Benchmarks

Wanless Mersenne +2
Forum rules
davidbam
General Bitchin'
General Bitchin'
Posts: 6371
Joined: Wed Aug 15, 2018 1:15 pm
Location: Huntly, Scotland
Contact:

#1 Benchmarks

Post by davidbam »

:geek:

Linux-only app so please clarify if your results are from a VM
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
Image
User avatar
Bryan
Boinc Brigadier
Boinc Brigadier
Posts: 2621
Joined: Thu May 21, 2015 6:18 pm

#2 Re: Benchmarks

Post by Bryan »

2990WX native 64t @ 3.3G

Code: Select all

34714481 	16490691 	4 Mar 2020, 5:48:46 UTC 	5 Mar 2020, 11:05:12 UTC 	Completed and validated 	3,077.13 	3,075.13 	32.14 	Random-base WEP Factorization v1.13
34714154 	16490528 	4 Mar 2020, 5:43:08 UTC 	5 Mar 2020, 11:04:49 UTC 	Completed and validated 	3,070.89 	3,069.42 	31.80 	Random-base WEP Factorization v1.13
34713207 	16490054 	4 Mar 2020, 5:36:11 UTC 	5 Mar 2020, 11:04:22 UTC 	Completed and validated 	3,071.90 	3,065.40 	32.25 	Random-base WEP Factorization v1.13
34712911 	16489906 	4 Mar 2020, 5:31:19 UTC 	5 Mar 2020, 10:54:28 UTC 	Completed and validated 	3,075.46 	3,069.99 	31.57 	Random-base WEP Factorization v1.13
34712596 	16489760 	4 Mar 2020, 5:30:14 UTC 	5 Mar 2020, 10:46:56 UTC 	Completed and validated 	3,073.42 	3,069.51 	32.11 	Random-base WEP Factorization v1.13
34712865 	16489888 	4 Mar 2020, 5:25:39 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,077.72 	3,071.66 	32.42 	Random-base WEP Factorization v1.13
34712753 	16489837 	4 Mar 2020, 5:24:29 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,078.13 	3,070.49 	32.44 	Random-base WEP Factorization v1.13
34712828 	16489870 	4 Mar 2020, 5:24:29 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,072.51 	3,068.67 	32.21 	Random-base WEP Factorization v1.13
34712791 	16489852 	4 Mar 2020, 5:19:32 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,076.85 	3,068.76 	32.30 	Random-base WEP Factorization v1.13
34712815 	16489864 	4 Mar 2020, 5:19:32 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,078.39 	3,073.04 	32.21 	Random-base WEP Factorization v1.13
34712720 	16489822 	4 Mar 2020, 5:13:06 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,069.70 	3,065.24 	31.40 	Random-base WEP Factorization v1.13
34712735 	16489828 	4 Mar 2020, 5:13:06 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,068.45 	3,065.52 	32.10 	Random-base WEP Factorization v1.13
34712622 	16489773 	4 Mar 2020, 5:10:38 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,077.34 	3,075.85 	32.35 	Random-base WEP Factorization v1.13
34712624 	16489774 	4 Mar 2020, 5:10:38 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,078.71 	3,076.60 	32.34 	Random-base WEP Factorization v1.13
34712661 	16489793 	4 Mar 2020, 5:10:38 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,066.12 	3,063.44 	32.04 	Random-base WEP Factorization v1.13
34712614 	16489769 	4 Mar 2020, 5:09:29 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,071.59 	3,067.44 	32.31 	Random-base WEP Factorization v1.13
34712616 	16489770 	4 Mar 2020, 5:09:29 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,077.49 	3,074.53 	32.33 	Random-base WEP Factorization v1.13
34712618 	16489771 	4 Mar 2020, 5:09:29 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,071.65 	3,070.36 	32.28 	Random-base WEP Factorization v1.13
34712657 	16489791 	4 Mar 2020, 5:09:29 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,073.38 	3,069.43 	32.10 	Random-base WEP Factorization v1.13
34712652 	16489788 	4 Mar 2020, 5:08:55 UTC 	5 Mar 2020, 10:43:12 UTC 	Completed and validated 	3,074.67 	3,071.59 	31.18 	Random-base WEP Factorization v1.13
Image
User avatar
Bryan
Boinc Brigadier
Boinc Brigadier
Posts: 2621
Joined: Thu May 21, 2015 6:18 pm

#3 2990wx at 3.3GHz

Post by Bryan »

Wattage calcs will be wrong
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
3077.13 3075.13 3257,7569,626Random-base WEP Factorization v1.13
3070.89 3069.42 3257,2619,543Random-base WEP Factorization v1.13
3071.90 3065.40 3258,0529,675Random-base WEP Factorization v1.13
3075.46 3069.99 3256,7629,460Random-base WEP Factorization v1.13
3073.42 3069.51 3257,7719,629Random-base WEP Factorization v1.13
3077.72 3071.66 3258,2489,708Random-base WEP Factorization v1.13
3078.13 3070.49 3258,2769,713Random-base WEP Factorization v1.13
3072.51 3068.67 3257,9689,661Random-base WEP Factorization v1.13
3076.85 3068.76 3258,0489,675Random-base WEP Factorization v1.13
3078.39 3073.04 3257,8589,643Random-base WEP Factorization v1.13
3069.70 3065.24 3156,5629,427Random-base WEP Factorization v1.13
3068.45 3065.52 3257,8479,641Random-base WEP Factorization v1.13
3077.34 3075.85 3258,1299,688Random-base WEP Factorization v1.13
3078.71 3076.60 3258,0859,681Random-base WEP Factorization v1.13
3066.12 3063.44 3257,7839,630Random-base WEP Factorization v1.13
3071.59 3067.44 3258,1669,694Random-base WEP Factorization v1.13
3077.49 3074.53 3258,0909,682Random-base WEP Factorization v1.13
3071.65 3070.36 3258,1119,685Random-base WEP Factorization v1.13
3073.38 3069.43 3257,7549,626Random-base WEP Factorization v1.13
3074.67 3071.59 3156,0759,346Random-base WEP Factorization v1.13
3,074.083,070.103257,7309,622Averages for Ryzen 2990wx, 250W, 64 WU per proc
Image
OldChap
Boinc Private
Boinc Private
Posts: 10
Joined: Sat Apr 27, 2019 11:49 pm

#4 E5-2670 v1 native vs. VM

Post by OldChap »

Native
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
4260.674231.203210,4703,794Random-base WEP Factorization v1.13
4267.604232.613310,5573,825Random-base WEP Factorization v1.13
4267.534231.773210,4503,786Random-base WEP Factorization v1.13
4268.534244.043310,5683,829Random-base WEP Factorization v1.13
4264.614233.963310,5673,829Random-base WEP Factorization v1.13
4277.404243.583310,5393,819Random-base WEP Factorization v1.13
4266.264236.133210,4143,773Random-base WEP Factorization v1.13
4270.464240.463210,4303,779Random-base WEP Factorization v1.13
4264.384235.653310,6523,860Random-base WEP Factorization v1.13
4269.704233.773210,4423,783Random-base WEP Factorization v1.13
4266.664237.773210,2583,717Random-base WEP Factorization v1.13
4278.514243.693310,6143,846Random-base WEP Factorization v1.13
4269.654234.133210,3963,767Random-base WEP Factorization v1.13
4264.314227.243210,3803,761Random-base WEP Factorization v1.13
4264.264227.463210,4063,770Random-base WEP Factorization v1.13
4268.574231.943210,4383,782Random-base WEP Factorization v1.13
4269.724240.413310,6493,858Random-base WEP Factorization v1.13
4264.814232.853210,4413,783Random-base WEP Factorization v1.13
4277.714234.903210,2703,721Random-base WEP Factorization v1.13
4261.434225.743310,6603,862Random-base WEP Factorization v1.13
4,268.144,234.973210,4803,797Averages for E5-2670 v1 @ 2.60GHz, 115W, 16 WU per proc

and in a VM
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
11789.4311773.71313,5791,297Random-base WEP Factorization v1.13
11806.4611788.04313,6031,305Random-base WEP Factorization v1.13
11819.0111804.28313,5781,296Random-base WEP Factorization v1.13
9754.899742.32243,4311,243Random-base WEP Factorization v1.13
11837.6211823.46303,5051,270Random-base WEP Factorization v1.13
11828.4611805.76303,5371,281Random-base WEP Factorization v1.13
11829.4911814.48313,5651,292Random-base WEP Factorization v1.13
11834.8411820.98303,5241,277Random-base WEP Factorization v1.13
11837.8911825.68313,5651,292Random-base WEP Factorization v1.13
11828.7411818.48313,5731,294Random-base WEP Factorization v1.13
11827.7311817.74313,5661,292Random-base WEP Factorization v1.13
11836.8711819.93313,6051,306Random-base WEP Factorization v1.13
11829.7611817.18313,5951,302Random-base WEP Factorization v1.13
11830.7711823.44313,6151,310Random-base WEP Factorization v1.13
11833.7811822.71313,5701,293Random-base WEP Factorization v1.13
11830.7111822.21303,5601,290Random-base WEP Factorization v1.13
11834.8911821.18313,5661,292Random-base WEP Factorization v1.13
11848.1211839.84303,5501,286Random-base WEP Factorization v1.13
11828.6111818.10313,5751,295Random-base WEP Factorization v1.13
11835.9111818.12303,5601,290Random-base WEP Factorization v1.13
11,725.2011,711.88303,5611,290Averages for E5-2670 v1 @ 2.60GHz, 115W, 16 WU per proc
Trotador
Boinc Recruit
Posts: 1
Joined: Sat Nov 09, 2019 4:25 pm

