home *** CD-ROM | disk | FTP | other *** search
- Path: ifi.uio.no!usenet
- From: ludvigp@ifi.uio.no (Ludvig Pedersen)
- Newsgroups: comp.sys.amiga.programmer
- Subject: CHIP RAM speed test results
- Date: 5 Mar 1996 13:50:25 GMT
- Organization: Dept. of Informatics, University of Oslo, Norway
- Message-ID: <938.6638T866T849@ifi.uio.no>
- References: <1067.6635T1081T1567@direktor.voima.jkl.fi> <352.6635T777T442@horus.co.jyu.fi>
- <734.6636T1248T1351@direktor.voima.jkl.fi> <38232801@kone.fipnet.fi>
- NNTP-Posting-Host: gymir.ifi.uio.no
- X-Newsreader: THOR 2.22 (Amiga;TCP/IP)
-
-
- There was some people here said that the Blizzard1230-IV/50mhz was able to
- copy 7 mb/s from FastRsm to ChipRam.
-
- Well, as my Blizzard1230-II/50mhz was only able to copy 4.9 mb/s I was
- supsious about this.
-
- So now I have performed a speed-test on a Blizzard1230-IV/50mhz.
-
- And the result is:
-
- There is NO-WAY that card can copy 7 mb/s to ChipRam.
-
- I was only able to copy 4.6-4.7 mb/s with no bitplane DMA.
-
-
- Btw:
-
- I was able to copy 6.8 mb/s from Fast to Chip with this loop:
-
- .loop:
- REPT 8
- move.l (a0),(a1)
- ENDR
- dbra d7,.loop
-
- I suppose this is the why people have said they can copy 7 mb/s!
-
- This loop doesnt work because of DATACACHE. As the loop reads from the same
- place in FastRam every time. The CPU uses its internal datacache instead of
- accessing the memory bus to fetch the longword from FastRam.
-
-
- <sb>Ludde - Amiga Demo Coder
- <sb>Virtual Reality & Official Be developer
- <sb>ludvigp@ifi.uio.no
-
-