F@H Web Client Dark Skin with PPD comparison by GPU / WU

[Gorgon]

2 hours ago, LAR_Systems said:

... and just disabling CPU reporting if you were using this approach. ...

sweet - just curious. Why would CPU reporting have to be disabled? All slots in use (GPUs & CPUs) would show up in the web client at the https://localhost:port URL.

[LAR_Systems]

4 hours ago, Gorgon said:

sweet - just curious. Why would CPU reporting have to be disabled? All slots in use (GPUs & CPUs) would show up in the web client at the https://localhost:port URL.

CPU reporting to the DB site would be disabled, because identifying the CPU comes from the browser, not F@H API... they pass the model of the GPUs, but not the CPUs in the web client, just a slot and the number of cores.   Which is too bad, since they know the CPU cause they show it in the log, but not the APIs.

[Justaphf]

@LAR_SystemsSeeing a weird issue with the research tab where it's showing the same row multiple times.  Seems to only happen with MG2+ and AAA+ causes.

 

 

 

Using Brave (latest stable) on Windows 10.

 

[LAR_Systems]

12 hours ago, Justaphf said:

@LAR_SystemsSeeing a weird issue with the research tab where it's showing the same row multiple times.  Seems to only happen with MG2+ and AAA+ causes.

 

 

 

Using Brave (latest stable) on Windows 10.

 

Fix for this will hopefully make it's way in to the public in the next 12 hours or so.

 

Thanks for letting me know.

[Justaphf]

On 11/30/2022 at 10:20 PM, LAR_Systems said:

Fix for this will hopefully make it's way in to the public in the next 12 hours or so.

 

Thanks for letting me know.

I can confirm this looks fixed now.

 

On a different note, where did you get the cause for these? I don't see them as cause options in the FAH API.  Was it just a manually created mapping based on reading the project description?  Have been working on a MR for HFM to add the cause by pulling from the API and these show up as Unspecified since that's how the API has them.

[LAR_Systems]

11 hours ago, Justaphf said:

I can confirm this looks fixed now.

 

On a different note, where did you get the cause for these? I don't see them as cause options in the FAH API.  Was it just a manually created mapping based on reading the project description?  Have been working on a MR for HFM to add the cause by pulling from the API and these show up as Unspecified since that's how the API has them.

I manually manage my own database and fill in the unspecified projects using their descriptions 

[Justaphf]

@LAR_SystemsI just started seeing the multi-row issue for the new Liquid Liquid-Phase Separation project type.  I did a F5 to refresh and it cleared all the duplicates but then I could watch it add a new line about every 5 seconds.  Running on Brave browser v1.50.119 and plugin version 2.06.

 

[LAR_Systems]

On 4/21/2023 at 6:18 PM, Justaphf said:

@LAR_SystemsI just started seeing the multi-row issue for the new Liquid Liquid-Phase Separation project type.  I did a F5 to refresh and it cleared all the duplicates but then I could watch it add a new line about every 5 seconds.  Running on Brave browser v1.50.119 and plugin version 2.06.

 

 

Thanks for the heads up, this is hopefully patched out as of mid day Saturday.

[Justaphf]

On 4/24/2023 at 7:53 AM, LAR_Systems said:

Thanks for the heads up, this is hopefully patched out as of mid day Saturday.

I can confirm this fix has gone through on my end now.  It was still happening until yesterday, but now all fixed.  Thanks for all the effort you put into the plugin and your stats site!

[Justaphf]

12 hours ago, Justaphf said:

I can confirm this fix has gone through on my end now.  It was still happening until yesterday, but now all fixed.  Thanks for all the effort you put into the plugin and your stats site!

Looks like I spoke too soon, seems like it wasn't duplicating while I was watching but when I came home it had a bunch of duplicates.  Will check again after this weekend.

[Justaphf]

11 hours ago, Justaphf said:

Looks like I spoke too soon, seems like it wasn't duplicating while I was watching but when I came home it had a bunch of duplicates.  Will check again after this weekend.

Actually, I think I see what the issue may be.  There are multiple entries for the liquid phase separation option, with various combinations of spaces or hyphens between words.  On the website I see

 

And in the plugin-in I see (this spelling is not duplicating).

 

Unfortunately I didn't take a screen capture of which spelling was having the duplication yesterday.  I'll let you know if it starts happening again.  Most of my boxes have switched back to Cancer WUs so not sure when the alternate spelling will come up again.

[LAR_Systems]

On 4/28/2023 at 9:16 AM, Justaphf said:

Actually, I think I see what the issue may be.  There are multiple entries for the liquid phase separation option, with various combinations of spaces or hyphens between words.  On the website I see

 

And in the plugin-in I see (this spelling is not duplicating).