#5 Platinum 8176 ES

Post by Trotador »

Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
5245.19 5224.15 3229,4817,445Random-base WEP Factorization v1.13
5259.67 5225.43 3229,7687,517Random-base WEP Factorization v1.13
5259.56 5223.73 3229,3097,401Random-base WEP Factorization v1.13
5260.59 5225.16 3229,3497,411Random-base WEP Factorization v1.13
5249.37 5225.37 3229,3387,409Random-base WEP Factorization v1.13
5255.54 5224.69 3229,3777,418Random-base WEP Factorization v1.13
5251.38 5224.29 3329,9537,564Random-base WEP Factorization v1.13
5259.42 5234.27 3330,1017,601Random-base WEP Factorization v1.13
5259.30 5224.55 3330,4517,690Random-base WEP Factorization v1.13
5270.68 5225.96 3330,4227,682Random-base WEP Factorization v1.13
5252.36 5223.62 3229,8197,530Random-base WEP Factorization v1.13
5262.57 5225.20 3229,5317,457Random-base WEP Factorization v1.13
5250.47 5223.53 3229,2677,391Random-base WEP Factorization v1.13
5258.58 5224.68 3229,6187,479Random-base WEP Factorization v1.13
5255.64 5224.41 3229,5617,465Random-base WEP Factorization v1.13
5251.52 5223.87 3229,5477,461Random-base WEP Factorization v1.13
5255.59 5224.51 3229,6357,484Random-base WEP Factorization v1.13
5256.51 5224.17 3330,2287,633Random-base WEP Factorization v1.13
5256.50 5223.02 3330,0357,585Random-base WEP Factorization v1.13
5260.48 5224.27 3229,7457,511Random-base WEP Factorization v1.13
5,256.555,224.943229,7277,507Averages for Platinum 8176 ES, 165W, 56 WU per proc
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#6 Re: Benchmarks

Post by Dirk Broer »

Odroid-XU4 octa-core (4x Cortex-A15 plus 4x Cortex-A7 in big.LITTLE) running Hardkernel-supplied Ubuntu Mate 20.04, estimated 13.2 Watt power draw, 8 WU's per proc -all cores running @1400MHz
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
14251.7113543.01301,4534,587Random-base WEP Factorization v1.15
14175.5813464.04301,4484,571Random-base WEP Factorization v1.15
14317.2713627.30301,4414,549Random-base WEP Factorization v1.15
14282.1413557.93301,4334,525Random-base WEP Factorization v1.15
13965.9913280.82311,5194,796Random-base WEP Factorization v1.15
13924.3813217.27291,4574,599Random-base WEP Factorization v1.15
14409.4313766.33301,4274,505Random-base WEP Factorization v1.15
14126.9213441.38301,4574,599Random-base WEP Factorization v1.15
14162.8213491.80291,3994,415Random-base WEP Factorization v1.15
14194.6813513.17301,4404,545Random-base WEP Factorization v1.15
13981.0613276.21291,4574,601Random-base WEP Factorization v1.15
14477.7613828.80291,4014,423Random-base WEP Factorization v1.15
14130.8013452.16301,4434,555Random-base WEP Factorization v1.15
14165.7613489.94291,4284,508Random-base WEP Factorization v1.15
14187.8613537.65291,4144,463Random-base WEP Factorization v1.15
14040.1213373.10291,4094,448Random-base WEP Factorization v1.15
14260.6413560.00291,3934,396Random-base WEP Factorization v1.15
14067.2013375.72291,4314,518Random-base WEP Factorization v1.15
14467.2713790.09301,4184,475Random-base WEP Factorization v1.15
14463.6613509.54301,4174,474Random-base WEP Factorization v1.15
14,202.6513,504.81291,4344,527Averages for Odroid-XU4, 13.2 Watt, 8 WUs per proc
This Odroid-XU4 is cooled with the passive heatsink of the XU4Q, on which a Noctua NF-A4x20 is mounted,
using the handy Noctua-supplied OmniJoin tool (comes with the fan) together with the original Odroid fan cable:
Image.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#7 Re: Benchmarks

Post by Dirk Broer »

Raspberry Pi 3 (4x Cortex-A53) running 32-bit Raspbian Pi OS Buster, real power draw at 3.7 Watt, 4 WUs per proc. Four cores running @1200 MHz
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
18494.1117513.84295476,157Random-base WEP Factorization v1.15
18522.9617520.55305606,308Random-base WEP Factorization v1.15
18550.0517521.78295446,128Random-base WEP Factorization v1.15
18532.6017522.84295466,145Random-base WEP Factorization v1.15
18499.7917516.97295496,185Random-base WEP Factorization v1.15
18515.2217520.71295496,184Random-base WEP Factorization v1.15
18511.7717519.65305516,204Random-base WEP Factorization v1.15
18470.4917514.39305636,338Random-base WEP Factorization v1.15
18470.2017511.57295496,180Random-base WEP Factorization v1.15
18454.7017510.01305636,337Random-base WEP Factorization v1.15
18462.4617512.91295516,202Random-base WEP Factorization v1.15
18438.9317508.74305566,267Random-base WEP Factorization v1.15
18483.2317511.86295506,188Random-base WEP Factorization v1.15
18478.0717512.40305556,247Random-base WEP Factorization v1.15
18484.7717515.23305546,241Random-base WEP Factorization v1.15
18486.9717515.56295366,040Random-base WEP Factorization v1.15
18483.4317515.67305546,241Random-base WEP Factorization v1.15
18513.4017520.52305526,216Random-base WEP Factorization v1.15
18539.2117519.41305526,220Random-base WEP Factorization v1.15
18520.2617520.28295496,184Random-base WEP Factorization v1.15
18,495.6317,516.24305526,211Averages for Raspberry Pi3, Power draw at 3.7Watt, 4 WUs per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#8 Re: Benchmarks

Post by Dirk Broer »

Raspberry Pi 4 (4x Cortex-A72) running 64-bit Debian Buster, 15W (5V/3A), 4 WUs per proc. Four cores running @2000 MHz, GPU @500 MHz
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
7303.946788.12291,3913,864Random-base WEP Factorization v1.15
7302.706788.13291,3853,846Random-base WEP Factorization v1.15
7283.346787.94291,3933,870Random-base WEP Factorization v1.15
7289.446787.89291,3953,876Random-base WEP Factorization v1.15
7289.206787.74291,3763,823Random-base WEP Factorization v1.15
7281.206788.66291,4003,888Random-base WEP Factorization v1.15
7289.236788.34301,4013,890Random-base WEP Factorization v1.15
7303.486789.53301,4023,893Random-base WEP Factorization v1.15
7283.726789.03291,3953,875Random-base WEP Factorization v1.15
7302.056788.65301,4033,898Random-base WEP Factorization v1.15
7292.266788.04291,3963,878Random-base WEP Factorization v1.15
7299.696790.16301,4013,893Random-base WEP Factorization v1.15
7306.966790.56301,3993,885Random-base WEP Factorization v1.15
7305.656789.35301,3993,887Random-base WEP Factorization v1.15
7307.396789.83301,4123,923Random-base WEP Factorization v1.15
7308.416790.72301,4143,928Random-base WEP Factorization v1.15
7302.576789.69301,3983,885Random-base WEP Factorization v1.15
7299.886789.41301,4063,906Random-base WEP Factorization v1.15
7296.246790.32301,4013,891Random-base WEP Factorization v1.15
7305.746789.32301,4013,892Random-base WEP Factorization v1.15
7,297.656,789.07301,3983,885Averages for Raspberry Pi4, 15W, 4 WU per proc
This Raspberry Pi 4 (8GB model) is cooled with a tower model 52Pi Ice Tower cooler.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#9 Re: Benchmarks

Post by Dirk Broer »

