NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/sys/amiga/misc-part1/502

Text File | 1996-08-05 | 1.7 KB | 45 lines

Newsgroups: comp.sys.amiga.misc
References: <49tus6$[e-mail]> <4cfvv3$[e-mail]>
X-Newsreader: MicroDot 1.8
Mime-version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-encoding: 8BIT
Path: news.tng.oche.de!tomate.tng.oche.de
X-Gateway: ZCONNECT UE utomate.tng.oche.de.tomate.tng.oche.de [PolyNet zTOr V4.901 Serie: "light"]
Subject: Re: OS features
Message-ID: <[e-mail]>
Date: Sat,  6 Jan 96  9:34:40 GMT
Organization: Developer
X-ZC-TELEFON: [phone]
X-ZC-POST: Mies-v-d-Rohe-Str.31 D-52074 Aachen
From: [e-mail] (Dietmar Eilert)

[e-mail] (Michael van Elst) schrieb am 04.01.1996:

ME> [e-mail] (James Burton) writes:
ME>
ME> >This is a common misconception. There is no significant performance hit
ME> >with virtual memory, the address translations are done in hardware
ME> >effectively as part of the RAM. If they couldn't be done within a RAM cycle
ME> >then somebody would find a faster way of doing it (read wider RAM).
ME>
ME> There is a noticable but small performance hit caused by the MMU
ME> if there is a translation cache miss. Nothing to be excited about,
ME> but you can't ignore it either.

There will be a performance loss due to the fact that memory protection
doesn't allow passing pointers to memory blocks in order to share data
between tasks. Special functions have to be used to exchange data (e.g.
memory mapped files).

* In a random sampling of GoldED users, 100% said that they'd rather use
  GoldED than be publically flogged, eaten alive by pirhanas, OR watch
  SeaQuest DSV.

/\/\ GED Support
/\/\ FAX [phone]
/\/\ FAX [phone]-(Pause)-22

-- MicroDot 1.8