<div dir="ltr"><div><div><div>Yes we benchmarked v5 and v4 in<br><br><a href="https://github.com/Project-OSRM/osrm-backend/issues/2211">https://github.com/Project-OSRM/osrm-backend/issues/2211</a><br><br>found some regressions and considerably improved v5 in<br><br><a href="https://github.com/Project-OSRM/osrm-backend/pull/2241">https://github.com/Project-OSRM/osrm-backend/pull/2241</a><br></div><div><br></div><div><br></div>Are you running the v5 release (and not a RC)?<br></div>What's your specific use-case?<br>If you enable steps and alternatives the response is larger than in v4 and especially assembling all the GeoJSON objects / variants has some allocation issues. If that's the case, try the Polyline format.<br><br></div>Daniel J H<br></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Apr 26, 2016 at 12:43 AM, James Grant <span dir="ltr"><<a href="mailto:james@grantzone.net" target="_blank">james@grantzone.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<font face="Helvetica, Arial, sans-serif">Hi there.<br>
<br>
Has anyone done any comparisons between version 4 and 5 of OSRM
speed wise when looking at the C++ examples (the simpleclient.cpp
and example.cpp)?<br>
<br>
I've now built version 5 and the example.cpp but on first
appearances, it is slower than version 4. I will have to recompile
the simpleclient.cpp with the same parameters as I've currently
got to directly compare speed. <br>
<br>
Regards<br>
</font>
</div>
<br>_______________________________________________<br>
OSRM-talk mailing list<br>
<a href="mailto:OSRM-talk@openstreetmap.org">OSRM-talk@openstreetmap.org</a><br>
<a href="https://lists.openstreetmap.org/listinfo/osrm-talk" rel="noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/osrm-talk</a><br>
<br></blockquote></div><br></div>