Odroid-N2+ (4x Cortex-A73 + 2x Cortex-A53) running 64-bit Armbian Buster, corrected for 6.2 Watt power draw, 6 WUs per proc, Four A73 cores running @2400 MHz, two A53 cores running @2016 MHz
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
9457.409249.39291,60810,808Random-base WEP Factorization v1.15
9463.309260.06301,62710,934Random-base WEP Factorization v1.15
9460.449257.78301,62310,908Random-base WEP Factorization v1.15
9467.489256.13301,63310,973Random-base WEP Factorization v1.15
9489.529254.14301,62010,885Random-base WEP Factorization v1.15
9491.489256.64301,62510,923Random-base WEP Factorization v1.15
9483.909236.74301,62310,910Random-base WEP Factorization v1.15
9517.159270.94291,60210,766Random-base WEP Factorization v1.15
9497.029254.51301,61710,869Random-base WEP Factorization v1.15
9482.899231.35301,63310,974Random-base WEP Factorization v1.15
9511.229262.20291,59610,725Random-base WEP Factorization v1.15
9502.059248.15281,55310,438Random-base WEP Factorization v1.15
9487.659236.03291,59010,686Random-base WEP Factorization v1.15
9514.529260.15291,60210,769Random-base WEP Factorization v1.15
9507.829252.91291,59810,736Random-base WEP Factorization v1.15
9499.459242.60291,60210,768Random-base WEP Factorization v1.15
9508.209257.88291,59310,706Random-base WEP Factorization v1.15
9505.179250.18291,59410,713Random-base WEP Factorization v1.15
9493.019238.43291,59710,735Random-base WEP Factorization v1.15
9488.049238.82291,59910,748Random-base WEP Factorization v1.15
9,491.399,250.75291,60710,799Averages for Odroid-N2+, 6.2 Watt, 6 WU per proc
On top of this the Odroid-N2+ stays cool at 37°C, due to the huge heatsink and (set to 'always-on') 80mm fan.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#10 Re: Benchmarks

Post by Dirk Broer »

nVidia Jetson Nano (4x Cortex-A57) running 64-bit L4T (Linux for Tegra), 3.8 Watt power draw, 4 WUs per proc. Four Cortex-A57 cores running @1400 MHz
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
11388.3810790.093193510,252Random-base WEP Factorization v1.15
11317.9710788.373092510,138Random-base WEP Factorization v1.15
10741.0210257.462994510,362Random-base WEP Factorization v1.15
11345.7710789.983092010,083Random-base WEP Factorization v1.15
11338.5810788.96309109,973Random-base WEP Factorization v1.15
11350.2210789.533193110,210Random-base WEP Factorization v1.15
11305.2010787.203092110,099Random-base WEP Factorization v1.15
11302.6210788.113092110,102Random-base WEP Factorization v1.15
11311.9810789.04298879,728Random-base WEP Factorization v1.15
11304.4110787.28309099,969Random-base WEP Factorization v1.15
11339.6610789.52309099,962Random-base WEP Factorization v1.15
11322.8610787.86309079,950Random-base WEP Factorization v1.15
11326.1110788.923091310,011Random-base WEP Factorization v1.15
11337.8810788.82298809,646Random-base WEP Factorization v1.15
11327.8310790.09309009,872Random-base WEP Factorization v1.15
11325.7710787.89298909,763Random-base WEP Factorization v1.15
11343.0910787.51308999,855Random-base WEP Factorization v1.15
11324.3510788.25298809,654Random-base WEP Factorization v1.15
11342.6310788.37308999,862Random-base WEP Factorization v1.15
11341.3710788.95309009,867Random-base WEP Factorization v1.15
11,301.8910,762.11309099,968Averages for nVidia Tegra X1 (ARM Cortex-A57 in Jetson Nano), @3.8 Watt, 4 WUs per proc
I replaced my Raspberry Pi 2's with Jetson Nano's -using the exact same 5V/3A power supplies. I could not find any performance differences between the 2GB and the 4GB model as yet
(though some projects will try to offer WUs beyond the capabilities of the 2GB model)
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#11 Re: Benchmarks

Post by Dirk Broer »

After the success over overclocking my 8Gb Raspberry Pi 4 to 2000 MHz and running 64-bit Debian Linux on it, I decided to overclock my two 4 GB models also to 2000 MHz.
Using the 52Pi low-profile cooler for both, i decided to use thermal paste on one and use the supplied adhesive thermal pads for the other. I also overclocked the GPUs of these two as described here how -and why. To my surprise the thermal paste only gave better temperatures, the performance crown went to the pads.

Raspberry Pi 4 (4x Cortex-A72) running 64-bit Debian Buster, 15W (5V/3A), 4 WUs per proc. Four cores running @2000 MHz, GPU @750 MHz.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
6881.076754.81311,5654,346Random-base WEP Factorization v1.15
6879.256754.41301,4934,146Random-base WEP Factorization v1.15
6884.416754.84301,4814,114Random-base WEP Factorization v1.15
6885.236754.93301,5224,227Random-base WEP Factorization v1.15
6883.646755.14301,5244,233Random-base WEP Factorization v1.15
6885.036755.34291,4734,091Random-base WEP Factorization v1.15
6878.636754.95301,4974,158Random-base WEP Factorization v1.15
6887.116755.90301,5154,210Random-base WEP Factorization v1.15
6882.866754.09301,4894,137Random-base WEP Factorization v1.15
6874.186753.38301,5264,240Random-base WEP Factorization v1.15
6875.546754.40301,4914,143Random-base WEP Factorization v1.15
6874.366752.65301,5114,196Random-base WEP Factorization v1.15
6871.196752.13301,5014,169Random-base WEP Factorization v1.15
6873.686752.14301,4994,165Random-base WEP Factorization v1.15
6871.966752.68301,5124,199Random-base WEP Factorization v1.15
6866.906752.33301,5104,194Random-base WEP Factorization v1.15
6870.416752.61311,5664,351Random-base WEP Factorization v1.15
6869.136752.99301,5084,190Random-base WEP Factorization v1.15
6870.086752.85301,4964,156Random-base WEP Factorization v1.15
6866.666751.72301,5104,196Random-base WEP Factorization v1.15
6,876.576,753.71301,5094,193Averages for Raspberry Pi 4, 15W, 4 WU per proc
Image
davidbam
General Bitchin'
General Bitchin'
Posts: 6371
Joined: Wed Aug 15, 2018 1:15 pm
Location: Huntly, Scotland
Contact:

#12 Re: Benchmarks

Post by davidbam »

"the thermal paste only gave better temperatures, the performance crown went to the pads"

Yes, that is rather surprising - wanless is very consistent in WU runtimes and that helps make comparisons
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#13 Re: Benchmarks

Post by Dirk Broer »

With all these ARM SBC's I have used -by lack of TDP- the wattage of the PSU for the Watt value, but that is way too much, compared to real life power consumption.
For the Raspberry Pi 4 I have found that, when you have the latest firmware installed, the real power draw is some 6.41 Watt, getting confirmed here.
But you'll say: "That applies to a stock speed Pi 4 (1500 MHz)", and right you are, I just overclock the sh%t out of the Pi 4, and that comes at a penalty, paid in Watts. The Wattage when running 2000 MHz comes at 11W, as shown here
What does this mean for e.g. the previous table?

Averages for Raspberry Pi 4 Model B 4GB, @11Watt power draw (corrected for overclocking), 4 WUs per processor running @2000MHz. GPU @750 MHz.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
6873.876752.59301,5225,765Random-base WEP Factorization v1.15
6873.636752.78301,5275,786Random-base WEP Factorization v1.15
1893.321864.2381,5155,739Random-base WEP Factorization v1.15
6869.876752.23301,5135,730Random-base WEP Factorization v1.15
6870.286752.68301,5065,703Random-base WEP Factorization v1.15
6878.016753.82301,5005,683Random-base WEP Factorization v1.15
6885.786754.42301,5065,703Random-base WEP Factorization v1.15
6881.076754.81311,5655,926Random-base WEP Factorization v1.15
6883.756754.89301,5075,707Random-base WEP Factorization v1.15
6879.256754.41301,4935,654Random-base WEP Factorization v1.15
6884.416754.84301,4815,610Random-base WEP Factorization v1.15
6885.236754.93301,5225,765Random-base WEP Factorization v1.15
6883.646755.14301,5245,772Random-base WEP Factorization v1.15
6885.036755.34291,4735,579Random-base WEP Factorization v1.15
4464.724388.67201,5215,762Random-base WEP Factorization v1.15
6878.636754.95301,4975,669Random-base WEP Factorization v1.15
6887.116755.90301,5155,740Random-base WEP Factorization v1.15
6882.866754.09301,4895,641Random-base WEP Factorization v1.15
6874.186753.38301,5265,782Random-base WEP Factorization v1.15
6875.546754.40301,4915,649Random-base WEP Factorization v1.15
6,509.516,391.43281,5105,718Averages for BCM2711 (ARM Cortex-A72 in Raspberry Pi 4) @2000 MHz, 11Watt power draw, 4 WU per proc
Image
davidbam
General Bitchin'
General Bitchin'
Posts: 6371
Joined: Wed Aug 15, 2018 1:15 pm
Location: Huntly, Scotland
Contact:

