Aside from being one of the pioneers in PC computing , this video shows Gary Kildall was likely the only one in the computer industry who understands how to comb their hair without looking silly...
Most balding, middle aged men in the 80s did some kind of comb over. My dad certainly did. I think that's why my generation grew up to shave their heads. They saw how ridiculous their dads and friends looked and we're determined not to do the same thing.
Chet conveniently forgot to mention that other manufacturers could use MCA, but they had to pay IBM steep royalties for that. By this time period, every company was rising their middle finger to IBM. MCA was dead before it arrived.
That card they are showing isn't an EISA card. EISA cards have a deeper AT style commendation, it doesn't have a third connection farther down the card. That was eventually the VLB (or something like it) bus.
@@tarstarkusz Well, the EISA wasn't fully specified back then. So, no surprise they don't have an actual EISA card there to demonstrate what *could* it look like...
@@cmhenator No, the specification was released in late 1988, but perhaps not by the time this episode was shot - hence my "not fully specified". Though you are right that the actual chipsets and machines didn't appear until 1989.
@@geoffreyoltmans4356 sometimes ideas are so bad that they wrap all the way around and become good ideas. Poor judgement overflow. If this isn't one of those cases, it's *really* close.
@@mattj65816 Haha, I'm not sure the "wet spaghetti across the head" look will ever be a good idea. How he got up in the morning, looked at himself in the mirror and decided "yeah, that looks good" and exposed that to the public is beyond all logical comprehension.
@@yellowblanka6058 oh yeah, it's definitely self destructive behavior. One day you're moving a couple noodles around to cover up a thin spot and a couple years later you're slopping the whole box across your head. But he did own all of PBS in that moment. Nobody watching that day looked away from the scene of that accident.
IIRC, companies licensing MCA not only had to pay licenses for new machines using MCA but had to pay IBM for all the PC compatibles they'd created up to that point. I think IBM felt they were in a stronger position than they were actually in.
Seriously, lol, who walks out of the house looking like that? It looks 10x worse than a bald pate and just looks very desperate. I thought Cheifet's comb-over was bad, but that one...wow.
17:26 Chet's insolent answer to this question is part of the reason why OS/2 and MCA generated such considerable resistance to broad adoption. You can't tell customers to just throw their old cards and software away - then ask them to buy a new computer, buy *new* cards, and buy *new* software to replace what they already have. The DOS [penalty] box on OS/2 versions 1.x, along with a bus that would not accommodate existing ISA expansion cards, ended up boosting the credibility and acceptability of competitors Compaq, Dell, AST Research, Zeos, Northgate, Gateway, HP, etc. And then the coup de grace was Microsoft introducing Windows 3.x running on DOS, and just like that... IBM was marginalized in the very personal computer industry standard that they had developed. It's the difference between marketing to consumers versus to businesses. IBM never quite understood the former, while Apple and Commodore never quite understood the latter.
And Tandy Died playing isolation and proprietary... A point Packard Bell fell to when they used proprietary motherboards and custom Riser ISA slots. Both paid for it. then dell came along and did the same, took a major hit and backed out
At least he can intelligently answer the question as many back in 1988 could. They had maturity and knowledge. Today, all you have as youngsters who cannot scratch the surface of the complexity of real technology and just parrot the marketing scripts.
@@trs-80fanclub12 Well, Apple did isolate themselves too and look where they are now. Yes, I know the company was virtually bankrupt at some point, but they are still here and selling overpriced products to hipsters.
23:24 tight spot when you're asked a question that everyone in the room *knows* has a correct answer, yet you're duty bound not to give the correct answer.
It wasn't until protected processing in operating systems like Windows NT, OS/2, *nix etc that the actual hardware interfaces became meaningless. At the time of this program, applications could directly address the hardware, so standards were really important. Now everything is abstracted to drivers. The software really has next-to-no-idea (or need to know) how the hardware talks to each other. Thank god for that.
