tag:blogger.com,1999:blog-3971202189709462152.post1822874891591164256..comments2024-03-11T12:50:02.036+01:00Comments on PyPy Status Blog: Introducing speed.pypy.orgCarl Friedrich Bolz-Tereickhttp://www.blogger.com/profile/00518922641059511014noreply@blogger.comBlogger20125tag:blogger.com,1999:blog-3971202189709462152.post-73430101426005102702010-03-17T18:42:06.636+01:002010-03-17T18:42:06.636+01:00I am sure your VM will get just as optimized. I am...I am sure your VM will get just as optimized. I am sure Google will pick you guys up soon. I have no doubt about it. Unladen Swallow seems a waste of time once yall get more credibility.large file transferhttp://www.filesdirect.com/planinfo.aspxnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-67721301482562886362010-03-09T19:02:41.858+01:002010-03-09T19:02:41.858+01:00It would be nice if the timeline had the date on i...It would be nice if the timeline had the date on it (only where the date changes, and the beginning + end).stuaxohttps://www.blogger.com/profile/08755227063937859112noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-24086645931716243222010-03-09T19:00:21.536+01:002010-03-09T19:00:21.536+01:00This comment has been removed by the author.stuaxohttps://www.blogger.com/profile/08755227063937859112noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-13970883544280776242010-03-09T18:09:12.347+01:002010-03-09T18:09:12.347+01:00@Luis
Hey. It's a valid option, but it's ...@Luis<br /><br />Hey. It's a valid option, but it's however at least hard (if not next to impossible). There is work planned on reducing warmup time, so it won't matter that much instead.<br /><br />Cheers,<br />fijalMaciej Fijalkowskihttps://www.blogger.com/profile/11410841070239382771noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-20296389534715482182010-03-09T02:28:58.531+01:002010-03-09T02:28:58.531+01:00Another silly question:
AFAIK, the benchmark impro...Another silly question:<br />AFAIK, the benchmark improvements seen lately are due to the way you measure avergages, by excluding warmup time. Seeing that warmup takes time that may be critical in some situations, I wonder if it's possible to somehow "save" the generated jited code so it can be reused after the first time it's generated.<br />This way, it would be possible to distribute programs already "warmed up", kind of a compiled version of them. Sorry if this doesn't make sense at all... for a clueless ignorant like me, it does!Luishttps://www.blogger.com/profile/01147433030878927988noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-16163063663566804792010-03-06T13:51:18.353+01:002010-03-06T13:51:18.353+01:00@della
yeah, such things are missing right now.
...@della<br /><br />yeah, such things are missing right now.<br /><br />An about page, and possibly an explanation (with links to the code) of each benchmark are probably going to be implemented. Currently there is only tooltip explanations for some.tobamihttps://www.blogger.com/profile/16388550213568324790noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-47372652148797485632010-03-06T09:55:35.426+01:002010-03-06T09:55:35.426+01:00Great work! BTW, could it be possible to also have...Great work! BTW, could it be possible to also have a quick link to the source code of the benchmarks in the website?dellahttps://www.blogger.com/profile/16873331731877660793noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-84681468110488267082010-03-05T05:17:36.637+01:002010-03-05T05:17:36.637+01:00I understand JVM is highly optimized.