#14 Re: Benchmarks

Post by davidbam »

I suggest you just delete the first table TBH, perhaps just noting the /kwH figure if TDP is used
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#15 Re: Benchmarks

Post by Dirk Broer »

davidBAM wrote: Sun May 23, 2021 12:39 pm I suggest you just delete the first table TBH, perhaps just noting the /kwH figure if TDP is used
Deleted the 6.41 Watt Pi4 version -and recalculated the Pi3 table for 3.7 Watt, the Jetson Nano for 3.8 Watt, the Odroid-XU4 for 13.2(!) Watt and the Odroid-N2+ for 6.2 Watt...
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#16 Re: Benchmarks

Post by Dirk Broer »

Averages for Odroid-C1, @5.1 Watt power draw, 4 WUs per processor, running @1540 MHz.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
19316.0318959.30295194,242Random-base WEP Factorization v1.15
18906.6318893.85295344,365Random-base WEP Factorization v1.15
19052.7218915.01295284,317Random-base WEP Factorization v1.15
13001.7712886.75205284,315Random-base WEP Factorization v1.15
19057.8618915.55285114,177Random-base WEP Factorization v1.15
19056.7118915.28295324,350Random-base WEP Factorization v1.15
19515.4518997.87295214,257Random-base WEP Factorization v1.15
19497.9119000.38295174,220Random-base WEP Factorization v1.15
19510.3218999.89305254,288Random-base WEP Factorization v1.15
19513.5218997.16295204,247Random-base WEP Factorization v1.15
19423.4318984.09295244,281Random-base WEP Factorization v1.15
19425.6118983.63305414,417Random-base WEP Factorization v1.15
19398.7118986.03305334,356Random-base WEP Factorization v1.15
19421.6918984.05295204,249Random-base WEP Factorization v1.15
19364.6318980.44295214,253Random-base WEP Factorization v1.15
19381.9618977.46295174,226Random-base WEP Factorization v1.15
19379.7318977.32295194,238Random-base WEP Factorization v1.15
19384.0418978.45295184,236Random-base WEP Factorization v1.15
19682.6819022.39295104,169Random-base WEP Factorization v1.15
19674.6119021.05295154,209Random-base WEP Factorization v1.15
19,048.3018,668.80295234,271Averages for Amlogic S805 (ARM Cortex-A5 in Odroid-C1), 5.1 Watt tdp, 4 WU per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#17 Re: Benchmarks

Post by Dirk Broer »

Averages for Banana Pi M2, @5 Watt power draw, 4 WUs per processor, running @1000 MHz.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
29124.7028086.37263062,551Random-base WEP Factorization v1.15
29594.2528201.29283272,728Random-base WEP Factorization v1.15
29272.6328199.08293372,807Random-base WEP Factorization v1.15
29254.1528198.56283342,785Random-base WEP Factorization v1.15
29272.5928198.91263102,581Random-base WEP Factorization v1.15
31538.5928227.91252772,311Random-base WEP Factorization v1.15
29588.6328201.40283262,716Random-base WEP Factorization v1.15
29598.0028202.03263072,559Random-base WEP Factorization v1.15
31552.0928230.09262842,363Random-base WEP Factorization v1.15
29576.2128202.35283272,728Random-base WEP Factorization v1.15
31537.5828228.44262802,330Random-base WEP Factorization v1.15
31553.1228228.39262832,361Random-base WEP Factorization v1.15
30,121.8828,200.40273082,568Averages for Allwinner A31s (ARM Cortex-A7 in Banana PI M2), 5 Watt tdp, 4 WU per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#18 Re: Benchmarks

Post by Dirk Broer »

Averages for Asus Tinker Board S, @5 Watt power draw, 4 WUs per processor, running @1800 MHz.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
12211.6812171.31298196,823Random-base WEP Factorization v1.15
12201.8612165.85298236,857Random-base WEP Factorization v1.15
11274.0711244.93298947,452Random-base WEP Factorization v1.15
12199.9312168.35308417,006Random-base WEP Factorization v1.15
11,971.8911,937.61298447,034Averages for RK3288 (ARM Cortex-A17 in Asus Tinker Board S), 5 Watt tdp, 4 WU per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#19 Re: Benchmarks

Post by Dirk Broer »

Averages for Banana Pi M2, @5 Watt power draw, 4 WUs per processor, running @1008 MHz under a Noctua NC-U6 chipset cooler with 60mm Noctua NF-A6x25 fan.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
32914.0924440.10313302,752Random-base WEP Factorization v1.15
32962.0624443.88313302,751Random-base WEP Factorization v1.15
32952.7224442.44313272,729Random-base WEP Factorization v1.15
32939.4124441.02313272,727Random-base WEP Factorization v1.15
35332.7624450.54323152,625Random-base WEP Factorization v1.15
35357.5424444.69323132,609Random-base WEP Factorization v1.15
35315.0724448.74323092,577Random-base WEP Factorization v1.15
32792.0324429.69313302,747Random-base WEP Factorization v1.15
32794.0924428.03313222,686Random-base WEP Factorization v1.15
32790.9924426.48313242,701Random-base WEP Factorization v1.15
32814.7824426.47303192,662Random-base WEP Factorization v1.15
32859.5824427.38303202,668Random-base WEP Factorization v1.15
32781.3424427.09303122,601Random-base WEP Factorization v1.15
32803.7424428.36293082,570Random-base WEP Factorization v1.15
32776.7824424.04303122,596Random-base WEP Factorization v1.15
34529.2924620.45292912,423Random-base WEP Factorization v1.15
34477.8824625.28292882,403Random-base WEP Factorization v1.15
34454.5324628.88292872,391Random-base WEP Factorization v1.15
34513.0724625.51292932,442Random-base WEP Factorization v1.15
34772.7524654.39282782,318Random-base WEP Factorization v1.15
33,646.7324,484.17303122,599Averages for Banana Pi M2 (SOC: Allwinner-A31s), 5Watt, 4 WU per proc
So the better cooling brought lower temperatures, but longer runtimes(!), albeit with less CPU time. Also slightly more credit on average per day/proc and a slightly better credit/kWH.
Image
Miklos M
Butterfly Whisperer / Milkweeder
Butterfly Whisperer / Milkweeder
Posts: 1452
Joined: Fri Aug 14, 2020 3:39 pm
Location: Florida, USA

#20 Re: Benchmarks

Post by Miklos M »

For SRBase what is a good running time if it is 2 up on a 2080 ti?
Image
Miklos M
Butterfly Whisperer / Milkweeder
Butterfly Whisperer / Milkweeder
Posts: 1452
Joined: Fri Aug 14, 2020 3:39 pm
Location: Florida, USA

#21 Re: Benchmarks

Post by Miklos M »

For me it is over 23 minutes and still running, seems too slow.
Image
davidbam
General Bitchin'
General Bitchin'
Posts: 6371
Joined: Wed Aug 15, 2018 1:15 pm
Location: Huntly, Scotland
Contact:

#22 Re: Benchmarks

Post by davidbam »

See the SrBase Benchmarks
viewtopic.php?f=259&t=19996&p=104991#p104991

- this area is for Wanless
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#23 Re: Benchmarks

Post by Dirk Broer »

What is the effect of putting a big Chinese HB-802 Northbridge cooler with a 5V Noctua 80mm fan on an Odroid-C1 on CPU time and running time? And perhaps more important: will the extra Wattage for the fan be justified?
Averages for Odroid-C1, @5.1 Watt power draw, 4 WUs per processor, running @1540 MHz, extremely cooled.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
18977.9718912.69336104,983Random-base WEP Factorization v1.15
18989.5718916.61346235,093Random-base WEP Factorization v1.15
18983.1318914.19346215,072Random-base WEP Factorization v1.15
18988.2818916.01346215,072Random-base WEP Factorization v1.15
18983.8918914.81325914,832Random-base WEP Factorization v1.15
18986.4918916.61325804,738Random-base WEP Factorization v1.15
18986.6318915.00325834,766Random-base WEP Factorization v1.15
18980.5418911.85325864,787Random-base WEP Factorization v1.15
18986.1318915.27325794,732Random-base WEP Factorization v1.15
18976.4518910.63325844,769Random-base WEP Factorization v1.15
18989.3118914.68315724,670Random-base WEP Factorization v1.15
18990.3018918.11315714,669Random-base WEP Factorization v1.15
18985.2518914.82315714,667Random-base WEP Factorization v1.15
18985.0418915.80325764,709Random-base WEP Factorization v1.15
18985.9418914.06315674,633Random-base WEP Factorization v1.15
18991.0118913.36315674,631Random-base WEP Factorization v1.15
18993.0918917.62315634,598Random-base WEP Factorization v1.15
18986.0518912.87315604,573Random-base WEP Factorization v1.15
18983.4218914.06315664,624Random-base WEP Factorization v1.15
18987.4818914.78315724,672Random-base WEP Factorization v1.15
18,985.8018,914.69325834,764Averages for Amlogic S805 (ARM Cortex-A5 in Odroid-C1), 5.1 Watt tdp, 4 WU per proc
Other than with the super-cooled Banana Pi M2 we see here less run time, but more CPU time. The Credit per kWH goes up from 4,271 to 4,764, the credit/day from 523 to 583.
Image
davidbam
General Bitchin'
General Bitchin'
Posts: 6371
Joined: Wed Aug 15, 2018 1:15 pm
Location: Huntly, Scotland
Contact:

