Jump to content

BOINC Pentathlon 2023

Go to solution Solved by GOTSpectrum,

So here are the winners of the LTTStore discount vouchers, these need to be entered during checkout: 

@Hyuumaru @QuantumSnow @texus @etv24 @OhBoy @blab blob @Lightwreather JfromN @Your_Stalker @UnauthorisedNoodle @miker07

The winners of the Stream giftcards are:
 

@Storm_007 @chaozbandit @Bruntosaurus

 

The Winner of the drawing from @Ithanul is:

@Cory74

To claim your prize you need to DM me here on the forum before 00:00 BST, May 30th, 2023, with a screengrab of your BOINCStats statistics showing you have points in the correct projects 

6 minutes ago, leadeater said:

Not all, depends on the project and application/task. It's up to the project owner and their requirements.

Should I stop creating new pending requests then at some point, because they might not all get validated in the next days? I'm up to over 90 tasks pending validation.

Link to comment
Share on other sites

Link to post
Share on other sites

12 minutes ago, adm0n said:

Should I stop creating new pending requests then at some point, because they might not all get validated in the next days? I'm up to over 90 tasks pending validation.

I have over 200 tasks pending. I’d keep on rocking, someone probably has the pending tasks in a bunker.

My PC Specs: (expand to view)

 

 

Main Gaming Machine

CPU:  Intel Core i7-14700K
CPU Cooler: Deepcool LT720
Motherboard: MSI PRO Z790-P WIFI
Memory: G.Skill Ripjaws S5 32 GB (2 x 16 GB) DDR5-6000

Storage 1: Samsung 990 Pro 2 TB

Storage 2: Crucial P3 Plus 4 TB
Video Card: EVGA XC3 ULTRA GAMING GeForce RTX 3080 10GB

Power Supply: Corsair RM850 850W
Case: Corsair 4000D Airflow
Case Fan 120mm: Noctua F12 PWM 54.97 CFM 120 mm (x1)
Case Fan 140mm: Noctua A14 PWM 82.5 CFM 140 mm (x2)
Monitor Main: MSI G274QPF-QD 27.0" 2560 x 1440 170 Hz
Monitor Vertical: Asus VA27EHE 27.0" 1920x1080 75 Hz

Link to comment
Share on other sites

Link to post
Share on other sites

Yep, just keep going and let the pending climb. Some of us have thousands pending at this point. Some people will drop large bunkers of completed tasks and clear out some pending, but others won't clear out until after the event.

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, adm0n said:

Oh, are all computations done twice in order to validate them? Thanks for telling me 🙂

All of Einstein@Home tasks have wingmen. IE two of the same task sent to two different users hosts.

 

Last I looked I was approaching 30k pending tasks.

BOINC Team Member of TeAm AanandTech

BOINCGames.com - BOINC Competition all year long.

Link to comment
Share on other sites

Link to post
Share on other sites

7 hours ago, adm0n said:

The special app on Linux is wicked fast. I can finish a single task in 90 seconds. But now lots of them don't get validated anymore. Are there too many completed tasks coming in?

depends on the nature of that you mean by "don't get validated anymore"

are these errors? (not necessarily normal, what are the specific errors if so)
are these invalids? (normal if you have a 40-series)
are these tasks stuck in pending? (normal during a competition with everyone bunkering)

 

are you using a 40-series GPU? these cards have high invalids. I think it's a driver bug with the 40-series, but that's just a hunch based on what I've seen. havent been able to correlate it with core/mem clocks, OS version, OS type, driver version, app version, or anything else. only 40-series or not 40-series when running gamma ray.

 

an aside,

Petri is a wizard with Nvidia GPU optimization. we have quite a few special apps 😉 when SETI was still running, our special app was like 5x faster than the project provided application. The Gamma ray app was his first OpenCL app attempt. He usually makes CUDA apps. and the early/beta iterations of gamma ray optimization were done without any source code and just man-in-the-middle intercepting the code and changing it on the fly LOL (some code in the binaries are in plain text). But he only makes stuff for Linux. back in the SETI days, several people tried to port his app to Windows but it never worked. something about specific driver features or functions that were in the code that didnt work in windows. don't remember the specifics.

 

there's probably more in the tank for gamma ray optimization. but it's close to being maxed out. it uses every bit of the GPU (full core utilization, full mem bus utilization if you have enough cores/GPU throughput, decently high VRAM use). we could probably squueze a few more % out of it by porting it to CUDA, but that would probably be an intensive code rewrite. Petri doesnt want to put more work into this app because Gamma Ray tasks will ultimately end. likely in less than a year unless the project drums up more work for it. but they have already said it's days were numbered. We actually thought it was going to run out like 6 months ago, but the project added some more. I'm not optimistic that that will happen again though.

