LTT Official Folding Month 2020!!!

[Arrogath]

24 minutes ago, Den-Fi said:

Idk if you've ever sat in a tree, but it's not super comfortable. Can you pick another place?

like the back of a Volkswagen?

[Guest]

1 minute ago, Arrogath said:

like the back of a Volkswagen?

I've come to the conclusion that it's not where that matters, it's who you're with that makes it comfortable.

The tree was fine.

[GOTSpectrum]

1 minute ago, Den-Fi said:

I've come to the conclusion that it's not where that matters, it's who you're with that makes it comfortable.

The tree was fine.

SEE!!! I told you, you would enjoy the tree 

[Egg-Roll]

1 hour ago, GOTSpectrum said:

SEE!!! I told you, you would enjoy the tree 

Or is the tree uncomfortable and the person they are with makes it bearable?

[LAR_Systems]

4 minutes ago, Arrogath said:

like the back of a Volkswagen?

[GOTSpectrum]

2 minutes ago, Egg-Roll said:

Or is the tree uncomfortable and the person they are with makes it bearable?

Shush you 

[Guest]

1 minute ago, Egg-Roll said:

Or is the tree uncomfortable and the person they are with makes it bearable?

There is no tree. There is nothing but that moment in time.

[chaozbandit]

14 minutes ago, Den-Fi said:

There is no tree. There is nothing but that moment in time.

If you take a selfie at a tree, but google drive/photos service is borked, was there ever really a tree?

[Egg-Roll]

1 hour ago, GOTSpectrum said:

Shush you 

There was a person

...

, and they all laughed at this post

The End

[Collin_S]

11 hours ago, GOTSpectrum said:

There is currently no support for folding on ARM based processors 

yes, but there is also the new Rosetta 2. No Idea how well that works yet though.

[Collin_S]

11 hours ago, LAR_Systems said:

The site is currently under rebuild in a development environment to add additional information.   

 

This would include when a GPU/CPU was added to the DB, and if the data samples are "NEW" aka... not scientifically significant but an indicator or "Mature" where there would be a threshold of enough samples to confirm your looking at a pretty solid average.    

 

This information while not on the site has been in the client for a few months in the blurb about the performance for the GPU on the WU.  

 

 

Regarding collection methods:

 

It comes from the web client, randomly sampled to see the fluctuations that users see in the client as calculated by F@H.   It does not get info from F@H regarding the final WU scores as they can be impacted by users pausing, stopping, upload issues etc. which have nothing to do with GPU / CPU performance.   That is extreme overclocks stats site and they do a great job as that info is only relevant to the individual user based on their final reported production captured at F@H.

 

Regarding RAM timings etc. as write in fields:

 

Been doing this a long time (development) user populated "scouts honor" data collection fields tend to be garbage data that requires moderation, algorithms for fraud entries etc.    

As the PPD varies the most by project, GPU/CPU and WU the significance of ram timings, bios settings etc. becomes less and less valuable to the overall community in the form of a database site as this is not a 1:1 benchmarking platform like gaming FPS on the same test over and over and you tweak settings as the project work load is always different by each work unit.

Your ram timings might make a specific project or WU run 1% better in one case, then it might make it 2% worse in the next WU case cause they are differing work loads on the same hardware and project.    So sharing those timings, your computers hardware configuration etc. is of little community value for the DB site as unlike a gaming benchmark, the "test" in the form of the WU and F@H core is constantly changing the results. 

For the majority of users you want to see "how good is this GPU/CPU" overall, or on a specific project, you want the average you can expect, and to see a trend over time because it's always changing with drivers, cores, etc.   

 

95% of users likely don't want to sift through a results drilldown of something like..

• A GPU model and brand running on project 1234
• that ran a specific WU PRCG
• those results sub divided by Corsair Dominator DDR 4 3200, CAS latency etc.
• then a list of configs with different ram timings,
• running on Ryzen CPU model / Intel CPU model,
• then a list of configs with different CPU clocks
• on a specific motherboard,
• the PCI lane sharing / bandwidth cause they did or did not have the following expansion cards installed,
• cooler(s) used,
• fan curves, thermal performance,
• ambient room temps...
• the machine is dedicated to folding or spends part of it's time under high user load... etc. etc.
• Operating systems, drivers, software installed / overhead   

 

The number of factors to capture for meaningful device level data is huge, and It's way to granular for the community at large to get a meaningful insight that they would try to replicate.   This information tends to only be of value to the individual user that would be submitting it.   

 

It's why gaming benchmarks are professionally done on open air test beds, running the same tests on a bare OS, with the same drivers when trying to compare hardware or settings run to run. 

You as an individual can look at your own PPD on your local configuration, keep track of variables with lots of different benchmarking / logging tools, and tweak your settings to get your best daily average PPD.   That can be validated in the logs or on EOC based on overall daily / weekly production for improvement, and use the web client to make adjustments in real time based on projects and WU level fold charting.

 

 

This and many questions like it, is why the new site rebuild also includes FAQ sections for CPU and GPU databases separately and another for the web client to centralize the various methods used and rationales where everyone can seem them.

Thanks! I can understand how so much data could be totally overwhelming to the average user. I don't think I noticed that part about the data on the web control tab before.

[Collin_S]

51 minutes ago, Egg-Roll said:

There was a person

...

, and they all laughed at this post

The End

I guess it was me...