#24 pi4 OC

Post by davidbam »

Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
3186.27 3167.24 373,96115,004Random-base WEP Factorization v1.16
3180.12 3165.30 373,97115,042Random-base WEP Factorization v1.16
3207.17 3168.65 363,88914,731Random-base WEP Factorization v1.16
2961.43 2933.81 343,90914,809Random-base WEP Factorization v1.16
3175.27 3166.01 343,73314,141Random-base WEP Factorization v1.16
3181.87 3166.38 353,81114,437Random-base WEP Factorization v1.16
3193.00 3167.30 343,64513,808Random-base WEP Factorization v1.16
3198.40 3168.13 373,94814,956Random-base WEP Factorization v1.16
3188.79 3166.09 373,96915,034Random-base WEP Factorization v1.16
3200.98 3167.28 353,80314,404Random-base WEP Factorization v1.16
3182.41 3165.17 363,93814,916Random-base WEP Factorization v1.16
3179.40 3164.74 353,83614,530Random-base WEP Factorization v1.16
3219.28 3168.73 353,77614,302Random-base WEP Factorization v1.16
3186.59 3166.01 363,89814,765Random-base WEP Factorization v1.16
3190.95 3167.45 373,97315,048Random-base WEP Factorization v1.16
3177.05 3164.34 353,85714,611Random-base WEP Factorization v1.16
3182.52 3166.30 373,97115,043Random-base WEP Factorization v1.16
3222.45 3170.21 363,90914,807Random-base WEP Factorization v1.16
3181.05 3165.98 363,92114,852Random-base WEP Factorization v1.16
3185.70 3165.66 363,95514,982Random-base WEP Factorization v1.16
3,179.043,155.04363,88414,711Averages for pi4oc, 11W, 4 WU per proc
I think this is fool-proof but could you just try it for me please? • There are 10 types of people in the world; those who understand binary, and those who don’t
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#25 Re: Benchmarks

Post by Dirk Broer »

I thought WEP-M+2 to have a bandwidth dependency, due to the much higher results when switching (how apt) from a 100 Mbit switch to a Gigabyte switch.
But the new benchmark figures show that it is far more likely that the Formula Boinc Battle between TSBT and P3D has lead to huge numbers of WUs to get verified against...
Averages for Asus Tinker Board S @1800MHz, 5 Watt (real) power draw, 4 WUs per processor.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
11241.7111187.64288687,237Random-base WEP Factorization v1.15
11245.3211184.64288677,225Random-base WEP Factorization v1.15
11248.5111187.32288607,164Random-base WEP Factorization v1.15
11242.0711186.05298807,332Random-base WEP Factorization v1.15
11242.5411185.72288587,152Random-base WEP Factorization v1.15
11246.0011184.19288557,124Random-base WEP Factorization v1.15
11246.9611186.63288667,216Random-base WEP Factorization v1.15
11244.1211186.89288727,269Random-base WEP Factorization v1.15
11238.3511184.84288627,181Random-base WEP Factorization v1.15
11246.2311184.49288687,234Random-base WEP Factorization v1.15
11248.4211186.33288567,131Random-base WEP Factorization v1.15
11245.3211184.94319537,939Random-base WEP Factorization v1.15
11245.0311187.03298787,320Random-base WEP Factorization v1.15
11244.7911185.76298887,402Random-base WEP Factorization v1.15
11241.9311185.05319658,039Random-base WEP Factorization v1.15
11235.5811185.45298927,431Random-base WEP Factorization v1.15
11225.6111183.56298987,481Random-base WEP Factorization v1.15
11219.5111183.59278447,031Random-base WEP Factorization v1.15
11220.1911184.02299057,541Random-base WEP Factorization v1.15
11219.6811183.15288587,146Random-base WEP Factorization v1.15
11,239.3911,185.36298807,330Averages for RockChip RK3288 (Quad-core Cortex-A17), 5 Watt powerdraw, 4 WUs per proc
Compared to DavidBAM's Pi4 benchmarks, the quad-core 32-bit flagship gets totally clobbered against its quad-core 64-bit opponent.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#26 Re: Benchmarks

Post by Dirk Broer »

But how does my 64-bit flagship do under the new switch?
Odroid-N2+ (Amlogic S922X with 4x Cortex-A73 + 2x Cortex-A53) running 64-bit Armbian Buster, 6.2 Watt power draw, 6 WUs per proc, four Cortex-A73 cores running @2400 MHz, two Cortex-A53 cores running @2016 MHz.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
3392.083327.34284,25528,593Random-base WEP Factorization v1.16
3464.203407.25284,17528,058Random-base WEP Factorization v1.16
3491.313426.19294,28728,808Random-base WEP Factorization v1.16
3454.003385.83284,18928,151Random-base WEP Factorization v1.16
3424.853360.78284,22328,381Random-base WEP Factorization v1.16
3401.863336.91284,23828,480Random-base WEP Factorization v1.16
3360.203297.30284,36929,362Random-base WEP Factorization v1.16
3612.363548.22284,04827,207Random-base WEP Factorization v1.16
3381.163307.83284,28128,768Random-base WEP Factorization v1.16
3493.173423.70294,23128,434Random-base WEP Factorization v1.16
3430.343361.81284,29328,853Random-base WEP Factorization v1.16
3513.403446.69294,24428,518Random-base WEP Factorization v1.16
3439.683366.53284,18428,117Random-base WEP Factorization v1.16
3428.283358.89284,24628,535Random-base WEP Factorization v1.16
3490.563420.98284,21028,296Random-base WEP Factorization v1.16
3430.413366.08284,25628,599Random-base WEP Factorization v1.16
3466.373401.61284,22228,373Random-base WEP Factorization v1.16
3492.783429.97284,19128,168Random-base WEP Factorization v1.16
3404.683338.30284,28928,825Random-base WEP Factorization v1.16
3493.883428.63294,24528,528Random-base WEP Factorization v1.16
3,453.283,387.04284,23428,453Averages for Amlogic S922X, 6.2W, 6 WU per proc
Note that it is running the 64-bit v1.16 application, instead of the 32-bit v1.15 in the older benchmark....
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#27 Re: Benchmarks

Post by Dirk Broer »

Benchmarks for Raspberry Pi 3 Model B (old model with small WiFi chip @1200 MHz) under 64-bit Raspberry Pi OS Bullseye:
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
9379.529055.97321,18513,340Random-base WEP Factorization v1.16
9408.649058.09331,20313,543Random-base WEP Factorization v1.16
9408.389057.67331,21013,626Random-base WEP Factorization v1.16
9393.799056.92331,23113,867Random-base WEP Factorization v1.16
9358.629055.12341,24714,039Random-base WEP Factorization v1.16
6561.246353.97231,22513,797Random-base WEP Factorization v1.16
9354.989054.05341,25614,149Random-base WEP Factorization v1.16
9405.069055.52341,24814,057Random-base WEP Factorization v1.16
9407.449056.13361,33915,079Random-base WEP Factorization v1.16
9407.899055.79341,25014,078Random-base WEP Factorization v1.16
9407.089055.32351,27614,364Random-base WEP Factorization v1.16
9384.149057.01361,31014,748Random-base WEP Factorization v1.16
9381.109055.93351,30214,657Random-base WEP Factorization v1.16
9380.149055.53341,23613,916Random-base WEP Factorization v1.16
9377.089055.48361,30914,746Random-base WEP Factorization v1.16
9558.129071.37331,20713,596Random-base WEP Factorization v1.16
9644.629073.13351,24213,982Random-base WEP Factorization v1.16
9683.089071.80341,21113,637Random-base WEP Factorization v1.16
9643.589073.88361,30814,726Random-base WEP Factorization v1.16
9414.119056.90341,25814,163Random-base WEP Factorization v1.16
9,297.938,924.28341,25314,106Averages for ARM Cortex-A53, 3.7 Watt tdp, 4 WU per proc
3.7 Watt tdp per
400% CPU load (stress --cpu 4) at 730 mA (3.7 W)
found here
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#28 Re: Benchmarks

Post by Dirk Broer »

Since I asked Bearnol for a 64-bit OS for 64-bit capable ARM cores -and he immediately replied- the field has changed, at least for Wanless.
Given the rising prices of electricity, can you really afford to run single- or dual-core 32-bit crunchers on this project?

