DayStar 68060 Accelerators Announced

[Trash80toHP_Mini]

Who'd like to bet that Apple didn't have anything to do with these becoming vaporware?

 

[beachycove]

There are many items of interest in this piece, but the one that stands out for me is the statement that there would be no incompatibilities: "Daystar says its 060 accelerators will run any software that's compatible with the 040, because the 060 uses exactly the same instruction set."

Daystar, which knew what it was up to, evidently did not envision the need for a System re-write to support the 060, since the product was to be precisely an upgrade board for people who wanted to run their old systems and their existing libraries of 68k software for another couple of years (rightly pointing out that a PPC would not be faster than an 040 Mac if running 68k software). Presumably a Control Panel and an extension were in the works, but that is scarcely the same as a major change in the OS itself — which is what we usually hear when the idea of an 060 in a Mac is mooted.

Why, then, has there been so much poo-pooing the idea of an 060 working in a Mac? Couldn't this, in principle, be done the Open Source way if (and here is where it falls apart) there were enough interest?

 

[Trash80toHP_Mini]

I find ALL KINDS of interesting goodies in my files, this was one of the better tidbits, it was marked 68060 with a stick-it, still in the magazine binding along with other choice tidbits! :beige:

It flies in the face of conventional wisdom, just like I like my info! }

I was hoping someone would get a kick out of the irony! [] ]'>

 

[CJ_Miller]

I don't have any reference links handy, but I have read comments from people who were developing 68060 accelerators that it was costing more than expected, and was more difficult to put into practice than hoped. It was deemed possible but not really worth bringing to market. Look at Czuba-tech's CT60, they made several 68060 Atari upgrades, but even in moderate quantity they were quite expensive and not entirely compatible (hence the bypass switch). It can be done! Maybe a retrocomputing philanthropist with lots of time, money, and brains will figure it out and I might be able to afford the end product.

 

[Bunsen]

the 060 uses exactly the same instruction set."

Well, see, that's just flat out wrong for starters.

 

[noidentity]

They probably meant that the user-level instruction set is a strict superset of the 68040's, so that user-level software will work without changes. The snag is that the Mac OS basically ran in supervisor mode all the time, even for user code, so that aspects of the processor normally only seen by an OS could be seen by everything. There are bound to have been some changes that affected the OS and other things that did low-level access.

 

[Trash80toHP_Mini]

I'm sure DayStar would have had no trouble at all doing it. The irony is that it was announced in May of 1994 and that Apple made the Clone announcement the following January 4th. One word from Apple and DayStar would realize that R&D was pointless, the extended utility of existing 680X0 machines would not be in DayStar's interest any more than Apple's!

Anybody want to take the side of the (rhetorical) bet I offered in the original post? [] ]'>

 

[Bunsen]

Why would it matter if they did?

 

[Trash80toHP_Mini]

It matters only because here at the 68kMLA and on 'fritter in its day, discussion was always about how great it would have been to have had a 68060 option, if only how to see how well it compared to PPC. Radius and DayStar were probably the only companies with the ability to pull it off . . . guess who got Clone licenses . . . Apple's announcement was about the Radius license!

Like I said . . . one word from Apple . . .

. . . and it's just interesting, that's all. [] ]'>

 

[CJ_Miller]

I'm sure DayStar would have had no trouble at all doing it.

Well, as I recall they said they *did* have lots of trouble doing it. Not impossible, but lots of trouble.

This is not to say that I don't share your enthusiasm! I would love an 060 Mac, I just can't afford to pay much for one. I don't even have an 040 yet. I would love to see new development of 68k hardware, and would whoop with the awesome of it all.

If I need to, I will put this on my wish list along with MacOS sourcecode, ColdFire logic board, AU/X for 9, Miller columns in 7-9, and lots of other theoretically possible things I don't expect anybody to do. Maybe we need to start a "Wakileaks" for disgruntled Apple coders to leak their commented source, this would get the ball rolling. :approve:

 

[johnklos]

An m68060 card would obviously have its own ROM to initialize things on boot. From a userland perspective, the instruction set is identical. The only program that I remember having trouble with the m68060 with all of the features enabled was Adobe Illustrator, so I'd have to switch off superscalar mode before running that.

The m68060 running Mac OS was definitely faster than the first generation of PowerPC 601 systems. Part of the issue, I'm sure, was the fact that m68060s might slow adoption of PowerPC, but another (and altogether important) reason was the cost. As you can see from the numbers, it'd cost almost as much for an m68060 accelerator as it would for a new machine. PowerPC accelerators for m68k Macs certainly didn't cost that much.

 

[Charlieman]

Apple also made it very appealing to trade in an old Mac for a PPC NuBus model. Here in the UK, Apple were not noted for deals. But there were some great ones if you were in education or had a good relationship with an AppleCentre.