Folding Community Board

[Bitter]

Well. I feel called out.

1650 is off, pulling it this weekend and tuning the P106 for max long term efficiency.

[Collin_S]

1 hour ago, Bitter said:

Well. I feel called out.

1650 is off, pulling it this weekend and tuning the P106 for max long term efficiency.

Being more efficient doesn't necessarily mean just having less power usage. It means to do more per unit of power. While other people with similar sized houses are using less energy, they are probably also not doing as much with it (like folding). More efficient folding can be good though.

Of course, your power company more than likely doesn't know about folding... Imagine if Power Company's had Folder's Discounts. @Den-FiWould LOVE that.

[GOTSpectrum]

2 hours ago, Bitter said:

Well. I feel called out.

1650 is off, pulling it this weekend and tuning the P106 for max long term efficiency.

The trick is to use so much they send you thank you notes, at least, that's the method @Den-Fiuses

[justpoet]

3 hours ago, GOTSpectrum said:

The trick is to use so much they send you thank you notes, at least, that's the method @Den-Fiuses

"Thank you for funding our new high power transmission line.  It was only possible because of your contribution."

[SGRacing]

OK, is anyone else having problems with CPU WU uploads. I just updated to the latest version and I'm about to go back because it keeps hanging up. I have 4 WUs stuck on my second client right now. The failing server is 130.237.11.145:80 also failing port 8080.

[Bitter]

3 hours ago, justpoet said:

"Thank you for funding our new high power transmission line.  It was only possible because of your contribution."

 

7 hours ago, GOTSpectrum said:

The trick is to use so much they send you thank you notes, at least, that's the method @Den-Fiuses

This is Illinois, they would just use it to lobby and bribe state politicians.

https://www.nbcchicago.com/news/local/chicago-politics/4-charged-in-alleged-comed-bribery-scheme-plead-not-guilty/2384654/

[LazyDev]

The work units that I've been getting, seems to be attempting to replicate the Tesla logo.

 

[GOTSpectrum]

WUs seem to be okay for me today, long may it last

[wall03]

What?! I usuallu get ~400k/day, not 1.15M!

 

Well, we'll see how it goes today. Need to get that 25M before the end of the year

[Happiness_is_Key]

Hi everyone! I have some questions about folding if you don't mind me asking. I've been using an oldish laptop for folding over the course of 2020 as I've basically been inside since March,, so I had to find something to do that also benefited someone other than me. I'm glad I found folding (from an LTT video)!

 

Anyways, I'm folding on a 2 core CPU (i5-5200U) and remote into when needed. I have almost 2,000,000 points from it (100,000 from another computer, but I've since stopped it on that one). If I were to change out the system in the future with a desktop system, what would be better for the cause and energy consumption GPU or CPU folding? What's the difference? I know quite a bit about CPUs, but I'm not familiar with GPUs all that much (I'm trying to be a sponge on the subject, but don't understand it yet). HWiNFO does say the CPU it limited by the power as the package power for cores 0 and 1 have been reached. The base clock is 2.20GHz and I'm between 2.46-2.49GHz usually.

 

Any tips on what I should do if anything? I average roughly 15,000 PPD with both cores firing 24/7. According to HWiNFO the max temps I get are 68ºC (154.4 ℉) with an average of 65ºC (149ºF). The fan runs at an average of 3,327 RPMs (over a period of about 1 hour). I have shucked the laptop from its black casing to make the temperatures better. It's just sitting on a self in the closet with no door so it has okay airflow. The room temperatures are kept from 21.11ºC (70ºF) to 23.89ºC (75ºF) throughout the day.

 

Let me know if you have any questions or things I can clear up.

 

Also, while I'm here, how do a get the badges that some people have? I saw something somewhere about requesting it from a staff member, but I have no idea where to do that (new to the forum). I don't have to have it, but I think it's pretty cool.

 

Thank you!

[LazyDev]

3 hours ago, Happiness_is_Key said:

Hi everyone! I have some questions about folding if you don't mind me asking. I've been using an oldish laptop for folding over the course of 2020 as I've basically been inside since March,, so I had to find something to do that also benefited someone other than me. I'm glad I found folding (from an LTT video)!