Winn Rosch of PC Magazine had to walk a tightrope with opinions between MCA and EISA, as print publications back then heavily relied on advertising revenue to stay in business. A verbal faux-pau could result with PC Magazine losing advertising from IBM or the other PC makers of that era.
21:08 Wells American, as they discussed, made the CompuStar, which supported both ISA/EISA and MCA. IBM sued them out of existence a few years later because MCA was proprietary, and Wells American didn't properly license (pay IBM) MCA to use in their computers.
I noticed that as well when he questioned proprietary, and it was available to everyone. From what i read MCA was pretty good, but licensing was expensive.
@@seani1473 The fee was a variable cost from 1 to 5% of gross sales. That is, a percentage of the total gross sales of the company, not just on computer sales but everything else if they sold other stuff, including service contracts. They had to also enter a licensing clause saying the company had to retroactively pay a smaller percentage back on prior PC clone sales, even machines that did not have MCA since they were sold pre-dated to that time period. And that was just the cost of licensing the technology, not for any type of hardware to go with it. Needless to say, nobody found this attractive in the least bit and only confirmed that IBM was the devil.
That was the secret mission of OS/2. They were going to make a play for the hardware lock in for OS/2 and Microsoft destroyed them over it. Yet people still look back at that garbage like it should have beat out Win 95. Go figure.
IBM wasn’t just trying to make it proprietary, they were trying to substantially improve it. A lot of how MCA works came from the busses IBM used in its mini and mainframe systems, and is much more similar to how PCI and now PCIe operate than how the S-100, Apple II, IBM PC and AT busses all operated.
25:59 When Lotus 1-2-3 3.0 finally did ship, it couldn’t fit in 640kiB of RAM, so expanded/extended memory was a must. This left a lot of existing users with older machines out in the cold, so Lotus brought out a new release of 1-2-3 2.xas well, to keep them happy.
I recall that era well. When someone expresses lament with saying: "They don't build them like they used to!" All I have to do is think back to the final decades of the 20th century when it came to PC computing, with the answer of: "Thank goodness!" The introduction of the various BUS(es) in that era made it a pain when it came to PC builds or upgrades
I felt that the battle over the BUS formats was a bit beyond my own capabilities at the time, so I decided to leave that to the manufacturer of the PC that I would buy around nine years later. I believe that I made the right decision when I bought a Pentium MMX Windows 95 computer in 1997. That PC served me well for around 9 to 10 years, when the industry standards state that PCs are supposed to be replaced around 3 to 5 years.
Not really. There’s not only sata (mentioned above) but also USB (used internally directly on traces, not just as ports on the back) and all sorts of legacy busses hanging off usb and pcie. That’s not even including things like memory channels and busses within and between various chips on the motherboard.
I think Gary Kildall was a better person than most of the people he competed against, and that’s why he ultimately lost. I don’t get the sense that he would have been as cutthroat as Bill Gates or Big Blue.
My friend, only idiots and savages get rich. Good people do not seek power and money, that's why they are good. Bad people get rich. Good, smart people are inside universities teaching, doing research, or working at "lower levels" inside companies. Gary was filfhy rich anyways. He had private plane and sports car.
@@PauloConstantino167 No, not all bad people get rich. It's a matter of planning and strategy. Now, GREED is saying "I have at least $1,000,000, but I want to create a monopoly to make more."
Real facts: He lost because he was a bad businessman. Big blue/microsoft didn't screw him. He wanted IBM to pay retail and refused their branding. MS did not. Gary didn't accept compromises needed for DRI to take off. IBM therefore charged more for DRI software because it was more expensive that MS. DRI made a better OS than Microsoft, but failed to proliferate it. DRI and MS, Beta tapes and VHS, Apple and PC: classic examples of superior tech dying in the air of the marketplace realities due to business missteps.
