uucibg@swbatl.UUCP (3929) (09/02/89)
As if this has been hashed to death in the past, please email to me rather than taking up net bandwidth. I'm thinking about going the 'blessed' route to 68030-ness (the Apple upgrade from Mac II to Mac IIx). The list on this is $2,199 for just the motherboard. As I understand it, this is just the board swap where they take (rip? :-) out your machines motherboard and put in a IIx motherboard and then move the SIMMs on your original motherboard to the new one. By the way, I am aware that the drive upgrade is $599 (and requires the IIx motherboard). What's the scoop on third-party boards? How compatible are they for things like A/UX, SIMM usage, SIMM speeds required, etc? On a bit of a tangent: does anyone out there have an idea of what sorts of speeds you'll need in your SIMMs to take complete advantage of a 25Mhz machine? I've got this sinking feeling that you'll need 80ns or better and all these 100ns SIMMS will induce wait-states or some other undesirable effect(s)... Thanks, -------------------------------------------------------------------------------- Brian R. Gilstrap ...!{ {killer,bellcore}!texbell, uunet }!swbatl!uucibg One Bell Center +---------------------------------------------------------- Rm 17-G-4 | "Winnie-the-Pooh read the two notices very carefully, St. Louis, MO 63101 | first from left to right, and afterwards, in case he had (314) 235-3929 | missed some of it, from right to left." -- A. A. Milne -------------------------------------------------------------------------------- Disclaimer: Me, speak for my company? You must be joking. I'm just speaking my mind.
ech@cbnewsk.ATT.COM (ned.horvath) (09/04/89)
From article <763@swbatl.UUCP>, by uucibg@swbatl.UUCP (3929): > I'm thinking about going the 'blessed' route to 68030-ness (the Apple > upgrade from Mac II to Mac IIx). The list on this is $2,199 for just the > motherboard. As I understand it, this is just the board swap where they > take (rip? :-) out your machines motherboard and put in a IIx motherboard > and then move the SIMMs on your original motherboard to the new one. > By the way, I am aware that the drive upgrade is $599 (and requires the IIx > motherboard). You should also consider installing a PMMU ("real" MMU, 68581); Apple wants around $500 list (including installation), the great Harvey Mudd MUG dodge is costing me $170 plus installation. Note that the '581 is a proper superset of the '030 MMU capabilities, so the only think you're giving away is the data cache. I believe Peripheral Land is marketing a superdrive that works with pre-IIx Macs, so that needn't be a gating factor. The "royal road" is a LOT more expensive in this case... =Ned=
adam@mit-amt.MEDIA.MIT.EDU (Adam Glass) (09/05/89)
In article <950@cbnewsk.ATT.COM>, ech@cbnewsk.ATT.COM (ned.horvath) writes: > You should also consider installing a PMMU ("real" MMU, 68581); Apple wants > around $500 list (including installation), the great Harvey Mudd MUG dodge > is costing me $170 plus installation. Note that the '581 is a... The PMMU (Paged Memory Management Unit) which allows "pages" of memory to be swapped in and out of memory (and to/from disk) is the MC68851, not the MC68581. Minor detail - or have I missed something, and there really is a MC68581? The MC68851 is a superset of the Apple MMU's functions. The MC68030 has the MC68851 built in, and thus, '030 users do not need the PMMU to do A/UX and virtual memory. By the way, this is more of a comp.sys.mac.aux question, but does A/UX do virtual memory? And is that the reason why you need an MC68851 chip to use A/UX, or is there another reason. Adam -- (adam@media-lab.media.mit.edu) No .sig today. I'm bored and too tired to think.
paul@taniwha.UUCP (Paul Campbell) (09/05/89)
In article <653@mit-amt.MEDIA.MIT.EDU> adam@mit-amt.MEDIA.MIT.EDU (Adam Glass) writes: >By the way, this is more of a comp.sys.mac.aux question, but does A/UX do >virtual memory? And is that the reason why you need an MC68851 chip to use >A/UX, or is there another reason. Yes it does have VM, but like most other Unixes, even if it didn't have VM (ie it swapped), it would still an MMU to do process address space switching and to protect processes and the kernel from each other. Paul -- Paul Campbell UUCP: ..!mtxinu!taniwha!paul AppleLink: D3213 "Free Market": n. (colloq.) a primitive fertility goddess worshipped by an obscure cult in the late 20th C. It's chief priest 'Dow Jones' was eventually lynched by an enraged populace during an economic downturn (early 21st C).
dwb@sticks.apple.com (David W. Berry) (09/07/89)
In article <653@mit-amt.MEDIA.MIT.EDU> adam@mit-amt.MEDIA.MIT.EDU (Adam Glass) writes: >By the way, this is more of a comp.sys.mac.aux question, but does A/UX do >virtual memory? And is that the reason why you need an MC68851 chip to use >A/UX, or is there another reason. A/UX implements demand paged virtual memory. That is one of the reasons it needs a '851. Opinions: MINE, ALL MINE! (greedy evil chuckle) David W. Berry (A/UX Toolbox Engineer) dwb@apple.com 973-5168@408.MaBell AppleLink: berry1
ts@cup.portal.com (Tim W Smith) (09/10/89)
< Yes it does have VM, but like most other Unixes, even if it didn't have < VM (ie it swapped), it would still an MMU to do process address space < switching and to protect processes and the kernel from each other. Since when do processes need protection from the kernel? Is A/UX that unstable or something? And how does the MMU accomplish this, since the kernel can reprogram the MMU? A kernel that is out to get a process ain't gonna be stopped by an MMU! :-) Tim Smith
paul@taniwha.UUCP (Paul Campbell) (09/11/89)
In article <22032@cup.portal.com> ts@cup.portal.com (Tim W Smith) writes: >< Yes it does have VM, but like most other Unixes, even if it didn't have >< VM (ie it swapped), it would still an MMU to do process address space >< switching and to protect processes and the kernel from each other. > >Since when do processes need protection from the kernel? Is A/UX that >unstable or something? And how does the MMU accomplish this, since >the kernel can reprogram the MMU? A kernel that is out to get a process >ain't gonna be stopped by an MMU! The kernel can use user mode access to a process (for example when a user does a read system call) to transfer data. If a user asks for a write to a read-only part of their address space (code for example) the system can do one of 2 things - either check every access page by page - or use the MMU to take a fault (from the kernel) to report the problem. Of cource this is not really protecting a process from the kernel, really from itself - I should have put up a smiley with my original comment, I assumed people would see the humor in it :-) A/UX is not unstable - this system (taniwha) has run for the past year almost continuously without any problems, (except when I am doing driver development :-). Paul -- Paul Campbell UUCP: ..!mtxinu!taniwha!paul AppleLink: D3213 "Free Market": n. (colloq.) a primitive fertility goddess worshipped by an obscure cult in the late 20th C. It's chief priest 'Dow Jones' was eventually lynched by an enraged populace during an economic downturn (early 21st C).