 

Anyways, I'm folding on a 2 core CPU (i5-5200U) and remote into when needed. I have almost 2,000,000 points from it (100,000 from another computer, but I've since stopped it on that one). If I were to change out the system in the future with a desktop system, what would be better for the cause and energy consumption GPU or CPU folding? What's the difference? I know quite a bit about CPUs, but I'm not familiar with GPUs all that much (I'm trying to be a sponge on the subject, but don't understand it yet). HWiNFO does say the CPU it limited by the power as the package power for cores 0 and 1 have been reached. The base clock is 2.20GHz and I'm between 2.46-2.49GHz usually.

 

Any tips on what I should do if anything? I average roughly 15,000 PPD with both cores firing 24/7. According to HWiNFO the max temps I get are 68ºC (154.4 ℉) with an average of 65ºC (149ºF). The fan runs at an average of 3,327 RPMs (over a period of about 1 hour). I have shucked the laptop from its black casing to make the temperatures better. It's just sitting on a self in the closet with no door so it has okay airflow. The room temperatures are kept from 21.11ºC (70ºF) to 23.89ºC (75ºF) throughout the day.

 

Let me know if you have any questions or things I can clear up.

 

Also, while I'm here, how do a get the badges that some people have? I saw something somewhere about requesting it from a staff member, but I have no idea where to do that (new to the forum). I don't have to have it, but I think it's pretty cool.

 

Thank you!

Welcome!

 

GPU's are far more efficient at folding. For example, the RTX 2080 Super can do at least 2.7M points per day. CPU folding on the other hand, is far slower, as shown from Lar_systems's database, https://folding.lar.systems/cpu_ppd/overall_ranks. Some GPU's are becoming obsolete, so again, you can look at the lar_system's database to find a suitable  GPU if you wish to give your contributions a boost. The RTX 2060 at this time has the best points per watt consumed, but as the new 3000 series comes out, that may change.

 

Bare in mind, using a GPU for folding, you have to watch out for the PCIe bandwidth, as the a CPU thread is also used for calculations for the GPU, so at least 1 thread is required (For Nvidia GPU's).

 

As for the badges, all the information that you need to know is on this thread, 

 

[Favebook]

4 hours ago, wall03 said:

What?! I usuallu get ~400k/day, not 1.15M!

 

Well, we'll see how it goes today. Need to get that 25M before the end of the year

You probably got awarded with lost points that were your all along but weren't awarded at the time. 

 

OR, someone folded for you .

[Bitter]

1650 is out, back into coolbits xorg.conf hell...

[phyrealarm]

Honestly, I forgot to check. I've only done 8 mil because I forget to leave the dang PC on to run. 

[Captainmarino]

Well, it looks like my R9 280 that has been limping along on 1 out of 2 fans is trying to keel over for good. It all of the sudden wants to creep it's temp up past 100 no matter what I do. I'm looking for a cheep/used card to replace it with. Thinking a GTX 970 or 1060 might be a good spot. Something at or below $100. I was hoping to get a bit more life out of the R9 before it crapped out on me

[GOTSpectrum]

Oh jeez I'm not been here for a few days. You guys are awesome though. We gotta keep it up and we will smash Curecoin and take the real number one spot! 

[Bitter]

1 hour ago, Captainmarino said:

Well, it looks like my R9 280 that has been limping along on 1 out of 2 fans is trying to keel over for good. It all of the sudden wants to creep it's temp up past 100 no matter what I do. I'm looking for a cheep/used card to replace it with. Thinking a GTX 970 or 1060 might be a good spot. Something at or below $100. I was hoping to get a bit more life out of the R9 before it crapped out on me

Re-paste it? Stick case fans on it?

 

I have 2 80x15mm fans on my P106-100 because it came without fans or shroud, they're held with screws jammed into the fins. I have a GTX 650 (not folding) that the original fan lost a blade on, so I shucked the whole fan/shroud and zip-tied a 92x15mm case fan to it. Both cards run cool and quiet this way, the GTX 650 runs cooler than it ever did before, the P106 I don't have a previous referenced for. What is nice though, on the P106 I used a video card to case fan adapter. If I wanted, I could connect the fans to the motherboard or an external fan controller for manual control to run them at a set speed or whatever since they're standard fans. The 650 had a weird connector so I just soldered/heat shrunk.

[Happiness_is_Key]

On 12/5/2020 at 11:31 AM, LazyDev said:

Welcome!

