LTT Folding Team's Emergency Response to Covid-19

[El_Psy_CoNGR00]

just had a bad gate way to the forum site.

 

me: hmmm forum is down.

 

Also me:   does this mean that papa linus is linking up the new server? MWAHAHAHAHAHAH!! YES WU'S FOR EVERYONE!!

 

mabey just me going fold crazy !!

[gmouz]

1 minute ago, El_Psy_CoNGR00 said:

just had a bad gate way to the forum site.

 

me: hmmm forum is down.

 

Also me:   does this mean that papa linus is linking up the new server? MWAHAHAHAHAHAH!! YES WU'S FOR EVERYONE!!

 

mabey just me going fold crazy !!

Same for me

 

[DelphinusVyse]

12 minutes ago, Valentyn said:

 

Simply and nice, as I like'em!

Thanks. I am planning to upgrade at some point in the not so distant future so I can move up to a much newer CPU, faster RAM, and other niceties like NVMe SSDs, but I continue to be surprised/impressed that I can still play modern games on a 3 year old graphics card and a 7 year old CPU (i7-3770K, albeit overclocked to 4.2GHz). It's certainly been a system I've more than gotten my monies worth out of, considering besides a larger SSD and replacing the PSU recently, the only performance related upgrade I made was going from a GTX760 to the 1080. I did as mentioned replace the case too, the Antec 902 seemed cool in 2013, but it's a pain in the butt to clean the filters and I'd had enough.

[Shigeru]

2 minutes ago, bmichaels556 said:

Well hey, with their processing power combined, you should still be working with some serious horsepower there!

 

How many points per day are you seeing with that GT 1030? Just curious. 

Well, will tell that tomorrow since I reseted my point thanks to a problem with a duplicated folding user name, but just got this WU now

[bmichaels556]

Hey gang. I was just curious why F@H is so bad when it comes to multiple cores? I've got 12 cores / 24 threads and only get a bit less hashrate using say, 12 threads, as opposed to all 24, or even 20. Clock speed seems relatively unaffected too with those different settings, but the numbers are what they are, but with vastly less power consumption. Why?

 

Could it maybe have to do with cache or something like that where there ends up being a big diminishing returns because you're already running close to max? Same kind of situation on my i7-4600u. 1 thread gets me like 5,000+ PPD. Two gets me like ~8,500. 3 gets about the same. And 4 actually seems lower! But clock speed and cache frequency do suffer after 3+ threads which is expected I guess.

[TheAmazingDave]

Here's one of my rigs. Stats are in my profile.

 

 

Other rig is a twin-Xeon PowerEdge T610, nothing special.

 

[Valentyn]

3 minutes ago, DelphinusVyse said:

Thanks. I am planning to upgrade at some point in the not so distant future so I can move up to a much newer CPU, faster RAM, and other niceties like NVMe SSDs, but I continue to be surprised/impressed that I can still play modern games on a 3 year old graphics card and a 7 year old CPU (i7-3770K, albeit overclocked to 4.2GHz). It's certainly been a system I've more than gotten my monies worth out of, considering besides a larger SSD and replacing the PSU recently, the only performance related upgrade I made was going from a GTX760 to the 1080. I did as mentioned replace the case too, the Antec 902 seemed cool in 2013, but it's a pain in the butt to clean the filters and I'd had enough.

 

I'm still running a 5820K, some older CPUs really just keep on chugging, and I love it.
 

Holding out for Zen 3, and I'll likely get an NVME drive myself then.

[HotaruMio]

1 hour ago, Yujah said:

Yes, those don't support double precision in/through OpenCL; if you study the log you will undoubtedly see being assigned a work server 192.0.2.1 for the GPUs which is in fact an unroutable address, reserved for documentation and examples, and F@h's way of telling you that the GPU is unsupported. It's not evident in the frontend that it is because in theory the queue's available for single-precision work units, but well, those don't actually tend to exist anymore in practice.

 

Of that generation of AMD only a few top-models in the HD 58xx, HD 59xx and HD 68xx, HD 69xx series support OpenCL DP. I do believe that anything "new" does but I couldn't really tell you where the break-off point is specifically. The Nvidia types you mention would to me also seem too old but frankly I haven't payed much attention to Nvidia for many, many years now (am a Linux user) so perhaps simply trying is the best method: if you're assigned 192.0.2.1 as a work server, then you know they're not going to fly either.

Understood, guess those units can only do CPU folding then. How about R9 M275X? it seems to be picking up jobs, but it was stuck at FAHCore, I try resetting the GPU value to 0, and it seems to be running, but it resets and returned back to ready. I already installed the latest driver from AMD.

[Cas_DS]

Hy, I'm new but I want to do my part. Can you run multiple computers under the same account on the same assignment with Folding@home?

 tnx

[Rune]

1 minute ago, Cas_DS said:

Hy, I'm new but I want to do my part. Can you run multiple computers under the same account on the same assignment with Folding@home?

 tnx

Yes! Edit: well, not the same WU, but the same type of WU yes.

[HotaruMio]

Just now, Cas_DS said:

Hy, I'm new but I want to do my part. Can you run multiple computers under the same account on the same assignment with Folding@home?

 tnx

Hello. You can definitely run multiple computers under your account for the team. But the assignments are assigned randomly. Just keep folding with whatever you are given. Anything counts.

[Shigeru]

3 minutes ago, Valentyn said:

 

I'm still running a 5820K, some older CPUs really just keep on chugging, and I love it.
 

Holding out for Zen 3, and I'll likely get an NVME drive myself then.

