Question about which tools to use, bugs, the best way to implement a function, etc should go here. Don't forget to see if your question is answered in the wiki first! When in doubt post here.
tom9876543 wrote:VESA VBE 2.0 is unfortunately another example of a pathetic terrible API design.
I'm not going to argue with that. However, there was no way that a 32-bit VESA was going to come out at that point.
I'd add that one of the problems they were trying to deal with is that legacy video card ROMs are limited to 64 KiB; and video card manufacturers were already struggling with that limit (don't forget video card ROM contains a bunch of font data, colour lookup table data, etc - it's not just code).
The "nice" thing about a 16-bit protected mode interface is that (with only a little care with segment register loads) almost all of the 16-bit code could be used for both real mode and protected mode, and a lot of code duplication could be avoided.
I'd be tempted to assume the same "64 KiB limit" is responsible for multiple other problems; like not being able to put VBE/AF (VESA's hardware acceleration standard) into ROMs, video cards that don't use EDID to determine which modes the monitor does/doesn't support, video cards that only offer a small number of video modes (e.g. common 4:3 modes and nothing else), video cards that don't support "text output" BIOS functions in graphics modes, etc.
Getting rid of the legacy ROM size limit (and all the problems it causes) is also one of the reasons for UEFI.
Cheers,
Brendan
For all things; perfection is, and will always remain, impossible to achieve in practice. However; by striving for perfection we create things that are as perfect as practically possible. Let the pursuit of perfection be our guide.
and even for us it is relevant only because the video card manufacturers are about as tight-lipped over their hardware APIs as KFC is about their eleven herbs and spices
Chortle.
Awesome write-up, though, Schol-R-LEA. You should maintain a blog or something where you just do rants like this. Highly enjoyable.
Hi Schol-R-LEA, thank you for correcting my memory....
As you said, PCI video cards didn't really exist.
Still, VESA could of defined a standard way for the Operating System to load a pure 32bit "VBE2 BIOS" into the 32 bit address space, for VLB Video Cards. This would totally bypass the existing VGA BIOS limitations (including 64KB limit). Motherboards could have multiple VLB slots, so it should of been possible to have multiple VLB Video Cards running independently at the same time.
An addendum could of been added for the PCI method to expose the pure 32 bit "VBE2 BIOS".
VBE 2.0 should of left standard ISA video cards and 286s in the dust. As you said, they were only selling a small number of 286s and they were low end machines.
If VESA started working on version 2 in 1991, and took 3 years, that is a sad example of design by committee failing miserably.
I would think there was only 7 - 8 companies manufacturing video cards e.g. ATI, nVidia, Matrox, Trident, Cirrus Logic etc. They should of been able to develop a pure 32 bit standard in 1993 - 1994. Very disappointing.
I will have to disagree with your statement OS/2 was "mainly a 16-bit p-mode". According to Wikipedia, OS/2 was mostly 32bit from 1992. If VESA organisation was smart, they would have got IBM to give them funding so the 32bit VBE2 API is implemented on most video cards, saving IBM the effort/expense of writing OS/2 drivers.
This was still the period when IBM was trying to throttle what they (and no one else) still called 'PC clones', and were still convinced that once everyone saw how nifty keen their 32-bit MCA was they would forget all about silly things like compatibility and industry standards. They had their own proprietary VGA extension, XGA, which was integrated hardware on their motherboards just like the original VGA (the 'VGA' everyone else sold was actually something else, reverse engineered - illegally, according to Big Blue's legal staff - to behave like IBM's version but using very different hardware), and were basically refusing to play ball with anyone, for either hardware or software - they only grudgingly gave Microsoft permission to port Windows NT to the PowerPC-based home and business systems they were selling at the time because the port of OS/2 was badly behind schedule, and even then, they only gave Billy Boy the keys to the x86 emulation layer, not the actual hardware.
The fact that in 1998 they were still trying to sell MCA licenses with the same crippling royalty agreements they had introduced over ten years earlier shows that they weren't interested in letting anyone have a piece of the PC pie, despite the fact that their own market share was down to around 5% at the time.
As for games, IBM wanted everyone to forget those existed. Last I checked, they still did.
You have to understand is that IBM never liked the PC monster they had created, and had the managers in 1980 thought it would be anything more than an adjunct to their mainframe business they would have throttled it in the crib - the whole point of entering the home computer market was to to smother companies like Tandy, Apple, Osborne, and Commodore before they could become a threat to the One True Path of Glorious Batch-Processed Socialism, a plan that backfired hilariously. Even as late as 2000 they were still trying to drag consumer PCs back into being glorified terminals, and OS/2 was geared for that purpose. It was until the old guard retired or died off that they admitted to themselves that the days of the centralized computing center had been dead and dust for over a decade.
Ever since Compaq won their infringement suit in 1983, Big Blue - still bitter over how DEC had 'stolen' the academic market from them ten years earlier, and distracted with trying to figure out how to quash the new high-end workstations from start-ups like Symbolics, Sun, and Apollo - had basically stuck their fingers in their ears and were waiting for all those other computer manufacturers to give up and let them have their monopoly back. They were certain that if they made everything proprietary again and pressed the marketing as hard as they could, the status quo ante would be restored. It wasn't really until much later that they woke up from that fantasy.
(Mind you, this applies to the top Midgets, not the engineers. The problem is, IBM is not and never really has been a technology company - their whole business model is driven by marketing, and the fact that they are primarily marketing computers is almost an afterthought. While they actually listen to their engineers and developers today, technical considerations still rate well below sales figures. The days where they could brainwash whole industries into marching in lock-step with their ideal of centralized computing are long gone, but the attitude remains, and the techniques they used back then are still widely imitated by the likes of Microsoft, Oracle, Apple, and shudder Salesforce.)
Rev. First Speaker Schol-R-LEA;2 LCF ELF JAM POEE KoR KCO PPWMTF Ordo OS Project
Lisp programmers tend to seem very odd to outsiders, just like anyone else who has had a religious experience they can't quite explain to others.
As for games, IBM wanted everyone to forget those existed. Last I checked, they still did.
I mean... considering how long it's been since they made a piece of consumer hardware I don't really see how this is something they 'still' would know or care about.
As for games, IBM wanted everyone to forget those existed. Last I checked, they still did.
I mean... considering how long it's been since they made a piece of consumer hardware I don't really see how this is something they 'still' would know or care about.
OK, I'll admit that this was hyperbole. The point was that they always saw computing as Serious Business, and thought using computers to play games was Just Plain Wrong. While they did ask MS for some games when the IBM PC was released, that was mostly because they assumed that most people weren't using home computers for anything real and that they needed at least one game to get the consumer users' attention.
Rev. First Speaker Schol-R-LEA;2 LCF ELF JAM POEE KoR KCO PPWMTF Ordo OS Project
Lisp programmers tend to seem very odd to outsiders, just like anyone else who has had a religious experience they can't quite explain to others.
