• 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."

Discussion PES | Assessing Power and Performance Efficiency of x86 CPU architectures

May 1, 2020
70
81
51
Dear Community,

so this is my first thread here as a long-time lurker - but I felt the desire to share a small hobby-project of mine from the last couple of months with you...

Performance Efficiency Suite - What is it about?
Most Reviewers solely focus on what they consider to be the most important aspect of modern CPUs - the absolute performance. But this is only one side of the equation. Today Power Efficiency is at least as important - or to be more precise: The amount of energy (Wattseconds or Joules) a CPU needs in order to accomplish a given workload. Sadly most Reviewers shy away from the extra mile it needs to assess this aspect. This suite measures the Total Package Power of a CPU while running the Cinebench R23 benchmarks first in single-threaded mode (1 run), then running in multi-threaded mode (for 10 minutes + whatever it takes to finish the last run). The results will be rendered in the provided Results.xlsx Excel file. To combine Efficiency and Performance there is also a score provided called Performance Efficiency Score (how amazingly inspired I am ;)).

In the meantime I was able to aggregate more than 80 samples from members of the 3DC & CB communities (see below).

How-To
  1. Unzip the latest release to wherever you want EXCEPT on your local OneDrive folder.
  2. Open Settings.txt and insert your local Cinebench23 Directory.
  3. Run PES Start - it will ask for Administrator rights as these are needed for measuring Package Power
  4. Wait until the Powershell finishes.
  5. Open the Excel file...
  6. Allow external connections (to the generated CSV-files with the data)
  7. Go to Data -> Refresh all
  8. Enjoy and share your results - just take a screenshot of what the Excel renders.
  9. If you want to do multiple measurements with different settings just copy the Excel file (inside the root-folder) before running and refreshing the data.

Some explanations about the Suite
  • This Suite has been made possible by Michael Möller and his amazing free and open-source Open Hardware Monitor and his .NET Library OpenHardwareMonitorLib.dll - Thanks a lot!!!
    Homepage: https://openhardwaremonitor.org/
    GitHub: https://github.com/openhardwaremonitor
  • The results for the Package Power look pretty accurate compared to the sparse data the internet provides. Seems, that the vendors are much more honest with those sensors than they are with temperature etc.
  • The suite basically consists some powershell scripts and an Excel file for presentation purposes
    • RunAsAdminWrapper.ps1
      This is needed to have a convenient relative path shortcut in the root folder and request admin-rights at the same time
    • Main.ps1
      • After setting up some stuff it basically starts the Cinebench R23 one at a time. It then checks for the "Cinebench.exe" process being active.
      • While this is true it queries the Package Power Sensor data with a lower bound of 10ms (in order to keep CPU-load of the script at bay).
      • After each run the aquired data gets pushed to CSV files located in the LogCsv subfolder.
    • Results.xslx
      • The Excel file basically just does some import, calculations and a hopefully nice presentation of the data.
      • Histogram
        The bold line shows a running average of the last 100 data-points which should be sufficiently accurate. The pale line shows each single data-point.
      • Calculation of Total Package Consumption
        To get that number we need the integral. That is why we first calculate the timeframe between two data-points and then multiply the measured value.
      • Everything else in that Excel is hopefully more or less self-explaining

Online Resources

Disclaimer
I am by no means a Powershell professional or a professional Reviewer. I was just sick of the lack of information and wanted to propose a low-effort solution. Any input for further improvement is highly welcomed. Please feel free to use/extend/rip-off this solution as you wish. But please share your findings to the world.
 
Last edited:
May 1, 2020
70
81
51
Performance-Consumption-Matrices (Single-Thread & Multi-Thread)
The following charts show the samples with total consumption on the x-axis and performance on the y-axis. It is scaled in a way that performance and efficiency are weighted 1:1.
The more to the left a sample is, the more energy efficient it is in processing the fixed workload.
The more to the bottom a sample is, the faster it processes the workload.



