Rise and Fall of the [H] team

Nathan_P

[H]ard DCOTM x2
Joined
Mar 2, 2010
Messages
3,430
/flex
When the opportunity was available, I made sure to take advantage of it.

It's neat to read about the history of the team as I never got seriously into it until more recently due to the COVID junk. What's up about the bigadv stuff? I'm familiar with what it is but I don't think it's discontinued. Never had to worry about it until recently.

To sum up the bigadv story, back in the day (2010-14)there were some monster sized cpu work units of 1m + atoms that give big PPD for the time, 250k-1m PPD depending on the hardware which was supposed to be 8 cores (not threads) or more. The hardware requirements changed a couple of times due to people spoofing the client to run on 2600k, Phenom x6, i7-920 etc that could complete the WU in time if they had a big enough overclock. Late 2013 it was announced that the hardware requirements were going to be changed twice within the next 4 months to keep up with the increasing core count of CPU's and stop the spoofing that was still going on, which meant that to run the projects you were going to need 24 then 32 physical cores to meet the deadlines. Older slower machines would drop back to SMP work units with a corresponding points drop. This caused uproar as people had sunk a lot of time and money into optimising rigs that were able to complete WU well in advance of the then current deadlines, would be able to under the new guidelines as well but couldn't due to the to new hardware requirements based solely on core count.

The team was raging at this and rightly so, many of them, including myself, had multiple 2p/4p rigs that cost thousands and couldn't be sold on or repurposed into a daily driver so people started to quit. The uproar continued on all the major teams that were cpu heavy and it was then announced that the bigadv project would be stopped due to the outcry over the new requirements at the end of January 2015 and that it, people just quit immediately.

A lot of people felt so bitter over it they vowed never to donate to Pande group again. The project lead and head of the entire folding project have both since moved on but the bitter taste remains. Bigadv finished 31st Jan 2015, the same WU with much lower points were available for most of 2015 then everything switched to SMP WU with much lower atom counts and points. WU with that many atoms have not be issued since.

Fun notes, I still have one of my bigadv 2p running - currently hosting a 1080 & 1070 plus a 20thread CPU slot giving 3 times the PPD it did back in the day. To find anything about bigadv on foldingforum means you have to go digging, the sub forums that existed have all been hidden and in some cases I suspect they have been deleted.
 

the901

[H]ard|DCOTM
Joined
Oct 11, 2008
Messages
198
To sum up the bigadv story, back in the day (2010-14)there were some monster sized cpu work units of 1m + atoms that give big PPD for the time, 250k-1m PPD depending on the hardware which was supposed to be 8 cores (not threads) or more. The hardware requirements changed a couple of times due to people spoofing the client to run on 2600k, Phenom x6, i7-920 etc that could complete the WU in time if they had a big enough overclock. Late 2013 it was announced that the hardware requirements were going to be changed twice within the next 4 months to keep up with the increasing core count of CPU's and stop the spoofing that was still going on, which meant that to run the projects you were going to need 24 then 32 physical cores to meet the deadlines. Older slower machines would drop back to SMP work units with a corresponding points drop. This caused uproar as people had sunk a lot of time and money into optimising rigs that were able to complete WU well in advance of the then current deadlines, would be able to under the new guidelines as well but couldn't due to the to new hardware requirements based solely on core count.

The team was raging at this and rightly so, many of them, including myself, had multiple 2p/4p rigs that cost thousands and couldn't be sold on or repurposed into a daily driver so people started to quit. The uproar continued on all the major teams that were cpu heavy and it was then announced that the bigadv project would be stopped due to the outcry over the new requirements at the end of January 2015 and that it, people just quit immediately.

A lot of people felt so bitter over it they vowed never to donate to Pande group again. The project lead and head of the entire folding project have both since moved on but the bitter taste remains. Bigadv finished 31st Jan 2015, the same WU with much lower points were available for most of 2015 then everything switched to SMP WU with much lower atom counts and points. WU with that many atoms have not be issued since.

Fun notes, I still have one of my bigadv 2p running - currently hosting a 1080 & 1070 plus a 20thread CPU slot giving 3 times the PPD it did back in the day. To find anything about bigadv on foldingforum means you have to go digging, the sub forums that existed have all been hidden and in some cases I suspect they have been deleted.

If you're completing the work units within the deadline, who cares. That's a bummer. Thanks for the history lesson.

I guess I still see the bigadv flag still available to use. Does that not change the WU assignments to the client, then?
 

Nathan_P

[H]ard DCOTM x2
Joined
Mar 2, 2010
Messages
3,430
That was our take on it later on but even we got irked by the single socket brigade in the beginning, especially when we were coughing up for 2p lga 1366 boards

Its still there but its redundant adding it to your slot config won't get you any work. Who knows it may come back at some point in the future, bigadv projects were working on flu virus proteins........
 

motqalden

[H]ard|DCOTM x3
Joined
Jun 22, 2009
Messages
1,623
Any of you GPU heavy guys that want to help crunch amicable numbers for the city run portion of the Pentathlon, you help would be greatly appreciated. This will be
Start: 07 May 2020 00:00 UTC
End: 12 May 2020 00:00 UTC

Project is Amicable numbers in the boinc manager. As has been mentioned these tasks take 8GB of system memory so if you run multiple GPU per rig you will want to watch for this.
 

