Jon Stokes used to be my favorite tech-vulgarization writer, and he always provided good analyses. I'm bummed that he stopped writing at Ars, but I hear he's doing great stuff elsewhere (but not writing as much I'm afraid).
All the people I cared about reading are gone. It's just another tech news site now. It used to be an original content site too. There is still some original content, but it's nowhere near the caliber of stuff Stokes was putting out, and some of the newer stuff has a political slant to it that I don't care for. Jon always kept it technical.
I hadn't frequented Ars for ages when I downloaded one of their podcasts last fall. I listened for a while before I realized the cacophony of giggling and memes wasn't pre-show foolishness - it was the whole show.
Now that I know the name and look back, pretty much everything I characterized Ars with was written by Jon Stokes.
When they did their last design, it was pretty obvious. They threw out their Microsoft and open source sections, and left the Apple section. These days I sarcastically refer to Ars as an Apple blog.
The first time I came across Ars I read an amazingly detailed, multi-part explanation of the Itanium/IA-64 architecture (must have been Stokes). It was fascinating. A few years ago I tried to find it in the archives and failed (maybe only subscribers have access?). It's a shame they let pieces like that die.
Do you remember who used to cover open source for them, and where they might be writing now? I really don't feel like Ars offers a comprehensive perspective of the industry any more.
So true. Jon Stokes also wrote the great book "Inside the Machine: An Illustrated Introduction to Microprocessors and Computer Architecture" while he worked at Ars Technica. I have it still in my bookshelf. http://www.amazon.com/Inside-Machine-Introduction-Microproce...
> The PS2 is such a bizarre and powerful beast that it took me many hours of poring over articles and slide presentations just to get my bearings with it
> When the programmers responsible for some of the greatest console games ever made say that the PS2's learning curve is steep, you know something's up.
Update this for 2006 with a s/PS2/PS3/g. I do remember articles about the Xbox and 360 touting its easy API interfaces, in stark contrast to Sony's.
Why did Sony keep creating such hard-to-code-for devices and yet end up with such a monstrous install base? It's the chicken and the egg, I suppose.
> Why did Sony keep creating such hard-to-code-for devices
While this was certainly true with the PS2 and PS3, Sony has largely moved away from this with the PS4. The Cell architecture of the PS3 was very difficult to effectively program. For the PS4, Sony spent a large amount of time interviewing game programmers asking "what would make it easier to development?" Hence the move to x86. Hence the simplified memory hierarchy, etc.
Ken Kutaragi who was the "father" of PlayStation and boss up to and including PS3 was known to be a big fan of developing exotic hardware. The PS4 is the post-Kutaragi machine.
In game development, the vast vast majority of your development expenses will be in art assets (graphics, textures, models, animations, music, sound effects, scripted events, etc.) and not in the engine development. Once the engine is built, most of the rest of the game is art production.
To add to that PS2 had a longer lifespan than the X Box which allowed companies to reuse exiting code for longer. Unfortunately most AAA games are cross platform making targeting the lowest common denominator which limits the value of unusual hardware.
PS: Xbox was released over a year after the PS2 yet MS released ther next console before Sony.
Anemic competition in the '90s (Nintendo and Sega's failed Dreamcast) made Sony dominant, and the PS2 kept them there. Microsoft then spent quite a bit of money taking them on throughout the 2000's, and eventually won (in the US, at least).
Also, consumers don't care about how difficult a console is to code for, they only care about the games. Game publishers don't really care either, they only care about the potential market (which was massive).
In what sense has Microsoft "eventually won"? Sony actually makes money on games, Microsoft more-or-less breaks even. The PS4, so far, is outselling the XBox One, even in the US -- despite Microsoft breaking down and offering a kinect-less cheaper version.
Really, the winner was Nintendo, which sold a buttload of very cheaply made consoles for close to the same price as far loss-leading higher-powered competitors.
Of these companies, only Microsoft was spending profits from other sectors to build its position in the console market.