@@yellowblanka6058 No. VHS won the format war for one simple reason - up to 8 hours recording time over BETA and it's 5 hours max with a L-830 tape recording at the slowest speed. People didn't care about video quality, tapes were expensive back then, and everyone wanted the most bang for the $$$
PS/2 machines were too expensive for most people anyway, MCA was basically a nieche. Most people never even touched or saw one of those PS/2 machines, businesses tied to IBM though did. Amazing how IBM basically killed with the better technology itself out of the market and taking OS/2 with it because it could not get their greed under control. If MCAs licensing fees would have been low or an open technology to begin with, things probably would have looked differently now. But PS/2s main existence was to kill off the clone market. I feel sorry for the engineers involved, they really made a kickass hardware which was in some aspects the pinnacle of what was possible at that time and all for nothing because of the greed and short sighedness of the IBM management!
SkuldChan42 my family had a PS/2 and I think the downfall was the lack of vendor support. There weren’t many options for cards, which meant the ones you could buy were expensive. IBM had no choice but to ship EISA after a while because otherwise they’d be selling MCA machines with no upgrade path.
@@Ojisan642 The expense of MCA cards comes from IBM's massive licencing fees for using their patents. Companies who made MCA cards without a licence got sued out of existence. That pretty much doomed it as no one was willing to buy or sell cards with a huge mark up.
Ooft, Gary wasn't a fan of MCA was he? He gave that ibm corporate speak a heavy dose of shade. "I really don't understand the term propriety" Gary: OK. "to me that would mean unavailable-" Gary: "correct."
ISA lasted into the late 90's. For a while it was like PCI.... with 1 slot hanging out ear the edge of the board for a period, just in case. And usually on full-size boards. And at that point EISA was a heavy standard in the server world. You could crack open a mid-90's AlphaServer and find plenty. I have old hardware I keep around as working museum pieces.. one my AlphaServer 1000 4/266 from 95 has 7 EISA slots. Have several EISA cards in it, as it was used as a file server it has 2 add on EISA Adaptec SCSI controllers (in addition to its onboard), an EISA 3Com Ethernet card.
Trying to restore this comb over in 2023. What bus is bro using? Can’t get anything working but remember seeing something about dude using 2nd party follicles to expand bandwidth
20:46 Not all cards must be high-performance. Especially special-purpose cards, a company isn't going to want to replace them when the old ones are plenty fast. With EISA you could use high-performance cards for things that mattered and still use the older cards for everything else. It also keeps the cost of those down because they have a wider market.
In a world where “IBM compatibility” was the holy grail, IBM introduced MicroChannel Architecture (MCA), possibly the least compatible architecture of the day. Add to that, the fact that MCA had to be licensed from IBM and MicroChannel was doomed from the outset.
Least compatible with existing hardware? I guess. But it wasn't for nothing, the platform had already outgrown its bus and needed something new. Instead it stuck to ISA for another 5+ years
Stewart clearly has no idea what anyone is talking about. Someone wrote his questions and he just sits through these segments with ill-timed “uh huh…. Ok…Ok….mmm hmmm… That one seems faster.”
MCA was completely dead by 1993. EISA remained a heavy standard (especially in the server space) well into the late 90's. While both are gone now, inarguable MCA was a massive failure.
@@tekcomputers Indeed, but I'm sure it helped pave the way to the more compatible PCI standard back in the day, as no one wanted to pay royalties to use MCA, but we needed 32bit interfaces to connect to. So maybe MCA wasn't "as useless" as it seems.
4:40 - Worrying about "learning" costs with multitasking, now there is a statement that did not age well. My yongest, (currently 6yrs old), was born with an innate ability to multitask on both computers and mobile devices. Just yesterday I caught him playing Minecraft while simultaneously working on a remote learning assignment live on Google Meet with his teacher. When I tried to tell him School Time is not game time the answer was, "Dad, I can do more than one thing at a time you know..." Well, ok then.
Uh no human can multitask as efficiently as single tasking. It’s backed by decades of research. But hey, you’re taking advice from your children so it’s clear the Apple didn’t fall far from the tree.
