[comp.sys.mac] SE accelerator comments?

erik@hpsad.HP.COM (Erik Kilk) (04/03/89)

I'm interested in comments about the various SE accelerators out there.
Especially the Radius 16 Mhz 68020 of which I received an ad recently.

Since I do mostly word processing, drawing, and data base applications,
I don't care about a FPP, but it would be nice to have 4x speed increase
when redrawing the screen.  The Radius 16 ad claims 4x.  Is this true? 

Also, how compatible and reliable are these with system upgrades?


Erik

siegel@endor.harvard.edu (Rich Siegel) (04/06/89)

In article <810008@hpsad.HP.COM> erik@hpsad.HP.COM (Erik Kilk) writes:

>Since I do mostly word processing, drawing, and data base applications,
>I don't care about a FPP, but it would be nice to have 4x speed increase
>when redrawing the screen.  The Radius 16 ad claims 4x.  Is this true? 
>
>Also, how compatible and reliable are these with system upgrades?
>
	We have a number of Radius Accelerators at the office in SE's; mostly
16MHz, and one  25MHz board.

	They're great. If you know your way around the inside of a Mac, they're
not too hard to install, and they seem to be compatible with just about
everything that will run on a Mac II. For things that won't work, you can
turn off the accelerator. There's also a way to configure the accelerator's
code and data caches - any way you like, as long as it's on or off. :-)

	Some hard disk drivers are balky on the accelerator, but I don't
blame Radius for that; they've done really well. I recommend the use
of SilverLining for hard disks that you want to connect; if you use
their fast handshake, your drives will run incredibly fast and will
work with the accelerator.

	I also have an Accelerator on my Mac Plus at home, along with an
FPD interface. I had an incredible bitch of a time squeezing all that junk into
a chassis not designed to hold it, but even with 4MB, the machine doesn't
run too hot. I recommend a fan to cool the Mac.

	You can't turn off the Mac Plus version - that's a function of the
Mac Plus design, not of the Radius board.

	If you have a 68882 it will drop right in place of the 68881, and go
really fast on floating-point stuff.

	Radius' tech support is really good; the times I've had to call, the
staff has been knowledgable and responsive.

	I think they might have developer pricing (I'm not sure); if you're
a Certified Developer or Apple Partner or whatever they call it, give them
a call.

	I strongly recommend the Radius accelerators. (Did I already say that?)

	-- and no, there was no money, no free product, and no sexual favors
	that prompted this; I'm simply a satisfied customer.

		--Rich



~~~~~~~~~~~~~~~
 Rich Siegel
 Staff Software Developer
 Symantec Corporation, Language Products Group
 Internet: siegel@endor.harvard.edu
 UUCP: ..harvard!endor!siegel

 "She told me to make myself comfortable, so I pulled down my pants
 and sat in the pudding." -Emo Phillips
~~~~~~~~~~~~~~~

fozzard@boulder.Colorado.EDU (Richard Fozzard) (04/08/89)

Erik Kilk writes:
>I'm interested in comments about the various SE accelerators out there.
>Especially the Radius 16 Mhz 68020 of which I received an ad recently.

I am using the Dove Marathon 020 in my SE.

It was cheap ($550 street), easy to install, and uses SE motherboard memory.
It also claims 4x speedup. This is bogus until you can afford 4Megs to add
to the accelerator board itself (I finally did and it's finally really 4x
for CPU only tasks - I have seen this on Word scrolls, Canvas and Illustrator
zooms/resizes, etc - remember disk access is another problem). I presume
this is because you need a true 32 bit data path for the full speedup.

I have had minor problems with system upgrades and because I am using a 3rd
party hard drive (Warp 9) but the tech support from Dove is good. Other than
the inability to initialize floppies, I have found very few programs (and I
use a LOT) that cant run on the 68020. These are usually old and still will
run if I boot the 68000 (try that on a Mac II!). My problem with the floppies
may be related to my hard drive (I know this sounds weird but I wont bore you
with the details).

Overall, I've been very pleased - and thoroughly spoiled. It's hard to use
unaccelerated machines now. But I do have a couple of recommendations:

	1. Before buying an accelerator, 1st get more memory and a fast,big
	hard drive - these make a much bigger difference in performance. I
	have tried the Quantum 80 - it'll knock your socks off. Wish I had
	bought it instead.

	2. The Radius uses only the SE memory - a 16 bit data path. They
	claim to get around it with a 32k cache, but this seems small in an
	era of megabyte programs, but maybe it's enough.

	3. Two things seem especially nice about the Radius over the Dove:
	Radius offers a version of SANE that uses the 68881 to speed up
	all numerical calculations (how important this is, I've no idea).
	Radius also lets you add their big screen video boards later. Dove
	promises this, but to date I know of no products.



========================================================================
Richard Fozzard
University of Colorado				"Serendipity empowers"
fozzard@boulder.colorado.edu

kierulf@inf.ethz.ch (Anders Kierulf) (04/12/89)

In article <810008@hpsad.HP.COM>, erik@hpsad.HP.COM (Erik Kilk) writes:
> 
> I'm interested in comments about the various SE accelerators out there.
> Especially the Radius 16 Mhz 68020 of which I received an ad recently.
> 
> Since I do mostly word processing, drawing, and data base applications,
> I don't care about a FPP, but it would be nice to have 4x speed increase
> when redrawing the screen.  The Radius 16 ad claims 4x.  Is this true? 
> 
> ...

I've had a Radius 25 Mhz accelerator in my SE for 2 months now and
I love it. For task that don't use floating-point extensively I
really get the 6x speedup that Radius claims; for those tasks it is
faster than a MacII.
  The only problem I've experienced so far that seems to be due to
the accelerator is that I cannot boot with MacsBug. Anybody know
how to fix that, or whether other debuggers work?

Anders Kierulf   ---   kierulf@inf.ethz.ch