Yes has been more reliable lately, I have had a run of problematic beta projects over the weekend that worked well in Linux and not so well in windows which dragged my daily total down a bit.

Hooray we are now number 33 in the team rankings and in a day or so number 32. Then we will be bumped down the ranks by a bunch of corporate, institutional and other new mass output teams probably 4 or 5 places. Oh well lots of science getting done and amazing effort by us as a bunch of strangers on the net.

It looks like we have about 36 hours to bask in the glory of being 32nd overall in the team standings... enjoy it while you can!

SAK917 wrote:It looks like we have about 36 hours to bask in the glory of being 32nd overall in the team standings... enjoy it while you can!

Yes! number 32 for a moment, probably a year to get back to this.

32nd spot ! Amazing and likely quite ephemeral.. Back to 33rd tomorrow and expect dropping 4 more spots in the next month by the predictions. Oh well, we have new team members and our output remains more than double our previous even with the ongoing workunit shortages.

So looks like CERN has 10x the daily output that we do. Their number one folding account is called CMS Experiment, so I did a little digging:

https://home.cern/news/news/cern/cern-contributes-computers-combatting-covid-19

CERN is contributing about 10 000 computer cores from its main data centre. However, this only accounts for about a third of the ‘work units’ the CERN team has completed for Folding@home. The rest comes from contributions made directly by LHC computing sites. Taken all together, the CERN and LHC computing team has been the 87th biggest contributor as of 21 April.

A little more info on what CMS actually is can be found here:
https://home.cern/science/experiments/cms

I will run out and buy a new GPU... that should keep us ahead of the CERN folks

10,000 computer cores and that only accounts for about a third of the WUs they produce? Do they have so much computing power that they can afford to throw all this capability at FAH and still do the work they need to support the LHC?

I just checked and it looks like CERN/LHC is down for now with "a gradual restart as of 18 May". They aren't planning to be fully online until mid-September, coronavirus permitting. Maybe we will be sucked into their slipstream as they pass us, though I was hoping the same for NVIDIA's Saturn V when they blasted past us... strangely it didn't happen?

Yes some big movers have overtaken us zoom.. all good for science. I found myself a cheap open box GTX1660ti twin fan which I squeezed in beside my GTX1080ti this a nice little card and manages about half the production of the 1080ti. I also updated this old system to SSD not HDD, (no MS2 slot) and windows 10 put the radiator for the 1080ti to expel through the top of the case and cleaned and reset the cpu cooler during the workunit shortage so that was something I had been putting off for a year or so.
Our teams production is still 4 x what it was at the beginning of the year and we have a new super folder called AJM cracking 5 million ppd as well as 18 members who do more than a million. This really extraordinary.

I see despite losing team members after the initial COVID-19 rush we are continuing to gain some. I popped a link to this thread on my signature on the F@H helpforum and wonder if that getting attention or it just people scrolling through the top 100 teams and thinking "hey that's me".

I can't remember what our active count was pre-covid (and the graph does not go back far enough). Do you remember what we were averaging previously? 485 (at current count) seem like a lot compared to what we might have had historically.

I quality for the "hey, that's me" classification...

It is an interesting question, I wonder if the person running the ExtremeOverclocking.com stats site would consider a request to have the Active Users chart go back 2-3 months instead of just one? I suspect it wouldn't be too hard a change to make and that others would be interested as well. Do you have any contact with that site kiore?

SAK917 wrote:I quality for the "hey, that's me" classification...

It is an interesting question, I wonder if the person running the ExtremeOverclocking.com stats site would consider a request to have the Active Users chart go back 2-3 months instead of just one? I suspect it wouldn't be too hard a change to make and that others would be interested as well. Do you have any contact with that site kiore?

I have interacted with Jason in the past, but you can go to his site https://folding.extremeoverclocking.com/ make suggestions/donations. I suspect it costs to hold info like that in server space.

DKeane wrote:I can't remember what our active count was pre-covid (and the graph does not go back far enough). Do you remember what we were averaging previously? 485 (at current count) seem like a lot compared to what we might have had historically.

I don't know that such a good measure as is distorted by config changes and core was from the days when 1 core meant 1 computer, now it probably means 1 slot. I for example have currently 1 slot= i5 quad 7200 laptop, and 1 slot= 16 core Ryzen 3700x desk top really not comparable. I think ppd still the best measure and we used to average just under 20mppd currently it is 82mppd

The F@H help forum is currently down is showing account suspended message but is actually everyone affected.

I get the same message on the F@H help forum kiore.

SAK917 wrote:I get the same message on the F@H help forum kiore.

It a general failure of that site and they are aware and trying to fix it.

Folding at home help forum is now back up.

Having become very tired of my RTX 2080ti's temperature issues and having to always power limit and manage the card to stay under 80c. So I bought myself an on sale open box hybrid conversion kit to liquid cool it. As open box it had no instructions, so went looking for online how to advice. Gulp.. first instruction is to remove 20x small screws.... An inauspicious first instruction. Feeling pretty confidant I can do the conversion looks like what you have to do to swap a cpu cooler with extra stuff to fix the GPU case so I will see this weekend. Bought some fresh thermal paste and hope that this conversion to liquid cooling will drop the full power temps from 80+ C to more like 60 C in time for the summer.

What temperature would your 2080ti reach if you didn't power limit kiore? I have a stock 2080ti (slightly overclocked from the factory) and it is humming away right now with the fans at 75% and the temps at 79-80C. Ambient temp in my office is about 21-22C. According to GPU-Z my GPU load is 91-96% which is about as high as I typically see in GPU-Z... other utilities like Task Manager or HWMonitor indicate the GPU load is a solid 100%, not sure why GPU-Z is slightly lower. This seems to be a good workunit for my card (more atoms?) as I can see the WU PPD estimate vary from about 1.8M to something like the current WU estimate of 3.4M PPD. So the temps I am experiencing right now are about worst case for me.

Good luck on the liquid cool install... will be curious to hear how it goes!

The RTX has always had cooling issues, I swapped it to a case with high airflow and 4 x140mm case fans plus run it at 80% card fans and even so when the ambient goes above 20c the card temp hits the throttling temps for the card at 85c by power limiting to 82% I can keep it under 80c . Better airflow in the room would improve things but I live in an apartment in Manhattan so space issues.