@@BlownMacTruck " no human can multitask as efficiently as single tasking." - And that matters why in this case again? The question/point has never been can a person do multiple tasks at peak efficiency simultaneously, since that really isn't required.
Isn't that "multi-bus" computer basically just a ridiculous oversized PC case that you can shove anything in? I suppose in that respect it is versatile.
I wouldn't say pointless. In the case of the x86 or 68k processors they are CISC, and in that they do not process either an instruction or the data involved in a single cycle. Which means that the bus is actually waiting more often than not. Now the bus is used by other devices in the system but in the case of the PC the CPU is king. So a 32 bit processor splits the data in to 2 consecutive 16 bit chunks to communicate over the bus, which will probably have enough time to handle this and still have time left over for the rest of the devices on the bus before the CPU is finished with it's next instruction. Lets not also forget that just because it's a 32bit processor not all instructions are in fact 32 bit with many being 16 bit commands. There is some loss, and thats down to architecture. Take a look at the Atari Falcon a 32 bit 68030 processor on a 16 bit bus compared to other similar 68030 systems it was vary comparable.
Aside from being one of the pioneers in PC computing , this video shows Gary Kildall was likely the only one in the computer industry who understands how to comb their hair without looking silly...
It helped that he wasn't balding do he didn't do one of those ridiculous combovers.
Most balding, middle aged men in the 80s did some kind of comb over. My dad certainly did. I think that's why my generation grew up to shave their heads. They saw how ridiculous their dads and friends looked and we're determined not to do the same thing.
I guess I've never really given it much thought. I've always been one of the rare men who hasn't lost any of my hairline.
@@sayyedal-afghani7896 I haven't lost my hair either but I've given it plenty of thought :)
@@ZagnutBarit definitely hurts to lose your hair, but I'd rather shave mine than look silly lol.
You could ask this guy from AST what he likes to have for breakfast and he'd say "breakfast is actually a natural evolution of the AT bus..."
@Jayce Boden hacking instagram passwords is a natural evolution of the AT bus
20:01 My man Winn here with the best comb-over ever.
Poor guy doesn't even look 30 and he's already bald.
This episode should be called The comb-over war!
23:20 That's a big freaking computer.
I'd love to get my hands on that pile of weird 32 bit cards on Stewart's desk!
Chet conveniently forgot to mention that other manufacturers could use MCA, but they had to pay IBM steep royalties for that. By this time period, every company was rising their middle finger to IBM. MCA was dead before it arrived.
That card they are showing isn't an EISA card. EISA cards have a deeper AT style commendation, it doesn't have a third connection farther down the card. That was eventually the VLB (or something like it) bus.
@@tarstarkusz Well, the EISA wasn't fully specified back then. So, no surprise they don't have an actual EISA card there to demonstrate what *could* it look like...
I was loving how both sides were being dishonest in there own way.
@@GeckonCZ EISA wasn’t specified at all then. It arrived around 1989-90.
@@cmhenator No, the specification was released in late 1988, but perhaps not by the time this episode was shot - hence my "not fully specified". Though you are right that the actual chipsets and machines didn't appear until 1989.
As soon as I saw that combover and paused the video and came to the comments section.
comb-over is never a good idea.
@@geoffreyoltmans4356 sometimes ideas are so bad that they wrap all the way around and become good ideas. Poor judgement overflow. If this isn't one of those cases, it's *really* close.
@@mattj65816 Haha, I'm not sure the "wet spaghetti across the head" look will ever be a good idea. How he got up in the morning, looked at himself in the mirror and decided "yeah, that looks good" and exposed that to the public is beyond all logical comprehension.
@@yellowblanka6058 oh yeah, it's definitely self destructive behavior. One day you're moving a couple noodles around to cover up a thin spot and a couple years later you're slopping the whole box across your head. But he did own all of PBS in that moment. Nobody watching that day looked away from the scene of that accident.
IIRC, companies licensing MCA not only had to pay licenses for new machines using MCA but had to pay IBM for all the PC compatibles they'd created up to that point. I think IBM felt they were in a stronger position than they were actually in.
