Allen Akin wrote:
On Wed, Jun 30, 2004 at 11:40:27AM +0900, Sasayama wrote: | How complete is it in regard to the OpenGL specification?
I've managed the development of a couple of OpenGL test suites (including the official conformance test suite for the first several years of its existence). A suite that aspires to full spec coverage needs about 20 engineer-years of effort to get started. An industrial-strength suite needs considerably more than that. Only a few graphics vendors have been willing to invest that much effort, and they aren't interested in open-sourcing the code because it would benefit their competitors.
Even the official conformance tests don't cover enough of the
specification to validate a commercial implementation. That's one of
the reasons vendors have developed test suites of their own.
Thank you for the description.
Glean consists of tests that have been contributed mostly by DRI andTwo more questions. First, is it clear on what is covered by each test of glean? If so, it is very useful for out needs.
Mesa developers, so it reflects their particular needs and interests. I
think that's the right approach for an Open Source test suite. Based on
past experience, I don't think it's possible to generate enough interest
from the graphics community to build a suite with full spec coverage,
but it is possible to cover a useful subset.
Second, is it easy to examine the output correctness of each test in glean? It is very nice if it could make a summary report automatically.
begin:vcard fn;quoted-printable:=E7=AC=B9=E5=B1=B1 n;quoted-printable:=E7=AC=B9=E5=B1=B1 email;internet:[EMAIL PROTECTED] x-mozilla-html:TRUE version:2.1 end:vcard