 

Unfortunately I didn't take a screen capture of which spelling was having the duplication yesterday.  I'll let you know if it starts happening again.  Most of my boxes have switched back to Cancer WUs so not sure when the alternate spelling will come up again.

Another update has been made, should resolve the duplicate on the site and the looping in the client.

 

Please let me know if you continue to see the issue.

[Justaphf]

21 hours ago, LAR_Systems said:

Another update has been made, should resolve the duplicate on the site and the looping in the client.

 

Please let me know if you continue to see the issue.

Seems to be fixed now.  I've at least run 16575, 16576 and 16572 in the last 24 hours and no duplicates.  My HFM hasn't been logging my Windows machine for the past 2 weeks it looks like so I can't tell any further in the past though, it's the only Windows box I run and where I originally noticed the issue.

[Justaphf]

@LAR_Systems Wanted to check to see if the stats for 4070 ti were updating correctly on your site.  Been running one under Linux for just under a week but don't see any recent Linux data points in the chart on your site for this card.  The plugin seems to be working fine on my machine so far as I can tell, but maybe it's not sending from my end for some reason.

[Justaphf]

13 minutes ago, Justaphf said:

@LAR_Systems Wanted to check to see if the stats for 4070 ti were updating correctly on your site.  Been running one under Linux for just under a week but don't see any recent Linux data points in the chart on your site for this card.  The plugin seems to be working fine on my machine so far as I can tell, but maybe it's not sending from my end for some reason.

Some additional information.  I'm running on latest version of Brave browser.  I did notice a CORS error listed in the debugging console, though not sure if it's related at all to this issue.  That CORS issue exists on all of my machines, not just the one with the 4070 Ti.  I also noticed that the 3060 Ti stats don't show any Linux results and I've been running one of those under Linux for over a year.

 

Also, due to wanting to be hands off with the machines and Brave pushing updates frequently and always wanting an app restart, I have a script which runs every 4 hours that shuts down the browser, apt updates Brave, then restarts the browser back to the plugin page.  Not sure if this could be affecting things, though it's been working flawlessly for a couple years now.

 

CORS issue:

 

Plugin info:

[LAR_Systems]

14 hours ago, Justaphf said:

Some additional information.  I'm running on latest version of Brave browser.  I did notice a CORS error listed in the debugging console, though not sure if it's related at all to this issue.  That CORS issue exists on all of my machines, not just the one with the 4070 Ti.  I also noticed that the 3060 Ti stats don't show any Linux results and I've been running one of those under Linux for over a year.

 

Also, due to wanting to be hands off with the machines and Brave pushing updates frequently and always wanting an app restart, I have a script which runs every 4 hours that shuts down the browser, apt updates Brave, then restarts the browser back to the plugin page.  Not sure if this could be affecting things, though it's been working flawlessly for a couple years now.

 

 

 

Can you show me a screen shot of the GPU slots in the plugin as that will help me identify the exact GPUs.

 

I ask as I can not see any Linux traffic for example on 3060 TI recently, but I see months worth on 3060 TI Lite Hash Rate version which could be part of the issue.

 

LHR 3060 Ti = https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/ga104_geforce_rtx_3060_ti_lite_hash_rate

3060 TI = https://folding.lar.systems/gpu_ppd/brands/nvidia/folding_profile/ga104_geforce_rtx_3060_ti

 

 

[Justaphf]

This is for the 4070ti.  I looked again today and it seems like some Linux data points are showing up now, though you can only see it by getting the tool-tip to pop-up due to the line overlap.  Yesterday when I looked it wasn't there even in the tool-tip.

 

For the 3060 Ti, must have been looking at the non-LHR graph by mistake .  I know that my card is LHR and when I looked at the correct graph today it is showing the expected data.  I had been popping the graphs up manually by scrolling down the overall PPD ranks table based on what cards I have running, rather than clicking on the link in the plugin for each slot, so that was my fault for opening the wrong graph.

 

Thanks for looking into that, seem with 4070ti data starting to show up now everything is functioning normally.

[LAR_Systems]

14 hours ago, Justaphf said:

This is for the 4070ti.  I looked again today and it seems like some Linux data points are showing up now, though you can only see it by getting the tool-tip to pop-up due to the line overlap.  Yesterday when I looked it wasn't there even in the tool-tip.

 

For the 3060 Ti, must have been looking at the non-LHR graph by mistake .  I know that my card is LHR and when I looked at the correct graph today it is showing the expected data.  I had been popping the graphs up manually by scrolling down the overall PPD ranks table based on what cards I have running, rather than clicking on the link in the plugin for each slot, so that was my fault for opening the wrong graph.

