home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!decwrl!deccrl!news.crl.dec.com!pa.dec.com!engage.pko.dec.com!riscee.pko.dec.com!wwlk
- From: wwlk@riscee.pko.dec.com (Bill Licea-Kane)
- Newsgroups: comp.graphics
- Subject: Re: NCGA Picture Level Benchmark (was: Time to put up or shut up)
- Keywords: benchmarks NCGA GPC
- Message-ID: <1992Sep2.183401.13257@engage.pko.dec.com>
- Date: 2 Sep 92 18:34:01 GMT
- References: <1992Aug20.172200.17418@dsd.es.com> <ortnla8@fido.asd.sgi.com> <1992Sep1.194132.2601@dsd.es.com> <pak809s@fido.asd.sgi.com>
- Sender: wwlk@riscee.pko.dec.com (Bill Licea_Kane)
- Organization: Digital Equipment Corporation, Maynard MA.
- Lines: 79
-
- In article <pak809s@fido.asd.sgi.com>, mtj@babar.asd.sgi.com (Michael Jones) writes:
- |>
- |> The PLB issue seems the perfect example of the Akin/Thomson dialogues
- |> on graphics programming interfaces. My understanding is that the PLB
- |> suite spends quite a bit of effort (thus time) "editing" the databases
- |> it is drawing.
-
- You understand wrong. Actually, the PLB is a *PICTURE* Level
- Benchmark. The geometry is STATIC. No editing during the
- timing loop to speak of takes place. (One or more matrixes
- are updated during the timing loop, and in the case of one
- benchmark, there is some control over what geometry is displayed
- each frame. There's also some minor bookkeeping. All the major
- bookkeeping is done outside of the timing loop.)
-
- |>
- |> (customer) "But why is this PLB application so much slower than
- |> all the similar applications I see on SGI machines?"
- |>
- |> (salesman) "Because we're using a PHIGS/PEXlib style API."
- |>
- |> (customer) "Oh. I guess it must be the extra data movement and
- |> redundant PHIGS structure editing."
- |>
- |> (salesman) "Exactly. Now let me show you the OpenGL version."
- |>
- |> (customer) "Wow!. No wonder they stopped making workstations."
-
- You again have a very common misperception, that the PLB is
- a PHIGS benchmark.
-
- While the BIF (Benchmark Interface Format) is generally
- derived from PHIGS clear text archives, and while the
- *SAMPLE* implementation uses PHIGS, there are multiple
- "ports" of the PLB. (Unlike SPEC, where you can generally
- expect to take source code, compile/link/run it, PLB
- could *NOT* begin to hope for such a goal when it was
- started. It brought too many limitations.)
-
- There are several existing ports of the PLB.
-
- DEC - DEC PHIGS
- E&S - ES/PEX (PHIGS binding)
- HP - HP-PHIGS
- - Starbase
- IBM - IBM graPHIGS
- - GL
- SGI - GL
- Sun - SunPHIGS
- - XGL
-
-
- SGI's port of the PLB does not use a PHIGS/PEXlib style
- API. It uses GL.
-
-
- Work is beginning to provide additional *SAMPLE* implementations
- of the PLB. I would hope there will be a suite of three widely
- used APIs covered by the sample implementations:
-
- PHIGS - already done, actively maintained.
- PEXlib - work started
- OpenGL - work beginning
-
- Proprietary (non-perjorative sense) APIs such as GL, Starbase
- and XGL are the responsibility of the individual vendors. But
- many people feel the industry and the user community will
- greately benefit from sample implementations of widely available
- and/or widely licensed APIs.
-
-
- Not to say that there aren't other legitimate things to be
- discussed about the PLB. But I first wanted to correct these
- misconceptions.
-
- -----
- Bill Licea-Kane (Do I really need to say "I speak for myself?")
- Digital Equipment Corporation
- (Representative to GPC)
-