|
Farmer Crack-rear end posted:What was it about UNIX workstations that made RISC CPUs particularly well-suited to them? UNIX and RISC also went well together for more “social” reasons: The UNIX workstation rose to prominence with 68K but higher performance was going to require new architectures, as BonHoward said. The UNIX workstation had also mostly come out of work done at universities, particularly Berkeley and Stanford. (Remember that SUN was originally an acronym!) RISC was a Berkeley-Stanford thing just like most of the UNIX workstations so it was natural for industry partnerships to form, especially when it came to the workstation companies that were university spin-offs. The other reason that UNIX workstations went with RISC was because UNIX was super-portable by the standards of the day. Not only was most of the kernel in C, so was the rest of the OS and its toolchain. So you could add support for your new architecture to the assembler, then the linker, then the compiler, and have something working in no time—running existing code with existing test suites. And any code generation improvements have global benefit.
|
# ¿ Jun 30, 2021 03:32 |
|
|
# ¿ May 14, 2024 20:54 |
|
Run NetBSD, OP.
|
# ¿ Jul 14, 2021 19:42 |
|
You can run a small office on terminals connected to a single 16-bit minicomputer with only a few hundred kilowords of memory and a few megawords of secondary storage, and a large office with not significantly more, scaling pretty linearly.
|
# ¿ Jul 15, 2021 09:29 |
|
PDP-8 had a good ISA. eschaton fucked around with this message at 19:27 on Jul 16, 2021 |
# ¿ Jul 16, 2021 19:18 |
|
Want to build your own? You can!
|
# ¿ Jul 16, 2021 19:32 |
|
I just built out a little RISC-V workstation!
It came right up from the SiFive SD card images, as one should expect. In the next year or so, I should be able to switch it to either Fedora or FreeBSD.
|
# ¿ Jul 26, 2021 01:42 |
|
You’re not crazy, it does not have either a front panel or USB header.
|
# ¿ Jul 30, 2021 10:30 |
|
In addition to H&P, “Computer Engineering: A DEC View of Hardware Systems Design” is also worth a read for some historical perspective.
|
# ¿ Aug 7, 2021 08:29 |
|
There’s also “Engineering a Compiler: VAX-11 Code Generation and Optimization” written by and about the VAX-11 PL/I compiler team’s work, for a look at the interplay between hardware and software for a brand new platform.
|
# ¿ Aug 7, 2021 09:06 |
|
Port NetBSD!
|
# ¿ Aug 9, 2021 01:59 |
|
corgski posted:Anyone here know much about Intel i960? I've used a bunch of equipment built around that chip, most of which was an upgrade from the company's previous hardware platform of multiple z80-based uCs (Hitachi HD64180) and a TI TMS320C20. It’s basically a RISCish 32-bit microcontroller completely unrelated to the i860. I think there were variants with MMUs but the only “computer” applications I’ve ever seen any i960 in were a couple of X terminals, otherwise they were mainly used for things like high-end storage controllers.
|
# ¿ Aug 9, 2021 20:02 |
|
Supposedly that’s SiFive’s bread and butter via Chris Lattner and others’ MLVM project.
|
# ¿ Aug 10, 2021 04:16 |
|
I now have an HP 3000-917LX minicomputer too. It’s a PA-RISC 1.1 system whose design is mostly shared hardware with the HP 9000-807 server, the big differences being that the 3000 runs MPE/iX and the 9000 runs HP-UX and there’s a different primary I/O board. 128MB of RAM and 4GB of disk, which isn’t too shabby for a system built in 1991. Unfortunately I haven’t been able to fully boot it yet, it’s pretty unhappy with the contents of the NVRAM/RTC. I need to figure out how to reset those to a reasonable state from the management console before bringing up the main CPU and booting MPE/iX. I might even have to reinstall everything from scratch. Hopefully I can find the necessary stuff to do so… MPE/iX on PA-RISC is interesting: The HP 3000 mini was a 16-bit system, and the PA-RISC variant of MPE uses a CPU emulator with mixed-mode support to run old code—including some parts of the OS! Very reminiscent of the way mixed-mode calling worked on the Power Mac, including some of the reasons why, except HP did it a few years ahead of Apple. One weird thing about MPE: Users and directories (with no nesting) are “in” accounts, directories are called “groups,” and all of files, users, groups, and accounts can have distinct passwords. Also, get this: Paths are backwards! The path POOPCHRT.POOPERS.BUTTHEAD indicates the file POOPCHRT in the POOPERS group in the BUTTHEAD account, which the user BEAVIS might log in to in order to update.
|
# ¿ Aug 30, 2021 10:19 |
|
priznat posted:Lol I was wondering what the ARM royalties are like for Apple, so they have fixed ones they have to re-up after x years or is it more fluid? They have got to be one of the biggest licensees of it. Apple was one of the original investors in ARM and it wouldn’t surprise me if they got a perpetual platform license out of it.
