Doom 3 Hardware Guide Debuts

Nosf3ratu writes "Over at HardOCP, the boys have teamed up with id software again to publish the Official Doom 3 Hardware Guide. As the guide states: 'With the prospect of so many new people being brought into gaming by DOOM 3, there will likely be a lot of questions regarding the computer hardware needed to support it.'"

Comments

Re:OSX version Needed

[Nosf3ratu]

Yes, they are.
Todd Hollenshead, acting CEO of id said in an email shortly after the announcement of the release date, of OSX and Linux:

"Mac and Linux: Unfortunately I don't have dates for either of these. However, Linux binaries will be
available very soon after the PC game hits store shelves. There are no plans for boxed Linux games.
More remains to be done for the OSX version of DOOM 3 and that will take some time. We won't release
the OSX version until it's just as polished as the PC version. The date for OSX DOOM 3 remains "when it's
done", but I can confirm that it's definitely coming. "

Re:SMP

[micr0c0sm]

It's confirmed that there is no SMP support. Most games do not have it. Check the [H]ardForums for more info.

Re:Heh heh

[moonbender]

From the article:

There is no doubt that DOOM 3s minimum system specifications can easily deliver a good gaming experience. We found it simply incredible that a system this old could run DOOM 3 at all, much less run it well. It may be hard to believe, but we can honestly recommend spending $50 on DOOM 3 if you have a system comparable to this. You can still have a very worthy DOOM 3 experience with it.

I know some of you are thinking that it would be tough to call gaming at 640x480 a good gaming experience, but the environments in DOOM 3 are very forgiving in terms of resolution as discussed in our IQ section. Do we suggest you use a higher resolution to place yourself in an even more immersive environment? Without a doubt, but it is hardly a requirement to really feel a part of the DOOM 3 story.

And in the conlusion:

All that aside, I'm still astonished by our experience on the min spec system. While we think you'll likely have the best experience at the resolution and quality levels we suggested, dont think for a second that those are the actual requirements for playing DOOM 3. If I had a 1.5GHz Pentium and GeForce 4MX system that was two years old, I would still purchase this game knowing I was in for an incredible experience.

So yeah, I think this article is fairly balanced. They don't pretend you need the absolute latest and greatest to play the game. I don't think it's very insightful, really - their conclusion seems to be "well it runs best on really fast systems, but we guess you can also run it on older systems in 800x600" which isn't exactly a surprising outcome. OTOH I guess they do test Doom 3 on a large range of systems which will give you a fairly good idea at what res and quality settings your own systems will be able to run it, so the article does have its uses.

Re:Autospooge

[th1ckasabr1ck]

Go pick up the latest PCGamer and read the review. Or go check out PlanetDoom and read the reviews in the forum section. All signs point to it being an amazing game that completely lives up to the hype.

Re:Hardware on GNU/Linux

[Nosf3ratu]

Here's the long and short answer:

ATI 3d support is spotty at best. Nvidia support is excellent. Their cards perform just as good (if not better, in some cases) in Linux as they do in Windows. I have been gaming with Linux and Nvidia exclusively for about three years (including all the recent titles, such as RtCW, Q3, UT2K3, UT2K4, and ET) and it performs very well. Pair up a big nVidia with a 64bit Athlon, and you've got an excellent gaming rig.

So, the answer: The optimal hardware is the same, no matter what the operating system (although you would require less RAM on Linux, for example, if you use a slim window manager, as compared to XP, etc, etc, ymmv, and all that rot.)

Minimum Required System

[spellraiser]

Strange that this hasn't been posted already ... well, here it is:

DX 9.0 compatible 3D card w/ 64MB RAM*
MS Windows 2000/XP
Pentium 4 1.5 GHz or Athlon XP 1500+
384 MB RAM
8x CD-ROM
2.2 GB of HD space
Broadband (for multiplayer)

*Supported 3D Graphics chipsets:
ATI: Radeon 8500, 9000, 9200, 9500, 9600, 9700, 9800
NVIDIA: GeForce 3, GeForce 4MX, GeForce 4 Titanium, GeForce FX, GeForce 6

Re:Systems

[DeltaSigma]

From this shot [hardocp.com], I would have to say, ATI looks nicer for quality of lighting. The blending seems more natural.

I couldn't find any real discernable difference between the ATI and nVidia side-by-side images, but I thought there must be at some microscopic level.

So when you pointed this one out I decided to test it.

I pulled the image into photoshop and copied the nVidia screen ontop of the ATI screen. I got them as best aligned as I could and then changed the nVidia layer's blending to "Difference." The thing is completely black (except for the gun, which occupies a different position and the text which was placed arbitrarily). Essentially this means there is almost no discernable difference in rendering quality.

What might make one look better than the other is probably a combination of a couple factors. 1: We're looking at a very hazy image. There's apparantly an additive transparent plane being cast from the light or gun or something. In the interest of making things look organic, the texture/shader on this haze probably moves or morphs in some manner, just to keep it from looking too static. 2: jpeg compression, combined with the minor differences between the two images has likely enhanced certain colors/shades to varying degrees.

Give it a try yourself though. I think you'll agree that there's very little difference between the rendering quality of these split screenshots.

Which isn't much of a surprise considering John Carmack's doing the programming.

Re:Sweeeeeet.

[Minna Kirai]

I routinely play Starcraft and Unreal Tournament 2004 on my Powerbook G4 at non-native resolutions and they look fine to me

If your laptop is a multiple of 640x480, like 1280x960, then the described blurriness wouldn't actually occur for you.

The effect is worst when you run a program a a resolution close to, but not identitical to, the fixed rez of the LCD. That produces some fairly painful scaling artifacts.

Re:Dither?

[Viking Coder]

No way. All of the differences you point out are completely masked by the JPG differences. (Those "cubes" you refered to are DCT 8x8 boundary artifacts.)

I look at computer-generated image quality differences all day long at my job, and comparing these two images, there basically aren't any. I've additionally used a JPG Artifact removal tool on the two images, and then done the image differencing trick, and you have to turn gamma up by about 1.5 before you see *any* differences, and I believe all of those are positional differences. Turning the gamma up by about 2.5 is even more revealing: the differences between the images are hugely dominated by JPG differences.

Look again.

We'd both need better source images to make any strong claims about one card being better or worse than the other. From where I'm sitting, they're nearly identical.