home *** CD-ROM | disk | FTP | other *** search
- Path: surfnet.nl!sun4nl!xs4all!usenet
- From: jtv@xs4all.nl (Jeroen T. Vermeulen)
- Newsgroups: comp.sys.amiga.misc
- Subject: Re: Speed: 68040 vs. 68060
- Date: Mon, 26 Feb 96 15:51:50
- Organization: Leiden University, Mathematics & Computer Science, The Netherlands
- Message-ID: <19960226.7B42F98.E8D9@asd06-03.dial.xs4all.nl>
- References: <4foi00$60t@gondor.sdsu.edu> <3125E74D.3390@gih.no> <19960223.425E10.10CBD@an100.du.pipex.com> <19960225.7AF9790.E534@asd10-22.dial.xs4all.nl> <19960226.477570.1832@an174.du.pipex.com> <4grotj$8q3@serpens.rhein.de>
- NNTP-Posting-Host: asd06-03.dial.xs4all.nl
- Mime-Version: 1.0
- Content-Type: text/plain; charset=iso-8859-1
- Content-Transfer-Encoding: 8bit
- X-NewsSoftware: GRn 2.1 Feb 19, 1994
-
-
- In article <4grotj$8q3@serpens.rhein.de> mlelstv@serpens.rhein.de (Michael van Elst) writes:
- > m.hendry@dial.pipex.com (Mathew Hendry) writes:
- >
- > >BTW, running the BYTEMarks in a multitasking environment is actually likely
- > >to show the Amiga in a good light, because AmigaOS has much lower task
- > >switching overheads than many other multitasking OSs...
- >
- > Right.
-
- Not necessarily in this case. What I meant is that the timing routines may
- include time spent on other tasks.
-
- As for task-switch overhead: You'll note that BYTE's comparison base has *no*
- task-switch overhead because it doesn't multitask. It's just a plain Intel box
- single-tasking in 32-bit mode with essentially no OS at all.
-
-
- > >: As for FP performance, I didn't look through the source all that closely but it
- > >: seemed to me that the FP tests happened to hammer mainly on the few FP
- > >: instructions that aren't implemented on the 040 (and are trapped by SW
- > >: emulation). Here too the Amiga could be getting results that can be said to be
- > >: artificially low by a very large factor.
- >
- > >Again, we're talking about the real world here. The algorithms were selected
- > >to mirror those common in real applications. Do you think that they are not
- > >representative?
- >
- > The binaries might be not representative. If compiled for the right CPU there
- > shouldn't be a huge emulation overhead.
-
- Whether emulation code is inlined or not, there is a lot of overhead involved
- simply in computing the expected results in software. I compiled for the 040 in
- both cases but that doesn't take away the fact that the tests may be focused on
- an extremely unlucky part of the 040 FPU instruction set.
-
-
- > >Does SAS support the inlining of the unimplemented instructions as well? I
- > >guess not.
- >
- > SAS either inlines FPU instructions or uses functions calls. For 040 you would
- > only inline the implemented instructions and use 040-aware functions for the
- > rest.
-
- AFAIK gcc inlines the emulation code for unimplemented instructions (or putting
- it differently, inlines those 040-aware functions). The difference for
- FP-intensive code should be noticable.
-
-
- --
- ============================================================================
- # Jeroen T. Vermeulen \"How are we doing kid?"/ Yes, we use Amigas. #
- #--- jtv@xs4all.nl ---\"Oh, same as always."/-- ... --#
- #jvermeul@wi.leidenuniv.nl \ "That bad, huh?" / Got a problem with that? #
-