 

GPU's are far more efficient at folding. For example, the RTX 2080 Super can do at least 2.7M points per day. CPU folding on the other hand, is far slower, as shown from Lar_systems's database, https://folding.lar.systems/cpu_ppd/overall_ranks. Some GPU's are becoming obsolete, so again, you can look at the lar_system's database to find a suitable  GPU if you wish to give your contributions a boost. The RTX 2060 at this time has the best points per watt consumed, but as the new 3000 series comes out, that may change.

 

Bare in mind, using a GPU for folding, you have to watch out for the PCIe bandwidth, as the a CPU thread is also used for calculations for the GPU, so at least 1 thread is required (For Nvidia GPU's).

 

As for the badges, all the information that you need to know is on this thread, 

 

Thank you! I've been a long time viewer (Watched probably half or so of the videos), but just decided to take the plunge.

 

Why are GPUs more so much more efficient than CPUs in this case? I've seen the F@H viewer, but I have no idea what F@H actually does other than max out your hardware and that it's good for contributing to society/science. Is it because it's actually a GPU process being translating into a CPU process as more people have CPUs than dedicated GPUs? I like to know the 'why' behind things. Haha.

 

I was getting ready to go for a 2060 SUPER, but I'm not sure. The prices of everything are nuts, so I might hold off for a bit. That 3090 though is insane. It does more in 6 hours then I've done in almost 6 months (I took the 200,000 out that my MBP did in 2 hours). Woah....

 

Lar Systems did a great job with that website. I also really like the extremeoverclocking.com folding subdomain website as well. Very cool! Both these I found here in the LTT Forums.

 

I'm not sure how threads work in a CPU. I've been researching it, but don't fully understand it quite yet. I thought PCIe that if you had (usually) 16x lanes for your GPU, you'd be fine on bandwidth. Is that incorrect?

 

I'll make sure to check out that thread you linked. Thank you!

[piratemonkey]

26 minutes ago, Happiness_is_Key said:

Why are GPUs more so much more efficient than CPUs in this case? I've seen the F@H viewer, but I have no idea what F@H actually does other than max out your hardware and that it's good for contributing to society/science. Is it because it's actually a GPU process being translating into a CPU process as more people have CPUs than dedicated GPUs? I like to know the 'why' behind things. Haha.

From what I understand, GPUs are much better at smaller tasks and folding is a lot of small things. That's why CPUs give lower points; they aren't as efficient. As for what folding is, it's simulating folding proteins. It's just your computer doing this or that, and seeing what happens. I'm not too sure on the specifics, so I don't know much else from that. 

Also, welcome to the forums and the folding team. Always nice to get new people

[LazyDev]

1 hour ago, Happiness_is_Key said:

I'm not sure how threads work in a CPU. I've been researching it, but don't fully understand it quite yet. I thought PCIe that if you had (usually) 16x lanes for your GPU, you'd be fine on bandwidth. Is that incorrect?

The more threads you are able to throw at a single work unit, the quicker the work unit can be processed.

 

In terms of the PCIe bandwidth, four lanes is typical sufficient. For example, I have a bifurcation card on my 16 lane slot, which has allowed me (Via the motherboard bifurcation support) to chain four GPU's onto one slot.

[Happiness_is_Key]

Quote

From what I understand, GPUs are much better at smaller tasks and folding is a lot of small things. That's why CPUs give lower points; they aren't as efficient.

 

Ah, that makes sense. That looks like a big margin for efficiency according to folding.lar.systems if you compare the two (yes, I do know those are based on user stats).

On 12/6/2020 at 11:19 PM, piratemonkey said:

As for what folding is, it's simulating folding proteins. It's just your computer doing this or that, and seeing what happens. I'm not too sure on the specifics, so I don't know much else from that. 

 

Okay, that's what I've heard as well. I do wonder what the success rate of that is meaning how many of the projects turn into something rather than eliminating something. To be clear, I'm don't mean to be critiquing anything. I just like to know how things work.

On 12/6/2020 at 11:19 PM, piratemonkey said:

Also, welcome to the forums and the folding team. Always nice to get new people

Thank you! It's a pleasure to be here, and I thoroughly enjoy it! 

 

[Happiness_is_Key]

1 hour ago, LazyDev said:

The more threads you are able to throw at a single work unit, the quicker the work unit can be processed.