Quad-core 32-bit SBCs compared
ModelCPUNumber of CoresArchitectureOSCredit/day
Raspberry Pi Model B+ARM11 @1000MHz (OC)Single CoreArmv632-bit40
BeagleBone BlackCortex-A8 @1000MHzSingle CoreArmv732-bit68
Banana ProCortex-A7 @1000MHzDual CoreArmv732-bit195
Raspberry Pi 2 Model BCortex-A7 @1000MHz (OC)Quad CoreArmv732-bit435
Banana Pi M2Cortex-A7 @1000MHzQuad CoreArmv732-bit440
Humming BoardCortex-A9 @1000MHzQuad CoreArmv732-bit495
Raspberry Pi 3 Model BCortex-A53 @1200MHzQuad CoreArmv832-bit550
Odroid-C1Cortex-A5 @1500MHzQuad CoreArmv732-bit580
Tinker Board SCortex-A17 @1800MHzQuad CoreArmv732-bit975
Quad-core 64-bit SBCs compared
ModelCPUNumber of CoresArchitectureOSCredit/day
Raspberry Pi3 Model BCortex-A53 @1200MHzQuad CoreArmv864-bit1130
Odroid-M1Cortex-A55 @1992MHzQuad CoreArmv864-bit2000
Jetson NanoCortex-A57 @1400MHzQuad CoreArmv864-bit2450
Raspberry Pi4 Model BCortex-A72 @2000MHz (OC)Quad CoreArmv864-bit3300
For the sake of comparison I split the tables into 32- and 64 bit, with the Raspberry Pi 3 Model B featuring in both.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#29 Re: Benchmarks

Post by Dirk Broer »

The new Odroid-M1, running Debian 11 with kernel 5.18 -installed straight from the internet:
Image
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
5448.675407.24322,04019,317Random-base WEP Factorization v1.16
5447.625406.30322,05219,435Random-base WEP Factorization v1.16
5453.065410.40322,00819,019Random-base WEP Factorization v1.16
5449.945408.13322,04419,354Random-base WEP Factorization v1.16
5450.075409.20322,03319,252Random-base WEP Factorization v1.16
5455.835415.33322,05219,435Random-base WEP Factorization v1.16
5455.815416.50352,19420,779Random-base WEP Factorization v1.16
5455.815415.79322,02919,214Random-base WEP Factorization v1.16
5456.725415.48332,09819,864Random-base WEP Factorization v1.16
5438.505409.90322,04619,371Random-base WEP Factorization v1.16
5438.495409.69372,33322,097Random-base WEP Factorization v1.16
5438.535409.55322,05819,491Random-base WEP Factorization v1.16
5436.265404.45332,07619,656Random-base WEP Factorization v1.16
5449.825402.78332,06519,559Random-base WEP Factorization v1.16
5449.835402.82322,03119,235Random-base WEP Factorization v1.16
5449.885403.33311,99418,886Random-base WEP Factorization v1.16
5446.415402.81322,03819,301Random-base WEP Factorization v1.16
5445.425402.75322,00018,938Random-base WEP Factorization v1.16
5445.435401.83322,03919,310Random-base WEP Factorization v1.16
5433.745392.86311,98918,834Random-base WEP Factorization v1.16
5,447.295,407.36322,06119,517Averages for Odroid-M1 with ARM Cortex-A55@1992MHz, 4.4Watt TDP, 4 WUs per quad-core processor
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#30 Re: Benchmarks

Post by Dirk Broer »

A more detailed comparison between the various 32-bit ARM SBC's -between the lines the reason why the single- and dual core boards are retired...

Image (click to enlarge)

The ARM-Cortex-A7 based boards may soon be retired in favour of the more recent 64-bit ARM Cortex-A55 boards (e.g. Banana Pi M5, Odroids C4 and M1, Rock Pi 3A and 3B), as the Odroid M1 gets around 2000 daily credits on WEP-M+2 at the same power draw. And you can run 32-bit apps on a 64-bit board, but not the other way around...
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#31 Re: Benchmarks

Post by Dirk Broer »

Those so-called 'super coolers' I use, do they help? Let's take -a closer look (Do I hear Chris Barnatt from explaining computers.com?)

First we tackle the Odroid-XU4 with its Xigmatek N881 Porter and 80mm Noctua fan
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
13326.8112697.72321,6655,257Random-base WEP Factorization v1.15
13313.0512667.75321,6465,197Random-base WEP Factorization v1.15
13254.9212608.98321,6625,246Random-base WEP Factorization v1.15
13475.0412865.10321,6515,210Random-base WEP Factorization v1.15
13362.7412739.61321,6395,174Random-base WEP Factorization v1.15
13478.0912900.08331,6735,282Random-base WEP Factorization v1.15
13586.0113002.26331,6805,304Random-base WEP Factorization v1.15
13194.5312585.16321,6865,321Random-base WEP Factorization v1.15
13802.8813257.41331,6325,153Random-base WEP Factorization v1.15
13671.8113113.30331,6445,190Random-base WEP Factorization v1.15
13729.1213145.19331,6455,192Random-base WEP Factorization v1.15
13065.9112445.38331,7605,556Random-base WEP Factorization v1.15
13139.8312562.30321,6845,317Random-base WEP Factorization v1.15
14113.0813531.27341,6805,304Random-base WEP Factorization v1.15
13614.1013056.22341,7335,470Random-base WEP Factorization v1.15
13484.3312863.96331,7085,391Random-base WEP Factorization v1.15
13315.6112713.01331,7375,484Random-base WEP Factorization v1.15
13417.3312739.95331,6825,311Random-base WEP Factorization v1.15
13662.4813026.44321,6415,180Random-base WEP Factorization v1.15
13193.7212535.01321,7025,373Random-base WEP Factorization v1.15
13,460.0712,852.81331,6785,296Averages for Odroid-XU4 with Samsung Exynos5422 (Cortex-A15-big/A7-LITTLE), Power Consumption 13.2 Watt, 8 WUs per proc
old values: 14,202.65 — 13,504.81 — 29 — 1,434 — 4,527 (Averages for Odroid-XU4, 13.2 Watt, 8 WUs per proc) under a XU4Q heatsink, aided with a 40mm Noctua fan (the thick 5V version). So, yes we can say that the sturdy Xigmatek does its job. Its USP (Unique Point of Selling) for Odroid-use is that the base is 4x4cm, just like Odroid heatsinks for Odroids-C1, -C1+,-C2,-C4 and -XU4/XU4Q -and the new Odroid-N2L!

How does that other Odroid do now, with its Chinese PCCooler HB-802 cooler, also aided by a 5V 80mm Noctua fan?
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
18916.2818902.47325894,815Random-base WEP Factorization v1.15
18933.0118902.44335974,878Random-base WEP Factorization v1.15
19517.0518990.15345994,890Random-base WEP Factorization v1.15
19912.5419054.68335754,701Random-base WEP Factorization v1.15
19926.0719056.59345884,804Random-base WEP Factorization v1.15
19743.7419028.66345914,831Random-base WEP Factorization v1.15
15861.7815379.40275894,815Random-base WEP Factorization v1.15
19754.3919033.40356114,988Random-base WEP Factorization v1.15
19827.5319042.83345954,857Random-base WEP Factorization v1.15
19824.9519043.33345974,877Random-base WEP Factorization v1.15
19882.7619049.05335734,678Random-base WEP Factorization v1.15
19896.5419052.41345904,822Random-base WEP Factorization v1.15
19892.1819055.58356014,913Random-base WEP Factorization v1.15
11526.8811023.09205904,821Random-base WEP Factorization v1.15
19903.7319053.51355994,894Random-base WEP Factorization v1.15
19585.2519003.77325694,645Random-base WEP Factorization v1.15
19586.3119005.12335814,746Random-base WEP Factorization v1.15
19841.4719046.23335754,699Random-base WEP Factorization v1.15
19876.7219048.61345834,764Random-base WEP Factorization v1.15
19853.1019044.61325594,570Random-base WEP Factorization v1.15
19,103.1118,440.80325884,800Averages for Amlogic S805, Power consumption 5.1 Watt, 4 WUs per proc
That was 19,048.30 — 18,668.80 — 29 — 523 — 4,271 (Averages for Amlogic S805 (ARM Cortex-A5 in Odroid-C1), 5.1 Watt tdp, 4 WUs per proc) in its original tiny heatsink configuration. It looks like a bigger cooling surface, combined with dito fan, does work in your favour.

