<div class="gmail_quote">On 4 November 2012 18:04, Tomasz Rola <span dir="ltr"><<a href="mailto:rtomek@ceti.pl" target="_blank">rtomek@ceti.pl</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
You mean this table?<br>
<a href="http://fah-web.stanford.edu/cgi-bin/main.py?qtype=osstats" target="_blank">http://fah-web.stanford.edu/cgi-bin/main.py?qtype=osstats</a><br clear="all"></blockquote></div><br>By now, the more or less 300 Tflops traditionally contributed by the Windows and Linux CPU communities each are down to 177 and 70 respectively, GPUs are also not doing too well, and the overall number of Tflops is <font face="'trebuchet ms', Arial, Helvetica, sans-serif">3677</font>, which is little more than *one third* of what was achieved in the past, and that in spite of the growing power of contributors' processors and the time GPU clients have been available!<br>
<br>Not exactly an exponential growth, in my book, in spite of Mr Pande's conferences about exaflops Real Soon Now.<br><br>So, even some <font face="'trebuchet ms', Arial, Helvetica, sans-serif">200,000 </font>processors are still on board worldwide (out of 8,5 millions who participated in time to the project), I am concerned that Folding@Home might be... just folding, after all.<br>
<br>Besides the loss for science, this of course tells us more on societal values and what really expects us than the raw power expected from a single Chinese supercomputer to come.<br><br>--<br>Stefano Vaj<br>