Folding Forum

I do remember that there was a warning in the installation instructions regarding having a properly cooled system before the migration to the new site. I can't find it right now.

My personal take is if you pick up a branded system from a shop, you shouldn't have to worry about cooling, or anything. If you want to run it under full load 24/7, it should be able to do it. If not, it's a faulty system and ask for RMA. The other side is for those that invest time and money into building their own system instead of buying a branded system. For that group, they would have enough PC knowledge to know what they are doing with their hardware.

Adding a warning was originally proposed in V7 but wasn't accepted due to the fact that a properly build system (branded or custom) should handle working under full load 24/7. It's the minority of systems that have cooling issues so there's no point in placing that warning when majority of systems will work fine.

I would also add, those that are dissuaded by hiccups are likely to quite sooner rather than later anyway. Some people just get riled up at problems and throw up their arms, others roll up their sleeves and figure it out for the joy of the thing.

I dont think F@H needs to solve for the former group, imo.

treckin wrote:I dont think F@H needs to solve for the former group, imo.

They do if they want as many contributors as possible. There are people who have the resources and the will to be a donor, but no time to actually fiddle with issues.

No software will damage a modern computer. Al cpus and modern gpus have thermal protections built in that throttle them and downclock them. An NVidia GPU for example will run at 80C for basically ever without breaking and the gpu's BIOS will keep it from going over that safe temperature. Same with cpus. They downclock if they get too hot.

The kind of people who need to be told that scientific software will heavily load their computer are also the kinds of people who ask questions instead of reading FAQ or who attack the project because their pride and joy overheats but plays all their games perfectly.

This project has been running for over fifteen years and now suddenly everyone wants to tell it how to run itself.

The kind of people who need to be told that scientific software will heavily load their computer are also the kinds of people who ask questions instead of reading FAQ or who attack the project because their pride and joy overheats but plays all their games perfectly.

Very Much True.

Running FaH again after several years in dormant, I find it funny and amused at the same time, individuals with medium to high end specs, cpu or gpu wise, bragging this & that but can't handle folding. Mainly because of too much gaming in their heads or have a faint heart to handle sustained work loads. Yet, these individuals for the lack of urgency (this current situation we're into) or common sense for that matter, are the ones who complains a lot or have an attitude to endorse a certain something that is meant to be "his is better than yours", lolz!

And come to think of it, how many times I have "endorsed", "challenged" that type of individual/s to try folding and make a difference. It's either they quit before even trying or after knowing data crunching is very different from their accustomed PC usage. Folding thus challenges your PC know-how to the max'ed degree, on hardware and software you're into.

AmunRa wrote:

The kind of people who need to be told that scientific software will heavily load their computer are also the kinds of people who ask questions instead of reading FAQ or who attack the project because their pride and joy overheats but plays all their games perfectly.

Very Much True.

Running FaH again after several years in dormant, I find it funny and amused at the same time, individuals with medium to high end specs, cpu or gpu wise, bragging this & that but can't handle folding.

And usually these same indivuduals are the ones who neglect their PCs the most. After a year of neglect, their heatsinks will look like they mistook their PC for a vacuum and cleaned their house with it. No wonder it overheats when they put a 24/7 load on it.

This is a developers project so in a way we all play a part in figuring things out. We are all feeling our way here and it is this board that brings that knowledge together. No point moaning about the software, learn how to make it run by experimentation and by asking questions. Keep in mind the good causes that we are helping to fight.

Cooling, oh yes, you will need cooling. You are going to floor the throttles on all your processor cores and they will need as much cooling as they can get. Which means you will need to check any bios settings on thermal config and set it to HIGH / MAX / ADVANCED, whatever can be found. You will need to service the machine regularly to clear the heatsinks, fans and airways of dust. You will need to get the machine low to the floor where the air is coolest, ensure plenty of place around the machine. Fluid cooled machines will need their heat exchangers checked regularly. You may need to upgrade your CPU cooler as you are pushing the machine where it wasn't originally designed to go.

My DL380 servers are designed to be thrashed, each have six high flow fans to cool the processors, RAM and onwards to the motherboard. They can hit 80% in the height of the day, which sounds like a B52 on flight idle and is damaging to the hearing. I actually don't run them when it is really hot, even though I have a large fan drawing outside air, through ducting, straight into the 19" rack and forcing it into the server intakes. Power, my system hits a good kilowatt so that's a lot of heat to get rid of and the room can get to 30 degrees C (Mid 80s F), even in good old Blighty.

I thought about the title The worst software I have ever seen.

I once had to support an application that tried to print a log that indicated that the printer was out of ink. (You could not turn this off as the software was designed to log security events)

You had to buy at least two ribbons, as the first ribbon would be used up printing that the printer was out of ink, over and over again.