 

Yes, thank you. I should've been more clear, my apologies. I do know what a CPU core is, but I'm not sure how CPU threads work. I know that it's normally double the cores (8 cores = 16 threads), but I'm aware as of why that is.

Quote

In terms of the PCIe bandwidth, four lanes is typical sufficient. For example, I have a bifurcation card on my 16 lane slot, which has allowed me (Via the motherboard bifurcation support) to chain four GPU's onto one slot.

 

It looks like I need to do some reading on that, but I understand the general idea. That's super cool! You must have a big case/server, or it must not be in one. Haha. So GPUs are generally only x4 lanes? For example, my AMD 560X uses x8 lanes. What's that based off of?

 

Edit: Just looked at your build. That's very nice! You made sure that every inch had something useful in it.

Edited December 7, 2020 by Happiness_is_Key

[LazyDev]

Just now, Happiness_is_Key said:

It looks like I need to do some reading on that, but I understand the general idea. That's super cool! You must have a big case/server, or it must not be in one. Haha. So GPUs are generally only x4 lanes? For example, my AMD 560X uses x8 lanes. What's that based off of?

GPU's come with 16 lanes by factory. However, they will only use the lanes that have been given, either by software (Bios) or by the connector/adapter that they're plugged into.

[piratemonkey]

36 minutes ago, Happiness_is_Key said:

Okay, that's what I've heard as well. I do wonder what the success rate of that is meaning how many of the projects turn into something rather than eliminating something. To be clear, I'm don't mean to be critiquing anything. I just like to know how things work.

I think (not too sure) fah helped create the spike model for covid19. Relatively speaking, folding is new (created around the start of 2010s maybe) so the results that may have come from it are probably not fully realized yet. I'm not aware of any other big results that have come folding. But, it is worth noting that the network was faster than the fastest supercomputer at its peak this year (around March; when the call to fold was put out by LTT, Nvidia etc.). It has since fallen quite a bit, but it's still really fast. Like one of the fastest computing projects. I'm sure further understanding of diseases and ailments bring researched will be realized. Several universities afaik use the network after all. And don't worry about offending anyone. Even if you do, it's for a good reason. 

[rkv_2401]

31 minutes ago, Happiness_is_Key said:

 

Ah, that makes sense. That looks like a big margin for efficiency according to folding.lar.systems if you compare the two (yes, I do know those are based on user stats).

 

Okay, that's what I've heard as well. I do wonder what the success rate of that is meaning how many of the projects turn into something rather than eliminating something. To be clear, I'm don't mean to be critiquing anything. I just like to know how things work.

Thank you! It's a pleasure to be here, and I thoroughly enjoy it! 

 

@piratemonkey

 

Welcome! It's super impressive that you've managed to score ~2 million points via just CPU folding. I was in a similar range when I resumed folding last year - my laptop's GPU only made 35k PPD running 24/7, and it seemed like even a million points would be practically undoable(I found it hard to sleep at night with the fan whine). But here I am now, having contributed just over 100 million points to the cause. :)

 

I'm a bit rusty on the topic, but I believe all the projects turn into something as the scientists only create projects based on compounds they need to predict and filter prior to experimental validation. I heard from a friend who works in the field that waiting for computers to finish processing is one of the bigger delays in such research, so they have a clear incentive to send WUs of the highest priority first.

 

For a single graphics card(like the 2060 Super you're considering), the no. of PCI-e lanes discussion isn't really relevant. Higher lanes means faster speeds/more bandwidth. One lane of PCI-e 3.0 can transfer data at upto 8 GT/s, so 4x can do 32GT/s, 16x can do 128GT/s, etc. It needs to be higher than 4x, yes, but there's no reason for you to set it to anything lower than the factory-default 16x. Lowering that may handicap your performance in other applications, such as gaming, anyway. These details are important when you're trying to create a machine with more GPUs than there are (high-speed) PCI-e slots, such as @LazyDev's F@h build. 

 

CPUs that have hyper-threading enabled have 2 threads for every CPU core, but a 2C/4T CPU isn't the same as having 4 actual cores. I believe the difference is that the extra threads can't all do floating point calculations at the same time, but the end difference is that the performance gain with HT on is around 30-40% compared to that with HT off for a workload where a CPU with 2x the cores would be roughly 2x faster.