 
Last edited:
May 1, 2020
70
81
51
Detailed sample data


CB23 Single-Thread results
Ref.PostCPUUserRemarkVer.PES STCons.Dur.STAvg.Pwr.ST
13DC #3R7 4700U (Renoir)CrazyIvanAC / Win: Best Perf. / HP: Recmd.v0.7.0143,1710432669,5715,58
23DC #6R5 3600 (Matisse)Lykav0.3.145,7632112680,547,19
33DC #7i7 1065G (IceLake)Naitsabesv0.3.1127,769839795,512,37
43DC #14R9 5950X (Vermeer)dosenfisch24v0.3.155,4135920502,4371,49
53DC #18R7 4750G (Renoir)Poekelv0.3.1153,8810352627,816,49
63DC #27R7 3700X (Matisse)TigersharkPBO onv0.3.151,830057642,346,8
73DC #29R7 4750U (Renoir)dosenfisch24v0.3.1137,8810396697,614,9
83DC #32R9 5950X (Vermeer)Sweepiv0.3.152,9437274506,7773,55
93DC #42R9 5900HS (Cezanne)MonkeyWin: Energy Savingv0.3.1111,7962391433,914,35
103DC #44R9 5900HS (Cezanne)Monkeyv0.3.1165,0910936553,8619,75
113DC #54i5 8365U (WhiskeyLake)MD_Enigmav0.3.188,2411657972,1511,99
123DC #69R5 PRO 4650G (Renoir)Tigersharkv0.3.1146,7410450653,1316,03
133DC #47R7 4750G (Renoir)Poekel25Wv0.3.1173,79122631,1214,45
143DC #3R7 4700U (Renoir)CrazyIvanBatt. / Win: Better Eff. / HP: Recmd.v0.3.1133,621016873613,8
153DC #38R9 5950X (Vermeer)Sweepiv0.3.15933870500,4267,68
163DC #65R9 5900HS (Cezanne)MonkeyWin: Best Perf.v0.3.1169,5510364569,1218,21
173DC #64R3 1200 (Summit Ridge)BlackArchonv0.3.131,132204998,3832,26
183DC #67R7 3700X (Matisse)TigersharkPBO offv0.3.155,0823918759,0731,51
193DC #68R9 5900X (Vermeer)Krischiv0.3.141,5545942523,9187,69
203DC #70R9 5950X (Vermeer)LeiwandErmanual Curve Optimizationv0.3.160,2933002502,5665,67
213DC #88R9 5950X (Vermeer)Lowkeyv0.5.062,6132182496,3264,84
223DC #90R9 5950X (Vermeer)misterhv0.5.063,9230783508,260,57
233DC #108i7 4820K (Ivy Bridge)Platos@4,5Ghzv0.3.117,45553731034,6453,52
243DC #102i7 1165G7 (TigerLake)misterhWin: Best Perf.v0.5.0172,4610777538,0620,03
253DC #94R9 5950X (Vermeer)misterh-0,1V Curve Optimizationv0.5.063,0428707552,5651,95
263DC #96R9 5950X (Vermeer)Sweepiv0.3.159,9733185502,5166,04
273DC #118R7 4750G (Renoir)Poekel20Wv0.5.1164,29800621,4315,77
283DC #129i7 5775C (Broadwell)MD_Enigmav0.5.155,0620078904,5922,2
293DC #133R5 4500U (Renoir)Poekelv0.5.1186,387582707,6810,71
303DC #134R9 5900HS (Cezanne)MonkeyWin: Better Eff.v0.5.0216,087445621,6511,98
313DC #135R9 5900X (Vermeer)harzer_knallerBalanced Power Planv0.5.160,1424336683,2335,62
323DC #136R9 5900X (Vermeer)Darkearth27v0.5.175,5725543518,0649,31
333DC #140R9 5900X (Vermeer)KrischiCTRv0.5.152,338103501,8475,93
343DC #141i7 2600K (Sandy Bridge)Tyrann@4,4Ghzv0.5.126,3838525983,8639,16
353DC #145R7 5800X (Vermeer)hq-hqv0.5.157,1334236511,2466,97
363DC #146i7 7500U (Kaby Lake)Tyrannv0.5.183,49110961079,3710,28
373DC #146Celeron N3450 (Apollo Lake)Tyrannv0.5.116,69181923293,495,52
383DC #148R7 5800X (Vermeer)patrock84v0.5.168,0628138522,1753,89
393DC #154i5 8600k (Coffee Lake)hq-hqv0.5.158,2527864616,0845,23
403DC #154i5 7500 (Kaby Lake)hq-hqv0.5.154,7420650884,6723,34
413DC #155i7 8700k (Coffee Lake)Bernman@5Ghzv0.5.161,5525887627,6241,25
423DC #156R7 5800H (Cezanne)Darkearth27v0.5.1168,7910124585,1817,3
433DC #160R9 5950X (Vermeer)GaryXv0.5.174,4426935498,7654
443DC #165R5 4600H (Renoir)GroschiWin 11v0.6.0158,598278761,7410,87
