Brenton, at 17:10 on Mon, Apr 10 1995, you wrote to Bill Grimsley ...

BG> Just like IBM says that OS/2 is usable in 4Mb?  Unless VLB technology has 
BG> made a recent quantum leap, 50MHz/0WS is not at all reliable IMO.
BV> Wanna make a bet ?
BG> We'll see.

BV> Yep. We will, won't we.

Uh oh, sounds like Brenton's about to chuck another wobbly.  |-)

BV> Patricia's machine is a VLB 80486DX-50 with a caching VLB-IDE
BV> controller and a VLB video card.
BG> Seems that filthy plutocracy must be contageous then.  |-)

BV> Isn't the contagious nature of filthy plutocracy the basic premise
BV> that helps our market driven economy survive and prosper ?

It's also what fucks the balance of trade, and drives interest rates
upwards at the same time.  Apart from those minor details, I'm all for it.

BV> It runs flat-strap at 50Mhz with 0 wait states and it has done so
BV> flawlessly for about 1.5-2 years. You do have to hand select the 
BV> boards, but you can do it and still have reliability.
BG> No, perhaps YOU can do it, but the average person just doesn't have
BG> the time or resources (nor indeed even the inclination) to keep 
BG> running back and forth to his dealer until he finds VLB cards which 
BG> will run happily at 50MHz.

BV> I'd hate to be perceived as a nit picking pedantic bastard,

Sure Brenton, I can just imagine how much you're hating it right now too.

BV> but I must correct you. 

At least you're much more tactful than Rod generally is.

BV> You clearly stated that 50Mhz/0WS VLB operation was not reliable. I was 
BV> merely offering my experience in the hope that it may assist you in 
BV> dispelling your unfounded belief.

I was generalising, you nit picking pedantic bastard.  |-)

BG> Sure, there are indeed some VLB cards which don't mind that speed, but the 
BG> bottom line is that most still don't, and you may even end up having to use 
BG> brands or chipsets that you normally would rather not.

BV> Perhaps, although I don't feel that this is the case at all. 

Maybe not for you, given your circumstances, but I still maintain that
unless the dealer does so for him, your average Joe either wouldn't bother,
or would eventually lose patience trying to find cards which ran reliably
at that speed.  I'm not suggesting that they don't exist, just that the
majority of brands don't work at 50MHz/0WS reliably.

BV> I have found that the best way to ensure reliability (regardless as to bus
BV> or CPU speed) is to pay a bit more for the quality components which
BV> are purchased from well known sources. 

Yep, that's absolutely correct, and therein lies the problem.

BV> If you feel that you don't wish to pay any more for quality components, 
BV> brands, or chipsets, and are prepared to accept lower levels of quality and 
BV> performance as the trade-off, then that is your prerogative.

It's matter of economics Brenton.  My dealer mate uses a video card which
not only runs at 50MHz, but also performs better under Windows that
anything else on the market.  And at $900, it'd bloody well want to!  If my
$120 Cirrus tops out at 40MHz, who cares?  Is the speed difference worth
the extra $780?  To a commercial graphic artist or architect, both of whom
can claim it as a tax deduction, it may be, but to me it's pissing money
down the drain.

BV> As an aside; Regardless as to the knowledge being offered by Mr. 
BV> Speed (or his wolf) Intel have never and will never manufacture an 
BV> 80486 CPU which will run 100Mhz internally whilst being driven by a 
BV> 50Mhz clock.

I have given up trying to stay abreast of CPU technology, so I wouldn't
know if it was AMD or Intel who had doubled DX/4s.  Nor do I really care. 
|-)

BV> If you have any doubt as to the validity of this, I suggest you call either 
BV> Intel Australia or their distributor NJS and enquire as to the part number, 
BV> price and availability of the device.

'Sokay Brenton, I'm quite happy to take your word for it.  We'll likely be
up to the Decium by the time I'll be able to afford a new processor anyway.

Regards, Bill
@EOT:

--- Msgedsq/2 3.05 alpha