Was an interesting experience, though all I remember now is how it required careful exploitation of the vector units in the SPEs to get any decent performance out of it, and how annoying it was to synchronize between the SPEs and the PPE.

For each assignment the prof would include a benchmark, so we could compare the performance in class. Was a huge difference between the students that spent time optimizing and those which did a basic implementation.

Here is the OS course: https://www.uio.no/studier/emner/matnat/ifi/IN4000/index-eng...

When I talk to people outside my home country it feels most don't do these kind of things at the bachelor level. But, like you, people who go on a masters then might do one of these things and go much more in-depth than I did.

That's "full stack" development :)

Also wrote a BBC Micro emulator, which was great fun.

Since I designed the instruction set I made some very quirky design decisions to make the test programs I was writing easier, instructions were 16bit but only 6 bits were used for the instruction so I had a lot of single-word instructions that did a lot of heavy lifting for my programs.

Did you really do that at the bachelor level? I feel like my uni was a bit abnormal on how many hardware classes and projects we had at a bachelor level for computer science.

I heard about the OS one, sounded super cool and also a lot of work from what I gathered. Heard about it too late and required a buddy. Oh well, can't have it all.

[1]: faculty of math and natural sciences

I can't remember exactly how many, but once you included lectures, scheduled labs and tutorials, was not that far from 9-5 (minus the standard university-wide Wednesday afternoon sport slot).

The friends would go quiet and say "oh, we have 12".

And that didn't even incude the "non-contact" hours in the labs grinding the coursework until security shooed us out late at night so they could lock the building. Good times (in retrospect, especially).

Plus in one term there was some miscommunication between the professors and they set us one experiment every week instead of every two weeks, I remember literally working like 12 hour days on the weekends to get all the data analysis and reports done.

I say used to, but still do. Even 10 years later me and my girlfriend (that studied at UiO) still tease each other about who is the smartest or had the hardest courses, or she mocks my ring, heh.

Part of your grade depended on the performance of your code, with a bonus for the fastest submission. It led to a very nice semi-competitive semi-collaborative atmosphere.

I did a little bit, but it ended up being a super PITA because I didn't want to put my PS3 on my desk and give up playing games on it on my TV.

Because I wanted to keep it with my TV I had tried to use it with Bluetooth KB/Mouse so I could use it from the couch.

And well typical Linux you always end up spending time on the most mundane thing. Bluetooth on the PS3 in Linux at that time didn't seem to work right.

That to me is like peak Linux right there. At some point I had said to myself "I wanted to play with the cell and instead I'm working on Bluetooth" and I lost interest cause of course I had real work too. Then eventually Sony disabled the whole thing.

This makes any third party trying to support them face nightmare after nightmare of unknowns, and a complex device like a computer of any sort is nearly always a complex web of interlinked mini-computers each doing their own thing. Like the damned bluetooth chip; which worse probably has government regulated transmission compliance stuff baked into it's software rather than hardware. That adds a whole other nightmare you might see with various firmware blobs and driver support, most often seen with other WiFi adjacent projects such as OpenWRT (and supported hardware).

It would be very nice if some law enforced free 'edge technical specification' for all products to anyone who buys them or a product containing them. That would help a level and fair market for competitors as well as right and ability to repair for consumers / device owners. That sounds a lot like effects desired by most (many?) Libre software licenses, like the GPL (and LGPL), and Creative Commons.

They did work great for mining ethereum though.

I think a later gen locked it right down.

However PS3 clusters did find niche success as a super computer. There are a handful of notable examples. The fastest one ever built was by the US Airforce, consisted of 1,760 PS3s, and was the 33rd fastest super computer at the time. It was used for satellite image processing.

The article references it.

As one goes back in console history, the performance to achieve contemporarily visually-stunning results required exotic architectures + hand-tuned code.

Over time, state of the art PC + GPU hardware improved, to the point there wasn't a functional difference vs exotic architectures. Additionally, the cost of developing leading-edge custom architectures exploded.

Microsoft really pushed this with the original "PC in a box" XBox, but you saw the console OEMs outsource increasing parts of their chip solutions after that.

