• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Milkyway@home TeAm competition - 10/15 to 11/15

Icecold

Senior member
Nov 15, 2004
447
130
116
Now that we're getting into better GPU crunching weather, I'm announcing an informal TeAm competition on Milkyway@home. The goal is simple: crunch as much Milkyway@home as you can from 10/15 through 11/15. I will post beginning stats from Free-DC on 10/15, we can all keep track along the way(if stats haven't been posted recently - feel free to post them if you can!), and we will check the stats on 11/15 and announce what everybody's total is. This is open to anybody and everybody, and I encourage as much participation as possible! No need to sign up - I will run the stats for all Milkyway@home contributors on TeAm Anandtech.

This is a fun, friendly push towards gaining ground in MIlkyway@home - and any crunching that happens is appreciated! There are some TeAm goals that are within sight on Milkyway@home as well - we are doing well on the Formula BOINC marathon on Milkyway@home and can likely pick up points there, as well as we're doing well in the overall Milkyway@home total points in BOINC, and could potentially move up there as well.

Some tips - Milkyway@home is unique in that it uses Double precision. What this means - there are some video cards that do quite well on it(even older ones), while what are generally some 'higher performing' newer cards do not based on the double precision performance being low. Even something as old as a Radeon 6950 can put up some decent points, Radeon 7950/7970/280X do quite well. The newer Radeon VII does extremely well, and there are some Titan cards that do as well. This makes it a unique opportunity for even members with older hardware to potentially be very high in the competition. If anybody has any information to add please chime in, and I'm happy to answer any questions I can!

Edit - one thing worth mentioning is the tasks you report will initially show as 'validation inconclusive' under the Milkyway@home account page. This happens if the task has not yet been validated by another user. There should not be any concern as long as the tasks do not show under the 'error' section in your tasks list.
 
Last edited:

emoga

Member
May 13, 2018
102
135
86
I wonder if we can reach 12th by the end of the challenge if we push it?

2020_10_13_18_15_49_Project_Team_Stats_for_TeAm_AnandTech_in_MilkyWay.png

Oh, and count me in ;)
 
  • Like
Reactions: Icecold

Icecold

Senior member
Nov 15, 2004
447
130
116
I think taking over 12th place is totally attainable! Thank you for posting the screenshot emoga!


aye-let-there-be-war.jpg
 

waffleironhead

Diamond Member
Aug 10, 2005
6,500
54
101
Im in, with as much power as my i can muster. Which isnt much tbh. My gpus are almost all integrated, with the exception of an rx550.
One surprising item in my mini fleet is the igp in my A12-9800 system, the IGP has 1/2 DP.
I may have to dust out my old x4 620 system with a hd6950 in it.
 
  • Like
Reactions: biodoc and Icecold

Fardringle

Diamond Member
Oct 23, 2000
8,768
298
126
It'll miss the first day or two, but I should have a 280X online and running for this sometime on Friday. :)

Not sure my old Nvidia cards are even worth putting on the project (750ti and 1060 3GB)..


I did a little bit of math, and assuming that the other teams keep their current production, if we can double our team's current output of about 10.2 million per day, we'll take over 12th place in just about 35 days, and could even move into 11th place just before the end of the year. That's assuming we can add that much, and that we keep that increase the whole time. Obviously, any Formula Sprint on a GPU project would change those numbers a bit...
 
