Відео

Man your vids are really fucking good. Your contemporarys imo are dwarfed by the engagement i have when watching your vids, including ones who are much more well known.

Lol, "All female spacewalk". WTF does that have to with innovation or progress? just other people doing what others have already done.

@UA-cam and Alphabet: A nearly 49 minute 1st of double ad at nearly 8 minutes into a content creator's nearly 59 minute video?!?!?!? GET £'d!!!!

The video is full of errors. I.e., author does not recognize VLIW and RISC. Itanium was VLIW, not RISC!!!

In the mid-80s I was able to use the University of London's Cray (not sure which model) to process data for a research project. I was at the University of Kent in Canterbury (UK). The data chugged up JANET (the Joint Academic Network - this was before the internet); that took 20 minutes. The processing in the Cray took seconds. The results took 20 minutes to chug back down JANET to UKC. How times have changed!

The heavy promotion of RISC architecture is misleading, CISC since the Intel Core has been heavily influenced by RISC with its micro ops, 😅 it’s the chip architecture that matters not the instruction set,

I think that where technology is going to start improving is in making new, specialized, discrete chips. We have his a wall for putting more transistors on the same chunk of silicon, and the only way forward is to start putting in more chunks of silicon that can have their own cooling. Look at how Bitcoin mining has shifted to GPU's and then to ASICs, AI has shifted to... GPUs and the push is now toward NPU's. I know that pointing to crypto and AI is a bit inflammatory, but these are both extremely computationally intensive systems and they both have shifted to more specialized hardware. I think that's where the trend will move to. Ironically, after decades of desktop PC's having fewer expansion card for discrete processes, now, IMO, I think that trend is going to start reversing. We've already been adding new specialized cores to existing architectures to try to give them an edge, and though we probably haven't hit the hard limit for how many new specialized cores can be crammed on the same chip it's probably likely that we're approaching a _practical_ limit, at least for consumer hardware. I can't predict what the new hot PCIe card is going to be the one that takes off, but most people have a few slots in their desktop computer corroding from disuse and technology has a way of tiptoeing laterally around these kinds of limits.

Used to be a data center tech for Dell, I would go into Cray in Chippewa and work on Dell storage. Walking around past the Cray racks was awesome!

I remember reading about Seymour Crays death, reading it in EETimes. A Jeep rollover, he died a few days later in hospital.

Livermore, CA is still a cool place though.

The Encarta cameo brought back some flash back memories. What a time to be alive. I am happy to see what the world looked like before mainstream Personal Computers. I think computers have really revolutionized humanity in many different ways, for better and for worse.

People who bought Sandy Bridge unironically paid less over time because of the significantly lower power draw lol

5:15 The 2600K idles at around 5W stock. Comparing the 8350's 99W idle power with Sandy Bridge's 95W TDP is purely misleading. Also, they got the math wrong. Whoever wrote that has no idea what they're talking about.

4:06 It most certainly didn't have a "limited amount" of cache. It had horrible cache latency but 2M of L2$ per core wasn't bad for its time.

I find it interesting that tech businesses have seldom made financial sense. Makes 1 Billion in sales, has to take minimum wage to save a project. Mind inflation. where was the money going to that was more urgent than developing the company's flagship? This story was masterfully done. I will watch it again for sure

Interesting, but i had to increase playback speed to 1.25 as the pauses in exposition are annoying.

At the 1 hour mark, you note that the i7-920 didn't often show a performance gain. Having been an early adopter of it, moving from the Q6600, I have to strongly disagree with that. The difference was obvious and noticeable the minute you did more than treat it as a "faster older PC". The multi-tasking performance was astounding and the overall responsiveness was a step up. I ended up buying 2 of them in 2009 to replace both of my Core2Quads at my office, the difference was noticeable enough.

And now with LLM algorithms, new experiments with electronic implants, and corporations vying for end product control, we are moving along towards a cyberpunk dystopia.

Man, quit hatin on x86-64. I love the straight backwards compatibility.

The Color Computer was built around 6809E

ps4/xboxone and zen literally saved AMD

Itanium wasn't RICS, it was EPIC, which is something different that RISC or CISC

Wasn't superscalar introduced in Pentium Pro, not the original pentium?

2:32 where the heck is the rest of the fricken computer?

