SOLVED

Timer interrupt experiment decreases performance

%3CLINGO-SUB%20id%3D%22lingo-sub-857987%22%20slang%3D%22en-US%22%3ETimer%20interrupt%20experiment%20decreases%20performance%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-857987%22%20slang%3D%22en-US%22%3E%3CP%3EOne%20of%20my%20friends%20told%20me%20that%20they%20found%20a%20flag%20that%20decreases%20performance%2C%20so%20I%20checked%20it%20out%20using%20browser%20benchmarks.%20The%20one%20in%20question%20is%20called%20Timer%20interrupt%20experiment.%20I%20put%20it%20through%20Motion%20Mark%20(%3CA%20href%3D%22https%3A%2F%2Fbrowserbench.org%2FMotionMark1.1%2F%22%20target%3D%22_blank%22%20rel%3D%22nofollow%20noopener%20noreferrer%20noopener%20noreferrer%20noopener%20noreferrer%22%3Ehttps%3A%2F%2Fbrowserbench.org%2FMotionMark1.1%2F%3C%2FA%3E)%20and%20found%20that%20it%20did%20perform%20slightly%20worse(on%20Beta).%20The%20flag%20claims%20that%20it%20reduces%20power%20consumption%2C%20but%20I%20was%20wondering%20if%20there%20was%20a%20way%20to%20do%20that%20without%20decreasing%20performance.%26nbsp%3B%3CA%20href%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fuser%2Fviewprofilepage%2Fuser-id%2F239638%22%20target%3D%22_blank%22%3E%40Elliot%20Kirk%3C%2FA%3E%26nbsp%3B%3C%2FP%3E%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20style%3D%22width%3A%20293px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Fgxcuf89792.i.lithium.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F132039i0E269B21E404CC68%2Fimage-dimensions%2F293x422%3Fv%3D1.0%22%20width%3D%22293%22%20height%3D%22422%22%20alt%3D%22motion%20mark%20throttled.png%22%20title%3D%22motion%20mark%20throttled.png%22%20%2F%3E%3C%2FSPAN%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20style%3D%22width%3A%20306px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Fgxcuf89792.i.lithium.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F132040iD80C5B113D03D4BC%2Fimage-dimensions%2F306x419%3Fv%3D1.0%22%20width%3D%22306%22%20height%3D%22419%22%20alt%3D%22motion%20mark.png%22%20title%3D%22motion%20mark.png%22%20%2F%3E%3C%2FSPAN%3E%3C%2FP%3E%3C%2FLINGO-BODY%3E%3CLINGO-LABS%20id%3D%22lingo-labs-857987%22%20slang%3D%22en-US%22%3E%3CLINGO-LABEL%3EFlags%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3EPerformance%3C%2FLINGO-LABEL%3E%3C%2FLINGO-LABS%3E%3CLINGO-SUB%20id%3D%22lingo-sub-859953%22%20slang%3D%22en-US%22%3ERe%3A%20Timer%20interrupt%20experiment%20decreases%20performance%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-859953%22%20slang%3D%22en-US%22%3E%3CP%3EHi%26nbsp%3B%3CA%20href%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fuser%2Fviewprofilepage%2Fuser-id%2F343543%22%20target%3D%22_blank%22%3E%40cjc2112%3C%2FA%3E%2C%20in%20general%20that%20is%20what%20we%20are%20trying%20to%20achieve.%26nbsp%3B%20We%20run%20these%20experiments%20periodically%20to%20see%20how%20these%20sorts%20of%20changes%20affect%20the%20system%20as%20a%20whole.%26nbsp%3B%20If%20the%20battery%20gains%20are%20good%2C%20and%20the%20performance%20hit%20is%20reasonable%2C%20we%20might%20offer%20this%20as%20a%20setting%2C%20or%20make%20it%20the%20default%20behavior%20when%20on%20battery.%26nbsp%3B%20But%20if%20the%20performance%20drops%20more%20than%20a%20reasonable%20amount%2C%20or%20the%20battery%20gains%20are%20negligible%20under%20real%20world%20conditions%2C%20then%20the%20flag%20disappears%2C%20and%20the%20feature%20never%20makes%20it%20into%20the%20product.%20Thanks%20-%20Elliot%3C%2FP%3E%3C%2FLINGO-BODY%3E
Highlighted
Super Contributor

One of my friends told me that they found a flag that decreases performance, so I checked it out using browser benchmarks. The one in question is called Timer interrupt experiment. I put it through Motion Mark (https://browserbench.org/MotionMark1.1/) and found that it did perform slightly worse(on Beta). The flag claims that it reduces power consumption, but I was wondering if there was a way to do that without decreasing performance. @Elliot Kirk 

motion mark throttled.pngmotion mark.png

1 Reply
Highlighted
Solution

Hi @cjc2112, in general that is what we are trying to achieve.  We run these experiments periodically to see how these sorts of changes affect the system as a whole.  If the battery gains are good, and the performance hit is reasonable, we might offer this as a setting, or make it the default behavior when on battery.  But if the performance drops more than a reasonable amount, or the battery gains are negligible under real world conditions, then the flag disappears, and the feature never makes it into the product. Thanks - Elliot