The Banana Pi M2 then, featuring Austria's chipset cooling pride, the sadly OOP (Out Of Production) Noctua NC-U6 combined with a Noctua 60mm fan
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
24833.5224176.14334573,809Random-base WEP Factorization v1.15
24839.8324177.04334643,866Random-base WEP Factorization v1.15
24860.6324183.54344703,920Random-base WEP Factorization v1.15
24801.7724164.98334553,790Random-base WEP Factorization v1.15
24808.0524170.51334573,809Random-base WEP Factorization v1.15
24833.5224176.14334573,809Random-base WEP Factorization v1.15
24839.8324177.04334643,866Random-base WEP Factorization v1.15
24860.6324183.54344703,920Random-base WEP Factorization v1.15
24801.7724164.98334553,790Random-base WEP Factorization v1.15
24808.0524170.51334573,809Random-base WEP Factorization v1.15
24833.5224176.14334573,809Random-base WEP Factorization v1.15
24839.8324177.04334643,866Random-base WEP Factorization v1.15
24860.6324183.54344703,920Random-base WEP Factorization v1.15
24801.7724164.98334553,790Random-base WEP Factorization v1.15
24808.0524170.51334573,809Random-base WEP Factorization v1.15
24833.5224176.14334573,809Random-base WEP Factorization v1.15
24839.8324177.04334643,866Random-base WEP Factorization v1.15
24860.6324183.54344703,920Random-base WEP Factorization v1.15
24801.7724164.98334553,790Random-base WEP Factorization v1.15
24808.0524170.51334573,809Random-base WEP Factorization v1.15
24,828.7624,174.44334613,839Averages for Allwinner-A31s,Power consumption 5 Watt, 4 WUs per proc
That was 30,121.88 — 28,200.40 — 27 — 308 — 2,568 (Averages for Allwinner A31s (ARM Cortex-A7 in Banana PI M2), 5 Watt tdp, 4 WUs per proc) in its original tiny heatsink configuration. Here yet again the Texas credo 'bigger is better' holds true: larger cooling surface, combined with bigger fan brings you lower temperatures and higher scores.

But....not high enough, when you compare the Credits/KWh values with those of 64-bit boards like the Odroid-N2+ or -M1. So the big coolers are going to move 64-bitwards, beginning with the Noctua NC-U6 and the PCCooler HB-802. Even the Xigmatek N881 Porter might find itself on an Odroid-C4 soon, as that only uses a quarter of the power that an Odroid-XU4 needs.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#32 Re: Benchmarks

Post by Dirk Broer »

How then does the Rock Pi 4B+ behave? This is as still under stock 4B speeds (1800 MHz for the two Cortex-A72 cores and 1400 MHz for the four Cortex-A53 cores)
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
6176.565249.57342,89012,040Random-base WEP Factorization v1.16
6111.105155.89352,95012,293Random-base WEP Factorization v1.16
6090.035141.26352,96312,346Random-base WEP Factorization v1.16
6169.145196.30363,01012,542Random-base WEP Factorization v1.16
6200.665250.66363,04812,701Random-base WEP Factorization v1.16
6057.845042.83363,08412,851Random-base WEP Factorization v1.16
6166.575189.27342,87411,976Random-base WEP Factorization v1.16
6054.805057.15373,14113,089Random-base WEP Factorization v1.16
6090.654994.31342,92012,168Random-base WEP Factorization v1.16
6138.795133.87342,91012,125Random-base WEP Factorization v1.16
6181.055156.19342,86911,955Random-base WEP Factorization v1.16
6239.005332.30363,01812,574Random-base WEP Factorization v1.16
6017.284920.32342,96812,366Random-base WEP Factorization v1.16
6242.055331.04363,02912,620Random-base WEP Factorization v1.16
6196.575204.92352,88612,026Random-base WEP Factorization v1.16
6308.855392.15352,90812,117Random-base WEP Factorization v1.16
6147.905091.78342,87811,991Random-base WEP Factorization v1.16
6199.445179.85342,86611,940Random-base WEP Factorization v1.16
6210.395210.67342,87911,996Random-base WEP Factorization v1.16
6214.495198.78332,74511,439Random-base WEP Factorization v1.16
6,160.665,171.46352,94212,258Averages for RK3399 (2 Cortex-A72/4 Cortex-A53), Estimated 10 Watt Power Consumption, 6 WUs per proc
Cooling by means of the Radxa 'floor' heatsink, coupled to a 53mm high black-aluminium Radxa case, the whole standing on an old Pentium Pro heatsink tied to the old Odroid-XU4Q heatsink, the lower stand cooled by a 40mm Noctua fan, while on top of the case a 80mm Noctua sucks out the heat. It doesn't get any cooler than that by air (CPUs at 35⁰C under 100% full load). The 10 Watt power consumption is just a wild guess, it might very well be lower. It is just that I haven't found any reliable figures on the net (I found as low as 6 Watt and as high as 19 Watt) and I am unable to measure it myself at the moment.
Overclocking -that is to say setting the RK3399 to OP1 speeds, 2000 MHz for the two Cortex-A72 cores and 1500 MHz for the four Cortex-A53 cores- brings the following values (CPUs at 44⁰C under 100% full load):
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
5603.954679.49333,08012,831Random-base WEP Factorization v1.16
5626.184718.68343,08912,873Random-base WEP Factorization v1.16
5618.124684.18373,41214,218Random-base WEP Factorization v1.16
5622.144728.07343,12913,036Random-base WEP Factorization v1.16
5744.834898.02302,70411,265Random-base WEP Factorization v1.16
5558.324600.75333,06412,766Random-base WEP Factorization v1.16
5516.834540.75333,10612,940Random-base WEP Factorization v1.16
5586.134655.48333,09712,903Random-base WEP Factorization v1.16
5864.895090.20342,96512,353Random-base WEP Factorization v1.16
5528.054576.42333,13013,043Random-base WEP Factorization v1.16
5766.714925.72322,87411,975Random-base WEP Factorization v1.16
5560.424582.37333,07612,815Random-base WEP Factorization v1.16
5584.514624.06333,10012,919Random-base WEP Factorization v1.16
5659.034753.14333,03512,645Random-base WEP Factorization v1.16
5634.494716.52333,05312,720Random-base WEP Factorization v1.16
5664.854781.76343,14313,098Random-base WEP Factorization v1.16
5803.434986.25332,97412,390Random-base WEP Factorization v1.16
5720.044896.70343,06212,760Random-base WEP Factorization v1.16
5502.854511.68333,11612,985Random-base WEP Factorization v1.16
5683.024783.51343,06712,778Random-base WEP Factorization v1.16
5,642.444,736.69333,06412,766Averages for rk3399, 10W, 6 WU per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#33 Re: Benchmarks

Post by Dirk Broer »

The nVidia Jetson Xavier NX is a beast, and the yet-to-come-out Jetson Orin NX will be even better:
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
2550.032373.05285,74123,921Random-base WEP Factorization v1.16
2550.812373.10285,61923,414Random-base WEP Factorization v1.16
2550.152373.38275,51522,979Random-base WEP Factorization v1.16
2549.502372.63295,86624,442Random-base WEP Factorization v1.16
2549.972372.50285,73923,913Random-base WEP Factorization v1.16
2553.262373.78316,31226,301Random-base WEP Factorization v1.16
2551.402373.80295,85424,390Random-base WEP Factorization v1.16
2550.252373.39295,82424,266Random-base WEP Factorization v1.16
2548.222372.75295,86124,421Random-base WEP Factorization v1.16
2551.172373.43295,85224,384Random-base WEP Factorization v1.16
2551.422373.36295,85824,407Random-base WEP Factorization v1.16
2548.962372.62295,89824,575Random-base WEP Factorization v1.16
2550.172372.94295,85424,394Random-base WEP Factorization v1.16
2548.962373.04295,86124,422Random-base WEP Factorization v1.16
2550.652373.21295,86624,440Random-base WEP Factorization v1.16
2548.842372.98295,86624,440Random-base WEP Factorization v1.16
2545.882372.38295,83624,316Random-base WEP Factorization v1.16
2547.392372.38306,12125,506Random-base WEP Factorization v1.16
2548.162371.84295,85124,379Random-base WEP Factorization v1.16
2548.962372.72306,02025,083Random-base WEP Factorization v1.16
2,549.712,372.96295,86124,420Averages for nVidia Jetson Xavier NX, 10Watt tdp, 6 WUs per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#34 Re: Benchmarks: The costs of overclocking

Post by Dirk Broer »