Why keep / buy entire chip design teams, when AMD will do it for you?

Especially if the performance difference is negligible.

The PS4 was sold at break even price and couldn't run unsigned code anyway.

It was retroactively removed when people started getting close to enabling full GPU access. This was previously crippled under OtherOS to prevent games developed without the PS3 SDK.

It was retroactively removed when Sony started worrying about "security" concerns - aka people running unlicensed commercial games, which is usually the killer app for any firmware that supports customization and programming.

It was also advertised on the original PS3 box.

Ultimately there was a lawsuit and some PS3 owners got $10 apiece.

Instead of following the Yaroze footsteps, and being a way for a new indies generation to learn how to create console games, the majority used PS2Linux for classical UNIX stuff and play games on emulators.

Note that PS2Linux had GPU access, via PSGL and an easier high level API for the Emotion Engine.

So, no GPU access, no emulators.

https://phys.org/news/2010-12-air-playstation-3s-supercomput...

>>About the 33rd largest supercomputer in the world right now is the US Air Force Research Laboratory's (AFRL) newest system, which has a core made of 1,760 Sony PlayStation 3 (PS3) consoles. In addition to its large capacity, the so-called "Condor Cluster" is capable of performing 500 trillion floating point operations per second (TFLOPS), making it the fastest interactive computer in the entire US Defense Department.

>>It will be used by Air Force centers across the country for tasks such as radar enhancement, pattern recognition, satellite imagery processing, and artificial intelligence research.

I got the impression that Sony encouraged this use of the console early on. They were probably aiming to establish themselves in the scientific computing niche. We also had nvidia-sponsored labs where they taught CUDA (or at least tried to - it was pretty difficult to do without support and most people took the class for access to computer labs equipped with amazing gaming machines). We all know nvidia won that war.

Made designing/porting games a massive headache though because nothing else was like it.

I think neither Sony nor anyone else expected the consumer CPU and GPU industry to evolve so quickly in that time, otherwise they would not have spent hundreds of millions developing anew architecture from scratch. Intel also made the same mistake back then when they tried to reinvent the industry with the new Itanium CPU architecture. Hindsight 20/20.

Those years were indeed wild for tech progress. You bought a new CPU, suddenly two years later you have 64 bit CPUs as the new thing. You then bought a 64 bit CPU to be like the cool kids and suddenly dual core CPUs are all the rage.

Now it's impossible to beat commodity SoCs on that front.

Then they tried the 2 Cells and software rendering approach, to try and keep it as an "in house" solution.

Finally they went to Nvidia and we got the final Cell + RSX solution.

I can't recall if if had all the details regarding adding a GPU to the PS3, but I remember it had stories about the awkwardness of having the IBM Cell team also working on Microsoft's Xenon CPU for the 360, which IIRC used modified Cell PPE cores.

I wonder how many PS3 games ignored the Cell processor and just developed for the CPU and GPU?

It would have been a library.

> I wonder how many PS3 games ignored the Cell processor and just developed for the CPU and GPU?

Essentially none.

Gabe Newell famously sounded off about how useless it was and wouldn't transfer to any other hardware, but he ended up being wrong. The architectural model forced on you by the SPEs is the model that the industry (not just games but any compute intensive work) has embraced for the heavy multicore but coherent system world.

Rather than the "thread per vague activity" model, we've embraced a one thread per core with a work stealing scheduler walking a DAG of chunked compute tasks.

For a modern view of this, this is exactly how Rust's rayon library works.

The whole thing reminds me of a lot of the complaints about the N64, a lot of which ended up being the fact that DRAM was no longer a single cycle away like it was for the SNES. Yes cache conscious code is more difficult to write, but that memory hierarchy architecture was more a harbinger of the new world rather than a one off weirdness of a single console.

To be fair the PS3 port of the Orange Box runs like absolute crap in many instances. Although I believe it was developed by EA, not Valve.

Maybe - the Orange Box at least was outsourced to EA and the performance was reportedly not great. I think Portal 2 was done in-house though and I don't remember anyone complaining about it.