If that's true, that certainly explains why no one licensed the architecture.
@@McVaio I think a couple companies did, but most took a pass and just expanded the old bus.
Oh boy Winn's combover is the best! Poor dude! He doesn't even look 30!
I cannot believe he went on television with that comb over! LOL you would think a friend or even an enemy would have stopped him before he done that
Seriously, lol, who walks out of the house looking like that? It looks 10x worse than a bald pate and just looks very desperate. I thought Cheifet's comb-over was bad, but that one...wow.
One wonders what Cheifet thought of it, and if it caused him to rethink his own.
@@joqqy8497 no he probably felt good about his own thinking at least it isn't that bad lol
17:26 Chet's insolent answer to this question is part of the reason why OS/2 and MCA generated such considerable resistance to broad adoption. You can't tell customers to just throw their old cards and software away - then ask them to buy a new computer, buy *new* cards, and buy *new* software to replace what they already have. The DOS [penalty] box on OS/2 versions 1.x, along with a bus that would not accommodate existing ISA expansion cards, ended up boosting the credibility and acceptability of competitors Compaq, Dell, AST Research, Zeos, Northgate, Gateway, HP, etc. And then the coup de grace was Microsoft introducing Windows 3.x running on DOS, and just like that... IBM was marginalized in the very personal computer industry standard that they had developed. It's the difference between marketing to consumers versus to businesses. IBM never quite understood the former, while Apple and Commodore never quite understood the latter.
And Tandy Died playing isolation and proprietary... A point Packard Bell fell to when they used proprietary motherboards and custom Riser ISA slots. Both paid for it. then dell came along and did the same, took a major hit and backed out
@@trs-80fanclub12 good points 👍🏽
At least he can intelligently answer the question as many back in 1988 could. They had maturity and knowledge. Today, all you have as youngsters who cannot scratch the surface of the complexity of real technology and just parrot the marketing scripts.
@@harryparker9452 No one is questioning his intellect. He and IBM talked down to PC customers, and they paid for it... _incurably._
@@trs-80fanclub12 Well, Apple did isolate themselves too and look where they are now.
Yes, I know the company was virtually bankrupt at some point, but they are still here and selling overpriced products to hipsters.
So interesting, great snapshot of the time. RIP EISA and RIP Gary Kildall, you are a legend.
Just to think now most is all inside the CPU or in other words SOC.
Incredible
For better or for worse.
23:24 tight spot when you're asked a question that everyone in the room *knows* has a correct answer, yet you're duty bound not to give the correct answer.
It wasn't until protected processing in operating systems like Windows NT, OS/2, *nix etc that the actual hardware interfaces became meaningless. At the time of this program, applications could directly address the hardware, so standards were really important. Now everything is abstracted to drivers. The software really has next-to-no-idea (or need to know) how the hardware talks to each other. Thank god for that.
Poor Stewart. He refuses to accept that he is going bald.
I feel like Winn is in more denial here.
Winn Rosch of PC Magazine had to walk a tightrope with opinions between MCA and EISA, as print publications back then heavily relied on advertising revenue to stay in business. A verbal faux-pau could result with PC Magazine losing advertising from IBM or the other PC makers of that era.
21:08 Wells American, as they discussed, made the CompuStar, which supported both ISA/EISA and MCA. IBM sued them out of existence a few years later because MCA was proprietary, and Wells American didn't properly license (pay IBM) MCA to use in their computers.
8:26 No mention of the cost to license the MCA bus from IBM or that they didn't want IBM holding them by the gonads.
I noticed that as well when he questioned proprietary, and it was available to everyone. From what i read MCA was pretty good, but licensing was expensive.
