Go Back   nV News Forums > Software Forums > Gaming Central

Newegg Daily Deals

Reply
 
Thread Tools
Old 03-24-04, 11:30 AM   #25
Nitz Walsh
Registered User
 
Join Date: Jul 2002
Posts: 622
Default Re: Console graphics vs Computer TV-OUT

Quote:
Originally Posted by NickSpolec
This is really a two part answer.

First, PC games are still in the relative dark ages when it comes to high polygon counts because PC GPU's have generally been lacking in high polygon output (and PC developers want to be able to cover a wide range of hardware, so they try and limitedly use high polygon scenes/models). Since there isn't this limitation in XBOX developement (one hardware configuration to exploit), developers can focus on the pure abilities of the machine. This is why most PC games (including newer games) still have less polygons per second in any given scene then the majority of PS2/XBOX games. PC developers are just too lazy, or just don't have the time to draw two different sets of models/scene (high and low polygon counts, which the use can select to use).
Of course, without actual polygon counts/scene from developers, this is guesswork.

Certainly, there are consoles games with excellent polygon throughput - or at least, appear to. But we have to keep in mind exactly what games we're comparing here, from my experience people seem to believe that since the DOA3 characters have lovely rounded breasts, and they don't see PC games with characters that bounce as well, that consoles have higher polygon capability.

The problem is, you don't see characters with that level of detail in many other consoles games either. PC games such as FPS/MMORGS have significantly more to worry about than 2-4 fighters in a very small arena, it's simply not possible to have characters of that detail in such a game. As well, eastern developers have generally favoured (at least IME) higher detailed characters models and less background detail, which especially for a console where your viewpoint is forced by the developer more than in PC games, is usually the wiser decision for certain types of games (opinions vary). This is a benefit of the larger console market, where production costs can be higher - you can simply get more focused artwork at times. In short, specific examples would be nice.

It can certainly go other ways - the Xbox version of RTCW has reduced polygon count on the characters (from what I've heard) compared to the PC version, and POP:SOT has reduced polygon count on the characters in the PS2/Gcube versions as compared to the PC/Xbox.
Quote:
Second, While the XBOX GPU may techically be a brother to NV2x, is still has some tricks. It's theorectical polygon output is over 200 million pps (polygons per second). It can substain about 100 some million (a second) when those polygons are fully lit and textures (though, this is only 1 light and 1 texture). Lets compare that (using the 3DMark2k1 polygon test) to the GeForce3 doing about 27 million per second with one light, the GeForce4 doing about 40 million, the Radeon 9700/9800 doing 75 million, and I'm not exactly sure about the GFFX abilites.
I'm well aware of the NV2A's "capabilities", the problem is you're comparing theoretical maximums of a chip to an actual, real-world benchmark that's over 2 years old, relies heavily on the CPU and uses an older API. It's about as nonsensical as comparing a CPU's MIPS rating to another using Winbench.

(ATI's specs indicate the Radeon 9800 has a "theoretical" maxium polygon output of 380 million/sec, for example.)

Due to the Xbox's memory architecture and the NV2A itself, the GPU can create polygons without the CPU involvement on the PC (which is why a PC with a Geforce4 and Celeron 733 would get whoped by an Xbox, not to mention the single platform-optimization theory). So that is a benefit in reducing the impact of having a slow central CPU, and why a PC needs to have a much faster one (which of course they do, I just get frustrated when people compare the Xbox to a P3-733 with a Geforce 3 as you likely do, they're _not_ the same thing). PCI express will apparently help this on PC, and DX10 will further transfer more of the load (and new GPU's at that time of course) from the processor, so this will improve on the PC side (if there's a PC gaming industry left at that point!).

Quote:
Also, the XBox has insane partical abilities. We're talking greated then 100 million particals a second. What are the partical abilities of PC GPU's, I'm not sure.
Again, theoretical maximums which are irrelevant in real gaming situations. And with no equivalent number to compare with on the point is rather moot, isn't it?
Nitz Walsh is offline   Reply With Quote
Old 03-24-04, 02:30 PM   #26
Edge
3d animator for hire!
 
Edge's Avatar
 
Join Date: Oct 2002
Location: Minnesota, USA
Posts: 3,564
Default Re: Console graphics vs Computer TV-OUT

Quote:
Originally Posted by Nitz Walsh
I'm not sure installing proper drivers for your system translates into a "lot of tweaking", ( ) but naturally that's a huge advantage for consoles, and that effects increases the marketing exposure as well. .
Well, considering I've tried every "tweak" in the book, and STILL can't get some things working, I'd say that it would certainly count as "a lot of tweaking". Also when some games work fine with certain drivers but other games don't work with those same drivers, and of course each different driver version offers different compatability, I'd say it can be a major hassle "tweaking" your system. I remember on my friend's computer we couldn't even install anything higher than the 2x.xx series of detonator drivers, or else all we got was a black screen, so if something didn't work right on anything except newer drivers he's kinda screwed. Not to mention I personally keep the 45.xx drivers on my system so that I have access to a larger number of options in Rivatuner (most notably the refreshrate fix), but of course Armed and Dangerous only works right with the 5x.xx series of drivers, but when I install that the official NV refreshrate overrider crashes when I select the refresh rate for a resolution, and Rivatuner doesn't support the refreshrate fix with the 5x.xx drivers, so basically the only thing I learned after screwing with it for an hour is that there is apperently no easy workaround for what I want to do, and that apperently the only way I can play Armed and Dangerous right with the 4x.xx series on my computer is to play the Xbox version of it (unless there's some magic fix out there that makes the game work right with the 4x.xx series). Yeah, I'd call that a lot of tweaking, and the funny thing is it still doesn't work the way I want it to.