Link to comment
Share on other sites

Link to post
Share on other sites

15 minutes ago, SkillzTA said:

All of Einstein@Home tasks have wingmen. IE two of the same task sent to two different users hosts.

 

Last I looked I was approaching 30k pending tasks.

only 30k? that's cute 😉

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, gsrcrxsi said:

are you using a 40-series GPU? these cards have high invalids. I think it's a driver bug with the 40-series, but that's just a hunch based on what I've seen. havent been able to correlate it with core/mem clocks, OS version, OS type, driver version, app version, or anything else. only 40-series or not 40-series when running gamma ray.

Not 40 series related but does ECC help at all with preventing the invalids? I've gotten a few come up as invalid that ran on my A40, but I guess it's more down to why specifically each one is invalid and I'd guess ECC doesn't really have anything to do with it.

Edited by leadeater
Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, leadeater said:

Not 40 series related but does ECC help at all with preventing the invalids? I've gotten a few come up as invalid that ran on my A40, but I guess it's more down to why specifically each one is invalid and I'd guess ECC doesn't really have anything to do with it.

I'm not sure. but an A40 is Ampere based and wont have the same invalids issue. seems specific to Ada. Enabling ECC is very likely to slow the task down a lot though, since they are so reliant on memory speed and bandwidth.

 

~4-5% invalids is normal. 40-series cards are seeing 10-20%.

 

invalids can happen for a lot of reasons. some apps just don't validate well with each other when they use very different math libraries. it's pretty normal that like-apps will validate with each other better than with other apps. but as long as they are relatively low, it's not a problem for the project. invalids get resent to a new user until one of them validates. it only sucks for the user who got the invalid and no points

Link to comment
Share on other sites

Link to post
Share on other sites

haha nice

image.png.bff1f08a52134d1031dd5d514a4b6d25.png

 

4 minutes ago, gsrcrxsi said:

invalids can happen for a lot of reasons. some apps just don't validate well with each other when they use very different math libraries. it's pretty normal that like-apps will validate with each other better than with other apps. but as long as they are relatively low, it's not a problem for the project. invalids get resent to a new user until one of them validates. it only sucks for the user who got the invalid and no points

Thanks, had a look at most of my invalids and they were different app versions or based on run time the optimized one.

 

I'd like to try out the optimized version but I had to assign the GPU back to app service that uses it, just does video transcoding of very long lecture recordings. Bit of a waste of an A40 really.

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, leadeater said:

haha nice

image.png.bff1f08a52134d1031dd5d514a4b6d25.png

😉

Link to comment
Share on other sites

Link to post
Share on other sites

You mean you guys aren't using your Ti 83 Plus calculators to do these tasks faster?

BOINC Team Member of TeAm AanandTech

BOINCGames.com - BOINC Competition all year long.

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, SkillzTA said:

You mean you guys aren't using your Ti 83 Plus calculators to do these tasks faster?

Too fancy, still using punch cards and post mail to send them off to get run and returned. Real pain if one of the cards isn't punched correctly.

Link to comment
Share on other sites

Link to post
Share on other sites

@SkillzTA Oh I gave up on running E@H on CPU btw. Multiple different CPU models and they were all going to take over a day to complete per task. I tried and failed.

Link to comment
Share on other sites

Link to post
Share on other sites

13 minutes ago, leadeater said:

@SkillzTA Oh I gave up on running E@H on CPU btw. Multiple different CPU models and they were all going to take over a day to complete per task. I tried and failed.

IIRC, you should only run gamma ray CPU tasks also. the gravitational wave tasks can need like 3+GB per task.

 

but really, CPU isnt worth it on Einstein, unless that's all you have. no use contributing CPU power to a project with both GPU and CPU tasks on the same subproject. just put your GPUs there, then use the CPUs for a CPU-only project. It's a much better use of the resources.

Link to comment
Share on other sites

Link to post
Share on other sites

25 minutes ago, leadeater said:

Too fancy, still using punch cards and post mail to send them off to get run and returned. Real pain if one of the cards isn't punched correctly.

How do you think Einstein@home is sending me my tasks? My internet is too slow.

BOINC Team Member of TeAm AanandTech

BOINCGames.com - BOINC Competition all year long.

Link to comment
Share on other sites

Link to post
Share on other sites

5 minutes ago, SkillzTA said:

How do you think Einstein@home is sending me my tasks? My internet is too slow.

I just presumed ticker tape 

My Folding Stats - Join the fight against COVID-19 with FOLDING! - If someone has helped you out on the forum don't forget to give them a reaction to say thank you!

