Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!zephyr!tektronix!tekcrl!tekgvs!keithe
From: keithe@tekgvs.LABS.TEK.COM (Keith Ericson)
Newsgroups: comp.sys.ibm.pc
Subject: Re: Vga Boards, and Seiko's new monitor
Keywords: vga,graphics,windows,video-7,wonder,fastwrite,ati
Message-ID: <5535@tekgvs.LABS.TEK.COM>
Date: 9 Jul 89 07:33:54 GMT
References: <2565@PEDEV.Columbia.NCR.COM>
Reply-To: keithe@tekgvs.LABS.TEK.COM (Keith Ericson)
Organization: Tektronix, Inc., Beaverton,  OR.
Lines: 55

In article <14995@ut-emx.UUCP> nghiem@walt.cc.utexas.edu (Alex Nghiem) writes:
<In an earlier article I wrote:
<>I and some co-workers evaluated the ATU Blunder card a couple of months ago.
<>Well, I think it has mostly to do with "Does The Test Use BIOS Calls" or
<>"Does The Test Access The Screen Directly?"  Our tests access the screen
<>memory directly.  
<
<The PC Magazine Benchmark tests video bios and video register compatibility 
<and speed. [It includes a Windows-based instruction benchmark test.]

It also uses drivers supplied by the vendors, which means that a
well-written driver may very well show it's associated card to (1) perform
better than another card & driver combination and (2) rank in a different
order of speed performance than it might in our direct video-display
accessing tests.

(By the way, I think I'll be able to get the MS-DOS binary for our test
sent of to Rahul pretty soon.  But don't write for it - I don't have it
(yet) and I'll be submitting it as soon as I get it.)

<I hope that the article was based on both types of access, not just one
<or the other. Also, I thought that accessing the registers directly was
<inherently faster than using the bios.

I'm afraid you may be hoping in vain, but I'm not sure...

<>In each and evey one of these tests the ATI card showed itself to be one of
<>the slowest cards we had ever used.

<Yikes! Which version of the ATI card did you evalute? Did you evaluate
<the ATI VGA Wonder 512K or did you evaluate the older ATI VIP card.
<How much memory did you have. Which specific operating system 
<did you use--SCO Xenix or Interactive, or something else?

Yep: ATI Wonder w/512K RAM.  MS-DOS 3.30A.  4 Megabytes of 32-bit-wide
system RAM.  We didn't benchmark it under any **IX.

<>Because of the difference in how "slow" is determined, and
<>because a UNIX driver may very well manipulate the display memory
<>direclty, I would be EXTREMELY reluctant to link DOS-based speed
<>measurements to UNIX, X-Windows or any other performance metrics.

<If a test is measuring pure register-level or bios-level response,
<why would the operating system make a difference? Isn't it risky
<to write a Unix VGA driver that accesses the registers directly?

The quality  of the video display driver provided by the **IX vendor will
greatly affect the performance of the video display system.  Said driver may
access the display directly, or may use the display card's BIOS.  Hence I
find it unrealistic for a test run under MS-DOS (or Windows) to necessarily
be an accurate reflection of that same display card's **IX performance.

Hope I've been a little clearer this time.

kEITHe