 

WAIT A MINUITE: Egg-Roll... Chicken Little... WHATS GOING ON HERE

[dogwitch]

23 hours ago, LazyDev said:

Sounds like something has fried, and now has a direct short.

yeah. so case and hs fan work. but nothing else does. hd might be thru. it was tiny one anyhow.

grock

this was before my second thread ripper build.

so said pc was on 24/7 crunching .  never got shut down unless a windows update.

[chaozbandit]

@LAR_Systemsan igpu on a cpu unit? 

 

Spoiler

 

https://folding.lar.systems/folding_data/gpu_ppd_by_project?project_id=16925

 

[Senzelian]

Badge(r) Badge(r) Badge(r) Badge(r) Badge(r) Badge(r) 

 

[Collin_S]

1 hour ago, chaozbandit said:

@LAR_Systemsan igpu on a cpu unit? 

 

  Hide contents

 

https://folding.lar.systems/folding_data/gpu_ppd_by_project?project_id=16925

 

I have that iGPU on my laptop. I am curious what OS people are on to be able to fold on it (and are they on Desktop?)

[Parideboy]

1 hour ago, chaozbandit said:

@LAR_Systemsan igpu on a cpu unit? 

 

  Reveal hidden contents

 

https://folding.lar.systems/folding_data/gpu_ppd_by_project?project_id=16925

 

hAcKs

[cbigfoot]

22 minutes ago, Parideboy said:

hAcKs

 that is a pretty crappy hack lol...that ppd is worthless lol..

[AlTech]

@GOTSpectrum It might just be me but I could have sworn the original post in the thread used to say 22nd instead of "15nd".

 

If it was previously 22nd then I think it would be unfair to have it as Nov 15 instead.

 

I relied on my belief that it said and was 22nd when deciding when to start folding since I had some electrical stability problems and I'd hate to not qualify for this event based on that factor.

[GOTSpectrum]

14 minutes ago, AluminiumTech said:

@GOTSpectrum It might just be me but I could have sworn the original post in the thread used to say 22nd instead of "15nd".

 

If it was previously 22nd then I think it would be unfair to have it as Nov 15 instead.

 

I relied on my belief that it said and was 22nd when deciding when to start folding since I had some electrical stability problems and I'd hate to not qualify for this event based on that factor.

It was corrected earlier in the event, furthermore it was said on a number of occasions that it is five weeks long. Plus, there's a reason we encourage people to fold as much of the event as they can. 

 

Errors happen and unfortunately I am very much a human like everyone else.  

 

You are correct though that it isn't ideal and I will in future be trying my best to double check things like dates to ensure we avoid this kind of confusion in further events. 

[AlTech]

1 minute ago, GOTSpectrum said:

It was corrected early in the event, furthermore it was said on a number of occasions that it is five weeks long.

I never saw the corrections nor the mention of the event being 5 weeks long

1 minute ago, GOTSpectrum said:

Plus, there's a reason we encourage people to fold as much of the event as they can. 

Okay but that has nothing to do the statement you just made above. Encouragement doesn't constitute the rules.

 

1 minute ago, GOTSpectrum said:

Errors happen and unfortunately I am very much a human like everyone else.  

 

You are correct though that it isn't ideal and I will in future be trying my best to double check things like dates to ensure we avoid this kind of confusion in further events. 

So there's no way with the time remaining that I'll qualify?

 

This may come off as being self-entitled but honestly the only fair course of action I see is, having recognized that you made a mistake and corrected it and also having seen your correction wasn't seen by others, for you to restore it to November 22nd to restore fairness.

 

Not all participants will have seen the corrections and a number of participants will be caught off guard if things stay as they currently are.

 

To not restore it to 22nd of November now would be exclusionary and punitive rather than trying to encourage participation. If I or others know that we can no longer qualify, what motivation do I or others have to continue since we won't be able to qualify?

[justpoet]

7 minutes ago, AluminiumTech said:

To not restore it to 22nd of November now would be exclusionary and punitive rather than trying to encourage participation. If I or others know that we can no longer qualify, what motivation do I or others have to continue since we won't be able to qualify?

Helping the human race solve immense medical challenges.

[AlTech]

Also @GOTSpectrumif in light of everything I've said you're not willing to change it back to what it was originally November 22nd (even if this date was chosen in error and later changed), then I have to ask who do I appeal this decision to? The mod team?

[AlTech]

Just now, justpoet said:

Helping the human race solve immense medical challenges.

Yes but a mistake was made by the Team Leader and fixing that mistake made the event unfair.

 

It has severely disadvantaged people or even disqualified them from the event when they joined the event knowing it ended on the 22nd but the Team Leader wanted to change it after the fact to the 15th like nothing ever happened.

 

The situation needs to be remedied back to the original end date for the event to remain fair, ethical, and for the event to not disadvantage, bias, or disqualify people who should objectively be allowed to qualify based on the rules at the start of the event.

[shaz2sxy]

@AluminiumTech I think you are making something out of nothing.   If the event finishes on the 15th or the 22nd it makes no difference.  If you don't qualify for the event because you don't meet the requirements. Then it is what it is. The date issue was brought up and rectified earlier.  If you missed it that's no one else's fault then your own.  It's even come up in the last few days on this chat.

 

What are you loosing at the end of the day?

 

The majority of people folding as @justpoet said do it for "Helping the human race solve immense medical challenges." 

 

To add the guys who do this, do it on there own accord, with there own time and nothing to do with LTT etc.