Incorrect. The Dreamcast had no OS. It has a proprietary firmware like all consoles of the time. Game developers had the choice to use a proprietary RTOS, a Windows CE based OS, or no OS at all. If they used an OS, it would ship on the same disc as the game. Very few games actually used Windows CE.
I remember articles discussing how Microsoft's experience developing DirectX for the Dreamcast helped them craft even better/more-developer-friendly ones for the XBox.
>Why did Sony keep creating such hard-to-code-for devices and yet end up with such a monstrous install base?
I'd say it's because when the developers learned how to make the best out of the PS2/PS3 hardware, the results were better than that of the competing hardware.
Since that's a heavily subjective thing to attempt to measure I would have to say it was more than just that. Marketing, exclusives, and brand loyalty would have much to do with it.
In fact that is the sole reason I own a PS3. After a couple weeks of rave reviews of The Last of Us I bought a PS3 and a copy of the game. I'd say it was worth it.
>'Why did Sony keep creating such hard-to-code-for devices and yet end up with such a monstrous install base?'
* Exclusive Titles: They managed to lock down some of best regarded franchises in each era.
* Backwards Compatibility: Each generation, at least initially, has been fully compatible the generation prior.
* Versatility: DVD capability in the PS2 was a fairly big deal at the time. Blu-ray in the PS3 was not, but it did look good in a bullet-point comparison against the 360.
* Free Online Features: The price of an XBL subscription is a deterrent for some.
* The direct problem of being hard-to-code-for always improves with time as developers gain experience and the toolset improves. At some point, you're only going to worry about content, not what's underlying.
A couple of those points are the sort that don't matter at all to 'smart' gamers and well-heeled adults who can own as many consoles as they like, but these are devices that ideally ship a few hundred million units to people of all ages and wide demographics.
By the way, the Cell [1] (PS3) is pretty interesting stuff if you're into computer architecture.
>>* Versatility: DVD capability in the PS2 was a fairly big deal at the time. Blu-ray in the PS3 was not, but it did look good in a bullet-point comparison against the 360.
While this statement is generally true, I'm sure the Blu-Ray was not an insignificant thing for some people.
Shortly after the PS3 was released, I got my first 1080p TV, and I wanted 1080p content like Planet Earth. The PS3 seemed to be the best value for a Blu-Ray player at the time, so I bought a PS3 as my HD content player. I ended up buying a bunch of games, but at the time I bought the PS3, the console games were gravy to me.
Brand loyalty left over from the PS1, and a strong launch line up with good exclusives promised down the line.
There was a lot of middleware being utilized. This helped keep dev costs down by letting devs reuse code from one platform to another, and not have to deal so much with the difficulty of the PS2.
Even if a game wasn't exclusive, it would quite possibly run on one multiplatform engine or another, which helped keep development costs down.
This brings back memories of working with PS2 dev boxes and reading Ars Technica. Back then it was a such a good source for this sort of information.
The PS2 was an incredible pain to develop for. Sony new it would be a hit no matter what they did. They really wanted it to have unique hardware. I also think they wanted it to be a machine that developers couldn't master in a singel development cycle. The way they released information about the VMUs later I think leads credence to the idea that they wanted the second and third life cycle of the device to show successive graphic improvements.
The big thing was just that the Saturn was supposed to be a badass 2D platform, with insanely powerful hardware sprite management. When Sega belatedly realized that 3D was the next big thing, they stapled on an extra CPU to give it enough power to handle 3D...but neglected to up the RAM or the CPU bandwidth. IIRC, it wasn't even theoretically possible to fully utilize both CPUs.
Very different situation than the Emotion Engine or Cell.
That's not true, it was meant to be a 3D console from the get-go.
However after they found out how powerful the Playstation was they quickly added extra power to make the console more comparable and that was where the complexity came from.
The one anecdote I know about the Saturn is that the programmers of the game Bug! just turned off the second CPU so they wouldn't have to deal with it.
I got that from this Dave Warhol inteview by Matt Barton, comes in around 2:00 -