The worst software I ever wrote had a bug in the error reporting code, so if there was any bug, it reported the one in the error reporting routine, not the one that caused you to go there.

I recall a Windows 3.11 for Networks error message that was particularly unhelpful at the time (back in the days of USR 14,400 fax/modems) along the lines of:

"Your modem driver is corrupted … Please download new driver"

Neil-B wrote:..."Your modem driver is corrupted … Please download new driver"

It was a future-proof error message since you can now have the same error message and use Wi-Fi or mobile data to get the new driver

iceman1992 wrote:

treckin wrote:I dont think F@H needs to solve for the former group, imo.

They do if they want as many contributors as possible.

And that is the root cause of all current problems. Too many people folding. This is a small project, run by a small number of people. Ideally suited to maybe 50% less folders than are currently running the software. At the 1 exaFLOPS level things were workable. Now its a complete SNAFU. No workunits most of the time, servers full or crashing. This situation will continue to get worse until it reaches the failure point and fully falls apart.

You have to remember they have one paid programmer / developer to deal with all the infrastructure and a bunch of scientists to deal with the work, not all of which are paid and all probably have families and other real life issues to deal with. So everyone should cut them some slack.

My suggestion is all the people who are moaning just shut down their folding instances, then it will improve for the rest of us in time. I've been doing this for a long time. im not some fly-by-nite parttimer that will be gone in 3 months or will get bored of things. but the whole lack of workunits does irk me a lot. i can also see the root cause of the problem (too many people folding) and the solution (shutdown registration of new folders). you can upgrade over and over, but this ship has been sinking for a long time, slowly in the past, rapidly in the present. I like to think someone will step up and stabilise the situation, but the realist in me says its unlikely.

v00d00 wrote:This is a small project, run by a small number of people.

Any chance of growing the project? I don't know how their funding works, grants or sponsorships maybe?
I just get really bothered thinking of the potential for growth and what they can do with more flops, only to be held back by the small team size - as you said only 1 paid dev. I haven't watched the recording of the dev chat, but if there's an opportunity to help out I will volunteer as a dev.

Its down to experimentation. For example. if you download the latest client then you get version 7.5.1. However if you download the Linux server version you get 7.4.4. So I have manually installed 7.5.1 on the Linux servers. No instruction to do this but I figured it out myself. Now I could have got on here and moaned as to why they don't fix their website. Well, actually I think they are better off trying to fix "Us" with their protein research, which I have absolutely no clue about. I do have a clue about Linux and I can get my machines to their highest performance, with a little help from the community. Some people need to start thinking about "how to fix" rather than "Who to blame".

Darmain wrote:Its down to experimentation. For example. if you download the latest client then you get version 7.5.1. However if you download the Linux server version you get 7.4.4. So I have manually installed 7.5.1 on the Linux servers. No instruction to do this but I figured it out myself. Now I could have got on here and moaned as to why they don't fix their website. Well, actually I think they are better off trying to fix "Us" with their protein research, which I have absolutely no clue about. I do have a clue about Linux and I can get my machines to their highest performance, with a little help from the community. Some people need to start thinking about "how to fix" rather than "Who to blame".

A few things:
1. I'm running on linux too, what's the difference between 7.5.1 and 7.4.4? I'm currently running 7.4.4
2. How do I get 7.5.1?
3. Yes I agree, I am also thinking about "how to fix" which is why I would love to help with the dev side of things, to contribute in more ways other than being a compute donor.

1. Changelog: viewtopic.php?p=301620#p301620

2. All installers are at https://foldingathome.org/alternative-downloads/ . If you've used the Advanced Installation steps from https://foldingathome.org/support/faq/i ... -advanced/, you can obtain the URLs for the 7.5.1 version of the installers. You can poke around at https://download.foldingathome.org/rele ... c/release/

3. A number of FAH components are on Github. However, please do check with the repo owner if they accept any major rewrites before you invest significant time.
Folding@home:
FAH Web Control - https://github.com/FoldingAtHome/fah-web-client
FAH Advanced Control - https://github.com/FoldingAtHome/fah-control
FAH Protein Viewer - https://github.com/FoldingAtHome/fah-viewer
FAH Screensaver - https://github.com/FoldingAtHome/fah-screensaver
FAH Windows Installer - https://github.com/FoldingAtHome/fah-cl ... -installer
FAH Client - Arriving soon (No ETA at this time)
FAH Science Cores (FAHCore) - Not available at this time
FAH Website - Not available at this time
FAH Forum - Not available at this time

Third party software:
HFM.NET - https://github.com/harlam357/hfm-net
FAHBench - https://github.com/fahbench/fahbench

Upstream science cores:
Gromacs - https://github.com/gromacs/gromacs
OpenMM - https://github.com/openmm/openmm