home *** CD-ROM | disk | FTP | other *** search
- Path: lou.teclink.net!usenet
- From: rad@teclink.net (rad)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: Processors
- Date: 28 Mar 1996 04:40:48 GMT
- Organization: TECLink Internet Services: info@TECLink.Net
- Message-ID: <3191.6660T1353T1798@teclink.net>
- References: <4ios8j$gtb@brachio.zrz.TU-Berlin.DE>
- NNTP-Posting-Host: tc1_25.teclink.net
- X-Newsreader: THOR 2.2 (Amiga;TCP/IP) *UNREGISTERED*
-
- On 20-Mar-96 12:04:35, Philipp Boerker <rawneiha@lyra.zrz.TU-Berlin.DE> wrote:
- X-SystemInfo: Teclink.net: comp.sys.amiga.programmer
- X-Message-No: 533 (database)
- From: Philipp Boerker <rawneiha@lyra.zrz.TU-Berlin.DE>
- Subject: Re: Processors
- Date: Wed, 20 Mar 96 12:04:00
- Message-ID: <4ios8j$gtb@brachio.zrz.TU-Berlin.DE>
- Path:
- lou.teclink.net!uunet!in2.uu.net!news.mathworks.com!zombie.ncsc.mil!nntp.coast
- .net!fu-berlin.de!zrz.TU-Berlin.DE!lyra.zrz.TU-Berlin.DE!rawneiha
- Newsgroups: comp.sys.amiga.programmer
- Organization: Technical University Berlin, Germany
- References: <1880.6651T550T1538@darwin.topend.com.au>
- <4ikc9l$t08@nadine.teleport.com> <1424.6652T807T2165@darwin.topend.com.au>
- NNTP-Posting-Host: lyra.zrz.tu-berlin.de
-
- envision@darwin.topend.com.au (travis) writes:
- X-SystemInfo: Teclink.net: comp.sys.amiga.programmer
- X-Message-No: 533 (database)
- From: Philipp Boerker <rawneiha@lyra.zrz.TU-Berlin.DE>
- Subject: Re: Processors
- Date: Wed, 20 Mar 96 12:04:00
- Message-ID: <4ios8j$gtb@brachio.zrz.TU-Berlin.DE>
- Path:
- lou.teclink.net!uunet!in2.uu.net!news.mathworks.com!zombie.ncsc.mil!nntp.coast
- .net!fu-berlin.de!zrz.TU-Berlin.DE!lyra.zrz.TU-Berlin.DE!rawneiha
-
- >envision@darwin.topend.com.au (travis) writes:
- >
- >>Here are the very sad results using the same scene,antialiasing etc.
- >> DX2/66 8Mb RAM , cheap video accel etc.................28 mins 1 sec.
- >>My A4000/040/40Mhz 28Mb on Warp 4Mb on mboard all 60nS....52 mins 1 sec.
- >>
- >>That's a very big difference considering the DX2/66's clock speed is
- >>considerably slower than the Warp Engine. ie 66Mhz compared to 80 Mhz.
-
- No, the integer unit on the 68040 is based on the BCLK (40Mhz) only the FPU
- uses the higher clock (80Mhz). Further most graphics calculations are done
- in integer instructions so you're barely if at all using any of the faster FPU
- instructions.
-
- Another major factor you seemed to have failed to consider is that graphics
- rendering is very memory intensive. Having a Level-2 cache as most 486's do
- would be a large advantage you won't have with your Warp Engine. (Having
- twice the memory on the DX2 may also be helping bias the results.) Another
- question you haven't addressed: Are you using the same revision software
- (and quality) on both systems?
-
- >>You must remember that Motorola considers the processor speed to be the bus
- >>speed. The 040 actually runs at twice the bus speed.
- >
- >The i486 DX2/66 has a bus clock of 33 MHz (that's where the "DX2" comes
- >from!), the usually non-L2-cached 040 has a bus of about 16 MHz (60 ns RAM
- >assumed). Compare apples to apples!
-
- You're misleading people here. The effective bus speed and the bus clock are
- not the same thing. What I think you're trying to say is that a 486DX/66 with
- level-2 cache has a much faster effective bus speed because of the level two
- cache.
-
- ---------------------------------------------------------------------------
- - Richard Deken EMail: (personal) rad@teclink.net -
- - VLSI Design Engineer (AuE) rad@aue.com -
- - Advanced Microelectronics PGP public key available -
- ---------------------------------------------------------------------------
-
-