[Sidefx-houdini-list] Standardized Render Test File?
craigleehoffman at aol.com
craigleehoffman at aol.com
Mon Aug 10 14:53:41 EDT 2009
Another useful idea is if someone could make and post production prman shaders as similar as possible to the mantra ones used for these tests so folks could do these render tests on their own machines and compare Renderman (or other "prman-like" renderers) vs. Mantra to help make an informed decision when deciding on a renderer.
In the next couple of months we are going to have the new cheaper Quad Core "Core i5" cpu's (pretty much the same as Core i7's but without "hyperthreading") coming out on cheaper motherboards, along with the much improved Windows 7 OS for PC's, and cheaper better performing SSD's (Solid State Drives), so this would be a great time to post a standard render test file for people to test with so they know what to ask Santa for.
From: Peter Bowmar <pbowmar at gmail.com>
To: sidefx-houdini-list at sidefx.com
Sent: Mon, Aug 10, 2009 11:28 am
Subject: Re: [Sidefx-houdini-list] Standardized Render Test File?
-single vs multi threading since often multi threading buys you little
-raytrace vs simple micropoly vs hybrid etc.
-production shaders (not SESI-supplied shaders) with lots of attribute
pickups from geometry driving shading looks etc
-instancing (generating the instances is a big factor so part of the
benchmark here is Houdini time)
-Polys vs SubD vs Nurbs
Likely more, no time right now :)
2009/8/10 David Johnson <daveinengland at gmail.com>:
> Please include at least one teapot. For old times sake.
> On Fri, Aug 7, 2009 at 10:03 AM, Michael Goldfarb <goldfarb at coredp.com>wrote:
>> I think it's a great idea and it's something we've talked about over at
>> od[force]...just a question of determining what the scene (or scenes)
>> should contain and then building it.
>> I know we'd be happy to host/manage the whole thing at od[force]...
>> craigleehoffman at aol.com wrote:
>> > All the recent talk on hardware and Mac vs. Windows vs. Linux, etc. made
>> > wonder if anyone had a standardized complex Render Test Hip File that
>> > folks could download and run to compare render speeds across platforms
>> > and hardware?
>> > I think this would help a lot with folks making new purchases like
>> answering questions such as:
>> > Does Quad Core make enough of a difference over Dual Core to warrant
>> > Would I benefit from a CPU with Hyperthreading?? (Core i7 vs. the
>> upcoming Core i5, etc.- I read that Maya's renderer was 16% faster when
>> Hyperthreading was enabled on a Core i7!)
>> > Will a SSD or faster Hard Drive make much difference?
>> > Should I bother installing Linux/Windows 7/Snow Leopard?
>> > Such a render test should deal with a good cross-section of complex
>> rendering in Mantra (perhaps with PBR as well as non-PBR ROPs) and also deal
>> with large complex models and test disk access by using large textures on
>> disk, etc.? (Also a good test hip file for Complex Dynamics performance
>> testing might be a good idea too.)
>> > Anyone have such a thing- or does something like this already exist?
>> > -Craig
>> > _______________________________________________
>> > Sidefx-houdini-list mailing list
>> > Sidefx-houdini-list at sidefx.com
>> > https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>> Michael Goldfarb ? ? ? ? ? ? ? ? ? ? ? ? ? ?email: goldfarb at corefa.com
>> c.o.r.e. digital pictures ? ? ? ? ? ? ? ? ? ? ? ?http://www.coredp.com
>> "Unless you're an astronaut, secret agent, vampire hunter,
>> or all three, you're probably a sellout; screw you."
>> - Maddox 2003
>> Sidefx-houdini-list mailing list
>> Sidefx-houdini-list at sidefx.com
> Sidefx-houdini-list mailing list
> Sidefx-houdini-list at sidefx.com
Sidefx-houdini-list mailing list
Sidefx-houdini-list at sidefx.com
More information about the Sidefx-houdini-list