Last edited:
  • Like
Reactions: Icecold

Icecold

Senior member
Nov 15, 2004
447
130
116
itstime.png


Starting stats(I'll only include members with points during the competition time moving forward, but this is a full list of all members on TeAm Anandtech with points as of today) - behind a spoiler tag since it's 200+ rows:

Name
Total Credit
1004717819​
802906827​
525964600​
346072284​
342746011​
333970173​
302964461​
283448179​
235762371​
208564571​
162063962​
118503745​
112501910​
103264391​
100721007​
100338213​
87743841​
Stig
76531966​
75837096​
57354132​
52458649​
mk
46121113​
42732038​
42130948​
36686962​
36378072​
36191533​
31424307​
31202122​
30084312​
28744562​
27806824​
26986003​
26369625​
25787059​
25201164​
23891055​
18466688​
18440410​
18280718​
17406495​
17183246​
17053886​
16029689​
15400984​
13961034​
Huge
13801154​
13109568​
12482468​
11220222​
Alyx
10991883​
9510186​
9499861​
9261830​
8877262​
8392455​
8255900​
7640690​
6970738​
6537212​
6349190​
6093078​
5800996​
5728890​
5702464​
4843534​
4739277​
4023255​
3974870​
3840927​
3793090​
3679178​
zzzs
3601807​
2373824​
IEC
2126890​
2007626​
1996711​
Mark
1569110​
1412898​
1339796​
1153390​
1114090​
1050969​
1006798​
973140​
864829​
730956​
716342​
612345​
515374​
513216​
510366​
495311​
449269​
399527​
396500​
348889​
346086​
336190​
315025​
312921​
308161​
307135​
284707​
281860​
268975​
260504​
231427​
Nick
217936​
191422​
186468​
179692​
170271​
zoic
166530​
164636​
157544​
147191​
138103​
122966​
lein
121742​
Nick
118173​
Ben
114003​
100326​
wiz
96808​
96039​
95788​
85641​
66241​
hca
62849​
Azam
57222​
54723​
52563​
52546​
G
51598​
50045​
48620​
Dia
41914​
41773​
39877​
38407​
37871​
37252​
mark
36821​
33284​
Mide
32213​
31994​
30829​
Dave
29759​
28593​
28070​
26485​
E.J.
23482​
21610​
15607​
15604​
Mike
15098​
14692​
11688​
11205​
9808​
8995​
8656​
8073​
8025​
7562​
7054​
6627​
6590​
6382​
6257​
6197​
5845​
4983​
4744​
4437​
3278​
3226​
3043​
2870​
2816​
2691​
2587​
1981​
1952​
1669​
1618​
Akos
1283​
1176​
1092​
1015​
983​
855​
745​
Kris
428​
428​
390​
307​
286​
228​
214​
214​
Ken
214​
205​
161​
130​
F35
104​
46​
Chad
10​
8​
1​
 

Endgame124

Senior member
Feb 11, 2008
391
228
116
Wow, switching to Milky Way added +50W to the power draw on the system (using integrated video in addition to the r250).

It’s completing 2 tasks at a time though, in a little less than 7 minutes each. Wild change from 2 days for a F@H task.
 

Assimilator1

Elite Member
Nov 4, 1999
23,505
202
106
Haven't run MW properly in ages! I'm in :).

Initially with just my RX 580, I may well fire up my 2nd rig with it's HD 7870 XT too, but I've got to be a little careful with heat now seeing as I have a couple of Gerbils living in my lounge now :). I don't want to cook my babies! ;). But I expect with the window open it'll be ok, I'll experiment with that tomorrow.

Currently drawing ~116w with 12T Rosetta, switching MW on (with Rosetta on 10T)....... it's drawing ~222w.
 
Last edited:

Endgame124

Senior member
Feb 11, 2008
391
228
116
I’m generating a fair number of inclusive results. Is this normal, or do I have a problem on my hands?
 

Icecold

Senior member
Nov 15, 2004
447
130
116
Haven't run MW properly in ages! I'm in :).

Initially with just my RX 580, I may well fire up my 2nd rig with it's HD 7870 XT too, but I've got to be a little careful with heat now seeing as I have a couple of Gerbils living in my lounge now :). I don't want to cook my babies! ;). But I expect with the window open it'll be ok, I'll experiment with that tomorrow.