In the design phase they were expecting almost 5ghz at launch, and to hit around 10ghz at the end of the product lifecycle. When the process guys came back later to let the architects know the reality we've sort of now all internalized (high end consumer electronics can expect ~3ghz with good cooling solutions) a lot of simple but fast pipeline designs stopped making sense. These are the exact same pressures that killed off the Netburst architecture too, making intel go back to a PIII design and make it smarter rather than clock markedly faster.

But, even that wasn’t as bad as it was made out to be. People rightly moan about the awkwardness of asynchronously moving data between main RAM and the SPE memory. What they don’t often mention is that the latency of those moves was about 500 cycles —the same latency as a cache miss on the PPE CPUs!

So, which was worse: implicitly waiting 500 cycles all over the place? Or, explicitly scheduling 500 cycle waits at specific points? Unsurprisingly, everyone preferred the first option :P

PS3 Cell could do about 154 GFLOPs across it's 7 cores (6 SPUs and 1 PPU, excluding the GPU), while Xeons were in the low double digits. What really was awkward about the Cell architecture was the DSP-style nature of it, which works great for something like shaders or audio processing with a clear pipeline, but not for general purpose computation. It was really a successor to the PS2 in this regard, where data flowed between specialized and memory constrained processors using a DMA.

> Sony nor anyone else expected the consumer CPU and GPU industry to evolve so quickly in that time

PS3 used an NVidia GPU. Though there was some clunkiness around sharing memory between the Cell and GPU compared to the UMA on PS4, but not especially so compared to a graphics card on a desktop machine.

That was the trickiest part - getting the data in and out of the thing. You had 6 SPE's available to you, 2 were reserved by the OS, and keeping them all filled was a challenge because it required nearly optimal usage of the DMA engine. Memory access was slow, something over 1000 cycles from issuing the DMA until data started coming in.

Back then, C++ was all the rage and people did their various C++ patterns, but due to the instruction size being so limited, we just hand-wrote some code to run on the SPU's which didn't match the rest of the engine, so it ended up gluing together two dissimilar codebases.

I both miss the cleverness required back then, but also don't miss the complexity. Things are so much simpler now that game consoles are basically PC's with PC-style dev tools. Also, as much as I complain about the PS3, at least it wasn't the PS2.

Actually, there is such a scattered supply chain for semiconductor chips, that any war that disrupts international trade in a material way, would bring that entire industry to its knees. So many critical suppliers in the semiconductor supply chain are single companies, or single country oligopolies.

In Europe, PS3 was widely popular especially in conjunction with Call of Duty because Xbox never really gained ground in EU, at least that was my impression growing up as an EU gamer. Wii was also popular af at the time but PS3 and Wii buyers were different audience, Wii was more casual type of gaming and it was suitable for family gaming while PS3 was for more serious gamers who played more hardcorish titles like COD, FIFA and PES and ofc there was and is PC master race for real hardcore gamers.

It seems an incorrectly strong wording by the interviewer.

And people seem to sleep on how powerful the xbox360 was - the CPU was a fair bit smaller than the cell, true, (~150m transistors vs 250m in the cell) not not really the multiples some articles seem to imply.

And the GPU in the xbox was arguably a fair bit more advanced - being a little larger (~300m vs ~335m transistors), and being a unified shader architecture that came to dominate GPUs from then on. And the more flexible architecture enabled more GPGPU-like tasks that the cell SPUs might have handled on the PS3. And these early GPGPU models were also a bitch to program for - like the cell SPUs were.

So it's not like one was simple and easy, the other crazy complex and advanced, but more one had the fancy experimental stuff on the CPU/Cell die, the other on the GPU die. Just the techniques used on the GPU continued to be developed and supplanted much of the Cell-style tech.

Interestingly, the Xbox 360 (Xenon) CPU cores were based on the Power Processing Element (not the Synergistic Processing Elements) from the Cell processor.

From [0]:

"When the companies entered into their partnership in 2001, Sony, Toshiba and IBM committed themselves to spending $400 million over five years to design the Cell, not counting the millions of dollars it would take to build two production facilities for making the chip itself. IBM provided the bulk of the manpower, with the design team headquartered at its Austin, Texas, offices. Sony and Toshiba sent teams of engineers to Austin to live and work with their partners in an effort to have the Cell ready for the Playstation 3's target launch, Christmas 2005.

