snoopy@sopwith.UUCP (Snoopy) (06/05/89)
In article <8081@killer.Dallas.TX.US> elg@killer.Dallas.TX.US (Eric Green) writes: |> on the few 68030 machines around (Next, Macs, Suns), it seems that one |> statement I have read (68030 == 68020+10/15%) is quite reasonable | |Figures I've seen indicate about a 20-30% improvement. Note that |performance gains can be sapped by high-overhead operating systems |(e.g. Mach on the NeXT) or by anemic memory systems (e.g. Mac IIx). |You've benchmarked 68030-based Suns??? Considering that they only |recently introduced them, you've been working fast! (the Sun's got |about the same overhead as a NeXT anyhow, except it doesn't have the |object-oriented layer to the user interface). I'm curious as to why you say that Mach is "high overhead" when it has better performance than 4.3 BSD. 15% faster compiling the kernel on a VAX, according to the article in the August '86 _Unix_Review_. Better/newer numbers welcome. _____ .-----. /_____\ Snoopy ./ RIP \. /_______\ qiclab!sopwith!snoopy | | |___| parsely!sopwith!snoopy | tekecs | |___| sun!nosun!illian!sopwith!snoopy |_________| "I *am* the next man!" -Indy
rang@cpsin3.cps.msu.edu (Anton Rang) (06/05/89)
In article <213@sopwith.UUCP> snoopy@sopwith.UUCP (Snoopy) writes: In article <8081@killer.Dallas.TX.US> elg@killer.Dallas.TX.US (Eric Green) writes: |Figures I've seen indicate about a 20-30% improvement. Note that |performance gains can be sapped by high-overhead operating systems |(e.g. Mach on the NeXT) or by anemic memory systems (e.g. Mac IIx). |You've benchmarked 68030-based Suns??? Considering that they only |recently introduced them, you've been working fast! (the Sun's got |about the same overhead as a NeXT anyhow, except it doesn't have the |object-oriented layer to the user interface). I'm curious as to why you say that Mach is "high overhead" when it has better performance than 4.3 BSD. Well, 4.3 BSD's performance is nothing to cheer about, actually. But empirically, running the same software, I find that the performance of software on the NeXT (using a 25 MHz 68030) is worse than the same software running on a Mac // (68020 16 MHz) let alone a Mac //x. I've found this with Mathematica and GNU C (under A/UX on the Mac, which I didn't expect to be a high-performance system). WriteNow seems to run at about the same speed on my Mac SE (8 MHz 68000) and the NeXT (though I don't have virtual memory of course). Then again, the operating system running on the NeXT is beta software. Maybe the release will be better.... +---------------------------+------------------------+ | Anton Rang (grad student) | "VMS Forever!" | | Michigan State University | rang@cpswh.cps.msu.edu | +---------------------------+------------------------+