Folding Community Board

[justpoet]

Spoiler

 

9 hours ago, Starman57 said:

Yes, I am aware that you just record what the FAH client reports, and have no say in the calculation.  Thanks for that.  Yes, WU calculations are not 100% consistent, but on each individual WU, they are close to being consistent assuming that the underlying computing loads are the same.  That's where a problem crops up, because if you are also using the computer at the same time, this will extend FAH compute times as it's fighting for computing resources.

 

Yes, my 6900XT falls into the "bucket" you specified.  I found out from FAH is that they "categorize" cards based loosely on the PCI ID, however they don't fully parse it hence the "bucket" approach.  As for the blend, yes of course, but I wonder what it would take in terms of performance from the "slower" members of the group to account for an "average" that rises from 1 hour 45 minutes to 2 hours 55 minutes?  It must mean that the slower 6xxx cards must be incredibly slow at FAH loads to account for an almost doubling of the time!  I looked at the FP32 performance of them, and they aren't that much slower, as the 6800 (the slowest) is only 30% slower than the 6900XT (fastest) in the group.

 

Quoted from the FAH forum where I raised this issue "FAH will continue to award the same baseline points for both AMD and nVidia."  As for folding architecture, FAH defines that.  The Nvidia and AMD GPUs aren't specifically designed folding cards.  They aren't ASICs.  As such, it should be a pretty easy correlation between cards based on the predominant compute instructions used, namely FP32 (as reported by FAH - "In fact, FP64 performance is quite different, too, favoring nV by a large factor but I don't use that number since it makes a much smaller contribution to overall performance than FP32 does.")

 

Looking at my FAH numbers as reported by Folding in the Dark, I'm surmising that a LOT of people are somehow running the FAH client (both GPU and CPU) really badly.  Correction, inefficiently.  Could it be they just install the program and that's it?  Windows does a lousy job of scheduling a specific heavy load, or at least the FAH load, on modern multicore CPU's which also impacts GPU's as well.  Witness the differences in performance with Linux which either has a smarter scheduler or less bloat to deal with.  I can only speak to my personal setup, and my contribution of Linux WUs on 5700XT GPUs for the last couple of weeks has changed the trend line quite noticeably.  As the days progress, I'm guessing the Linux line will deflect even more based on the PPD numbers my 2 machines are locally reporting.  Take a look.

 

https://folding.lar.systems/gpu_ppd/brands/amd/folding_profile/navi_10_radeon_rx_5600_oem5600_xt57005700_xt

 

As for help, you are already doing monumental work by providing Folding in the Dark and it's website database.  Kudos.   I just asked the basic question of client "rebalancing" just in case you were somehow plugged into the goings on behind the scenes (pending change lists) as to client changes and API calls.  Since you were blind-sided by the latest API difficulties, I'm guessing not.  Never hurts to ask.

 

Anyway, I'll still plod along with the 6900XT at 15-25% higher than the average.  Now that I've finally got my 5950x, I'm getting 75-100+% higher PPD:PLP number.  Example?  WU 16959.  5950x CPU average: 23,442.  My current: 49,124.  I've had the same effect on my 3950x's which possibly contributed to the skewed results comparing the 3950x to the 5950x.  Ever notice the 3950x is quite a bit higher than the 5950x?  That should change over time as I know based on local results that the 5950x is quite a bit faster than the 3950x (my guess based on FAH is about 20%).  Whether it will ever show up in your database after being watered down by inefficient results and averaged remains to be seen.

 

Again, thanks for the work you do.

Something I believe you're not accounting for is the rapid return bonus, which makes points not linear based on performance, but closer to exponential.  A lot of the low end 6000 cards also don't have coolers capable of keeping them running at advertised speeds when under a long constant workflow like folding.  When you combine those two factors, I wouldn't be surprised to see the 30% performance difference you mention be around 50% points difference in practice.

[Starman57]

1 hour ago, justpoet said:

  Reveal hidden contents

 

Something I believe you're not accounting for is the rapid return bonus, which makes points not linear based on performance, but closer to exponential.  A lot of the low end 6000 cards also don't have coolers capable of keeping them running at advertised speeds when under a long constant workflow like folding.  When you combine those two factors, I wouldn't be surprised to see the 30% performance difference you mention be around 50% points difference in practice.

Could be.  But that's for points, not for time.

