Jump to content

User talk:Mikegt

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Mikegt (talk | contribs) at 06:07, 30 October 2007 (macintosh 128K). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Duplicate images uploaded

Thanks for uploading Image:Psp big.JPG. A machine-controlled robot account noticed that you also uploaded the same image under the name Image:Psp big.jpg. The copy called Image:Psp big.jpg has been marked for speedy deletion since it is redundant. If this sounds okay to you, there is no need for you to take any action.

This is an automated message- you have not upset or annoyed anyone, and you do not need to respond. In the future, you may save yourself some confusion if you supply a meaningful file name and refer to 'my contributions' to remind yourself exactly which name you chose (file names are case sensitive, including the extension) so that you won't lose track of your uploads. For tips on good file naming, see Wikipedia's image use policy. If you have any questions about this notice, or feel that the deletion is inappropriate, please contact User:Staecker, who operates the robot account. Staeckerbot 05:48, 7 June 2007 (UTC)[reply]

Hi Mike,

I don't see why comparing the built-in memory is so important. If you were selling Apples in 1984, or if you'd read a few of the articles linked from the 128K, you would know that the Lisa was introduced in early 1983 with a 1 MiB ceiling and that the 512K was introduced in September 1984 shortly after the AT. And that link you gave shows the IBM Portable PC in a comparable form factor and price with 256K RAM introduced just days after.

If you look at the schematics for the 128/512K, you can see that the non-upgradeability is a result of the video circuitry being hard-wired using 7400 series chips to double as RAM refresh logic. Getting Macintosh quality graphics was REALLY HARD for Apple at that price point, so if you wanted a meg you needed to drop $10K.

So the sweet graphics cost not money, but upgradeability. Why make such a big deal of complaining? Potatoswatter 05:28, 28 October 2007 (UTC)[reply]

Hi... If you don't mind keeping the thread on one page...
I did offer one important fact to show why Apple couldn't have socketed the RAM, which is that the graphics DMA controller is used to refresh ram, and that it is implemented in 7400 series logic. RAM upgrades became possible with the Mac Plus, which had a more sophisticated RAM controller implemented using a PAL. See Macintosh 128K/512K technical details. Potatoswatter 04:27, 29 October 2007 (UTC)[reply]
You can contact me using your own talk page, since I'm watching it. I'd respond on mine, but I'm not sure you're watching mine...
The link I gave you shows a 512K design and says there is a legend that they designed it to carry 512K at the outset. The video DMA engine refreshes the RAM as a side effect of scanning the screen bitmap. This is documented, and it is evident in the two 74LS393's in the schematic and Image:OriginalMacintoshLogicBoardPre6000SerialNumber.jpg. To access the correct addresses that regularly exercise the RAM, the video buffer has to be at a particular location in memory. This location was different for every single RAM configuration of the early Macintoshes [1]. Because in the 128K & 512K this address was hardwired to the 74393's, RAM could not be upgraded.
Do you have evidence of someone actually being able to use additional RAM that was simply soldered in? That would put the framebuffer right smack in the middle of the application memory space, which would seem to hurt compatibility.
So yes, I do have complete design information and a Bachelor's in EE. Potatoswatter 01:06, 30 October 2007 (UTC)[reply]

"Hard-wired" - do you mean via the board traces, or in the DMA controller chip? If it is in the chip, then it seems like replacing the controller chip would fix the problem (assuming it was socketed). In any case, whether the design fault lies with soldered versus socketed, or somewhere else, the bottom line is that the only way to upgrade the memory was to replace the motherboard. The essential controversy remains unchanged; for everyone who bought the original Mac, they had to pay $1,000 or more so that their less-than-one-year-old system could keep up with the latest Mac software. Apple's engineers knew in 1983 that 128 K wasn't going to be adequate; the essential point here is that Apple sold a system for eight months that it *knew* would be very quickly obsolete. They still could of made good by selling the upgrade board at a reasonable mark-up, but instead sold it to the early Mac owners at five times their manufacturing cost (plus dealer labor for installation).

For the sake of the Mac 128 article, we have to leave out what Apple intended, which is speculative, but we can and should document the facts of what happened to the purchasers of this system as a direct result of it's design. Unless you can think of a good reason why this information should be withheld ? If you are willing to add this info, I will leave it up to you as to the wording.

BTW, I am not trying to pick on Apple; I think any of the manufacturers design choices that negatively impacted consumers should be documented. For example, the Wikipedia article on the IBM PC-AT does mention the stuff IBM did to specifically lock the CPU speed at a slow frequency (apparently they were trying to protect their mini-computer products).

Mikegt 05:59, 30 October 2007 (UTC)[reply]