45CB #4R9 5900X (Vermeer)Asghan@95Wv0.6.058,1533913507,0766,88
46CB #5R5 5600X (Vermeer)mesohornyv0.6.090,0621193523,9240,45
47CB #9R7 3700X (Matisse)Puffer0815Outlier?v0.6.0101,2915775625,8425,21
48CB #10R5 3500U (Picasso)TenferenzuOutlier?v0.6.0147,3666191025,226,46
49CB #13R7 3700X (Matisse)Hardy72@95Wv0.6.069,3122812632,536,07
50CB #14R7 3700X (Matisse)Jon Dohnson@PBOv0.6.082,8819421621,2731,26
51CB #20i5 8250U (WhiskeyLake)Rabrogov0.6.0107,3910395895,7411,63
52CB #36i7 4800MQ (Haswell)DrAgOnBaLlOnEv0.6.040,92241291012,9123,82
53CB #49R5 3500U (Picasso)AsghanThinkPad E495v0.6.091,9790721198,557,57
56CB #57i7 9750H (Coffee Lake)Blende Upv0.6.0104,6513860689,4120,1
57CB #60i7 3770K (Ivy Bridge)Blende Upv0.6.035,72270731034,0926,28
58CB #60i5 4300U (Haswell)Blende Upv0.6.058,95133791267,910,55
59CB #39R5 2600X (Pinnacle Ridge)HasseLadebalkenv0.5.141,7430535784,5738,92
60CB #63i5 3320M (Ivy Bridge)noplan724v0.6.037,38189661410,713,44
61CB #83R5 3500U (Picasso)andi_scoOutlier?v0.6.043,45195681239,3214,98
62CB #102i7 2600 (Sandy Bridge)raser0248v0.6.028,37302921163,8226,03
63CB #102i3 6157U (Skylake)raser0248v0.6.0112,0369871277,455,47
64CB #112R7 3700X (Matisse)FabianoPBO off?v0.6.054,0729485627,2447,01
653DC #190R9 5900HS (Cezanne)MonkeyWin: Energy Savingv0.6.02565293737,977,17
663DC #204R7 5800X (Vermeer)patrock84v0.7.077,2224558527,3346,57
67CB #132R5 3500U (Picasso)TenferenzuThinkPad E495 defaultv0.7.0180,545863944,686,21
68CB #118i9 11980HK (TigerLake-8C)JeanLegior 11900H (Eng. Sample)v0.6.0147,4812519541,6223,11
69CB #137R5 3500U (Picasso)andi_scov0.7.035,34206031373,3815
70CB #140R5 3500U (Picasso)AsghanThinkPad E495 cool and quietv0.7.0144,3767171031,196,51
71CB #143i7 9750H (Coffee Lake)Blende Upv0.7.0111,0713063689,2418,95
72CB #149R7 2700X (Pinnacle Ridge)Tanzmusikusv0.7.050,2225952767,2833,82
73CB #152R5 3500U (Picasso)TanzmusikusWin = Balancedv0.7.078,0913745931,7314,75
743DC #205R5 4500U (Renoir)Poekelv0.7.01907302720,7810,13
753DC #212R5 2500U (Raven Ridge)Tiberiusv0.7.0126,4977991013,617,69
76CB #173R5 5600X (Vermeer)Freiherausv0.7.094,9220058525,2238,19
773DC #234R7 5800H (Cezanne)Fondnessv0.7.0210,668085587,1813,77
783DC #241R5 5600X (Vermeer)Scotyv0.7.078,3823969532,3145,03
793DC #242P Silver N6000 (JasperLake)Tralalakv0.7.295,02857712276,99
803DC #244Celeron N5100 (JasperLake)y33H@v0.7.265,8595051597,645,95
81CB #178R3 4300G (Renoir)Lord Maikiv0.7.0188,446350835,727,6
82CB #181i7 1165G7 (TigerLake)mkl1v0.7.0155,8411590553,6720,93
83CB #184i5 11500 (Rocket Lake)Freiherausv0.7.283,4720987570,8336,77
843DC #257i7 11700K (Rocket Lake)Triskainev0.7.283,9723459507,6446,21
85CB #186i5 11400F (Rocket Lake)zymotic-95mV offsetv0.7.2106,6416480568,9928,96
863DC #261R5 5600X (Vermeer)Holgiv0.7.275,8724717533,2246,35
873DC #279TR 1900X (Whitehaven)BlackArchonv0.7.226,6348597772,6162,9
88CB #214R9 5900X (Vermeer)Verangryv0.7.289,8923661470,1750,32
89AT #8R9 5950X (Vermeer)JoeRambo@4,4Ghz noSMTv0.7.294,3319142553,8234,56
90CB #218R9 5900X (Vermeer)Verangry@Stockv0.7.271,4326897520,4951,68
91AT #17i5 4690k (Haswell)zebrax2@Stockv0.7.240,9328989842,7434,4