 

The average reported is 2h55m (175 minutes).  You'd need a much larger difference to make up the loss from 105 minutes (6900XT) to 175 minute average.  Since it's an average, the slowdown has to be higher to offset the 6900XT time of 1h45m.  So for every 6900XT return in 105 minutes, the 6800 would have to return it's WU in 245 minutes ((245+105)/2=175).  Based on a 30% slower FP32 performance than a 6900XT, a 6800 should do that same WU in 137 minutes, which is a lot lower than 245 minutes.  This would mean that the 6800 would have to be almost 1.8x slower than the FP32 performance would indicate.  What would that do to the points awarded, including the rapid return bonus (maybe not called rapid then?).

 

The same average problem arises when you mix in the 6800XT which is slower than the 6900XT, but faster than the 6800 as all of them are in the same "bucket".

 

Something is way off on the average?  Something is way off on the points?  Something is way off on the reported completion times?  All of them?

[rkv_2401]

2 hours ago, Starman57 said:

Could be.  But that's for points, not for time.

 

The average reported is 2h55m (175 minutes).  You'd need a much larger difference to make up the loss from 105 minutes (6900XT) to 175 minute average.  Since it's an average, the slowdown has to be higher to offset the 6900XT time of 1h45m.  So for every 6900XT return in 105 minutes, the 6800 would have to return it's WU in 245 minutes ((245+105)/2=175).  Based on a 30% slower FP32 performance than a 6900XT, a 6800 should do that same WU in 137 minutes, which is a lot lower than 245 minutes.  This would mean that the 6800 would have to be almost 1.8x slower than the FP32 performance would indicate.  What would that do to the points awarded, including the rapid return bonus (maybe not called rapid then?).

 

The same average problem arises when you mix in the 6800XT which is slower than the 6900XT, but faster than the 6800 as all of them are in the same "bucket".

 

Something is way off on the average?  Something is way off on the points?  Something is way off on the reported completion times?  All of them?

Your time calculation could be derailed by a lot of simple things, such as :

i) The 6800 samples not boosting/not achieving their rated clock speed

ii) The users doing something else in the background, say, playing a less-demanding game like Fortnite or Minecraft or watching 4K HD videos

iii) The users turning off their computer at nights, without noticing(or caring) that the WU hasn't been finished yet

iv) Some other problem with the computer causing the Folding core to shut down and restart multiple times in the time it takes to finish the WU.

[joshmaze]

Hi all!  I have been watching the Linus Media Group's youtube stuff for awhile, and I've also been contributing to Folding@Home for awhile.  I just recently figured out how to set my team, and I finally signed up for a forums account.  I had 9,500,000 points, but I guess they don't transfer over to the team when I switch.  Oh well, starting from scratch!

[LazyDev]

Since I installed my 1080 in my rig, I keep getting the odd "video scheduler internal error" BSOD. The 1080 passed all benchmarking and has been running for several days straight, without issue. The BDOD comes when I touch the memory on the card, even by just +1Mhz, or the GPU task suddenly stops. This was also noted on my test system.

 

Could the riser (4x4x4x4) be contributing?

 

Any ideas? It's a reliable card otherwise.

[justpoet]

14 hours ago, joshmaze said:

Hi all!  I have been watching the Linus Media Group's youtube stuff for awhile, and I've also been contributing to Folding@Home for awhile.  I just recently figured out how to set my team, and I finally signed up for a forums account.  I had 9,500,000 points, but I guess they don't transfer over to the team when I switch.  Oh well, starting from scratch!

You keep your existing points, they just don't count for the team.  Future points count for both you and the team.

Welcome to the LTT Fold!

[LAR_Systems]

16 hours ago, rkv_2401 said:

Your time calculation could be derailed by a lot of simple things, such as :

i) The 6800 samples not boosting/not achieving their rated clock speed

ii) The users doing something else in the background, say, playing a less-demanding game like Fortnite or Minecraft or watching 4K HD videos

iii) The users turning off their computer at nights, without noticing(or caring) that the WU hasn't been finished yet

iv) Some other problem with the computer causing the Folding core to shut down and restart multiple times in the time it takes to finish the WU.

To add too this, users that lower power settings for their card to save money folding.

The Facepalm of people I see trying to fold using old mining rigs on 1X risers.  

Combine these with any of the above and brutal PPD.

[maximumzero]

Really wish I could get a cheap GPU to toss into my Folding rigs, but even “cheap” GPUs these days have gotten ridiculous. Ugh.

[Gorgon]

With the resurgence of mining perhaps now is the time to request separate sub forms for mining and distributed computing.

 

even if we have to walk away from all the existing content getting rid of the “noise” world be worth it IMHO

[LazyDev]

1 hour ago, Gorgon said:

With the resurgence of mining perhaps now is the time to request separate sub forms for mining and distributed computing.

