home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga,comp.sys.amiga.hardware,comp.sys.amiga.programmer,no.amiga
- Path: newsfeed.ed.ac.uk!hwcee!news
- From: Jon Harrison <ceejh>
- Subject: Re: FWD: Fate of 68080
- Content-Type: text/plain; charset=us-ascii
- Message-ID: <DLoyEB.17J@cee.hw.ac.uk>
- Sender: news@cee.hw.ac.uk (News Administrator)
- Content-Transfer-Encoding: 7bit
- Organization: Dept of Computing and Electrical Engineering, Heriot-Watt University
- References: <4d3c27$n6c@rs18.hrz.th-darmstadt.de> <1141.6593T1100T790@norconnect.no>
- Mime-Version: 1.0
- Date: Wed, 24 Jan 1996 15:15:46 GMT
- X-Mailer: Mozilla 1.12 (X11; I; HP-UX B.10.01 9000/712)
- X-Url: news:1141.6593T1100T790@norconnect.no
-
- kenneth@norconnect.no (Kenneth C. Nilsen) wrote:
- >>> I heard some rumours about a 700MHz PowerPC. This has very little backing,
- >>> but is it true or not? I know there have been difficulties to get more than
- >>> 300MHz so this sounds not too probable. Anyone?
- >
- >>The current/near future top-of-the-line is the PPC 604@150 MHz.
- >>The 300 MHz was foreseen for the 64-bit PPC 620, but the situation about this
- >>beast is unclear since the projected performance gain is not enough these
- >>days. Instead a design rework will be done, maybe they call it 630 or so.
- >>Anyway the race for MHz sounds silly to me, since the memory can't keep pace
- >>with it, except for large (and expensive) caches. 700 MHz is science fiction
- >>I'd say.
-
- A recent article in Electronics Weekly quoted an Intel (Yeuk!!) source as
- stating that by 2010 they would be knocking on or exceeding 1GHz clock rates,
- using a 0.07um process (very close to the brick wall).
-
- [snip]
-
- >But wouldn't it be possible to write data parallell to
- >memory, I mean, using double memory bus writes so that in principle you can
- >write not 1 byte, but 2 bytes simultainously or even 4 bytes etc. (hope you
- >get the picture) ?
-
- The pentiumPRO (Sorry again) uses three 64bit busses (cache, fpu, external RAM)
- at the moment.
-
- Many high performance systems use Interleaved memory architectures at this time
- to boost access times. e.g. in a 4-way interleave you have 4 x-bit wide memory
- banks so that access 1 goes to bank1, 2 to 2, 3 to 3 etc.
-
- as total memory access time i.e. the 70ns you see on your SIMMS is made up of
- approximately 20ns setup and 50ns write/refresh you can get the speedup shown
- below:
-
- # - setup time where processor needs to be on Bus
-
- @ - memory write/refresh time where processor does not
-
- BANK1 #### @@@@@@@@@@ #### @@@@@@@@@@
-
- BANK2 #### @@@@@@@@@@ #### @@@@@@@@@@
-
- BANK3 #### @@@@@@@@@@ #### @@@@@@@@@@
-
- BANK4 #### @@@@@@@@@@ #### @@@@@@@@@@
-
- rather than :
-
- #### @@@@@@@@@@ #### @@@@@@@@@@ #### @@@@@@@@@@ #### @@@@@@@@@@
-
- See, more than twice as fast. Unfortunately this magnitude of speedup is not
- usually possible with very fast SRAMS, as the ratio of # : @ is much less.
-
- jon
-
-