X-Box Limitations (Hemos Is Dumb) (Yes, I am)

Fervent writes: "Daily Radar has an interesting article with Michael Abrash, one of the lead XBox technological designers. What's fascinating about this article is not what the XBox can do, but what it can't do. Abrash talks about programming limitations, HDTV, and goes against the NVidia ratio quote (the one where Gates said the GPU would be 3 times as fast as current NVidia hardware). Get your fill of the talk here." Update: 10/03 03:54 PM by CT : hemos was out of town all weekend. He missed this story when we posted it the first time HAHA! Update: 10/03 07:33 PM by H : /me hangs head in shame.

Comments

Re:Wrong market?...

[64.28.67.48]

but that moderate percentage of pro-Linux anti-M$ people just ready to jump on the X-box and hack it _must_ have M$'s attention, at the very least.

You're right. Microsoft is smart enough to know how to court developers -- they will encourage the cottage-industry guys, not stop them. If you hack PSX, you get a cease-and-desist letter. If you want to hack X-Box, here's some free tools! For a few bucks, here's a whole development kit. Philosophical differences aside, you give a toy like X-Box with the tools to do whatever you want with it, and hackers/developers will go wild with it.

Does anyone know whether Lego uses child labor? Or maybe they ruthlessly ran the Bric Blocs people out of business. Who cares? They make cool inexpensive toys and let me do what I want with them. And if it's the same with X-Box, you'll see a lot of people say, "well, they're not all bad"...

-------------

Re:Wrong market?...

[Darren Winsper]

Oh dear, another clueless idiot bashing the Mozilla project for no good reason. Oh well, someone has to dispell the myths...

Did you know that the vast majority of the code in Mozilla was written by somebody with an @netscape.com address? If you did, then you're just slagging off Netscape for no good reason. If you didn't, you are a clueless moron who should not be making such comments as you did.

Now, let's also not forget that Netscape have generously given us so much free code. Thanks to Mozilla now also being under the GPL (Or soon will be), a lot of open source projects will be able to benefit (Nautilus or Galeon anyone?).

Next time, please operate the strange device known as your brain before posting.

Do you guys think stuff like this is biased?

[Operandi]

For example, would such effort have been put into finding holes in DreamCast's ability if it were not spearheaded by Microsoft? While I dislike M$ as much as anyone else, I do like being my own devil's advocate... good for keeping from becoming narrow minded I believe.

Regards

Re:Do you guys think stuff like this is biased?

[iceT]

I'm sorry, but 1/2 a dozen EasterEggs in Microsoft's code can NOT account for over 180MB of program being installed, nor the seconds it takes to load a program and all it's DLL's.

The things that slow MS products down ARE design decisions. You'd be hard pressed to convince me that the #1 design consideration for Microsoft is to always choosing the user experience over execution speed. Example:

- Displaying a HTML file called 'blank.htm' (that requires rendering, with graphics, no less) when a user stops a page from loading in IE, instead of not displaying anything.

- Dynamic, self-modifying menus in Office2k that 'redraw' less popular items after a fixed amount of time.

- Menu pop-ups that fade in and out by default, or 'roll-up'/'roll-down', instead of just appearing (Win2k).

- Transparent drop shadows for cursors (Win2k)

- a Web Server that needs to have a specific web browswer installed before you can install the server (IIS 4 under NT4)

These are not 'programmer addons'. They are product features, designed in from the beginning...

Re:Do you guys think stuff like this is biased?

[Evangelion]

Goddammit people - the flight sim in Excel was put there as an easter egg by the programmers, so thier names would show up *somewhere* in the goddamn product. This was not a Microsoft thing, this was not in the design document. It was a nifty little extra put in by the programmers so they could actually leave thier names in the prodcut. For fuck's sake, if you're going to attack MS, at least attack them for something they did.

--

Deja v�

[Troed]

or what do you think?

...to the Metal

[Boone^]

As a hardware guy, I kind of enjoyed Michael's comments about taking the time to understand things down to the metal.

It seems that in the PC gaming world, there are many "disjointed" efforts that haphazardly come together to make a game; programmers optimizing their code (or not) for the latest in OpenGL or Direct3D, then you've got the API handlers written by NVIDIA, ATI, 3dfx, et al translating them as best as possible to the graphic chipsets' native language.

And, of course, all of this works on top of Microsoft's OS. That's 3 pretty big things that are unable to be tuned properly. They must have generic interfaces due to the plug-n-play nature of the PC business. The solution has always been to say stuff like "Pentium II 300MHz, 64MB RAM, 3D Card w/16MB required". With the Xbox, it seems like the designers will have control of 2 of the 3 items listed above, and with a standard set of hardware, optimizing 3d engine/game code has got to become a lot easier. Suddenly the requirements can easily transform from a PII 300 to a Pentium 166, the 64MB RAM turns into 16 MB RAM, and the Video Memory gets to drop considerably as well considering the target is NTSC/PAL output.

Of course, like the Dreamcast, we'll be seeing VGA output boxes so we can play the newest games on our 21" monitors. And since NTSC resolution is hard on the eyes on a 21" monitor, the Xbox will need variable resolutions, forcing faster processors, bigger 3D cards, and more RAM, bringing us full circle to where we started. :P I think the Xbox will be wildly successful if users treat it as what it is: a closed-box console used for gaming, not general applications + games.