 

Thanks for looking into that, seem with 4070ti data starting to show up now everything is functioning normally.

Thanks for the report regardless, the 4070ti data was delayed due to a processing script running poorly so I fixed that up and now should be every 24 hours again which is why you did not see the update of your work this week until yesterday.

[enhanced08]

@LAR_Systems Greetings! After randomly stumbling across this extension on another website, doing some googling I finally came across this post. Ive been a subscriber of LTT but never ventured into the forums, but decided to register just to post here.

Thank you for creating this tool, Ive just installed it and will be playing around with it more in the coming days. 

 

Some background on me, Ive been folding for... awhile. Back in the 2005-2006 time frame I had an idea to build a PPD database to compare PPD of different CPU and GPUs. The website is since long gone but required manual entry of data. Im no programmer but met a guy who was willing to work with me and we created YURT which automated the data collection. The developer sorta took over the project and I lost most all access to it. It appears to be totally shutdown these days but its great to see someone picked up the idea and ran with it.   Fantastic!

Good luck and continued success!
Happy Folding! 

[LAR_Systems]

Hello Everyone, it's been awhile so here is an update.

 

In anticipation of the eventual general release of the new Folding@Home version 8.1.18 client currently in Beta https://foldingathome.org/beta/?lng=en I have created a new web client that will work with the new web sockets approach that F@H dev's seem to be headed in.

 

This is a ground up redesign of the web clients code, functions and related folding database systems to accommodate the large number of differences between the current F@H Version 7 client vs their new Version 8 one.

 

This means that what you're getting right now is buggy and an Alpha build from my perspective as the F@H Beta client has bugs and issues IMO that I hope are being ironed out, which means that my best efforts to account for those and give a similar experience to what exists in my current Chrome Extension for F@H Version 7 has been quite the pain in the ass. 

 

That said sharing this broadly now for people to start messing with if they are running the F@H Beta 8.1.18 client as I have got the new client to where it's looking half decent and "working" in terms of the basics to pull up device / project info and submit your PPD to the database for inclusion.

 

Additional features will be worked on as I have time to integrate the client deeper features that existed in the V7 version as well as new integrations with the https://folding.lar.systems/league/ folding league team and user stats DB that has been added to my site.

 

So, if you would like to join the "alpha" of the new client and you are running F@H Version 8.1.18 BETA on your system you can navigate directly to https://webclient.lar.systems/ which will load the new client in ANY BROWSER as it is not dependent on Chrome extensions like the F@H V7 client system is!!!

 

Otherwise, if you're not running F@H Version 8.1.18 BETA you can ignore all this as the Version 7 client is far more polished as is F@H V7 as this time, so remember you're a ALPHA/BETA user if you move up to the latest right now and things will be buggy.

 

Cheers,
LAR_Systems

 

Example: F@H Version 8.1.18 BETA - Folding.LAR.Systems Redesign https://webclient.lar.systems/

Replaces what F@H provides by default at https://app.foldingathome.org/ in version 8

 

 

[LAR_Systems]

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,

[LAR_Systems]

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.

 

[Justaphf]

@LAR_Systems for the League statistics on your site, how are you determining which league a long dormant return folder is in? I was trying to trace down the top current producer and couldn't find them initially because they are ranked as Minor league even though have been in the top 2 global producers for at least 12 days running and their total monthly production puts them well within the top 10 from the All Stars page.  Looks like they were dormant from June '23 till middle of this month.

 

Looking from a few different angles, it seems like there may actually be some kind of bug because the team 24-hour PPD page actually lists this user as All Stars, but everywhere else they seem to be listed as Minors.  A few screen captures below in case things change since I posted.

 

 

 

[Justaphf]

5 minutes ago, Justaphf said:

Looking from a few different angles, it seems like there may actually be some kind of bug because the team 24-hour PPD page actually lists this user as All Stars, but everywhere else they seem to be listed as Minors.

Looks like the 24-hour PPD league difference is intentional, seems there is a daily league and a monthly league.

 

Still curious about monthly league calculation though.

[LAR_Systems]

On 11/25/2023 at 5:52 PM, Justaphf said:

@LAR_Systems for the League statistics on your site, how are you determining which league a long dormant return folder is in? I was trying to trace down the top current producer and couldn't find them initially because they are ranked as Minor league even though have been in the top 2 global producers for at least 12 days running and their total monthly production puts them well within the top 10 from the All Stars page.  Looks like they were dormant from June '23 till middle of this month.

 

Looking from a few different angles, it seems like there may actually be some kind of bug because the team 24-hour PPD page actually lists this user as All Stars, but everywhere else they seem to be listed as Minors.  A few screen captures below in case things change since I posted.

 

 

 

 

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.