But a funny thing happened along the way: A new "partner" entered the picture. In late 2002, Microsoft approached IBM about making the chip for Microsoft's rival game console, the (as yet unnamed) Xbox 360. In 2003, IBM's Adam Bennett showed Microsoft specs for the still-in-development Cell core. Microsoft was interested and contracted with IBM for their own chip, to be built around the core that IBM was still building with Sony.

All three of the original partners had agreed that IBM would eventually sell the Cell to other clients. But it does not seem to have occurred to Sony that IBM would sell key parts of the Cell before it was complete and to Sony's primary video-game console competitor. The result was that Sony's R&D money was spent creating a component for Microsoft to use against it."

( From the WSJ )

[0] https://web.archive.org/web/20150129121858/http://www.wsj.co...

If they just asked for a 3-core SMP version of a PowerPC and the Cell project never existed, I suggest it's likely it would still look very similar to what was released in the xbox. VMX already existed, POWER5 already had SMT.

I don't see there being a huge amount of new technology made for the Cell's PPE for the Xbox project to import in the first place.

I'm pretty sure the PPE was an entirely new core design. It used the PowerPC ISA and had features similar to earlier designs, but it was new. From [0]:

"although the central processor is based on IBM's PowerPC architecture, it's a new design, not an existing PowerPC core. Cell's central processor is similar to the current PowerPC 970 chip, although it's not an exact match. "

> If they just asked for a 3-core SMP version of a PowerPC and the Cell project never existed, I suggest it's likely it would still look very similar to what was released in the xbox. VMX already existed, POWER5 already had SMT.

I don't think at the time IBM had anything else that Microsoft could've used. It's PowerPC 7xx wouldn't have been an improvement (or not enough of one) over the Pentium 3 in the original X-Box. The PowerPC 970 was hot / power hungry - the dual-core models that Apple used requiring liquid cooling and weren't out until 2005.

The POWER5 with SMT was in development, but it was designed for IBM's servers and existing customer workloads. I don't know if it could've modified it to add another core, plus the VMX128 stuff that Microsoft needed, if IBM would've been willing to do that, or if that chip would've been too expensive for a console.

[0] https://www.embedded.com/a-glimpse-inside-the-cell-processor...

As far as I understood, Halo was the killer app for Xbox but I suppose it was the killer app only in US, because like I said, from my EU gamer's point of view, Call of Duty was all the rage first on PC and then afterwards on PS plus FIFA and PES were huge on PS as well.

Considering the absolute dominance of PSX and PS2, it was definitely not what they were hoping for. Not quite as dramatic as Nintendo with the market loss from SNES to the N64, but I’m sure Sony had a lot of similar conversations. You also have to consider that Xbox was the distant third place console in the previous generation, so barely losing to Sony at the last minute was definitely considered a huge success. Especially after the disaster that was the RROD. If that hadn’t happened Microsoft likely would have beaten Sony’s sales.

When you buy a gaming pc you can play all prior generations of pc games.

* Yeah, Yeah, I know this is not really all that true/over simplified. But you have to understand I am the type who enjoyed the fight, almost more than the game. the fight? you know first you fight the pc a bit and if you win you get to play the game. Really, if you just want to sit down and game get a console.

https://en.m.wikipedia.org/wiki/List_of_backward-compatible_...

I was amazed that I am actually able to play PS3 games on my (older, not at all top of the line) laptop thanks to the brilliant RPCS3 emulator.

A PS5 is way more powerful than my laptop, so in theory SONY could just make all those PS3 games available through emulation.

The Wii GPU is very much the GameCube one with some tweaks and faster clock speeds.

The 360 one is absolutely an ATI custom Radeon part.

At the time, UT was building TRIPS[0], a processor capable of > 1 teraflop / sec.

In the middle of his talk (with several TRIPS members present), he made an off-hand remark "I think I could build a petaflop processer...without too much trouble,"* which caused a small stir in the audience!