Why I love redundant /. stories

[Mtgman]

I can just go back to the older story, cut and paste all the high-scoring comments, sit back and watch the karma just roll in. Thanks Hemos!

Steven

Wrong market?...

[BrK]

I'm not much of a gamer, so my comment may be totally off-base...
Doesn't it seem like the X-box is going to be marketed directly toward a crowd with a large population of anti-MICROS~1 people? Sure, there's lots of people that have never heard of Linux, and think that Bill Gates is a visionary, but that moderate percentage of pro-Linux anti-M$ people just ready to jump on the X-box and hack it _must_ have M$'s attention, at the very least.

The thing that worries me, I remember when M$ release the first version of IE, and thinking "there is NO WAY this thing can be a threat to Netscape". I certainly don't want M$ to become the dominant set-top box company...

Re:Wrong market?...

[nihilogos]

I'm not much of a gamer either (although I like to play lots of Quake 3) but I think the market they're aiming at is the Playstation and like game consoles. These people aren't really noted for anti MS sentiments and would probably go for anything as long as the rendering looks nice. I like nice 3D stuff, MS platforms currently kick Linux's butt in this department, and I have a Win98 partition solely for 3D studio and Quake 3.

Misinformation

[CaseyB]

It's an interesting story, but the lead-in above is entirely misleading.

Abrash has nothing but good things to say about the new hardware. Granted, he works for the company, but he has more than enough credibility outside of the Microsoft arena for me to listen when he speaks.

He talks about the constraints that ALL hardware-level developers have to deal with, but he says nothing that indicates the X-Box hardware is especially limited.

To wit: "the bottom line is that this is the most powerful chip I could imagine anyone getting into a console in 2001"

"Ratios" in processing power are not mentioned anywhere in the article. Apparently some overenthusaistic PR guy (probably not Gates) said it was 3 generations ahead of current parts, and Abrash says that's a bit of an overstatement. It's merely 1.5 or 2 generations ahead. Wow, that really sucks. :)

Re:Misinformation

[tjwhaynes]

"Ratios" in processing power are not mentioned anywhere in the article. Apparently some overenthusaistic PR guy (probably not Gates) said it was 3 generations ahead of current parts, and Abrash says that's a bit of an overstatement. It's merely 1.5 or 2 generations ahead. Wow, that really sucks. :)

This 1.5/2 generations makes perfect sense given what we know of NVIDIAs processor roadmap. Given that they have new processor releases, we will see at least the successor to the GeForce 2 before the XBox hits the shelves, and possibly another incremental improvement on that as well (like GeForce2 -> GeForce2 Ultra). So 1.5 -> 2 generations is entirely in line with what we are likely to see on the NVIDIA cards in our PCs in the same time frame.

Cheers,

Toby Haynes

Duplicate/really old stories WAS: Re:Deja v�

[milkman1]

It seem like there is a redudant story post every day or two these days.
Might I sugest that some kind of story warning system be implemented wherein the story would be showen to say 100 randomly selected readers, who loaded the main slashdot page at the right time, before it is actually posted. I would suggest a system where the main page is randomly replaced with just the prospective new story. Comment posting would be disabled. There would also be several options for moderating the story. I would suggest:
Redudant
Ancient
OT (Not relevent to slashdot)
Great story

It would also have a box for explantion (forinstance to link to the older story)
The results of the moderation would be fed to a real time display shown to the poster of the story. This would allow them to cancel or delay unneeded redudant/otherwise bad postings.

Xbox Beyond the limitations of TV displays

[CandyMan]

I would like to point out that NTSC has a total of 640*480*30fps=9,216,000 pixels per second, and Pal has 720*576*25fps=10,368,000 pixels per second. (Please don't knock me over the specifics, I might be slightly wrong about NTSC -being European, I've never actually worked with it- and I know that Pal has a "square pixel" mode where horizontal resolution is 768, right? Just trying to give you nice thousands here.) With that in mind, and neglecting overdraw, you don't need more than 12,000,000 polygons/sec anyway. If your rate is steady, that is.

This is why I think Abrash's words are very revealing:

MA: It's impossible to tell what performance developers will get until people are actually programming the hardware. It's also hard to evaluate because the chip is so programmable; how do you compare 125 mtris/s with 1 texture to, say, 12.5 mtris/s with a custom lighting model, along with 4 textures doing reflective bump mapping and a combiner program doing custom shading, plus shadows done on a second pass? It's not a matter of raw polygons anymore, but rather of the impact on image quality of the intersection of many factors: polygons, vertex shading, multitexture, texture lookups, pixel combiners, antialiasing, and multipass.

I am not a graphics überhacker, and don't have the answer on that comparison, but the second option (the way the Xbox design team have taken) sure sounds nicer to programmers. And you don't really need any more triangles anyway. Hmm. It will take Playstation II hackers many headaches to do what will come naturally to the programmers of this simpler-yet-more-complex approach.

On a related note: In a recent interview, John Carmack revealed that the Doom 2000 engine will have eight texture passes per polygon. (I am adding the emphasis). What, the Xbox can only do four? It is clear that id wants us PC gamers to keep our leer on when talking to those lowly conlosers. Hah!