@@seani1473 The fee was a variable cost from 1 to 5% of gross sales. That is, a percentage of the total gross sales of the company, not just on computer sales but everything else if they sold other stuff, including service contracts. They had to also enter a licensing clause saying the company had to retroactively pay a smaller percentage back on prior PC clone sales, even machines that did not have MCA since they were sold pre-dated to that time period. And that was just the cost of licensing the technology, not for any type of hardware to go with it. Needless to say, nobody found this attractive in the least bit and only confirmed that IBM was the devil.
15:05 That's the biggest card I've ever seen. That thing is twice the height of a normal one. Amazing.
come on Chet. Everybody knew back then that PS2/PS1 and MCA was just IBM attempting to make the PC proprietary and kill off the clones.
I think IBM's plan was to keep control over the development of MCA while at the same time license it to others and make mucho dollares.
Mikey you could tell that Chet knew it too, he sounded so defensive there. Gary and Stewart went easy on him by moving the conversation along.
That was the secret mission of OS/2. They were going to make a play for the hardware lock in for OS/2 and Microsoft destroyed them over it. Yet people still look back at that garbage like it should have beat out Win 95. Go figure.
@@aliren6118 OS/2 *Warp* was objectively better than Win 3.11, and could be argued better than Win 95.
IBM wasn’t just trying to make it proprietary, they were trying to substantially improve it. A lot of how MCA works came from the busses IBM used in its mini and mainframe systems, and is much more similar to how PCI and now PCIe operate than how the S-100, Apple II, IBM PC and AT busses all operated.
25:59 When Lotus 1-2-3 3.0 finally did ship, it couldn’t fit in 640kiB of RAM, so expanded/extended memory was a must. This left a lot of existing users with older machines out in the cold, so Lotus brought out a new release of 1-2-3 2.xas well, to keep them happy.
I recall that era well.
When someone expresses lament with saying: "They don't build them like they used to!" All I have to do is think back to the final decades of the 20th century when it came to PC computing, with the answer of: "Thank goodness!"
The introduction of the various BUS(es) in that era made it a pain when it came to PC builds or upgrades
I felt that the battle over the BUS formats was a bit beyond my own capabilities at the time, so I decided to leave that to the manufacturer of the PC that I would buy around nine years later. I believe that I made the right decision when I bought a Pentium MMX Windows 95 computer in 1997. That PC served me well for around 9 to 10 years, when the industry standards state that PCs are supposed to be replaced around 3 to 5 years.
Now, we only have 1 bus for internal components. PCIe, now at version 4 on products.
nope cause there is sata bus as well
Not really. There’s not only sata (mentioned above) but also USB (used internally directly on traces, not just as ports on the back) and all sorts of legacy busses hanging off usb and pcie. That’s not even including things like memory channels and busses within and between various chips on the motherboard.
For consumer products. In the high-performance world we have NVLink, Quick Path/Ultra Path Interconnect, OpenCAPI, potentially CXL.
I think Gary Kildall was a better person than most of the people he competed against, and that’s why he ultimately lost.
I don’t get the sense that he would have been as cutthroat as Bill Gates or Big Blue.
My friend, only idiots and savages get rich. Good people do not seek power and money, that's why they are good. Bad people get rich. Good, smart people are inside universities teaching, doing research, or working at "lower levels" inside companies. Gary was filfhy rich anyways. He had private plane and sports car.
@@PauloConstantino167 No, not all bad people get rich. It's a matter of planning and strategy. Now, GREED is saying "I have at least $1,000,000, but I want to create a monopoly to make more."
Real facts: He lost because he was a bad businessman. Big blue/microsoft didn't screw him. He wanted IBM to pay retail and refused their branding. MS did not. Gary didn't accept compromises needed for DRI to take off. IBM therefore charged more for DRI software because it was more expensive that MS. DRI made a better OS than Microsoft, but failed to proliferate it. DRI and MS, Beta tapes and VHS, Apple and PC: classic examples of superior tech dying in the air of the marketplace realities due to business missteps.
@@user-lk7cv8vg7r Pretty sure in the case of VHS it had to do with the proliferation of pornography on VHS tapes.
@@yellowblanka6058 No.