 

even if we have to walk away from all the existing content getting rid of the “noise” world be worth it IMHO

Was considering requesting a "Mining Community Board", as this is obviously missing. I think it's time that we request one.

[LazyDev]

17 hours ago, maximumzero said:

Really wish I could get a cheap GPU to toss into my Folding rigs, but even “cheap” GPUs these days have gotten ridiculous. Ugh.

I've started to repair my own GPU's. Out of 5, I've brought 2 back from the dead (No detection). Other 3 have shorts, which I'm waiting for funds on a thermal camera to identify.

 

My current 1080 that running 24/7, just had a blown fuse on the 12 volt rail. I could sell it now, and make £160 profit, if I wanted.

[NetoriusNick]

3 hours ago, LazyDev said:

I've started to repair my own GPU's. Out of 5, I've brought 2 back from the dead (No detection). Other 3 have shorts, which I'm waiting for funds on a thermal camera to identify.

 

My current 1080 that running 24/7, just had a blown fuse on the 12 volt rail. I could sell it now, and make £160 profit, if I wanted.

Dumb Q
why not test R to ground? I'm thinking that because to don't have a super sensitive meter

[vsteel]

8 hours ago, Gorgon said:

With the resurgence of mining perhaps now is the time to request separate sub forms for mining and distributed computing.

 

even if we have to walk away from all the existing content getting rid of the “noise” world be worth it IMHO

Is there really that much to go through?  It seems like 80% of the traffic for folding is in this thread or the request for badges thread.  

[TVwazhere]

7 minutes ago, vsteel said:

Is there really that much to go through?  It seems like 80% of the traffic for folding is in this thread or the request for badges thread.  

Requests for badges are down I'd guestimate 80% from a few months ago. 

 

Red is 5 pages of requests between November and December, Blue is only 3 pages of requests between January and Now (April)

[NetoriusNick]

This could be a red herring as the badges changed end of December no?

6 minutes ago, TVwazhere said:

Requests for badges are down I'd guestimate 80% from a few months ago. 

 

Red is 5 pages of requests between November and December, Blue is only 3 pages of requests between January and Now (April)

 

[RollinLower]

2 minutes ago, NetoriusNick said:

This could be a red herring as the badges changed end of December no?

 

that's also around the time we hold our largest F@H event.

[TVwazhere]

19 minutes ago, NetoriusNick said:

This could be a red herring as the badges changed end of December no?

 

17 minutes ago, RollinLower said:

that's also around the time we hold our largest F@H event.

He's right. The Folding Event ended in November and a lot of people were submitting their requests for badges for the points they acquired (Event badges are handed out automatically based on eligibility) There's a typically a surge of requests at the end of events (based on Official event 2019 and 2020; sprints to a lesser degree)

[Captainmarino]

I think F@H may be going for round 2 on updating things again.

[NetoriusNick]

Just now, Captainmarino said:

I think F@H may be going for round 2 on updating things again.

DING DING 
ANNNNNNNNNDDDDDD
FIGHT!

[justpoet]

2 hours ago, Captainmarino said:

I think F@H may be going for round 2 on updating things again.

And just like that, I made the massive leap and was tied for first place output of the team!

[sazrocks]

16 hours ago, Gorgon said:

With the resurgence of mining perhaps now is the time to request separate sub forms for mining and distributed computing.

 

even if we have to walk away from all the existing content getting rid of the “noise” world be worth it IMHO

I asked that this be done during the last mining boom and the admins/mods really didn't seem to like the idea. Maybe they'll change their minds this time.

[Kilrah]

18 hours ago, TVwazhere said:

He's right. The Folding Event ended in November and a lot of people were submitting their requests for badges for the points they acquired (Event badges are handed out automatically based on eligibility) There's a typically a surge of requests at the end of events (based on Official event 2019 and 2020; sprints to a lesser degree)

Also the requirements for badges have gone up, so it's kinda harder to get one now...

[RollinLower]

so far it seems the site is still not quite ready 

 

wait nvm, it's working now!

[Captainmarino]

37 minutes ago, RollinLower said:

so far it seems the site is still not quite ready 

 

wait nvm, it's working now!

What is the url they are using now? I couldn't get to my new stats page (only the "classic" one using classicstats...)

[RollinLower]

3 minutes ago, Captainmarino said:

What is the url they are using now? I couldn't get to my new stats page (only the "classic" one using classicstats...)

https://stats.foldingathome.org/donor/RollinLower gets you to mine