CB23 Multi-Thread results
Ref.PostCPUUserRemarkVer.PES MTCons.Dur.MTAvg.Pwr.MT
13DC #3R7 4700U (Renoir)CrazyIvanAC / Win: Best Perf. / HP: Recmd.v0.7.02656,062410156,2215,43
23DC #6R5 3600 (Matisse)Lykav0.3.11386,39722399,8672,33
33DC #7i7 1065G (IceLake)Naitsabesv0.3.1885,223912288,7713,55
43DC #14R9 5950X (Vermeer)dosenfisch24v0.3.14779,3624233,52186,22
53DC #18R7 4750G (Renoir)Poekelv0.3.12637,56526272,0573,03
63DC #27R7 3700X (Matisse)TigersharkPBO onv0.3.12058,48637776,1883,71
73DC #29R7 4750U (Renoir)dosenfisch24v0.3.13599,632029136,9214,82
83DC #32R9 5950X (Vermeer)Sweepiv0.3.15760,71450738,52117,02
93DC #42R9 5900HS (Cezanne)MonkeyWin: Energy Savingv0.3.13815,051738150,8511,52
103DC #44R9 5900HS (Cezanne)Monkeyv0.3.13481,64408570,358,11
113DC #54i5 8365U (WhiskeyLake)MD_Enigmav0.3.1656,664575332,8513,75
123DC #69R5 PRO 4650G (Renoir)Tigersharkv0.3.11818,77578595,0560,86
133DC #47R7 4750G (Renoir)Poekel25Wv0.3.14670,05222796,1723,15
143DC #3R7 4700U (Renoir)CrazyIvanBatt. / Win: Better Eff. / HP: Recmd.v0.3.12586,762649145,9418,15
153DC #38R9 5950X (Vermeer)Sweepiv0.3.15578,81456139,3116,04
163DC #65R9 5900HS (Cezanne)MonkeyWin: Best Perf.v0.3.13498,15383174,6351,33
173DC #64R3 1200 (Summit Ridge)BlackArchonv0.3.1262,613138289,8645,32
183DC #67R7 3700X (Matisse)TigersharkPBO offv0.3.12787,1440481,4854,05
193DC #68R9 5900X (Vermeer)Krischiv0.3.13983560744,78125,22
203DC #70R9 5950X (Vermeer)LeiwandErmanual Curve Optimizationv0.3.15295,16563333,52168,04
213DC #88R9 5950X (Vermeer)Lowkeyv0.5.05945,36435638,61112,84
223DC #90R9 5950X (Vermeer)misterhv0.5.04834,19590235,05168,38
233DC #108i7 4820K (Ivy Bridge)Platos@4,5Ghzv0.3.1237,5920531205100,15
243DC #102i7 1165G7 (TigerLake)misterhWin: Best Perf.v0.5.01438,783774184,1820,49
253DC #94R9 5950X (Vermeer)misterh-0,1V Curve Optimizationv0.5.05167,06533236,3146,87
263DC #96R9 5950X (Vermeer)Sweepiv0.3.16103,75435437,63115,69
273DC #118R7 4750G (Renoir)Poekel20Wv0.5.14760,572005104,7919,13
283DC #129i7 5775C (Broadwell)MD_Enigmav0.5.1560,079308191,8348,52
293DC #133R5 4500U (Renoir)Poekelv0.5.11839,933342162,620,56
303DC #134R9 5900HS (Cezanne)MonkeyWin: Better Eff.v0.5.03936,18301084,4135,66
313DC #135R9 5900X (Vermeer)harzer_knallerBalanced Power Planv0.5.14414,66415154,5776,08
323DC #136R9 5900X (Vermeer)Darkearth27v0.5.14461,23518843,21120,07
333DC #140R9 5900X (Vermeer)KrischiCTRv0.5.13945,77576044130,92
343DC #141i7 2600K (Sandy Bridge)Tyrann@4,4Ghzv0.5.1269,6118669198,6893,96
353DC #145R7 5800X (Vermeer)hq-hqv0.5.12347,02750856,75132,29
363DC #146i7 7500U (Kaby Lake)Tyrannv0.5.1384,595226497,5510,5
373DC #146Celeron N3450 (Apollo Lake)Tyrannv0.5.135,61129202173,785,94
383DC #148R7 5800X (Vermeer)patrock84v0.5.11876,01790267,46117,13
393DC #154i5 8600k (Coffee Lake)hq-hqv0.5.1739,3112266110,27111,24
403DC #154i5 7500 (Kaby Lake)hq-hqv0.5.1336,4210055295,6134,02
413DC #155i7 8700k (Coffee Lake)Bernman@5Ghzv0.5.1925,561201789,91133,65
423DC #156R7 5800H (Cezanne)Darkearth27v0.5.13171,28451669,8364,67
433DC #160R9 5950X (Vermeer)GaryXv0.5.16668,05414936,14114,8
443DC #165R5 4600H (Renoir)GroschiWin 11v0.6.01878,683886136,9928,36