VHS won the format war for one simple reason - up to 8 hours recording time over BETA and it's 5 hours max with a L-830 tape recording at the slowest speed. People didn't care about video quality, tapes were expensive back then, and everyone wanted the most bang for the $$$
That dude has the definition of a comb over
Sometimes, I'm really happy my entry into PC computing world happened after the bus war.
But we got a fair share of IT wars after that too ;)
PS/2 machines were too expensive for most people anyway, MCA was basically a nieche.
Most people never even touched or saw one of those PS/2 machines, businesses tied to IBM though did.
Amazing how IBM basically killed with the better technology itself out of the market and taking OS/2 with it because it could not get their greed under control. If MCAs licensing fees would have been low or an open technology to begin with, things probably would have looked differently now. But PS/2s main existence was to kill off the clone market. I feel sorry for the engineers involved, they really made a kickass hardware which was in some aspects the pinnacle of what was possible at that time and all for nothing because of the greed and short sighedness of the IBM management!
Sales person: Just buy a new card.
Drink every time someone says 'bus'
I think the nail in MCA's coffin as it were (asides from PCI) was the fact that eventually IBM shipped EISA machines themselves.
SkuldChan42 my family had a PS/2 and I think the downfall was the lack of vendor support. There weren’t many options for cards, which meant the ones you could buy were expensive. IBM had no choice but to ship EISA after a while because otherwise they’d be selling MCA machines with no upgrade path.
I think MCA was dead well before PCI...
IBM shipped EISA machines, as EISA became a standard especially in the minicomputer server space well into the mid-90's
@@Ojisan642 The expense of MCA cards comes from IBM's massive licencing fees for using their patents. Companies who made MCA cards without a licence got sued out of existence.
That pretty much doomed it as no one was willing to buy or sell cards with a huge mark up.
The year this aired, I ran into a sign in a CompSci lab at my university, "Bus error - exact change required"
Ooft, Gary wasn't a fan of MCA was he? He gave that ibm corporate speak a heavy dose of shade.
"I really don't understand the term propriety"
Gary: OK.
"to me that would mean unavailable-"
Gary: "correct."
ISA was still around when I got my first PC in 94. Didn't last long after that. It was a AST too. I still have it.
ISA is still around for industrial computers, they just call it PC/104 now.
en.wikipedia.org/wiki/PC/104
ISA lasted into the late 90's. For a while it was like PCI.... with 1 slot hanging out ear the edge of the board for a period, just in case. And usually on full-size boards.
And at that point EISA was a heavy standard in the server world. You could crack open a mid-90's AlphaServer and find plenty. I have old hardware I keep around as working museum pieces.. one my AlphaServer 1000 4/266 from 95 has 7 EISA slots. Have several EISA cards in it, as it was used as a file server it has 2 add on EISA Adaptec SCSI controllers (in addition to its onboard), an EISA 3Com Ethernet card.
Trying to restore this comb over in 2023. What bus is bro using? Can’t get anything working but remember seeing something about dude using 2nd party follicles to expand bandwidth
2022: RTX 4090 is huge!
1988: hold my processor module (22:22)
21:46 Welcome to modern computers :D
PC magazine guy has the worst hairstyle ever
His combover is desintegrating by the minute.
I call it Crossfeed.
20:46 Not all cards must be high-performance. Especially special-purpose cards, a company isn't going to want to replace them when the old ones are plenty fast. With EISA you could use high-performance cards for things that mattered and still use the older cards for everything else. It also keeps the cost of those down because they have a wider market.
And none of them survived.
I have seen the future and it's vesa local bus.
24:38 that's one hell of a donald trump comb over🤣🤣🤣
Check out out cool interchangeable bus ($$$) machine. It's got this bus, it's got that bus, and then it's got this proprietary bus in it. Ooops.
Desqview was running on that computer!
There's a bomb on a BUS. Once the BUS hits 33Mhz, the bomb is armed. If it drops below 33Mhz, it blows up! What do you do, hotshot!?