Currently drawing ~116w with 12T Rosetta, switching MW on (with Rosetta on 10T)....... it's drawing ~222w.
I definitely don't want the friendly competition / push on Milkyway to cause any harm to any gerbils, so be careful on that :)

I'm glad to hear you're in - I see your benchmark thread all the time so in my mind you're kind of 'the Milkyway guy'.

I’m generating a fair number of inclusive results. Is this normal, or do I have a problem on my hands?
That is normal. It will show them as validation inconclusive until the other person that has the same task returns the same result. The verbiage is confusing - I had the same concern when I first ran it, I would have expected it to show validation pending instead. I'll add that to the initial post.
 
  • Like
Reactions: Assimilator1

Icecold

Senior member
Nov 15, 2004
447
130
116
Having ran the numbers, we as a TeAm have a really solid chance of hitting position 12 overall - so as much as this is a fun competition between the TeAm it is just as much us coming together as a TeAm for a common goal, and I think we can reach it! I really appreciate all the participation we've had so far! I will post the stats tomorrow morning.

battle.jpg
 

lane42

Diamond Member
Sep 3, 2000
5,425
338
126
Endgame, Validation inconclusive" means your result didn't match with your wingman's result, so another copy will be sent out as a "tie breaker". If that third copy agrees with your result, then the two of you will get credit.
 

Fardringle

Diamond Member
Oct 23, 2000
8,768
298
126
Having ran the numbers, we as a TeAm have a really solid chance of hitting position 12 overall - so as much as this is a fun competition between the TeAm it is just as much us coming together as a TeAm for a common goal, and I think we can reach it! I really appreciate all the participation we've had so far! I will post the stats tomorrow morning.
Yep. If we can double our current production, we'll take over the #12 spot overall in 35 days and move from 8th into 6th place in the Formula BOINC marathon in 22 days. That's a lot of production to add, but certainly possible.
 

biodoc

Diamond Member
Dec 29, 2005
5,654
910
126
Endgame, Validation inconclusive" means your result didn't match with your wingman's result, so another copy will be sent out as a "tie breaker". If that third copy agrees with your result, then the two of you will get credit.
That's true on all boinc projects except MilkyWay. As Icecold mention, "validation inconclusive" means "validation pending". Weird huh?

Here's my tasks status:

State: All (16148) · In progress (456) · Validation pending (0) · Validation inconclusive (1777) · Valid (13915) · Invalid (0) · Error (0)
 

Icecold

Senior member
Nov 15, 2004
447
130
116
Taking foot out of Butt, sorry Endgame:oops:
That seems to be one of the oddities of Milkyway, and I would have mentioned it in the original post had I remembered that. Your advice was very informative and appreciated - Milkyway just handles it different for whatever reason :)
 

Fardringle

Diamond Member
Oct 23, 2000
8,768
298
126
So my "new" R9 280X was delivered a full 30 hours before USPS said that it would be, so it's up and running on MilkyWay already! :D

It's too big to fit in the second PC I was going to put it in, so it's in the 4X PCIe slot on my X750 Aorus Elite motherboard next to my GTX 1060 (16X slot) for now. I'm letting them both run MilkyWay to see how badly the 1060 does before decide whether to leave it on MilkyWay for the race or move it back to SRBase.

Also need to determine if the 280X isn't getting enough bandwidth for MilkyWay on the 4X slot so that I need to swap the cards...
 

Skillz

Member
Feb 14, 2014
132
110
116
Wow, switching to Milky Way added +50W to the power draw on the system (using integrated video in addition to the r250).

It’s completing 2 tasks at a time though, in a little less than 7 minutes each. Wild change from 2 days for a F@H task.
I think the Radeon VII can do them in less than a minute (4 or 5 at a time) lol

With that being said. I wont be crunching when the competition starts, but I should be in a few days after it does start.

@fardingle, 4x should be fine for MW. I don't think it comes close to saturating a 8x or 16x pipe line. You will definitely pull that 1060 off MW when you see just how badly it does. lol
 
  • Like
Reactions: Icecold

ASK THE COMMUNITY