45CB #4R9 5900X (Vermeer)Asghan@95Wv0.6.04388,11486846,82103,97
46CB #5R5 5600X (Vermeer)mesohornyv0.6.01843723075,0596,34
47CB #9R7 3700X (Matisse)Puffer0815Outlier?v0.6.02569,91544471,4876,15
48CB #10R5 3500U (Picasso)TenferenzuOutlier?v0.6.01538,342529257,019,84
49CB #13R7 3700X (Matisse)Hardy72@95Wv0.6.02268,8620171,0887,23
50CB #14R7 3700X (Matisse)Jon Dohnson@PBOv0.6.02738,85527769,1976,26
51CB #20i5 8250U (WhiskeyLake)Rabrogov0.6.0838,175030237,221,21
52CB #36i7 4800MQ (Haswell)DrAgOnBaLlOnEv0.6.0451,858981246,4436,44
53CB #49R5 3500U (Picasso)AsghanThinkPad E495v0.6.0935,443335320,5310,41
56CB #57i7 9750H (Coffee Lake)Blende Upv0.6.01370,416345115,0155,16
57CB #60i7 3770K (Ivy Bridge)Blende Upv0.6.0447,2111190199,8356
58CB #60i5 4300U (Haswell)Blende Upv0.6.0184,89015600,2215,02
59CB #39R5 2600X (Pinnacle Ridge)HasseLadebalkenv0.5.1768,8211691111,26105,08
60CB #63i5 3320M (Ivy Bridge)noplan724v0.6.0177,2710172554,5618,34
61CB #83R5 3500U (Picasso)andi_scoOutlier?v0.6.0458,585880370,8815,85
62CB #102i7 2600 (Sandy Bridge)raser0248v0.6.0226,4417714249,3171,05
63CB #102i3 6157U (Skylake)raser0248v0.6.0388,054965519,019,57
64CB #112R7 3700X (Matisse)FabianoPBO off?v0.6.01920,89736270,72104,1
653DC #190R9 5900HS (Cezanne)MonkeyWin: Energy Savingv0.6.04673,21253084,5829,91
663DC #204R7 5800X (Vermeer)patrock84v0.7.02341,54677763,01107,56
67CB #132R5 3500U (Picasso)TenferenzuThinkPad E495 defaultv0.7.01709,412399243,849,84
68CB #118i9 11980HK (TigerLake-8C)JeanLegior 11900H (Eng. Sample)v0.6.02564,763825101,9437,52
69CB #137R5 3500U (Picasso)andi_scov0.7.0443,886048372,5216,23
70CB #140R5 3500U (Picasso)AsghanThinkPad E495 cool and quietv0.7.01517,622129309,466,88
71CB #143i7 9750H (Coffee Lake)Blende Upv0.7.01535542912045,24
72CB #149R7 2700X (Pinnacle Ridge)Tanzmusikusv0.7.01502,87762087,3287,26
73CB #152R5 3500U (Picasso)TanzmusikusWin = Balancedv0.7.0590,895238323,1116,21
743DC #205R5 4500U (Renoir)Poekelv0.7.02061,892724178,0615,3
753DC #212R5 2500U (Raven Ridge)Tiberiusv0.7.01216,692588317,628,15
76CB #173R5 5600X (Vermeer)Freiherausv0.7.02098,99587081,1672,33
773DC #234R7 5800H (Cezanne)Fondnessv0.7.03492,77377575,8449,77
783DC #241R5 5600X (Vermeer)Scotyv0.7.02001,77604282,773,08
793DC #242P Silver N6000 (JasperLake)Tralalakv0.7.2512,3937035277,03
803DC #244Celeron N5100 (JasperLake)y33H@v0.7.2287,184550765,235,95
81CB #178R3 4300G (Renoir)Lord Maikiv0.7.01513,554075162,1325,14
82CB #181i7 1165G7 (TigerLake)mkl1v0.7.01136,335208168,9930,82
83CB #184i5 11500 (Rocket Lake)Freiherausv0.7.21480,216750100,0967,44
843DC #257i7 11700K (Rocket Lake)Triskainev0.7.21887,59824164,28128,21
85CB #186i5 11400F (Rocket Lake)zymotic-95mV offsetv0.7.21485,51798184,3494,63
863DC #261R5 5600X (Vermeer)Holgiv0.7.21924,72616784,2573,19
873DC #279TR 1900X (Whitehaven)BlackArchonv0.7.2771,771469388,2166,6
88CB #214R9 5900X (Vermeer)Verangryv0.7.25170,32484439,93121,33
89AT #8R9 5950X (Vermeer)JoeRambo@4,4Ghz noSMTv0.7.25254,59441243,14102,27
90CB #218R9 5900X (Vermeer)Verangry@Stockv0.7.24236,1527444,76117,82
91AT #17i5 4690k (Haswell)zebrax2@Stockv0.7.2260,3616486232,9870,76