Indeed! I used for years an Intel E2180 and still missing that old machine... now my oldest running one is an AMD A8 and after dealing with its termal issues is just on all day

 

And that make me wonder if its worth to try and Old Athlon64+.... could be worse that the old Intel Atom? Uhmm.... and a GeForce 8400 GS? (I dont think what that one is support, and would melt anyway XD )

[Shigeru]

14 minutes ago, Shigeru said:

Well, will tell that tomorrow since I reseted my point thanks to a problem with a duplicated folding user name, but just got this WU now

Updated now to more credits? Well, still new in this XD

[DelphinusVyse]

1 minute ago, Shigeru said:

Indeed! I used for years an Intel E2180 and still missing that old machine... now my oldest running one is an AMD A8 and after dealing with its termal issues is just on all day

 

And that make me wonder if its worth to try and Old Athlon64+.... could be worse that the old Intel Atom? Uhmm.... and a GeForce 8400 GS? (I dont think what that one is support, and would melt anyway XD )

Heh, I remember those days. I had an Athlon 64 X2 and an 8800GTS 512. Ended more or less giving the 8800 to a friend who was building a PC on a very tight budget and when the thermal paste finally crapped out and the thing cooked itself, EVGA ended up sending us a GT650 as a replacement. Man I miss those lifetime warranties.

[HaRdLy007]

This is my contribution.

CPU Cores: 40

GPUs:          2

 

Still trying to figure out how to run the client in Manjaro Linux without having a term window open, as well as configuring the client to allow remote tracking.

 

Cheers!

 

Jeff

 

[ViscountStyx]

50 minutes ago, J1mjam said:

Jesus, I've managed to bag a CPU WU that is estimated to take 10 cores of a 9900k 13 hours to complete 

 

EDIT: Its not a COVID one. It's for project 13870 and has a MASSIVE 800 base credit.

Think I just got the same!

 

8.7% through - stats are for an OC 8700k

 

[Shigeru]

1 minute ago, DelphinusVyse said:

Heh, I remember those days. I had an Athlon 64 X2 and an 8800GTS 512. Ended more or less giving the 8800 to a friend who was building a PC on a very tight budget and when the thermal paste finally crapped out and the thing cooked itself, EVGA ended up sending us a GT650 as a replacement. Man I miss those lifetime warranties.

That made me think in my EVGA GeForce 9600 GSO 768mb... used that card for years in the E2180 until I retired it because the cooler was failing... maybe I could try and repair that.

[bmichaels556]

1 minute ago, HaRdLy007 said:

This is my contribution.

CPU Cores: 40

GPUs:          2

 

Still trying to figure out how to run the client in Manjaro Linux without having a term window open, as well as configuring the client to allow remote tracking.

 

Cheers!

 

Jeff

 

Really impressed by the RX 5700 series' performance from the numbers I've seen. 

 

I do wonder what a Fury or Radeon VII could do, but I'm not sure whether F@H can benefit from memory the way crypto mining for example can..

[J1mjam]

53 minutes ago, J1mjam said:

Jesus, I've managed to bag a CPU WU that is estimated to take 10 cores of a 9900k 13 hours to complete 

 

EDIT: Its not a COVID one. It's for project 13870 and has a MASSIVE 800 base credit.

So that super slow WU I got, I decided to reduce the number of cores in that slot to 4, and create a new 6 core slot for something else to run at the same time.

 

Turns out reducing it from 10 to 4 cores has had ZERO impact on its TPF...like what?

 

  EDIT: for visibility for @ViscountStyx

2 minutes ago, ViscountStyx said:

Think I just got the same!

 

8.7% through - stats are for an OC 8700k

 

[DelphinusVyse]

1 minute ago, Shigeru said:

That made me think in my EVGA GeForce 9600 GSO 768mb... used that card for years in the E2180 until I retired it because the cooler was failing... maybe I could try and repair that.

Yeah, I really wonder what they'd deem as an equivalent replacement at this point if you called them. Haha.

[bmichaels556]

7 minutes ago, Shigeru said:

Updated now to more credits? Well, still new in this XD

Yeah, the reported numbers are a bit weird. I'm not sure exactly how it comes up with that PPD number. Current work unit? Processing power over the last x-amount of minutes? I've found it hard to tell when changing a setting actually shows the correct number on the client.  

 

Also, that number is crazy low for a GT 1030. At least, I think it is? But it might just be that the full number hasn't shown yet.

[bmichaels556]

4 minutes ago, J1mjam said:

So that super slow WU I got, I decided to reduce the number of cores in that slot to 4, and create a new 6 core slot for something else to run at the same time.

 

Turns out reducing it from 10 to 4 cores has had ZERO impact on its TPF...like what?

 

  EDIT: for visibility for @ViscountStyx

Yeah, see that's what bugged me, so it doesn't seem fully dependant on CPU cores at least.

 

Are you seeing ~2x performance splitting your cores into two jobs, as opposed to the small gains using them all on one job?

[Shigeru]

2 minutes ago, bmichaels556 said:

Yeah, the reported numbers are a bit weird. I'm not sure exactly how it comes up with that PPD number. Current work unit? Processing power over the last x-amount of minutes? I've found it hard to tell when changing a setting actually shows the correct number on the client.  

Indeed I have noticed what even the ETA changes sometimes at random points... I suppose what that depends in how the calculations are doing?

[GOTSpectrum]

Update time!

Averaging more than 100M Point Per Hour during this update!

 

[ViscountStyx]

Oh nice just noticed we're up to 5th on the teams overall (all time):