Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!lll-winken!taco!hobbes.catt.ncsu.edu!kdarling From: kdarling@hobbes.catt.ncsu.edu (Kevin Darling) Newsgroups: comp.sys.amiga.advocacy Subject: Re: 486 clone vs. Amiga 3000 Message-ID: <1991May28.012343.18630@ncsu.edu> Date: 28 May 91 01:23:43 GMT References: <1991May24.223604.19390@crash.cts.com> <00675215513@elgamy.RAIDERNET.COM> Sender: news@ncsu.edu (USENET News System) Organization: North Carolina State University Lines: 25 In <00675215513@elgamy.RAIDERNET.COM> elg@elgamy.RAIDERNET.COM (Eric Lee Green) writes: > [...] that the VGA controller you're talking about is the bottom of > the line 640x400x16 VGA with about 60 waitstates for each access to > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > its onboard memory (no, I'm *NOT* joking, with a fast processor it > really DOES need that many wait states)... Or even more. OR far less, if he was talking about a better VGA card than you are. But this kind of comparison opens a real can of worms: Consider. In the same 640x400x16 Amiga mode, even a stock 7Mhz 68000 cpu can be forced to sit through, ummm what? up to 320 waitstates to access video ram? Or over a thousand waitstates on a 25Mhz 68030. Sure, half of the overall time waitstates can be minimal (perhaps none to a dozen). But the above slowdown could occur 12,000 times a second. Or in other words, only about 800K possible cpu->vidram accesses per second can hit the minimum waitstate scenario. Not super impressive. The point is: Hmm, no real point ;-). Except perhaps that I agree: > Raw CPU speed is only one part of what makes a system a system. Yep, especially with video. regards - kevin