A bit late to the party, but here are some pedantic notes: * AIRESEARCH should be pronounced "air research" with the two Rs kind of running into each other, BUSICOM is a portmanteau of "business" and "computer" so it should be pronounced "busy-com", MITS is generally pronounced "mitts", and TAITO is "tie-toe". * It's splitting hairs slightly, but the Atari VCS used a 6507 not a 6502. * Apple were not the last manufacturer of 680x0 computers; even if you exclude the TecToy-manufactured Sega Genesis that was still being produced until 2023 there were oddities like the DraCo, Emerson produced a bunch of 68040 and 68060 VME bus SBCs way into the 2000s, and they're still heavily used in industrial control systems and telecoms equipment. * Not sure by what metric the BBC was "one of the best selling home computers of the period". C64 12-17 million, MSX 9-4 million, Apple II 6 million, ZX Spectrum 5 million, Atari 8-bit 4 million, CPC 464 2 million, BBC 1.5 million. Essentially, apart from abject failures like the Dragon 32 and Oric Atmos it was one of the *worst* selling micros! Source: I'm old.

So in short; when mad at management, leave and found your own company.

That was a very informative series; thank you.

Good vid. You should have way more subscribers. Might I humbly suggest you change your channel name to something a little more unique? It reads like one of the many, many AI channels atm.

Although these super computer companies produced a lot of inovative wild tech, now your pocket computer (mobile phone) has more processing power and memory than the last generation of super computers from the 80s. Super computing now is now about parallelisation because we've pretty much reached the physical limit of what can be acheived with semi conductors.

This is an EXCELLENT video. That was really "the time to be in the computer industry," that's for sure. When it was still about the clever technology more than about manufacturing economies of scale. Modern supercomputers are... boring by comparison, frankly. They're amazing, but just not inspiring in the same way.

I played pong. I used a Commadore Pet in 7th grade math class and had an IBM 8088 Excellent presentation. I enjoyed this video very much.

I wath your video from India

Found this through a news letter to old CDC graduates. Brought back memories for me of many of the early players. A wonderful introduction of the importance of the labs and Sid Fernbach to early computing. LRL Livermore (Sid Fernbach) bought the first CDC 3600, the first CDC 6600, the first CDC 7600 and the first CDC Star 100. Seymour Cray was unique in the computer industry: First Transistor computer, first RISC, first multi functional unit design, first instruction stack, then pipeline, first short vector approach, and always the most aggressive packaging and his ability to completely stop one design approach and go to a new one. Congratulations on a well done journey from concepts in the 50's to the last super computers of the early 90's. I enjoyed 29 years on that journey until 1989.

Really good. About the only quibble is that the 37 cent transistors Cray used in 1960 would have been in TO-5 cases, like a pencil eraser. The flat epoxy transistors you show are from 1967 or so.

What program did you use for these graphics?

Well done, an amazingly researched and engaging presentation. Thank you.

That story is Cray

Does Cray make any other computers and why did other manufacturers not make S computers ? In the mid-90's I read about S computer maker by Hitachi that was used to track the flow of water through a coffee filter !.

in the 70ies and 80ies I worked for a machine shop in Hopkins, Mn making spindles for the disc drives for those computers. These things were massive as a 5 mg memory disc was the size of a record album. we made them by the thousands for control data and MPI.

I got to stop in the Air&Space Smithsonian in DC not too long ago. They have a CRAY-1 on display. Pretty bad ass, lots of covers are removed so that you can see wiring, circuit boards, and the power supplies.

So cray invented the computer. By putting invention over capitalism. While Singapore was responsible for his success. Without the chips invented in Singapore; cray's computer would be a pipe dream.

ARM is the future

Feels like this video needs an addendum now that Apple have released a few versions of their Arm desktop/laptop processors? Is Arm still the future or is RISC-V a better long term bet? Also would’ve been good if you’d covered a few of the problems various manufacturers had, such as the Pentium floating point bug and spectre/meltdown!

Great trip, congrats! A small addition: P3 Tualatin (next Pentium M), better than P4 and used for Intel Core instead of netburst.

Very nice job on this video.

"the first bomb was developed with mechanical computers" at the time of Los Almos, they had punch card processing machines. These IBM machines were capable of running through a deck of computer cards and perform a single math operation on them, programmed by a pluggable electronics board in the machine. One of the scientists at Los Almos figured out how to perform complex calculations on large data sets by running the decks multiple times and changing the electronics board and running it again, over and over. Thus, before a true computer was available, complex calculations were possible.

How odd that you didn't mention either ATI or AMD. Anti-Aliasing and Anisotropic Filtering also deserved a mention. They were god-send for gorgeous 3D gaming, in an era of horrible jaggedness.