Hagrid

[H]F Junkie
Joined
Nov 23, 2006
Messages
9,002
I have the 5960x and 1700 running now on BOINC. So a new run is coming for Rosetta?
 

motqalden

[H]ard|DCOTM x3
Joined
Jun 22, 2009
Messages
1,623
Rossetta will continue the full duration of the Pentathlon, but it is only one of the disciplines and is CPU only. In two days the event for Cross Country starts which is the BOINC project amicable numbers. This is not Rosetta but a completely different project which you can run on just your GPU if you select only AMD GPU or NVIDA GPU from the project preferences. In this way you can run Rosetta on your CPU and then keep the GPU's busy on Amicable Numbers.
 

Gilthanis

[H]ard|DCer of the Year - 2014
Joined
Jan 29, 2006
Messages
8,197
Just make sure you have a boat load of RAM since Rosetta work can ask for a lot (recently was told up to 4GB each) and Amicable work units want 8GB Ram each.
 

Hagrid

[H]F Junkie
Joined
Nov 23, 2006
Messages
9,002
Just make sure you have a boat load of RAM since Rosetta work can ask for a lot (recently was told up to 4GB each) and Amicable work units want 8GB Ram each.
Ouch! My 1700 has 32gb which isn't bad, but the 5960x is at 8gb. Maybe I can buy faster ram for my main and put the 2 other sticks in the 5960x.
 

Xilikon

[H]ard|DCer of the Year 2008
Joined
Oct 12, 2004
Messages
14,523
To sum up the bigadv story, back in the day (2010-14)there were some monster sized cpu work units of 1m + atoms that give big PPD for the time, 250k-1m PPD depending on the hardware which was supposed to be 8 cores (not threads) or more. The hardware requirements changed a couple of times due to people spoofing the client to run on 2600k, Phenom x6, i7-920 etc that could complete the WU in time if they had a big enough overclock. Late 2013 it was announced that the hardware requirements were going to be changed twice within the next 4 months to keep up with the increasing core count of CPU's and stop the spoofing that was still going on, which meant that to run the projects you were going to need 24 then 32 physical cores to meet the deadlines. Older slower machines would drop back to SMP work units with a corresponding points drop. This caused uproar as people had sunk a lot of time and money into optimising rigs that were able to complete WU well in advance of the then current deadlines, would be able to under the new guidelines as well but couldn't due to the to new hardware requirements based solely on core count.

The team was raging at this and rightly so, many of them, including myself, had multiple 2p/4p rigs that cost thousands and couldn't be sold on or repurposed into a daily driver so people started to quit. The uproar continued on all the major teams that were cpu heavy and it was then announced that the bigadv project would be stopped due to the outcry over the new requirements at the end of January 2015 and that it, people just quit immediately.

A lot of people felt so bitter over it they vowed never to donate to Pande group again. The project lead and head of the entire folding project have both since moved on but the bitter taste remains. Bigadv finished 31st Jan 2015, the same WU with much lower points were available for most of 2015 then everything switched to SMP WU with much lower atom counts and points. WU with that many atoms have not be issued since.

Fun notes, I still have one of my bigadv 2p running - currently hosting a 1080 & 1070 plus a 20thread CPU slot giving 3 times the PPD it did back in the day. To find anything about bigadv on foldingforum means you have to go digging, the sub forums that existed have all been hidden and in some cases I suspect they have been deleted.

And not only that but even before that in 2010-2011 when some projects is getting half the points of others projects and people was trying to argue with PG about how they calculate. There is a lot of units dumping... this is about when I got fed up with the reply of the PG and quit. The Bigadv situation is like playing in a open wound and that's understandable...
 
  • Like
Reactions: wurmy
like this

The_Heretic

Certified [H]
Joined
Jun 22, 2001
Messages
13,358
Any of you GPU heavy guys that want to help crunch amicable numbers for the city run portion of the Pentathlon, you help would be greatly appreciated. This will be
Start: 07 May 2020 00:00 UTC
End: 12 May 2020 00:00 UTC

Project is Amicable numbers in the boinc manager. As has been mentioned these tasks take 8GB of system memory so if you run multiple GPU per rig you will want to watch for this.


Ahhh this is where I got confused and started the Amicable project early. (I'm old now and confuse easily)


Ultimately it just started my machines running Amicable a few days early, no worries. :)
 

skylaar24

Weaksauce
Joined
Oct 5, 2017
Messages
65
I recall running -bigadv on my 920 and it made me cut my gaming back to try and hit deadlines while increasing the bonus. That same 920 is tuned down some for its senior years to 3.2GHz but is currently on Rosetta for the Marathon at 6 threads.
 

EngrChris

Gawd
Joined
Dec 28, 2008
Messages
847
My workstation competed a handful of early bigadv units on-time with stock clocks within the deadline.
For science, it would be nice to have large atom count projects again.
 

motqalden

[H]ard|DCOTM x3
Joined
Jun 22, 2009
Messages
1,623
Ultimately it just started my machines running Amicable a few days early, no worries. :)

Dang that's twice now i got my project mixed up. Well the good thing with having started it early is that about half your points would have been tied up waiting for validation and wouldn't have been awarded till the D day anyways.
I will try harder not to mix important details like this up going forward :facepalm:
 
Top