15th Annual Folding@Home Holiday Season Race: Mark's Marauders win!

biodoc

Diamond Member
Dec 29, 2005
6,256
2,238
136
Welcome to the 15th Annual intra-team Holiday Race between members of Team Anandtech. Participation is voluntary and the goal is to have a bit of fun while contributing to scientific goals of the Folding@Home project.

The race starts December 1, 12:00 UTC and ends January 1, 12:00 UTC
  • Please post in this thread if you intend to take part (see below for more info).
  • If you need help installing or running the F@H app go to: http://folding.stanford.edu/start-folding/ or ask for help here. The TeAm AnandTech number is 198.
  • If you don't have a passkey, get one here. If you have a passkey you'll get bonus points.
  • You can join any time during the race and I'll decide which team you are assigned.
For those who want to participate:
Please post your forum name and your Folding@Home name in this thread and estimated PPD.
Please also indicate if you will participate for the full month or just part of the month.
I will put together the initial composition of the teams just before the start of the race.

@Pokey is the captain of the Moonshot Gang team
@Markfw is the captain of Mark's Marauders team

Enrolled participants (forum name - Folding@home name - Est. PPD - participation estimate (days or full month):
biodoc - biodoc - 5 million -full month
Pokey - Pokey_TA - 35 million - full month
Assimilator1 - [TA]Assimilator1 - 0.4 million - full month
Endgame124 - Endgame124 - 3 million - full month
crashtech - crashtech - 12 million - full month
TennesseeTony - 10esseetony - 3 million - full month
Markfw - Mark_F_Williams - 35 million - full month
emoga - emoga - 0.5 million -full month
Desslok - Desslok - 4 million - full month
petrusbroder - TAS-petrusbroder - 1.1 million - full month
salvorhardin - salvorhardin - 0.5 million - full month
VirtualLarry - VirtualLarry - 5 million - full month
sswingle - Scott_Swingle - 1.9 million - full month
StefanR5R - xii5ku - 4.2 million - full month
Icecold - Icecold - 1 Million - Full Month
Neurodog - Neurodog - 1 Million - Full Month
Mike_F - Mike_F - 7.3 million - Full Month
Grumble_D - Grumble_D - 2.5 million - Full Month
blckgrffn - blckgrffn
IEC - IEC
geecee - geecee
 
Last edited:

Assimilator1

Elite Member
Nov 4, 1999
24,118
507
126
<----Forum name ;)
[TA]Assimilator1, unknown ppd (I'll look at my stats for when I ran it a few months ago, last year I ran it on my i7, this year it'll be on my Ryzen) [edit] Ok I've looked here and my output is all over the place, so I haven't a foggiest. If it helps I'll be running it 24/7 on my Ryzen 5 3600 which under load runs at about 3.7-3.8 GHz as I've reduced it's PPT, and I'll be running it on my RX 580 which I've underclocked to RX 480 levels.
Full month :)
Will they have enough WUs though??
 
Last edited:

StefanR5R

Elite Member
Dec 10, 2016
5,441
7,681
136
I'll be running it 24/7 on my Ryzen 5 3600 which under load runs at about 3.7-3.8 GHz as I've reduced it's PPT, and I'll be running it on my RX 580 which I've underclocked to RX 480 levels.
The Ryzen will be a lot more productive in other projects. Check out the PPD difference between CPU slot and GPU slot in FahControl, and consider the PPD/Watt difference...

If you don't have a current favorite CPU project: Some of us are active in the projects which have points listed in this table: Formula BOINC 2020 marathons. (And next week, many of us are going to run WCG of course. Or are already stockpiling results, to be reported next week.)

Please post your forum name and your Folding@Home name in this thread and estimated PPD.
Please also indicate if you will participate for the full month or just part of the month.
I have yet to check out what my GPUs can do, now that the CUDA application core is in production — and I will need to make up my mind with how much heat I am ready to put up with in December. I intend to keep my dual-processor computers fully engaged in FB marathons until December 31, 23:59 UTC, and they alone produce far more heat than required to heat the apartment on normal winter days.
 

biodoc

Diamond Member
Dec 29, 2005
6,256
2,238
136
I have yet to check out what my GPUs can do, now that the CUDA application core is in production — and I will need to make up my mind with how much heat I am ready to put up with in December. I intend to keep my dual-processor computers fully engaged in FB marathons until December 31, 23:59 UTC, and they alone produce far more heat than required to heat the apartment on normal winter days.
There's no pressure to join in since many of us have other commitments for at least part of the month. This is meant to be a fun event with maybe a little trash talk to spice things up a bit. :) Oh and then there's the science support too. :)
 

Assimilator1

Elite Member
Nov 4, 1999
24,118
507
126
Welcome new guy! :D

The Ryzen will be a lot more productive in other projects. Check out the PPD difference between CPU slot and GPU slot in FahControl, and consider the PPD/Watt difference...