And overtime...I understand JVM is highly optimized.<br />And overtime and once yall have more momentum industry funding I am sure your VM will get just as optimized. I am sure Google will pick you guys up soon. I have no doubt about it. Unladen Swallow seems a waste of time once yall get more credibility. <br /><br />Even then I do expect Dynamic scripting capabilities to perform slower the Java. <br /><br />I am just hoping that eventually the non-dynamic parts of python will perform on par with Java.<br /><br />And we can all program in just Python and C. :-))Sarvi Shanmughamhttps://www.blogger.com/profile/16978361731038460021noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-38352748956969043542010-03-04T14:14:37.454+01:002010-03-04T14:14:37.454+01:00@sarvi: reaching the speed of Java is a really non...@sarvi: reaching the speed of Java is a really non-trivial goal, because Sun's JVM has really been highly optimized over many years. I guess it will take us a long time (if at all) to reach such levels of performance.Carl Friedrich Bolz-Tereickhttps://www.blogger.com/profile/00518922641059511014noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-18534015837126392612010-03-04T13:20:53.394+01:002010-03-04T13:20:53.394+01:00@Neil
a view showing all timeline graphs at once i...@Neil<br />a view showing all timeline graphs at once is also planned.<br /><br />About the average, of course you can not take from it that pypy-c-jit is nearly 3 times as fast as cpython. Because it depends on the particular choice of benchmarks, which right now is not at all representative of actual real-world usage.<br /><br />Regardless, it is there so that a developer gets an overall <i>feeling</i> for how a given revision change has affected performance across <i>all</i> benchmarks.<br /><br />We can't avoid the risk of people reaching wrong conclusions, but that is always the case with statistics, averages and benchmarks ;-)tobamihttps://www.blogger.com/profile/16388550213568324790noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-15297527505349551272010-03-04T13:08:13.302+01:002010-03-04T13:08:13.302+01:00Neat! I still like the original graphs though, it&...Neat! I still like the original graphs though, it's nice to see the history for all the benchmarks together.<br /><br />I think the 'average' is pretty meaningless - it implies that a simple average of all the benchmarks will correspond to the typical real-world speed up you will get using pypy with your existing python code, which I don't think is true.Neilnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-3509568912404258522010-03-04T10:54:30.869+01:002010-03-04T10:54:30.869+01:00@Philip, Anonymous
The first features are caterin...@Philip, Anonymous<br /><br />The first features are catering to trunk development, which was the most urgent thing.<br /><br />But my plan all along was to implement a third tab for comparing implementations (among other things. See mailing list announcement for details).<br /><br />So your wish should come to pass :-)tobamihttps://www.blogger.com/profile/16388550213568324790noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-72229676370196615992010-03-04T10:46:34.303+01:002010-03-04T10:46:34.303+01:00@Reid: maybe I can help you out setting it up. You...@Reid: maybe I can help you out setting it up. You could actually even begin saving results to speed.pypy.org right away with minimal configuration changes (though I understand you may prefer to have your own site and DB).tobamihttps://www.blogger.com/profile/16388550213568324790noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-85375256217017101532010-03-04T05:59:12.680+01:002010-03-04T05:59:12.680+01:00With the JIT would a script that does not use the ...With the JIT would a script that does not use the dynamic aspects of python be able to match the speed of Java?Sarvi Shanmughamhttps://www.blogger.com/profile/16978361731038460021noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-13399777857808269912010-03-04T00:56:10.812+01:002010-03-04T00:56:10.812+01:00@matt
Ok. So I've seen those feature request ...@matt<br /><br />Ok. So I've seen those feature request often enough. These benchmarks are not good for tracking memory usage - they'll simply measure the amount interpreter allocates at the beginning. If you provide better ones, we'll do it.<br /><br />Cheers,<br />fijalMaciej Fijalkowskihttps://www.blogger.com/profile/11410841070239382771noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-19158585636851650672010-03-04T00:40:53.775+01:002010-03-04T00:40:53.775+01:00Awesome work. One more feature request: Track me...Awesome work. One more feature request: Track memory usage.matt harrisonhttp://panela.blog-city.comnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-41035371591076139122010-03-04T00:09:35.347+01:002010-03-04T00:09:35.347+01:00This is great! It's excellent to see the fruit...This is great! It's excellent to see the fruits of the pypy jit work so clearly. <br /><br />I'd also like to see this in place for other Python implementations.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-25240907193441199012010-03-03T23:49:00.414+01:002010-03-03T23:49:00.414+01:00I would love to see this become a Python implement...I would love to see this become a Python implementation shootout, a single place where we could compare the speeds of CPython/PyPy/Unladen/Jython/IronPythonPhilip Jenveyhttps://www.blogger.com/profile/17437958274873977298noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-91323472120481189852010-03-03T20:48:43.270+01:002010-03-03T20:48:43.270+01:00In my mind PyPy with its JIT will/should eventuall...In my mind PyPy with its JIT will/should eventually get us close to matching or beating Java performance for the non-dynamic subset of python. Would that be a fair statment? If so is there some bench mark that allows us to compare that. What that be usefull?Sarvi Shanmughamhttps://www.blogger.com/profile/16978361731038460021noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-16845058794725349162010-03-03T18:12:45.797+01:002010-03-03T18:12:45.797+01:00Excellent! We really ought to deploy this for unl...Excellent! We really ought to deploy this for unladen, too. Unfortunately, I don't think I'll have the time to get that going. :(Reid Khttps://www.blogger.com/profile/10131127807440335781noreply@blogger.com