[0] https://www.cs.utexas.edu/users/cart/trips/

* I might be misremembering the exact quote, but this was the gist of it.

I was not running codes that would utilize even a small fraction of roadrunner so my projects quickly moved to the GPU based test beds for the next gen computer so I didn't get to see these research coalesce. My understanding from people who stayed on roadrunner is that for the most part people didn't adopt too much fanciness for roadrunner as the rewrites were too deep and GPUs were on the horizon. There was a lot of vision about making generic heterogeneous compute frameworks but too my knowledge they didn't pan out and just writing against CUDA, OpenMP, and MPI won the day.

The 6502 was descended from the 6800 - one of a few ur-microprocessors - and somewhat clumsily evolved into the 65816 before its brilliant but highly specialized design (hard to target for HLLs and which fundamentally assumed memory could be accessed roughly as fast as the CPU was running) couldn't keep up with the times. As far as I know that's pretty much it for its branch of the CPU family tree.

Meanwhile - and I admittedly know less about this - as I understand it the PowerPC is ultimately descended from the IBM 801, which was a high-end workstation and server chip. There might've been some influence from Motorola, and Motorola made the 6800 - the 6502's father - so in that sense the 6502 and the PowerPC might be very distant cousins. Still, that's pretty tenuous.

My understanding is that the original ARM team was directly inspired by the 6502 - so if its legacy lives on anywhere, maybe it's there.

There is still so much value in this ancient machine.

I recall various news stories about governments buying up PS3s because they could be used for cheap and effective computation at scale.

Lets start from the end, game consoles these days are basically optimized PCs. They are effectively nothing like the first, second, or even third generation consoles that had fairly unique hardware, desirable exclusives and so on.

Cell may have been difficult to program well for, but it speaks volumes about the current state of consoles that devs worry about being able to be lazy with ports. The idiot consumer will buy whatever anyway. The side effect of the first generations' relatively unique hardware profile is that games actually had to be optimized and tailored to the hardware. No lazy ports ( I still get pissy sometimes over ps3 dragon age port ). Exclusives had to be good or at least showcase what the console could do.

So is that what makes Cell infamous? The thing made news when could be run in a cluster[1].

I have a different take. Cell is the remnant of the consoles, when they were still worth one's time.

[1]https://en.wikipedia.org/wiki/PlayStation_3_cluster

It’s not about being lazy with ports. It’s that the asset budgets are so large that making games is way too expensive. For any normal company you basically _have_ to launch on multiple systems if you want a financial payback.

And every time graphics get better the problem gets worse. More models, more textures, more detail, more everything. $$$

Unless you have a monster hit (Zelda, Last of Us) you may be unable to make your money back on one machine. Even if you could, why not just make more money by adding in the other machines?

There’s a reason pretty much only 1st party studios do exclusives.

Totally. Thomas Was Alone is one of the best games I’ve ever played and it’s just a bunch of boxes. It’s basically high resolution Atari.

Great graphics are nice and can certainly help, but the be all end all race to always look the best is not working well.

Some kind of reckoning has got to be coming. We’re getting to the point where it’s like 5+ years between sequels to big titles just cause that’s how long it takes. Much like mega budget Hollywood movies no one‘s willing to take a risk when it costs that much.

You have your actory voice intro the game, set the stage of the plot, show some brief segments of critical moments in cut scenes, and then occasionally intersperse a half second or so of actual gameplay. No idea why this gets some people fired up, but it seems to work reliably enough. I suppose it's just targeting a demographic that isn't you nor I!

Indies don’t have this problem. But no big companies seem to be able to not spend a ton on graphics.

Today it’s just expected many games will have voice acting. It’s rare that there is something terribly special about it.

There are only two I can think of.

First Mass Effect having two voice actors for Shepard stood out. Usually you’d get only one or a silent protagonist.

The second is Disco Elysium: The Final Cut because they voiced _every_ line in the game no matter how small. And that game had an insane number of lines.

We need to get back to AA and maybe even A games. Cheaper, faster to make, still fun.