The costs of overclocking: Go big, or go home. I decided not only to let my 8GB Pi4 run from an SSD, but also to overclock it to 2200MHz, AND to upgrade the kernel to 6.1 via sudo rpi-update next.
I was greeted by a black screen that was cured with un-commenting the lines in /boot/config.txt
hdmi_force_hotplug=1
config_hdmi_boost=4
and changing
dtoverlay=vc4-kms-v3d into dtoverlay=vc4-fkms-v3d -note the added 'f'
After that I got video, but no task panel. Changing the HDMI port cured that.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
2908.252879.99273,18610,793Random-base WEP Factorization v1.16
2908.262879.81273,18710,796Random-base WEP Factorization v1.16
2908.262879.62273,18710,796Random-base WEP Factorization v1.16
2908.562879.80263,10710,526Random-base WEP Factorization v1.16
2909.562880.97273,16810,731Random-base WEP Factorization v1.16
2907.662880.68273,15510,686Random-base WEP Factorization v1.16
2908.342879.91263,12210,575Random-base WEP Factorization v1.16
2909.352880.03273,19910,837Random-base WEP Factorization v1.16
2909.362880.81273,17010,740Random-base WEP Factorization v1.16
2909.352880.56242,8949,803Random-base WEP Factorization v1.16
2909.462880.88252,9209,891Random-base WEP Factorization v1.16
2914.802880.63263,12810,596Random-base WEP Factorization v1.16
2914.792880.69263,12710,592Random-base WEP Factorization v1.16
2912.422881.08263,11710,560Random-base WEP Factorization v1.16
2911.422880.22242,8839,767Random-base WEP Factorization v1.16
2905.462879.17273,15510,686Random-base WEP Factorization v1.16
2906.482880.06273,15310,682Random-base WEP Factorization v1.16
2906.482879.10273,15810,698Random-base WEP Factorization v1.16
2896.222878.34263,13410,615Random-base WEP Factorization v1.16
2896.222879.25263,06210,372Random-base WEP Factorization v1.16
2,908.042,880.08263,11110,537Averages for arm-cortex-a72, 12.3W, 4 WU per proc
The 12.3 Watt needed isn't what I bargained for...
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#35 Re: Benchmarks

Post by Dirk Broer »

I've started a new chapter in benchmarking: I bought myself a power consumption meter, to be able to have the exact wattage that a system runs on.
Image
A Brennenstuhl PM 231 E
So far we had the exact values for runtime, CPU time, and credits and we were able to give the exact Per day/proc score. The Credit per KWh however was an estimate based partly on the wattage, which in practice meant the TDP value of the CPU -if known, such as e.g. with ARM boards.
But in the future I will now be able to report the exact wattage of e.g. an ARM system when running Wanless -or any other project/BOINC platform.

The first competition will be between my best 32-bit boards: The Odroid-C1, the Odroid-XU4, and the ASUS Tinkerboard S on one side versus my worst 64-bit boards on the other side: the Raspberry Pi 3B's (both original version, not 3B+). All losers will be retired in favour of additional better performing 64-bit boards. Pretty expensive buggers though: the 8GB nVidia Jetson Orin Nano devkit and the 16GB Radxa Rock 5B are the most likely candidates.
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#36 Re: Benchmarks

Post by Dirk Broer »

The Odroid-C1, with its Amlogic-S805 eating ~3.8 Watt from the wall, running both BOINC and 5V 80mm Noctua fan cooling its four threads.
How efficient is it?
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
18961.9718898.08315566,101Random-base WEP Factorization v1.15
18963.7618899.51315726,275Random-base WEP Factorization v1.15
18970.8118901.63315686,230Random-base WEP Factorization v1.15
18971.0518900.04305486,008Random-base WEP Factorization v1.15
18966.3418900.41315626,166Random-base WEP Factorization v1.15
18977.7718902.63305526,056Random-base WEP Factorization v1.15
18969.1518900.67325776,327Random-base WEP Factorization v1.15
18974.5618901.19305556,089Random-base WEP Factorization v1.15
18966.9518902.36315706,254Random-base WEP Factorization v1.15
18969.0018900.91315726,273Random-base WEP Factorization v1.15
18971.8118901.54305546,070Random-base WEP Factorization v1.15
18966.4618900.95315696,238Random-base WEP Factorization v1.15
18971.4018900.48315736,278Random-base WEP Factorization v1.15
18967.7218901.42305556,087Random-base WEP Factorization v1.15
18965.7718900.98305546,074Random-base WEP Factorization v1.15
18968.5118901.12315686,225Random-base WEP Factorization v1.15
18972.6318899.99315626,166Random-base WEP Factorization v1.15
18965.7218901.38325906,470Random-base WEP Factorization v1.15
18968.4618899.94315586,113Random-base WEP Factorization v1.15
18972.6018902.17315656,194Random-base WEP Factorization v1.15
18,969.1218,900.87315646,185Averages for Odroid-C1 with Amlogic S805 (ARM Cortex-A5), Real Power consumption 3.8 Watt (including 5V Noctua 80mm fan), 4 WUs per proc
Despite the lower -but real- power usage parameter, we still see it fall short of the 10,000 Credits/kWh mark by quite a long way...
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#37 Re: Benchmarks

Post by Dirk Broer »

Argh! -As the pirate p4 said-, they're not gonna like me at ASUS headquarters...The power consumption of their Tinker Board S negates its otherwise good (32-bit-wise) daily scores. Their marketing department claimed 1A was all that was needed, which coupled to a 5V PSU would give 5 Watt (Volt x Ampère = Watt), only it is 9.3 Watt, so 1.86 Amps.... :D
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
12352.2912283.83339244,139Random-base WEP Factorization v1.15
12345.4612268.21339304,167Random-base WEP Factorization v1.15
12323.9912251.64339144,096Random-base WEP Factorization v1.15
12315.4412240.30339174,108Random-base WEP Factorization v1.15
12315.4512250.17339164,104Random-base WEP Factorization v1.15
12321.8612255.97339344,186Random-base WEP Factorization v1.15
12341.5312275.76339204,123Random-base WEP Factorization v1.15
12424.6412352.31339174,106Random-base WEP Factorization v1.15
12435.4212368.61349474,243Random-base WEP Factorization v1.15
12465.4512398.07359624,310Random-base WEP Factorization v1.15
12505.4812439.87359584,290Random-base WEP Factorization v1.15
12524.9512453.15359724,355Random-base WEP Factorization v1.15
12556.9112481.52349434,226Random-base WEP Factorization v1.15
12574.3612499.64339114,082Random-base WEP Factorization v1.15
12577.1312504.25359494,253Random-base WEP Factorization v1.15
12546.2912483.03359604,302Random-base WEP Factorization v1.15
12519.3812450.03359534,268Random-base WEP Factorization v1.15
12438.8412356.67339204,122Random-base WEP Factorization v1.15
12420.6712344.14359854,412Random-base WEP Factorization v1.15
12418.4812345.68339254,144Random-base WEP Factorization v1.15
12,436.2012,365.14349384,202Averages for Asus Tinker Board S (Quad Cortex-A17 RK3288), 9.3 Real Watts Tdp (including 40mm 52Pi Ice Tower cooler), 4 WUs per proc
Image
User avatar
Dirk Broer
Corsair
Corsair
Posts: 1962
Joined: Thu Feb 20, 2014 11:24 pm
Location: Leiden, South Holland, Netherlands
Contact:

#38 Re: Benchmarks

Post by Dirk Broer »

As my Odroid-M1 lost connection when the Tinker Board was decommissioned, I grabbed the chance to measure the M1.
Run TimeCPU TimeCreditPer day/procCredit per kWHApplication
5437.585399.97332,10912,208Random-base WEP Factorization v1.16
5439.705400.91332,09312,115Random-base WEP Factorization v1.16
5436.575398.51332,08812,081Random-base WEP Factorization v1.16
5432.495394.96342,16912,550Random-base WEP Factorization v1.16
5432.145394.21332,10412,176Random-base WEP Factorization v1.16
5432.575394.76332,08012,039Random-base WEP Factorization v1.16
5431.035394.19332,09712,134Random-base WEP Factorization v1.16
5431.685394.26342,19212,685Random-base WEP Factorization v1.16
5431.885392.79332,12412,290Random-base WEP Factorization v1.16
5430.885393.01332,10412,175Random-base WEP Factorization v1.16
5423.215395.48342,13512,354Random-base WEP Factorization v1.16
5435.455396.93342,13112,334Random-base WEP Factorization v1.16
5436.855398.20342,15012,445Random-base WEP Factorization v1.16
5435.485397.13332,12712,308Random-base WEP Factorization v1.16
5437.065399.17332,11812,257Random-base WEP Factorization v1.16
5417.595395.49352,25313,035Random-base WEP Factorization v1.16
5432.635394.50332,12412,292Random-base WEP Factorization v1.16
5417.205395.85332,07612,014Random-base WEP Factorization v1.16
5419.475397.55332,09612,130Random-base WEP Factorization v1.16
5419.085396.80342,15712,482Random-base WEP Factorization v1.16
5,430.535,396.23332,12612,305Averages for Odroid-M1 with RK3568 (quad-core Cortex-A55), 7.2 Watts real Tdp, 4 WUs per proc
So, that's better -but not as good as I had hoped for the M1....

Edit: Just before I disconnected the Brennenstuhl-o-meter I discovered that the watts for the M1 had stabilized to a 'mere' 6.9 Watt, still a far cry from the figures quoted by the Hardkernel marketing staff (4.4 Watt)...
Image
Post Reply Previous topicNext topic

Return to “Wanless”