Open main menu

CDOT Wiki β

Changes

Firefox Performance Testing Lab Fall 2010 dacallow

2,411 bytes removed, 12:51, 21 September 2010
Results
=== Minefield Build ===
*Mozilla/5.0 (Windows NT 6.1; WOW64; rv:2.0b7pre) Gecko/20100914 20100916 Firefox/4.0b7pre
=== Chromium Build ===
*Chromium 7.0.525528.0 (59405Developer Build 59770
=== Results ===
 
[http://www.megaupload.com/?d=S4I3GQCR Perfmon Results]
<table border='1' cellpadding='5' cellspacing='0'>
</tr>
<tr>
<td>7191</td> <td>[http://www.chromeexperiments.com/detail/browserwater-pongtype/ Browser PongWater Type]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1517/09/2010</td> <td>Not much Speeds seems compairable to be said of speed. Both browsers seemed to be identicalChrome</td> <td>A little jerky on window redraw Small pause during beginning when moving "paddle" compared waves are first spreading out, less noticable in Chrome. Noticable pause after a moment if you click down mouse to Chromiummake waves and slide it around in smooth lines.. likely related to garbage collection.</td> <td>Both very responsive, utilized same processor amount (4.6% on Minefield, 5.8% on Chromium)Equal to Chrome</td> <td>"Ball" window rendered considerably taller Not sure which is better, but Chrome looks more blurry / maybe anti aliased, then firefox. Bug observed in Chromiumboth browsers- creating waves across the bottom edge will sometimes leave a persisting vertical glowing bar. Closing parent window orphans paddle and ball windows, they remain dead (but interactivable) on screen</td>
</tr>
<tr>
<td>7292</td> <td>[http://www.chromeexperiments.com/detail/2d-cloth-simulationtunnelers/ 2D Cloth SimulationTunnelers]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1517/09/2010</td> <td>Fairly fluid drawing on both Minefield and Chromium. Can slow on Minefield if draw trackers (dots) are added while curtain is in high state of fluxEqual to Chrome</td> <td>SmoothEqual to Chrome</td> <td>Minefield was a little slow Equal to pick up on mouse clicksChrome</td> <td>Processor utilization same on both (avg Minefield was 34%, avg Chromium was 37%). Chromiummuch lighter on memory</td>
</tr>
<tr>
<td>7393</td> <td>[http://www.chromeexperiments.com/detail/animatedcheloniidae-harmonographlive/ Animated HarmonographCheloniidae Live]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1517/09/2010</td> <td>Drawing of complex patterns Won't run on Minefield would sometimes cause centre to "wobble"my build.</td> <td>Redraw very slow Won't run on Minefield, sub-16 FPS. Chromium very fast and smoothmy build.</td> <td>Both slow to pick up Won't run on mouse clicks, Firefox slightly slowermy build.</td> <td>Minefield slowly leaked memory (8MB over 10 minutes), CPU was pegged at 100% (Chromium's avg was 86%)</td>
</tr>
<tr>
<td>7494</td> <td>[http://www.chromeexperiments.com/detail/jscanvasbikeplane-deformations/ jsCanvasBikePlane Deformations]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1517/09/2010</td> <td>No difference foundAt home running must faster in Mindfield, 50 fps (25 in Chrome). At school Mindfield is slower, 12fps vs Chrome at 20fps.</td> <td>Minefield seemed a little less Very smooth when the bike suddenly changed speed. Smoothness generally went down as prolonged periods of speed went up.</td> <td>Movements responded well to inputHighly responsive. CPU pegged during long periods user input (such as holding down keys to accelerate or balance on one wheel) At home movement seems delayed in Chrome but not Firefox</td> <td>Minefield guzzled 160MB of memory over 4 minutes of constant drivingNot sure which is better, Chromium's would remain constant with short, periodic spikesbut Chrome looks a bit more blurry then firefox.</td>
</tr>
<tr>
<td>7595</td> <td>[http://www.chromeexperiments.com/detail/venetianizationplasmatree/ VenetianizationPlasmaTree]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1517/09/2010</td> <td>No problemsSame speeds.</td> <td>No problemsSame speeds.</td> <td>N/AEqually responsive</td> <td>10% higher on processor in Minefield than in Chromium(73.496% avg compared to 60% avg), with much more resources (140 MB compared to 80 MB)</td>
</tr>
<tr>
<td>7696</td> <td>[http://www.chromeexperiments.com/detail/catch-itcrashhaus/ Catch It!Crashhaus]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1517/09/2010</td> <td>Very quick on Running faster in Firefox, great fun!</td> <td>Smooth redraw, felt like PC gameSeems to be the same</td> <td>Very responsiven/a</td> <td>Sluggish and near-unplayable in Chromium. Less CPU in Minefield (4-12%, compared to 9-15% in Chromium) but more memory usage.</td>
</tr>
<tr>
<td>7797</td> <td>[http://www.chromeexperiments.com/detail/liquid-particlestrail/ Liquid ParticlesTrail]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1617/09/2010</td> <td>Occasionally Minefield would not render more than 4 particle dots (should be around 100).Faster in Firefox</td> <td>Very choppy on Minefield when drawing both particles and the letters versionSame</td> <td>Sometimes unresponsive for up to a second due to high CPU usageSame</td> <td>Incredibly CPU intensive. Perfmon shows CPU usage as 600% at some points. Chromium and Minefieldboth ran letters version choppy (Minefield slightly more so), though Chromium could handle particle version very well</td>
</tr>
<tr>
<td>7898</td> <td>[http://www.chromeexperiments.com/detail/harmonygraphycalc/ HarmonyGraphyCalc]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1617/09/2010</td> <td>Very smooth renderingRuns the same</td> <td>The Graphics were very smooth.Runs the same</td> <td>Very responsive during normal drawing. CPU spike when converting drawing to a PNG imageRuns the same</td> <td>Both Minefield and Chromium ran this well. Chromium used much more CPU on average than Minefield</td>
</tr>
<tr>
<td>7999</td> <td>[http://www.chromeexperiments.com/detail/realtime-video-ascii-conversionmagnetic/ Realtime Video->ASCII ConversionMagnetic]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1617/09/2010</td> <td>Rendering was great for both, ASCII characters don't take much to drawTiny pause during initial load. Same speeds after load.</td> <td>Has harder time converting at higher "resolutions" (ASCII character size decreases) than at higher scales (larger canvas)Same</td> <td>Became quite unresponsive over time, especially at higher resolutions and scalesSame</td> <td>Chromium handled higher resolutions noticeably better than Minefield, though Minefield handled higher scales slightly better than Chromium.</td>
</tr>
<tr>
<td>80100</td> <td>[http://www.chromeexperiments.com/detail/internet-graffiti-boardflowerpower/ Internet Graffiti BoardFlowerPower]</td> <td>[http://zenit.senecac.on.ca/wiki/index.php/User:SWeerdenburg Steven Weerdenburgdacallow Kaitlyn Callow]</td> <td>1617/09/2010</td> <td>Quick on bothSame</td> <td>SmoothSame</td> <td>Input was very responsive.Same</td> <td>Both Minefield and Chromium ran this test very well. Both had a bug in that it was possible to mess up the "drag and drop" functionality of the sketch pad: likely a programming error and how the onMouseOver, onMouseDrag and onMouseOut events are handled.</td>
</tr>
</table>
7
edits