home *** CD-ROM | disk | FTP | other *** search
- Path: nntp.teleport.com!sschaem
- From: sschaem@teleport.com (Stephan Schaem)
- Newsgroups: comp.sys.amiga,comp.sys.amiga.hardware,comp.sys.amiga.programmer,no.amiga
- Subject: Re: FWD: Fate of 68080
- Followup-To: comp.sys.amiga,comp.sys.amiga.hardware,comp.sys.amiga.programmer,no.amiga
- Date: 31 Jan 1996 12:03:27 GMT
- Organization: Teleport - Portland's Public Access (503) 220-1016
- Message-ID: <4enlqf$jcc@maureen.teleport.com>
- References: <4ctbjv$js6@bcrkh13.bnr.ca> <867.6583T1061T1835@norconnect.no> <judas.0hjs@tomtec.abg.sub.org> <4e7rhi$4fo@maureen.teleport.com> <md94-tar.822613553@nada.kth.se>
- NNTP-Posting-Host: kelly.teleport.com
- X-Newsreader: TIN [version 1.2 PL2]
-
- Tomas Arvidsson (md94-tar@nada.kth.se) wrote:
- : In <4e7rhi$4fo@maureen.teleport.com> sschaem@teleport.com (Stephan Schaem) writes:
-
- : > A 133mhz pentium is near 2 time slower then a 133mhz 604.
-
- : I wonder where you got that information from...
-
- I will repeat for the third time... byte magazine... integer and float
- benchmark on a 133mhz 604 mac vs a 133mhz pentium pc.
-
- : As far as I can see and as far as you can trust benchmarks (where I
- : rate SPEC as one of the more reliable ones) the processors are about
- : equally fast. (Note that the SPEC numbers for the 604 is the highest
- : reported so far and that they fall below the estimated 200/200
- : (SPECint/fp-92)). See the table below:
-
- : *******************************************************************
- : Processor Clock (MHz) SPEC-92 SPEC-95 Note
- : int fp int fp
- : M68040 25 21 15 (0)
- : M68060 50 ~60 ~45 (0)
- : Pentium 133 191 121 4.1 3.1 (1)
- : Pentium Pro (P6) 150 276 200 6.1 5.4 (2)
- : Pentium Pro (P6) 200 366 283 8.1 6.7 (2)
- : PPC 604 133 176 157 4.1 3.3 (3)
-
- Now from spec 92... the 604 is slower then the p5 (int) :)
-
- : > Now to equal PPC performance you would need at least a 266mhz
- : > pentium... (This from some past byte issue comparing the new macs
- : > with top of the line PC)
-
- : So what they did was comparing one "top of the line" PC (I wonder what
- : they considered was "top of the line"), probably running Windows or
- : Win95, and one PowerMac running System 7.5 or some such. Now, what
- : conclusions could you make about processor performance from such a
- : test? The answer is *none*.
-
- Stop assuming for your own sake! Yes, like they are goingto measure
- how fast you can open window to report the integer/float speed.
- Get some info on the Byte Performance Benchmark then you can claim
- it measure system7 os unix performance and not raw cpu + compiler.
-
- They showed the FASTEST 133mhz pentium PC to be near 2 time slower then a
- 133mhz 604 mac for integer benchmark and float benchmark... So could
- conclude instead that if:
- You run the same 3D renderer on a aux 132mhz 604 mac it will render near 2
- time faster then on a 133mhz penitum PC running linux...
-
- : > The P6 is not showed to be a major leap forward at all ...
-
- : No, it isn't but it is still more than twice as fast as *any* existing
- : PPC today (see table above).
-
- How much for a pentium pro? is the 133mhz still priced at ~600$?
- How much for a ppc 604?
-
- : > so for the next few years (maybe forever) PPC will have the edge mhz
- : > for mhz against any x86 incarnation.
-
- : They will probably not have such an advantage (they don't have it now)
- : but they will probably have a (much) better price/performance ratio
- : even if Intel probably could lower their prices significantly if they
- : felt like they need to
-
- They have it now by maybe a factor of ~2, so better price, design and
- performance. The intel serie deserve to die on the stop and let the
- crap be emulated by a ppc! :)
-
- Can I compile my app with 64bit int on the Pentium PRO?
-
- Stephan
-