Excel-Version
 
Last edited:

JoeRambo

Golden Member
Jun 13, 2013
1,246
1,183
136
Nice work there, lovely scripting.

One thing of note is that it will run >600s in MT test, so people don't get mad waiting for MT tests to finish.

And it seems Excel 2021 seems to be unable to open and calculate results:
1633779449898.png

:(

After some digging - excel file seems awfully small and empty - maybe due to Excel being not actually available on machine where the tests ran ?
 
May 1, 2020
70
81
51
Nice work there, lovely scripting.

One thing of note is that it will run >600s in MT test, so people don't get mad waiting for MT tests to finish.

And it seems Excel 2021 seems to be unable to open and calculate results:
View attachment 51148

:(
Might have to do with localisation. Could you please attach the CSVs?
And what CPU SKU with which settings (TDP, UV, OC) did you run?
 
May 1, 2020
70
81
51
After some digging - excel file seems awfully small and empty - maybe due to Excel being not actually available on machine where the tests ran ?
Now I get it: You tried it at machine A and wanted to open the Excel on machine B?
No problem. Just copy the whole folder to machine B. The Excel file expects the CSVs to be in the relative folder \LogCsv.

Explanation: The PS script writes to the CSVs. The Excel needs to read the measurements from them.
 

JoeRambo

Golden Member
Jun 13, 2013
1,246
1,183
136
Now I get it: You tried it at machine A and wanted to open the Excel on machine B?
No problem. Just copy the whole folder to machine B. The Excel file expects the CSVs to be in the relative folder \LogCsv.

Nope, i already did it, still does not work on Excel 2021 :( Can someone process my csv files?
 
May 1, 2020
70
81
51
Forum does not allow attaching cvs or zip files. I ran it on custom undervolted 5950x 4.4Ghz with HT off.

https://file.io/TUdlpzLm6CEa

zip file with both csv files
Your link doesn't work. Please upload them again.
I also would like to understand why you are having trouble. Could you please explain the steps you took? Did you download the Zip-file of the release v0.7.2 our did you accidentally download the sources (which does not work)?
Direct link to the Zip: https://github.com/BrsVgl/PerformanceEfficiencySuite/releases/download/v0.7.2/PerformanceEfficiencySuite.v0.7.2.zip
 
Last edited:

JoeRambo

Golden Member
Jun 13, 2013
1,246
1,183
136
Your link doesn't work. Please upload them again.
I also would like to understand why you are having trouble. Could you please explain the steps you took? Did you download the Zip-file of the release v0.7.2 our did you accidentally download the sources (which does not work)?
Direct link to the Zip: https://github.com/BrsVgl/PerformanceEfficiencySuite/releases/download/v0.7.2/PerformanceEfficiencySuite.v0.7.2.zip
Uploaded it again, i think it auto deletes after several downloads.

