Thanks for the kind words Thomas... they are appreciated. I agree that all of these engines can render objects correctly ( some better than others and your right it does depend on the engine at that point ). But IMHO there is a lot of nuance between how the engines implement themselves and I Papervision when you compare it in a general way offers the most advantages. I'm not knocking the other two engines. I think there great and just the fact there is a choice out there for the community is awesome and kudos to those devs! I also believe that those engines will improve over time and perhaps my opinion may change. But for now, I do believe that overall Papervision offers the most advantages.
Btw... that's a cool picture thanks for posting! Cheers, :D Sam! On Aug 2, 2007, at 3:40 AM, Thomas wrote: > Hi people, > > please Samuel, do not talk too fast :) I don't blame you, but I think > you answered a bit too fast above. > > I don't want to create a big debate, but each engine have their own > advantages. It is up to you to pick what is really important for your > project. Ask to their respective mailing list, or do some fast tests > illustrating the important point of your project (good performance, > immerssion, good z-sorting, etc..) > > I think mrDoob's test shows that each engine are almost as fast, and > as Sandy is missing, I just show you that Sandy is also capable to > render correctly that amount of polygons : > http://kiroukou.media-box.net/blog/mes-recherches-sur-flash/118- > sandy-30-mrdoob-benchmark.html > > Good work with your project :) > Thomas > > 2007/8/2, John Grden <[EMAIL PROTECTED]>: >> PV3D 1.5 was about a 40%+ increase in speed. Away3D was based on >> PV3D's >> original engine, but I've heard they're doing a restructure of the >> engine, >> so I don't know if they'll adopt some of the techniques 1.5 has or if >> they'll go a different direction. >> >> I don't know about Sandy to be honest. >> >> Ricardo did some tests with PV3D and Away3D, but this was *before* >> the 1.5 >> release - I would have to assume he would see a significant >> increase in >> speed with 1.5: >> http://mrdoob.com/?postid=393 >> >> then, he did what we now call a "doobie" test with Ralph >> (triangles rendered >> per second is a doobie count - the higher the #, the better >> obviously): >> http://mrdoob.com/?postid=394 >> >> Learning curve: >> PV3D has a really good wiki (wiki.papervision3d.org), tons of >> tutorials, >> sample files and blog entries ( blog.papervision3d.org, >> www.rockonflash.com/blog/, www.unitzeroone.com) >> >> there's also the new PV3D CS3 Components that lower the entry bar >> even lower >> than ever. New features like code generation help you get started >> even if >> you're not a coder - and it all works at DESIGN time. >> >> I can't really speak for Sandy or Away3D on the learning curve, >> but I can >> say that all 3 API's are very similar, and the approach is nearly >> identical >> from a usage stand point. >> >> I *can* speak on PV3D's focus: to provide a very fast, lean >> consistent >> engine. Features don't come as quickly as we'd like, but the the >> engine is >> very tight and some things we're doing have to come first before >> we can add >> certain features. >> >> Away3D's team have done an impressive job on their features and >> version of >> the PV3D engine - really outstanding stuff going on there. >> >> Anyway, hope that helps some. >> >> John >> >> >> On 8/1/07, Johannes Nel <[EMAIL PROTECTED]> wrote: >>> i think as misguided as his "factual statements" are he touched >>> on one >> important point in relation to the question: >>> -which one of the three would add the LEAST amount of code for a >>> simple >> 'cube rotating' effect + -which one of the three has the smallest >> learning >> curve (i have a tight deadline to meet)==> >>> means from whom can i copy and paste the easiest. and there is an >>> example >> cube rotating in pv3d's examples library and more people blog >> about it. >> hence more vigarous copy-paste application. >>> >>> for the rest the statements carry as much relevance as >>> creationism and he >> can never offer sufficient proof of any of these claims. >>> >>> >>> >>> On 8/1/07, Mariano Cerrutti < [EMAIL PROTECTED]> wrote: >>>> Samuel: >>>> Same as Martin, please give some references for your comments. >>>> >>>> cheers to all >>>> Mariano >>>> >>>> >>>> >>>> On 8/1/07, Martin Wood-Mitrovski < [EMAIL PROTECTED]> >>>> wrote: >>>>> >>>>> >>>>> Samuel Agesilas wrote: >>>>>> no no no no!! This is not the case.... Papervision3d is far more >>>>>> superior than Sandy and away3d. Papervision performance competely >>>>>> blows away sandy and is much Much better than Away3d. >>>>>> >>>>>> To answer your question more directly. >>>>>> >>>>>> 1. Papervision - Why? The papervision API is really clean and >> simple. >>>>>> Something like a rotating cube can be accomplished with very VERY >>>>>> little code ( even if you want to texture map the sides ) >>>>>> 2. Papervision - Why? It is the more mature of all 3 of those >>>>>> platforms. Papervision3d has a very small learning curve and >>>>>> there >>>>>> are lots of examples out there to analyze and learn from. >>>>> >>>>> can you give a bit more evidence to back up those claims? >>>>> >>>>> some code, some links, some benchmarks for example. >>>>> >>>>> you could well be right, but i want proof :) >>>>> >>>>> thanks, >>>>> >>>>> Martin >>>>> >>>>> _______________________________________________ >>>>> osflash mailing list >>>>> [email protected] >>>>> >> http://osflash.org/mailman/listinfo/osflash_osflash.org >>>>> >>>> >>>> >>>> _______________________________________________ >>>> osflash mailing list >>>> [email protected] >>>> http://osflash.org/mailman/listinfo/osflash_osflash.org >>>> >>>> >>> >>> >>> >>> -- >>> j:pn >>> http://www.memorphic.com/news/ >>> _______________________________________________ >>> osflash mailing list >>> [email protected] >>> http://osflash.org/mailman/listinfo/osflash_osflash.org >>> >>> >> >> >> >> -- >> [ JPG ] >> _______________________________________________ >> osflash mailing list >> [email protected] >> http://osflash.org/mailman/listinfo/osflash_osflash.org >> >> > > _______________________________________________ > osflash mailing list > [email protected] > http://osflash.org/mailman/listinfo/osflash_osflash.org _______________________________________________ osflash mailing list [email protected] http://osflash.org/mailman/listinfo/osflash_osflash.org