The bigger issue with current consoles imo, specifically the PS5 and XB, is their input homogenization and stagnation. Consoles have historically been forward-looking via their controllers to enable new ways to play games:

[NOT including optional accessories, only out of the box abilities]

- N64: Analog controls, 4 controller ports, controller expansion ports

- PS1: Dual analog controls

- PS2: Pressure-sensitive buttons, built-in rumble

- Dreamcast: ...

- Xbox: ...

- Gamecube: ...

- Wii: Motion controls

- PS3: ...

- Wii U: Gamepad screen

- 360: ...

- PS4: [Correction!] Touchpad

- Switch: Detachable and sharable controllers

- XB1: Kinnect

- PS5: ...

- XB1X: ...

I'm definitely not implying that every new input type was successful (See Wii U, kinnect), but consoles up until about PS3/360 had something new to offer regarding ways to experience games. This has all but stopped, with Nintendo being the only one to consistently introduce new ways to play that cannot easily be replicated on a PC. Plugging a PS5/Xbox controller into a PC yields the same or better experience as buying a dedicated console these days, so why buy the console at all?

Putting console exclusives aside, I can build a PC that does everything I need it to do but play some new games + a console for less than it costs to build & maintain a "gaming rig". Also, my PS4 is still working after ~10yrs. I would have to upgrade my PC more frequently than that to keep up with new games.

You're also forgetting the two-point capacitive touch pad on the PS4 controller. It's OK, almost all the video game developers did too.

It’s easier. I don’t have to fiddle with drivers. Or settings. Or choosing which card to upgrade to when. Or debugging mismatches. I don’t have to have the OS + Steam/Epic/whatever + games.

The games are written to what I have and I don’t need to worry about it. The OS is completely built around playing games and making it easy for me to acquire them and I don’t have to think about any of it.

I had the years of building gaming PCs and messing with all of that and doing all the research. And I had fun. And at this point in my life I just don’t wanna deal with it I just wanna play a nice game.

Before haptics became standard, game designers could not base any significant gameplay experiences on the presence of haptics. Early games like Ocarina of Time had neat optional side quests enabled by haptics, but in no way required because they couldn't rely on all customers buying an optional accessory.

The presence of the Switch's HD Rumble (And novel controllers in general) enabled new types of play on games like 1-2-Switch, and Nintendo's Labo series that use the rumble to control cardboard toys' movement.

The same cannot be said for the PS5's more refined rumble or triggers.

I do like the way they feel and I think they add to immersion. But honestly I’m not sure you can really enable new kinds of play with any kind of rumble past a basic hot/cold system of how strong it’s running.

While the switch is capable of something more is it really used anywhere besides those two games? A big part of the problem is I’m not sure you can feel it very well unless the joycons are detached. If you keep them attached to the system or use a different controller than you seem to lose the benefit.

This is what GP was arguing about regarding lazy ports on current hardware.

>While the switch is capable of something more is it really used anywhere besides those two games?

The success of the feature is kind of orthogonal to its inclusion into the console though, since console makers won't know how successful it'll be until it's developed and released. I agree with you that it didn't really take off, but that's true of other things like button pressure sensitivity, or Kinnect-style controls.

I'd also argue that HD rumble is but a small portion of a much larger and game-impacting package that is the Switch controller, while the PS5 controller is a refinement of the same controller they've had for decades now.

Playstation and Xbox have seemingly abandoned novel hardware that enables new gameplay experiences.

The Dreamcast VMU ate multiple CR2032 batteries for lunch, and not everyone bought their console at launch. There were plenty of dumb memory options throughout its lifespan.

I am not sure how you are counting them, but the second generation were the Atari 2600 and its peers, and the third generation were the NES and the Master System. Consoles had unique hardware until the 7th generation (PS3 and X360).

This is totally wrong. MI300x was released Dec 2023, just a few months ago. It is a fantastic product, but not into the hands of enough people, yet.

Give it time.

The only bugs we have run into so far are the lack of virtualization and the documented ones in the AMD rocm docs.

If you're just running some popular pretrained models out of the box, sure.

But you're saying that you develop new model architectures on them? And it works reliably? That's not at all what we saw.

