First visualised Core 17 workunit

This forum contains information about 3rd party applications which may be of use to those who run the FAH client and one place where you might be able to get help when using one of those apps.

Moderator: Site Moderators

Post Reply
ChristianVirtual
Posts: 1596
Joined: Tue May 28, 2013 12:14 pm
Location: Tokyo

First visualised Core 17 workunit

Post by ChristianVirtual »

I guess some fellow folder realised that I love my iPad and try to make it as much as possible involved in my personal folding experience. And for every one who want too ...

In addition it always bothered me that Core 17 WU couldn't be visualised properly in any tool available for FAH. But here it comes the outcome of a day work: the first ever in public visualised Core 17 Workunit :D

Image

Project 13001 (R:140, C:7, G:23)
Atoms: 4351
Bonds: 3597

The data are mainly available in the working directory; I first hoped it was as easy as copying the viewerTop.json and the related viewerFrame1.json files.
But those files contain not all relevant information like all bonds or indicators for alpha Carbon.
:idea: So I had to write a little Python script serving as server and listen to a port. Once received a command it will read the system.xml and combine it with the viewerTop.json file to produce the required output as defined by 3rd part API. And skipping all the water in the system.xml file. :egeek:

Two issues I would like to suggest being raised in new bug fix reports or added to existing one: :!:
1) The bonds in the the viewerTop.json file are only about 50% of the required bonds; need to combine segments "Constraints" and "Bonds"
2) The viewerTop.json don't contain important information about AlphaCarbon, charge, radius, mass of each atom or position in the side chain (only one-digit chemical symbol H, C, N and O, ...).

Right now the python script is not yet robust enough in case of multiple connections or reconnections. So would be a bit immature to share in the current state; but will hopefully soon share here and/or on github. :ewink:
Last edited by ChristianVirtual on Tue Oct 14, 2014 5:06 am, edited 1 time in total.
ImageImage
Please contribute your logs to http://ppd.fahmm.net
PS3EdOlkkola
Posts: 184
Joined: Tue Aug 26, 2014 9:48 pm
Hardware configuration: 10 SMP folding slots on Intel Phi "Knights Landing" system, configured as 24 CPUs/slot
9 AMD GPU folding slots
31 Nvidia GPU folding slots
50 total folding slots
Average PPD/slot = 459,500
Location: Dallas, TX

Re: First visualised Core 17 workunit

Post by PS3EdOlkkola »

Nicely done ChristianVirtual!

Thoughtful, logical and incredibly creative technical skill -- the foundation of truly great engineering talent.

I'm not easily impressed, but what you've created is really fantastic. I'm looking forward to its release on github!
Image
Hardware config viewtopic.php?f=66&t=17997&p=277235#p277235
ChristianVirtual
Posts: 1596
Joined: Tue May 28, 2013 12:14 pm
Location: Tokyo

Re: First visualised Core 17 workunit

Post by ChristianVirtual »

Thanks PS3EdOlkkola,

great to hear. Its just a hobby, powered by positive feedback :D Unlikely there is an any engineering talent anymore growing in me. Too old for that (never say no, though).

The github is published under https://github.com/ChristianVirtual/FAH ... 7Trajectoy

But I need to reduce the expectation a bit: that is only a script transforming the data from Core17 (as a workaround) into something any generic viewer can use ( Jesse_V , still around ?). To really enjoy the visual impression of Core17 I need to submit a change to Apple AppStore for my iPad app (and finish my Android version of the viewer). That might take another week or two or three.

Just as a teaser ... and in hope that the official developer team eventually will enable the official software to provide the data on regular way.
ImageImage
Please contribute your logs to http://ppd.fahmm.net
ChristianVirtual
Posts: 1596
Joined: Tue May 28, 2013 12:14 pm
Location: Tokyo

Re: First visualised Core 17 workunit

Post by ChristianVirtual »

... and another image on how it look in the context of the UI:

Image

Finally I also to managed to analyse the backbone and side chain properly:

Image
(One question: someone with chemical background and detailed knowledge about WU 13001 could confirm the collection of side chains ? That would be fantastic ! :) )
ImageImage
Please contribute your logs to http://ppd.fahmm.net
Jesse_V
Site Moderator
Posts: 2851
Joined: Mon Jul 18, 2011 4:44 am
Hardware configuration: OS: Windows 10, Kubuntu 19.04
CPU: i7-6700k
GPU: GTX 970, GTX 1080 TI
RAM: 24 GB DDR4
Location: Western Washington

Re: First visualised Core 17 workunit

Post by Jesse_V »

ChristianVirtual wrote:The github is published under https://github.com/ChristianVirtual/FAH ... 7Trajectoy

But I need to reduce the expectation a bit: that is only a script transforming the data from Core17 (as a workaround) into something any generic viewer can use ( Jesse_V , still around ?). To really enjoy the visual impression of Core17 I need to submit a change to Apple AppStore for my iPad app (and finish my Android version of the viewer). That might take another week or two or three.
I'm still around. I lurk a bit from time to time.

As long as you transform the data into a more readable format that's fine. If you transform it into the format that FAHClient puts out, that's even better since then any third-party visualizer doesn't have to have two separate parsers. From your code, it looks like that's what you're doing, so awesome! I'll see it I can put out some updates to my viewer, I've got a bit of market pressure now that you've gone and put your app in the Apple store. :)

Official Folding@home development is quite slow, which is disappointing at times, but I'm really excited to see these third-party add-ons.
F@h is now the top computing platform on the planet and nothing unites people like a dedicated fight against a common enemy. This virus affects all of us. Lets end it together.
ChristianVirtual
Posts: 1596
Joined: Tue May 28, 2013 12:14 pm
Location: Tokyo

Re: First visualised Core 17 workunit

Post by ChristianVirtual »

The main difference between the standard behavior and my script is: Slot vs workunit. Else should be replaceable in format. I'm also a lazy guy don't want two parser. (I might add a zip-mode though to save bandwidth on mobile connection).

Standard API based on slot ID while my reading of the file structures are based on WU ID. Was partially too lazy to do the mapping inside the script from slot to WU; there would be not an reliable way except login remote again and get the value.

The viewer should differentiate between the standard port 36330 (for slot based trajectory) and as foreseen by the script via port 36331 for WU based data. Again: once the standard Core/clients is fixed that solution should disappear anyway.
ImageImage
Please contribute your logs to http://ppd.fahmm.net
artsci2
Posts: 2
Joined: Fri Nov 21, 2014 4:34 pm

Re: First visualised Core 17 workunit

Post by artsci2 »

Thanks for taking this project up! I would love to keep a 3D molecule portfolio with before and after versions of "my" molecules.
Post Reply