If you don't have a current favorite CPU project: Some of us are active in the projects which have points listed in this table: Formula BOINC 2020 marathons. (And next week, many of us are going to run WCG of course. Or are already stockpiling results, to be reported next week.)

Ah yea, good point about CPUs and F@H, although I do normally switch it to F@H just to eek out that bit more for the race ;).
Atm, I'm running Rosetta on it.

Biodoc
I did already mention I'd been in it for the full month :)
 

Endgame124

Senior member
Feb 11, 2008
954
669
136
Hopefully we won't have 3 paper launches in a row.
Per Kyle at the [H] forum:


What you are going to see will be different from what NVIDIA has done.

AMD MBA cards will be in short supply up front. However, where you saw NV front-load almost all its cards, AMD is backloading. Once AIB cards hit, you will see 5 to 7 times the inventory released than what NV has shipped total in NA at this time. AMD will be heavily back-loaded into the channel. AMD is shipping ASIC BOM kits nearly daily to AIBs at this point and it will continue to happen throughout Q4. There will be a decent supply through Q4.

Now let me say this again as it is worth hearing. There is no way that when AMD was doing its projections for supply, that it could have even fathomed that NVIDIA would totally shit the bed on 3080/90 supply for the quarter, effectively abandoning the $500 to $1000 GPU TAM. NVIDIA has left such a gaping hole in the market that there is no way it can be filled currently. From my information and analysis, AMD projected to grow its portion of the TAM and will have inventory to do that. The sad fact is that NV simply abandoned its customer in that market and there is no possible way for AMD to fill the void.
 

Assimilator1

Elite Member
Nov 4, 1999
24,118
507
126
Err, can we have some translation? lol ,what the heck are back and front loading, ASIC BOM, AIBs and TAM? :p
 

lane42

Diamond Member
Sep 3, 2000
5,721
624
126
TAM = Maybe, Time and Materials or Total Available Market .
Just guessing.
 

Endgame124

Senior member
Feb 11, 2008
954
669
136
Err, can we have some translation? lol ,what the heck are back and front loading, ASIC BOM, AIBs and TAM? :p
ASIC BOM
Direct translation:
Application-specific integrated circuit Bill of Materials.

This is, essentially, the GPU and other needed materials to make a video card. Kyle says AMD is shipping a lot of these with shipments arriving daily.

AIB:
Add in Board Partners
These are the companies, besides AMD, that make video cards, such as Sapphire, XFX, etc.

TAM:
Total Available Market
This is essentially the "Demand" at a particular price point. For example, even if you had an unlimited supply of cards, you would only sell so many at a particular price point over a quarter. For example, this may be a total of 250,000 cards that would sell at $500 over a quarter

I'm a little less certain on the specifics of Front loading vs back loading sales. I believe it goes like this.
Front loading - Company can't make very many of item X, so it focuses on building a stock and sells it all at once, trickling in a small number of additional units over time.

Back Loading - Company has focused on getting manufacturing set so it can build a lot of item x. It starts selling them before building inventory, though, so the initial sales sell out quickly, with a medium to large number of additional units coming over time.
 
  • Like
Reactions: Ken g6

Endgame124

Senior member
Feb 11, 2008
954
669
136
A little more translation:

Nvidia essentially failed a providing supply of the 30X0 series of cards, leaving a huge hole in market at the $500+ cards. NVidia is approximately 80% of the market, so if 250,000 total $500+ cards would sell in a quarter in the US with unlimited supply, we would expect Nvidia to sell 200,000 of them and AMD to sell 50,000 of them. Per another post from Kyle, Nvidia has, to this point, made fewer than 10,000 3080s and 3090s available to the US market.

AMD expected to claim some market share because they think that the 6X00 series is very good, so they planned on selling more like 75,000 total cards instead of 50,000 cards.

Unfortunately, in this example, even though AMD increased its production 50%, there will only be 85,000-100,000 total cards available to the market in Q4, depending on how many more cards Nvidia can make, and a demand for 250,000 cards. AMD would be in a prime position to sell even more cards, but when they were planning how many to make a year ago, they didn't expect a complete failure on the part of Nvidia to supply cards.
 
  • Like
Reactions: Assimilator1