|
# ¿ Sep 4, 2021 19:38 |
|
doing it in hardware is how you get the massive amount of bandwidth needed for the transaction processing loads mainframes handle a guy who I chat with on Twitter sometimes has been doing IBM’s free online mainframe courses and has shared some of the insane stuff they have now, like not just parity or error correction for RAM but straight up RAID-style hot-swapping
|
# ¿ Jan 2, 2022 07:36 |
|
Kazinsal posted:I have made a terrible, awful, wonderful mistake. Congratulations! You can run Ultrix and BSD and VMS! It’ll run VMS pretty nicely, too.
|
# ¿ Feb 13, 2022 09:49 |
|
Z80 was how your spelled x86 before the IBM PC
|
# ¿ Mar 10, 2022 12:24 |
|
I emulated a little 68000 VME board, the Xycom XVME-600 it was pretty straightforward since it’s a 68000 CPU, 68681 DUART, 68172 “transparent” VMEbus controller, 128KB SRAM, and 64KB EPROM one of the things I haven’t figured out is how to make MAME map ROM to 0x000000 for the first few bus cycles after reset like the hardware does so another thing that I’ve found is that I need to pretend it has a ton of RAM rather than treat everything between 0x020000 and 0xFBFFFF as VMEbus address space, I think it’s trying to probe a peripheral on startup somewhere in 0xF70000 and if I don't do this it causes a bus error loop some pics of the FORTH words available are in the imgur gallery
|
# ¿ Aug 15, 2022 04:30 |
|
KozmoNaut posted:PC clones became a thing after Compaq did a clean room reimplementation of the IBM PC BIOS and opened the market for competition, letting people buy PCs from any vendor, in comparison to the previous microcomputers, which were all proprietary. Early on in the microcomputer world CP/M on S-100 systems ruled, first 8080 then Z80, and were largely compatible. Microsoft was actually huge in this market with their languages and productivity applications, and even told IBM to talk to Digital Research about using CP/M-86 as their OS instead of licensing one from Microsoft. But Digital Research fumbled and Microsoft bought a CP/M clone for 8086 to take to IBM—without exclusivity. So in a way, the IBM PC and its open software architecture were a continuation of what was already happening with most microcomputer use in business; open hardware was in some ways just an almost inevitable side effect. There were MS-DOS compatible systems before Compaq; what made the “PC market” not the “MS-DOS market” was that IBM PC games really only ran on 100% hardware compatibles, providing an incentive to start from something that could run MS-DOS and end with something that could run Flight Simulator. The only reason that games even mattered was that the PC was getting into homes because it enabled people to take their work home, and then once home they were used for other things too (or mostly). Minicomputers were used extensively for office automation in the 1970s through the 1980s but there were very few games for those environments: Unless you had both a personal computer or terminal and a modem at home and your work had a modem, you were only using the office minicomputer when you were physically present in the office, and then only for work because that’s all it was set up to do. Of course people actually wrote plenty of games for minis, but outside those created in academia few were sufficient to make people want to access them from home—they were just office diversions. Whereas people were buying PC-compatibles to play games, with the excuse that they could do work at home…
|
# ¿ Aug 15, 2022 09:24 |
|
You’d be surprised what’s common in process control.
|
# ¿ Aug 15, 2022 20:36 |
|
phongn posted:It did strongly influence Postscript, right? Definitely, they’re both stack oriented, as is SPL, the HP System Programming Language for the HP 3000 that was a contemporary of FORTH. The differences between FORTH and PostScript are substantial, though, it’s kind if like looking at BCPL and then looking at C. HP’s RPL language for the high end calculators like the HP-48SX is similarly advanced relarive to FORTH.
|
# ¿ Oct 5, 2022 06:05 |
|
hobbesmaster posted:It’s not a strawman, it’s an example of how a big push for a radical change from x86 was unlikely to gain market acceptance. That’s not because it was a big push for a radical change from x86. That’s because it was exceptionally difficult to write well-optimizing compilers for Itanium. If Itanium had been a 64-bit RISC, or even a 64-bit equivalent of i860, it probably would have taken off. Instead it was The Bizarro CPU and while it was eventually able to get some serious throughput (my Itanium 2 VMS box does pretty well running FORTRAN) the compiler problem was grossly underestimated as a factor. x86-64 and ARMv8/AArch64 succeed in part because they only go a little way afield rather than trying to radically rethink everything in cutting-edge ways. eschaton fucked around with this message at 10:10 on Jan 20, 2023 |
# ¿ Jan 20, 2023 10:06 |
|
phongn posted:Wasn't i860 also also a VLIW processor with a bunch of compiler-dependent scheduling and pipelining voodoo? DeMone at RWT wrote 22(!) years ago that the promises of IA64 reminded him of Intel's overblown ones for i860 years before. My impression has always been that i860 was more RISC-ish than VLIW-ish, but that compilers for it did turn out to be a Hard Problem. That’s why the majority of use outside massively parallel supercomputing was… massively parallel graphics! The Silicon Graphics RealityEngine treated the i860 as a building block, just like supercomputing systems did. (I have a bunch of VME i860 boards that I’m looking for configuration details on… And a DEC TURBOchannel graphics card with one too, for my AXP 3000-400.) Eventually in the early 1990s compilers for the i860 got pretty decent and you could actually achieve some of the theoretical throughput. Most of that though was obviated by the uses that it wound up being put to, since something like the SGI RealityEngine will have each of its many CPUs running custom assembly that fits in the instruction cache to serve its purpose in the render pipeline.
