home *** CD-ROM | disk | FTP | other *** search
- Path: informatik.tu-muenchen.de!fischerj
- From: fischerj@Informatik.TU-Muenchen.DE (Juergen "Rally" Fischer)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: CHIP RAM speed test resul
- Date: 4 Apr 1996 23:05:39 GMT
- Organization: Technische Universitaet Muenchen, Germany
- Distribution: world
- Message-ID: <4k1kk3$i2q@sunsystem5.informatik.tu-muenchen.de>
- References: <4j6jv0$1im@serpens.rhein.de> <5827.6659T112T770@mbox.vol.it> <1996Apr2.234528.8971@scala.scala.com>
- NNTP-Posting-Host: hphalle5.informatik.tu-muenchen.de
- X-Newsreader: TIN [version 1.2 PL2]
-
- Dave Haynie (dave.haynie@scala.com) wrote:
- : In <5827.6659T112T770@mbox.vol.it>, bizzetti@mbox.vol.it (Fabio Bizzetti) writes:
-
- : >As far as I recall my A1200 chipram is 80ns; while it's exploited as if it was
- : >only 140ns. The rest of ns is wasted because Alice holds control of everything,
- : >and Alice is slower than the CPU.
-
- : You don't understand dynamic RAM. The 80ns rating on a DRAM is the row
- : address access time. That's usually (depending on the part and the
- : memory controller), the fastest that data can be accessed from that
- : part. However, DRAM also has a row address precharge time, which is
- : usually about 60-80% of the access time. So while you can access a
- : word in 80ns from the start of the memory cycle, you can't
- : continuously run random access cycles any faster than around 140ns,
- : for an 80ns part.
-
- ok, so why my 020 needs _12_ cycles , i.e. _846_ ns (!!!!) to load a
- byte/.w/.l from chipmem ?
- why AGA got the curious quality that _any_ kind of chipmem acess is
- just 2 times slower than it is in fastmem ?
- that's unlogic, because any acess should be delayed by a fix amount
- of time. but: load 6 -> 12 cycles (difference: 6), store 4 -> 8 cycles
- (difference: 4).
-
- it's not chipram speed, no offence, but you're imho "hiding" the
- true reasons.
-
- BTW imho it should cost almost nothing to add a A3000-alike chipmembuffer,
- did you do it in walker ? I really hope so. 4 longwords would even
- be better for 8plane modi, which seem to be unideal on LORES,
- i.e. it blocks a storing cpu more than 2 planes SHRES.
-
- : >It would be sufficient to modify the interface CPU <->chipram<-> Alice.
-
- : take it. The chip bus is a synchronous bus, and if you were to burst
- : four words out of it for a CPU cycle, you would either need a CPU that
- : perfectly aligned with the chip bus timing, or you would need a FIFO
- : device to store the fetched data for when CPU could take it.
-
- again, beeing no expert at all, I can't stand the feeling that this FIFO
- would be just another $0.2 TTLs. again, what about walker ?
- I see there's no way for a compatible AGA+ _now_, but adding some TTLs
- should really not be that problem for a company.
- Am I missing something ?
-
- : Alice has a 14MHz bus clock.
- And the cpu can start acess in 4 cyle parts (14 mhz cycles) and
- store in 8 cycle parts (funny thing).
- exact same timings like on OCS.
-
- : Oh by the way, "24Mb/sec" means "24 megabits per second". I think you
- : mean 24MB/s in all your talk here...
-
- please just give us 7mb/sec _copy_ speed in walker.
-
- : Dave Haynie | ex-Commodore Engineering | for DiskSalv 3 &
- : Sr. Systems Engineer | Hardwired Media Company | "The Deathbed Vigil"
- : Scala Inc., US R&D | Ki No Kawa Aikido | info@iam.com
-
- : "Feeling ... Pretty ... Psyched" -R.E.M.
- ------------------------------------------------------------------------
- fischerj@Informatik.TU-Muenchen.DE (Juergen "Rally" Fischer) =:)
-
-
-