIs in the service do not provide the CPU information, only the GPUs.
  22. LAR_Systems
    They know it's out there, I have made them aware on their forum... official word last time I looked was they only care about the core client, no side projects or PPD related stuff... just the hard science which makes sense. If that's changing now, it would be cool.
  23. LAR_Systems
    Just as a follow up, some changes have been made to how the charts are implemented for when you toggle them on and off. The F@H "slot id" id dynamically assigned and the client when you tried only supported up to the number 8... issue is F@H if you add and remove devices in the same session can increment the number to be larger than 8 and then you don't get chart even if there is only one device folding. So little re-work on the code and now to account for this it can dynamically create charts up to 32 slot IDs for those that end up with the high IDs from trying stuff in the client or if you have a banging system with sooooo many GPUs Anyways try it now to see if the charts show, Cheers,
  24. LAR_Systems
    The people reporting numbers in the 2 Mil range are currently running on target, looks like PPD reward / WUs have changed over the past two weeks for the card which happens when projects that reward heavy are replaced by something else.
  25. LAR_Systems
    Hi, There are multiple 1070s and other GPUs with duplicate records because they will be slight variations on the model as identified by F@H client (and shown under the card name) so the records are kept separate as the different revisions, mobile units etc. behave differently with different firmware/hardware. In terms of the ranks, they are based on real-time data that is collected form users using the extension and are averaged across different work units and projects which have different PPD rewards that can and do change. As for you card reporting low PPD. The image in your post appears to be broken so can't see what card / issue you are showing... but if you are seeing dramatically lower PPD and you are new to folding you may not have an API key setup which will give you far less PPD than those values reported in the database as they only record PPD values for users with a F@H API key. Otherwise I would need to see the issue / know what hardware and work unit you are on to review further. Cheers,
×