 

The only true wisdom is in knowing you know nothing. - Socrates
 

Please put as much effort into your question as you expect me to put into answering it. 

 

  • CPU
    Ryzen 9 5950X
  • Motherboard
    Gigabyte Aorus GA-AX370-GAMING 5
  • RAM
    32GB DDR4 3200
  • GPU
    Inno3D 4070 Ti
  • Case
    Cooler Master - MasterCase H500P
  • Storage
    Western Digital Black 250GB, Seagate BarraCuda 1TB x2
  • PSU
    EVGA Supernova 1000w 
  • Display(s)
    Lenovo L29w-30 29 Inch UltraWide Full HD, BenQ - XL2430(portrait), Dell P2311Hb(portrait)
  • Cooling
    MasterLiquid Lite 240
Link to comment
Share on other sites

Link to post
Share on other sites

21 minutes ago, gsrcrxsi said:

unless that's all you have

Sadly it is so wanted to give it a try.

 

22 minutes ago, gsrcrxsi said:

you should only run gamma ray CPU tasks also

Thanks, I'll try and remember if I ever have to try it again.

Link to comment
Share on other sites

Link to post
Share on other sites

7 hours ago, adm0n said:

Docker utilizes the host OS, in this case windows, to run programs in an virtual environment, without actually running another OS. 

Just fyi, docker on macOS just runs Linux 

"The most important step a man can take. It’s not the first one, is it?
It’s the next one. Always the next step, Dalinar."
–Chapter 118, Oathbringer, Stormlight Archive #3 by Brandon Sanderson

 

 

Older stuff:

Spoiler

"A high ideal missed by a little, is far better than low ideal that is achievable, yet far less effective"

 

If you think I'm wrong, correct me. If I've offended you in some way tell me what it is and how I can correct it. I want to learn, and along the way one can make mistakes; Being wrong helps you learn what's right.

 

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, SkillzTA said:

You mean you guys aren't using your Ti 83 Plus calculators to do these tasks faster?

I use my TI-85 that I had in high school. (In the mid 90's.....) 

Not much of a Gamer….. But I have thing about F@H that may be a little over the top.   See my builds here

Link to comment
Share on other sites

Link to post
Share on other sites

E@H sure does like AMD GPUs it seems. Both my mine are finishing WUs in under 5 mins.

My Folding Stats

My BOINC Stats

 

 

VelosterN:

AMD Ryzen 9 5950X - Asus ROG Strix X570-E Gaming - Corsair Vengeance RGB Pro 3600Mhz 32GB - Asus ROG Strix Gaming 6750 XT OC

Corsair Crystal Series 680x RGB - Samsung 970 Evo Plus 250GB NVMe - Samsung 970 Pro 512GB NVMe - Samsung 860 Pro 256 GB 2.5" SSD X2

EVGA P2 80+ Platinum 850Watt PSU - BenQ XL2730Z 27.0" 2560x1440 144 Hz - be quiet! Dark Rock 4

Corsair K70 LUX - Logitech G502 Proteus Spectrum - Sennheiser HD599 - Blue Yeti Mic

Windows 11 Professional Version 22H2

BettyBoop:

AMD Ryzen 5 2600X - Asus ROG Strix B450-I Gaming - Corsair Vengeance LPX 3000Mhz 16GB - Asus ROG Strix Gaming 5500 XT - Fractal Design Core 500

Samsung 860 Pro 512GB Sata - EVGA 550GM 80+ Gold 550Watt SFX PSU - be quiet! Dark Rock 4

Windows 11 Professional Version 22H2

HTPC:

Intel i7 6700K - Asus Maximus Hero VIII - Corsair Vengeance LPX 3000MHz 16GB - MSI RX480 Gaming-X 8GB - Cooler Master 932 HAF

Seagate 250GB HDD - EVGA G2 80+ Gold 650Watt PSU - Corsair H100i

Windows 10 Professional Version 21H2

 

Link to comment
Share on other sites

Link to post
Share on other sites

36 minutes ago, HaRdLy007 said:

E@H sure does like AMD GPUs it seems. Both my mine are finishing WUs in under 5 mins.

My 4070ti's are doing them in 3:08 and the 4080's in 2:20. But there could be other variances going on. 

Not much of a Gamer….. But I have thing about F@H that may be a little over the top.   See my builds here

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Mxyzptlk said:

My 4070ti's are doing them in 3:08 and the 4080's in 2:20. But there could be other variances going on. 

😭 my old rx480 is taking 30 mins to finish a job...😭

Link to comment
Share on other sites

Link to post
Share on other sites

Guest
This topic is now closed to further replies.


×