Talk:Macintosh 128K/512K technical details

POV

This article reads like a personal point of view at times. Like when talking about Apple not compromising features and how the 68000 was designed for ease of use and speed. Those are opinions and not facts. cbmeeks 18:52, 1 December 2017 (UTC)

IEEE Computer magazine, January 1985

According to a letter in the May 1985 edition of MacTech (volume 1, issue 6):

"The January issue of IEEE Computer[] has a very comprehensive technical/ design article on the logic board of the Mac that I am sure most people would enjoy reading."

The online archive only goes back to 1988. Does anyone have access to this article? —204.42.16.169 04:17, 13 May 2006 (UTC)

I have access, but I don't see that article. They do mention what might have been the first DOS compatibility product in the news section.

The really relevant info was in the Inside Macintosh and Macintosh Hardware Reference books, published a little later, which I happen to have ;v) . Potatoswatter 03:33, 12 February 2007 (UTC)

Scope

I'd like to see this article finished out, but just to give my vision...

The old Macs were really great machines for understanding the basics of PC architecture. They did all the multimedia in basically the minimum number of transistors. But newbie readers won't jump into the transistors side of things. A strategy for presentation would be to introduce the functionality first, and the functionality of the component chips, and then show how it all comes together.

Precise details about memory maps are probably not as important as the broader ideas of how the different kinds of chips interfaced with the bus — all connected right in parallel to the 68000's pins. In particular the VIA was well utilized, connecting to the sound, mouse, and video systems yet still controlling the keyboard. Also the DMA engine is worthy of in-depth explanation. There was no RAM refresh overhead as audio/video DMA accesses refreshed every DRAM line as a side effect.

If someone's going to write an OS, they won't do it from this article alone. We should be very heavy on references, linking to good primary source docs for the 6522, SCC, and prolly the original SCSI controller as well. The Linux port for the Mac Plus has most of the memory map you really need. And most readers are a lot more casual to start with. Potatoswatter 03:33, 12 February 2007 (UTC)

Self-Reference

The whole intro breaks policy, not to mention all article on Wikipedia are a work in progress. I think I'll rewrite the into and mark this as a stub. I'm also going to remove the merge tag, it's been there forever. TrevorLSciAct 20:50, 15 March 2007 (UTC)

Thanks, and welcome to this little project! What is this about a stub? Potatoswatter 01:26, 16 March 2007 (UTC)

Yea, I guess it isn't...so i didn't tag it as such. TrevorLSciAct 16:38, 17 March 2007 (UTC)

Component Location

You know what would be great? The logicboard location reference as it pertains to the components in a table. (e.g. Zilog chip D-13)--Mac128 (talk) 19:40, 2 March 2008 (UTC)

Huh? What we have is a high-resolution photo of the logic board and a schematic diagram indexed exactly like that. How can you have indexes anyway without knowing locations? The chips are on a grid. Potatoswatter (talk) 14:15, 3 March 2008 (UTC)

"Each fetch took two cycles out of eight per implying a memory bandwidth tax of 25%"

I'm struggling to find the proper reference now, but I believe this statement to be false. It's true of the Macintosh SE but the 128k and 512k reserve 50% of bandwidth for video during pixel output periods, not 25%. You can find weak evidence of this in the various repeated claims that the SE is 10, 15 or 20% faster than the Plus, despite both requiring the same video bandwidth and using the same clock, but I'll update if and when I can find the official Apple word. — 209.249.190.124 (talk) 13:59, 23 September 2019 (UTC)

I found a reference for this, and updated the article. — 209.249.190.124 (talk) 14:44, 25 September 2019 (UTC)