It has nothing to do with availability or pricing. You can't rent single MI300x today cause ROCm doesn't support virtualization. PCIe passthrough doesn't work yet. AMD knows about the issue and is working on it.

We will have 8xMI300x for rent soon. Once AMD fixes their stuff, we will be able to rent out one at a time.

Problem is: no one wanted it. So none of that panned out. If it had gone on to be something like Intel’s Core 2 Duo that showed up everywhere in hindsight it would look smart.

Apple wasn’t a great example but it was the best I could think of. It took Apple 10 years to go from the A4 in the first iPhone with an Apple chip to the M1 in the first Mac. They also were just using Arm, the same instructions that they had been using before and something well understood.

They didn’t suddenly release something people weren’t really prepared for like a Transputer and just declare “this will be everywhere within three years“. They let the switch take its time as necessary.

Sony had the arrogance to do both of those.

Also, the PS3 did the same for PS2. Sony eventually did away with that to cut costs, but the launch models had it.

Apple switched to (and largely invented) ARMv8/AArch64 in the middle of that, which is very different from the ARMv7 they started with.

Previous PlayStation hardware was just as odd. Ps2 reused old surplus Ps1 processors just for audio, and had coprocessors just like cell.

Then out of nowhere Xbox 360 comes out, Halo that was a macOS title suddenly a main and a well selling IP on Xbox. Not just that, it also picked HD-DVD as its format posing a threat to Blu-ray.

Sony had to put something out.

The Xbox 360 released its HD-DVD drive a few days before the PS3 hit store shelves in North America. Sony had established the PS3 would use Blu-Rays a year before the Xbox 360 was even publicly announced.

I think there were a number of things Sony wasn't expecting about the Xbox 360, but HD-DVD or Halo wouldn't have been any of those things.

No it didn't. The PlayStation 3 outsold the Xbox 360. Have a look at the 7th generation charts on VGChartz:

https://www.vgchartz.com/

It's true the Wii sold the most.

The 8th generation didn't go well for Nintendo. The Wii U was a flop. But they're back on top with the Switch.

It's true the devs have become more comfortable develeoping games as the console matured but that's about it.

Xbox was always flop comparing to PS

For PS3 vs 360, the PS3 sold 87.40M units to the 360's 85.73M.

In this case, better to reduce early sales which are sold at a loss (production being expensive, significant R&D, scale being low), maximise at the end of the life when i assume they are maximising profits per console due to scale and R&D already being covered.

So, no -- it is never better to reduce early sales. You want to sell as many units as possible, as many games as possible, as many accessories as possible, as soon as possible.

(Also, as the same console does becomes cheaper to manufacture over the years, the price tends to go down as well.)

The PS3 started outselling the Xbox 360 for the year in 2007 and kept steadily gaining on the Xbox in total sales for the generation until it overtook it.

Xbox's one year head start kept it level with the PlayStation 3 for quite a while but it lost out in the end.

Look at the global yearly sales figures on VGChartz. From 2007 the PS3 outsold the Xbox 360 every year except for 2008:

https://www.vgchartz.com/yearly/

The charts you link to show game sales that lag way behind the xbox. For example Skyrim sold 2.6 million in 2011 on the xbox and only a million on the ps3.

I only say this because I got my ps3 for free with the purchase of a Sony television at a Sony store (remember those?) in late 2009 and never purchased more than a bare handful of games which I almost never played.

After the cost-reduced redesign in 2009 Sony gave away a free ps3 with every television purchase in Sony stores for almost an entire year.

Skyrim was one the games I bought and it was so buggy, even for Bethesda, on the ps3 due to ram limitations (I've read) that I purchased it again for pc.

Used my ps3 to watch a ton of blurays and Netflix, though. My ps3 was, and based on those yearly charts a lot of other peoples' ps3s were also, extremely powerful streaming boxes and I only retired mine when I upgraded to a 4k tv and got an appletv 4k.

You have to consider total software sales, not individual titles. And for the PS3 you have to consider its impact on blu-ray movie sales (which aren't in those charts).

Blu-ray also launched in 2006 and the PS3 was one of the most useful players available because it could do more than just play movies.