https://file.io/Wd8O8X7l4wxZ

I downloaded simply distribution from github from releases, ran it, it generated csv folder and xls + odt files. I have copied whole folder to my PC with excel 2021, ran it, it asked for external sources and then when i tried data -> refresh all it throws excel error as shown above.
 

zebrax2

Senior member
Nov 18, 2007
933
42
91
The problem in excel seems to be a localization problem. For the English version of excel, graph sheet cell T2 formula should be "filename" instead of "dateiname"

Intel Core i5-4690k (stock with aftermarket air cooler)
Screenshot 2021-10-10 141110.png
 
May 1, 2020
70
81
51
@itsmydamnation
Your 35w TDP result looks weird - especially in the ST area. Seems like there were some other (background) processes doing some lifting as well. It also took 45s longer than your 25w run. I think you should repeat that one. May I ask what Notebook SKU this is?

In general: Please be aware that it is not possible to measure per process consumption. So it is really important that the system has already "settled" after a reboot.
 

itsmydamnation

Platinum Member
Feb 6, 2011
2,249
1,925
136
@itsmydamnation
Your 35w TDP result looks weird - especially in the ST area. Seems like there were some other (background) processes doing some lifting as well. It also took 45s longer than your 25w run. I think you should repeat that one. May I ask what Notebook SKU this is?

In general: Please be aware that it is not possible to measure per process consumption. So it is really important that the system has already "settled" after a reboot.
its a lenovo ideapad 5 pro 16".

yeah i will try 35watt again, i wasn't paying that much attention at the time , could have easily been some updates etc going on.
 

itsmydamnation

Platinum Member
Feb 6, 2011
2,249
1,925
136
Its better then last time but still not as good as 25watt single thread. Given that i cant control voltage only TDP/base / boost wattage i think its suffering from hotspot / overvoltage and thus limiting frequency. my Zen3 single core in CB23 only consumes around 9 watts peak single thread regardless of what i set TDP to , it must hit max heat density at that point.

it probably explains why AMD have never gone the intel route in Laptops of allowing single cores to consume ~ 20-25watts , because it will actually lead to worse performance for AMD.


1633988269355.png
 
May 1, 2020
70
81
51
No, this seems to be some weirdness from Lenovos side wrt thermal management. I have seen that kind of behaviour from other Lenovo samples das well. For a good reference for a 5800H just look at line 77 from the ST results table in post #3.
 

Det0x

Senior member
Sep 11, 2014
549
772
136
First post don't really say if its stock only systems, or if tweaked ones also are allowed... But i have done some runs on my 5950x :)

5950x pbo curve optimizer limited to 80w:
  • ST Efficiency Score = 88,98
  • MT Efficiency Score = 7395,84
1634010055739.png

5950x pbo curve optimizer limited to 100w:
  • ST Efficiency Score = 89,92
  • MT Efficiency Score = 7516,68
1634010097330.png


5950x pbo curve optimizer limited to 120w:
  • ST Efficiency Score = 89,94
  • MT Efficiency Score = 7345,68
1634010178593.png

5950x Hydra optimized to 100w:
  • ST Efficiency Score = 97,04
  • MT Efficiency Score = 8844,92
1634010244221.png

5950x Hydra optimized to 120w:
  • ST Efficiency Score = 102,43
  • MT Efficiency Score = 7842,27
1634010304097.png

Think these are decent results considering this is a 16 core "HEDT cpu" with a IOD/soc using upto and above 30w alone.
Could maybe get a higher MT Efficiency Score if i tried hydra optimized for 80w, but that will have to be for a other day :)

*edit*
All my results are done with high performance powerplan in windows.. could maybe gain some points there also (?)
 
Last edited:

itsmydamnation

Platinum Member
Feb 6, 2011
2,249
1,925
136
No, this seems to be some weirdness from Lenovos side wrt thermal management. I have seen that kind of behaviour from other Lenovo samples das well. For a good reference for a 5800H just look at line 77 from the ST results table in post #3.
I can control max package temp target via ryzen controller.
what i can tell you is my 25watt sustains higher average clocks then 35watt, maybe other laptops have better cooling and thus the extra voltage allows for higher sustained clocked, but for my 5800H the lower voltage leads to better performance.
 

ASK THE COMMUNITY