|
# ¿ Jan 21, 2023 07:31 |
|
can you imagine if people had actually used the WDS-1600 chipset in personal computers in the mid-1970s instead of the 8080/8085/Z80? (that’s what’s in the LSI-11, and WD also used it with different microcode to make the Pascal Engine)
|
# ¿ Jan 30, 2023 10:38 |
|
except they set it up byte-swapped to be compatible with little-endian S-100 is only an 8-bit bus, but on the other hand the M68000 had no A0 line, only /UDS and /LDS (upper and lower byte data strobe) to indicate which part of a word on the data bus mattered so you could build a little state machine to generate A0 and input or output either or both halves of a word to the S-100 bus and get a pseudo-little-endian linear address space you know, if you didn’t want to just treat S-100 as an even-byte-only 128KB window in the loving 24MB (soon 2GB with 68012, and 4GB theoretical) address space like any sane hardware developer would then again these were the people who pioneered backup to VHS and their OS was essentially a TOPS-10 clone for 68000 so who the hell can say, insanity abounds
|
# ¿ Jan 31, 2023 10:59 |
|
I know a bunch of people who have recapped the PSUs. The VAXen themselves typically don’t need it.
|
# ¿ Feb 15, 2023 04:45 |
|
we still need a YOSVAX running somewhere
|
# ¿ Feb 15, 2023 05:15 |
|
Nice!!
|
# ¿ Feb 15, 2023 05:36 |
|
Why don’t they pay someone to port OpenWRT to it?
|
# ¿ Jul 6, 2023 11:28 |
|
Hasturtium posted:It looks like a number of motherboards with Loongson CPUs are appearing on AliExpress and the like at sub-$400 price points. With the knowledge that the CPU is not performance-competitive with x86 and subject to various limitations and frustrations endemic to being outside of more widespread ISAs, what are these things supposed to be like? should run NetBSD pretty well
|
# ¿ Sep 6, 2023 07:36 |
|
BobHoward posted:sorry for the delayed reply. Here's a decent writeup someone did a while ago which covers a lot of it: wish rjmccall would collect his RISC-V criticism like this, be good to point folks to
|
# ¿ Sep 6, 2023 07:37 |
|
what we need is a way to trick software people into writing something that looks like “normal” software but which is actually an FSM that can be formally verified
|
# ¿ Oct 17, 2023 05:26 |
|
POSIX with Chinese Characteristics
|
# ¿ Nov 10, 2023 09:17 |
|
what we really need are open source VHDL for a T805 Transputer, C004 link switch, and C011 link interface
|
# ¿ Nov 13, 2023 00:09 |
|
BobHoward posted:There is no way to stop China from practicing. Even if it was necessary for them to practice with the best commercially relevant ISA (it's not), they could always just buy an Arm architectural license. they don’t even need to do that, they could just implement the instruction set, since that’s not protectable IP in and of itself of course if they wanted the best ISA they’d be doing a 64-bit extension of the Motorola 68000…
|
# ¿ Dec 25, 2023 01:03 |
|
also they think the only thing Pinochet did wrong was let the Chicago Boys run the Chilean economy instead of Oxbridge economists
|
# ¿ Jan 19, 2024 08:33 |
|
I found some more Inmos T805 Transputer CPUs in a box in my office along with an Inmos VMEbus board fully decked out with CPUs-less TRAMs also a couple decked out with TRAMs that do have CPUs installed I think it may be a signal from past me to write some Occam-2 or something
|
# ¿ Feb 26, 2024 09:06 |
|
I just set up an m88k-interest mailing list for people still interested in the Motorola 88000 series
|
# ¿ Mar 7, 2024 21:44 |
|
BobHoward posted:If wikipedia is to be believed, m88k was a product for just 3 years, 1988 to 1991. 1991 was about when the AIM alliance formed up, and yes, PowerPC absolutely killed m88k - it hadn't gotten much adoption and PowerPC had a built-in volume customer. Especially since the 601 was specifically intended to be a design replacement for the 88110, to the point of using the same bus protocol. Swap the part in your design, spin the board for the pinout, recompile/rewrite your firmware, done. Something that surprises me is that Data General switched to Intel instead of PowerPC; either they really wanted nothing more to do with Motorola or wouldn’t have anything to do with IBM, because they had a breadth of m88k designs they could have turned into 601 designs quickly. Maybe they were worried IBM would price the part to ensure no DG workstation or server was cheaper? (But IBM was still IBM then, and IBM never cared about price…)
|
# ¿ Mar 10, 2024 10:35 |
|
|
# ¿ May 14, 2024 20:54 |
|
Like a buddy has an I think 8-CPU AViiON minicomputer running DG-UX with 1.5GB of RAM, and I have single-CPU tower and pizzabox AViiON workstations, and both designs would have worked with a PPC601 dropped in. They’d have been the first to market, with a decent and secure SVR4 using IXI X.desktop for its UI…
|
# ¿ Mar 10, 2024 10:41 |