I think it wasnt a big deal in the rest of the world. Then Microsoft console focused on "doning more that just play" and flopped worldwide.

As you write later, Skyrim sucked on PS3 (like other multiplatform games like Bayonetta). It is not strange that it sold much less since it was a much worser game on the PS3.

Snark: Nintendo’s market is kids, people with kids, and people who were previously kids. Xbox and PlayStation are fighting over the market of people who’d be better served by a PC but don’t want to deal with all that. /Snark

That is to say, I still think there's significant competition for sales.

And if Nintendo didn't exist, it would be easier for Sony and Microsoft to take in the market for cute, whimsical, child-friendly, instead of having to distinguish themselves by being the more serious or more powerful option.

And handhelds; Nintendo has more or less stolen that entire segment, Steam Deck still being relatively niche and PS Portal being a different kind of thing. Sony's handheld ambitions are moribund and not for lack of trying.

As, what, an ornament? People buy games consoles to play games on (or occasionally for movies or some such), and I think very few of the games that people are buying a Switch to play are even available on Xbox/PS, nor vice versa (or at least, the Switch ports of the games that people buy Xbox/PS to play are not on a level where those people would buy a Switch to play them).

The target market for the Switch are people who play Animal Crossing, Super Mario Galaxy, and Zelda games.

There's some amount of crossover there, but there's also a large amount of separation in those markets. I know lots of people who play a ton of Animal Crossing and Zelda but would never want to play something like Borderlands or CoD or Darksouls.

Put another way, if Nintendo had a Sega-like market failure, then Sony and Microsoft would be the heirs apparent to the kingdom of cute. I'm sure they'd savor the opportunity. Do you think they wouldn't take these customers right now if they could?

And again, there's going to be a class of customers that could go either way but won't buy into two or more consoles, because they're expensive. They are directly competing for them.

But hey, the future is unknown, and my crystal ball is on the charger at the moment.

And there's gamers like myself, who mostly play games on PC but still have a Switch because while I can play a lot of those console games on my PC there's no offical way to play Zelda or Mario or the wide range of Nintendo first-party titles on PC. I think if Nintendo had a SEGA level implosion and started only being a publisher, whichever platforms Nintendo publishes for would become the heir to those consumers. I know that would be the case for me.

Assuming Nintendo steps away from the hardware biz, would they publish games exclusively to one console? Would they just start making mobile games? Would they sell on all consoles and PCs? ¯\_(ツ)_/¯

That's a pretty damn large market (= all adults). I think you meant to say "people who were previously kids playing Nintendo"

But yeah, I 100% agree that the overwhelming draw of Nintendo consoles is their exclusives, so it's mostly orthogonal to Xbox/PS/PC. If you want to play Mario or Zelda or Pokemon or Fire Emblem, you've got one choice. The common idea that the Steam Deck is competing with the Switch is almost entirely wrong.

Of course, as a woman engineer to make it this far she’s used to not getting the limelight in a male dominated industry - which may also explain some of AMDs recent success.

The concept of multiprocessors was already well understood around 1963, i.e. the multiplication of the performance of a computer by the multiplication of the CPU hardware.

On the other hand, the purpose of SMT is to enhance the throughput of a computer by ensuring that the existing execution units do not stay idle, by executing instructions from multiple program threads in order to keep them busy.

The modern use of SMT has been started by a research paper published in 1992, as a proposed improvement to the superscalar out-of-order CPUs already popularized by IBM POWER, but the name SMT (simultaneous multi-threading) has been coined only later, in 1997 (in the IEEE Micro from September/October 1997).

IBM had already used SMT in the experimental Advanced Computer System in 1968 (and they have filed for SMT patents in 1971: US 3,728,692 & US 3,771,138), but that project has been canceled and its superior architecture has been mostly forgotten for several decades.

CPUs that could execute instructions from multiple program threads, but not during the same clock cycle, like in SMT (hence the "simultaneous" of SMT), are older than the IBM ACS, for instance such a processor with fine-grained multi-threading was the peripheral processor of CDC 6600 (in 1963). SMT can exist only in superscalar CPUs, because only there multiple instructions are initiated in the same clock cycle.