I'd ask what BUS it is
Insert wait states. Next question.
Pull out liquid nitrogen pour it continuously on cpu while running heavy benchmarks continously 🫣🧐
In a world where “IBM compatibility” was the holy grail, IBM introduced MicroChannel Architecture (MCA), possibly the least compatible architecture of the day. Add to that, the fact that MCA had to be licensed from IBM and MicroChannel was doomed from the outset.
Least compatible with existing hardware? I guess. But it wasn't for nothing, the platform had already outgrown its bus and needed something new. Instead it stuck to ISA for another 5+ years
Stewart clearly has no idea what anyone is talking about. Someone wrote his questions and he just sits through these segments with ill-timed “uh huh…. Ok…Ok….mmm hmmm… That one seems faster.”
And then MCA failed anyway, even EISA...
Geee Thanks PCI!!
MCA was completely dead by 1993. EISA remained a heavy standard (especially in the server space) well into the late 90's. While both are gone now, inarguable MCA was a massive failure.
@@tekcomputers Indeed, but I'm sure it helped pave the way to the more compatible PCI standard back in the day, as no one wanted to pay royalties to use MCA, but we needed 32bit interfaces to connect to. So maybe MCA wasn't "as useless" as it seems.
NuBus was not really a participant in the "bus wars" because it did not compete in or affect the PC world.... Only Apple computers.
USB-C 2023
The need for coprocessors
4:40 - Worrying about "learning" costs with multitasking, now there is a statement that did not age well. My yongest, (currently 6yrs old), was born with an innate ability to multitask on both computers and mobile devices. Just yesterday I caught him playing Minecraft while simultaneously working on a remote learning assignment live on Google Meet with his teacher. When I tried to tell him School Time is not game time the answer was, "Dad, I can do more than one thing at a time you know..."
Well, ok then.
Uh no human can multitask as efficiently as single tasking. It’s backed by decades of research. But hey, you’re taking advice from your children so it’s clear the Apple didn’t fall far from the tree.
@@BlownMacTruck " no human can multitask as efficiently as single tasking." - And that matters why in this case again?
The question/point has never been can a person do multiple tasks at peak efficiency simultaneously, since that really isn't required.
@@looneyburgmusic You may or may not agree with me but your original statement is unclear and makes it sound like you believe him over reality.
@@BlownMacTruck I don't think you understand what "multitask" means.
It doesn't require 100% efficiency between tasks.
@@looneyburgmusic Sure. And if you think not giving 100% attention to education is important, that explains pretty much everything about your replies.
Isn't that "multi-bus" computer basically just a ridiculous oversized PC case that you can shove anything in? I suppose in that respect it is versatile.
wow!
check out error at 0:11
Greasy
Yeah it's pointless to have a 32bit proccesor in your pc if the bit bus is only 16bit,like the 68000,wich is even found in the sega genesis.
Unless the bus runs at double the clock and you have a 32 bit to 16 bit converter between. But that makes the entire architecture more complicated.
Addressing is still an issue however.
I wouldn't say pointless. In the case of the x86 or 68k processors they are CISC, and in that they do not process either an instruction or the data involved in a single cycle. Which means that the bus is actually waiting more often than not. Now the bus is used by other devices in the system but in the case of the PC the CPU is king. So a 32 bit processor splits the data in to 2 consecutive 16 bit chunks to communicate over the bus, which will probably have enough time to handle this and still have time left over for the rest of the devices on the bus before the CPU is finished with it's next instruction. Lets not also forget that just because it's a 32bit processor not all instructions are in fact 32 bit with many being 16 bit commands. There is some loss, and thats down to architecture. Take a look at the Atari Falcon a 32 bit 68030 processor on a 16 bit bus compared to other similar 68030 systems it was vary comparable.
and that children is how hyperthreading was born..... :) .
It has nothing to do with hyperthreading. What you see there is early busmastering and asymmetric multiprocessing...