Oh, and it's apperently a lot more complex than simply "installing proper drivers", because from what I can tell I have to take these steps to get my computer running at "optimal" efficience: Install Win2k Service Pack 4, install Blasterworm fix (is that included in SP4? I didn't see a mention of it), install DX9.0b, install the newest Nvidia drivers and pray the refreshrate fix doesn't crash, install your motherboard AGP drivers, install Rivatuner, and finally customise your AA and Anisotropic settings, not to mention pre-draw limit and v-sync options, for every game (2xAA/lvl8 aniso with no V-sync are great for NS, same options with Farcry turns it into a slideshow). I have never once started playing a PC game and spent less than a couple minutes customising options and seeing what settings I can use with it and still make it look good.

Quote:
What the hell does mouse control have to do with a space flight sim? Are you the one who actually though the mouse control in Freelancer was a good idea? Cripes, that dumbass decision completely ruined the game. I don't see how this example translates to "developer laziness" - use the control system the game was intended for. It's a space flight sim, not an FPS. You get a flightstick to pilot it, like you do with the majority of flight/combat sims. A gamepad is horrid by comparison for these types of games
Umm, maybe because every other flightsim I played worked great with the mouse and I found the joystick controls to be very screwy? Seriously, the only way I've really been able to play a space flight game well is either with an analog game pad, or with a mouse. Playing it with a keyboard (which is what I tried to do with Starlancer) simply sucked...horribly. Hell, even Colony Wars was playable with a DIGITAL pad. The developer of Starlancer was simply too lazy to offer mouse control like EVERY OTHER GAME IN THE GENRE DID. And yes, the mouse control in Freelancer was quite screwy, hopefully they'll make a console version of it so that it works on an analog pad

Quote:
Bear in mind some of the games that claim to support 720p/1080i are simply "upscaled" (ie: the pixels are doubled) - you're not always getting the true resolution.
I think Enter the Matrix is the only one that actually upscales a low-res image to fit the screen, from what I've read all the other ones run at true high-resolution.
Edge is offline   Reply With Quote
Old 03-24-04, 03:31 PM   #27
OWA
...
 
OWA's Avatar
 
Join Date: Jul 2002
Location: Virginia
Posts: 9,481
Default Re: Console graphics vs Computer TV-OUT

Quote:
Originally Posted by Nitz Walsh
I believe you're find that installing a significant, beta upgrade not meant to be used on production systems can have negative effects at times on your target system.

Go figure.
You're reading more into my post than should be there. In the part where he talked about PCs being very customizable but kind of screwy I replied saying that was one of the reasons I considered a console.

He then mentioned he recently wiped his disk clean and started over. I then replied I had just done the same thing mainly b/c of SP2 that I recently started testing not working as well as I had hoped. That issue, to me, was separate but I guess you wouldn't know that unless I stated it explicity. I started considering an Xbox long before I decided to try SP2. SP2 or using any sort of beta software (unless you consider the drivers we have to use as always being in a somewhat "beta-like" state) really has nothing to do with why I started thinking about a console.
OWA is offline   Reply With Quote
Old 03-24-04, 07:42 PM   #28
NickSpolec
 
NickSpolec's Avatar
 
Join Date: Aug 2002
Location: Burning Inferno, Arizona
Posts: 371
Default Re: Console graphics vs Computer TV-OUT

Quote:
I'm well aware of the NV2A's "capabilities", the problem is you're comparing theoretical maximums of a chip to an actual, real-world benchmark that's over 2 years old, relies heavily on the CPU and uses an older API. It's about as nonsensical as comparing a CPU's MIPS rating to another using Winbench.
No, no, NO!

Why does everyone think this? Just because it's a test in 3DMark2k1 DOES NOT MEAN IT RELIES HEAVILY ON YOUR CPU AND OR FSB

It's as simple as a test gets when it comes to testing the polygon abilities of a GPU. And it's almost COMPLETELY GPU LIMITED. It is NOT CPU limited. The CPU plays a minimal role in the test. It does not, in your words, "rely heavily" on the CPU. Not in the least.

Proof? Look at this test I did. (All tests were performed at the 3dMark2001 default settings)


First test, only adjusting CPU/FSB speed.


Quote:

GeForce 3 (220c, 460m)
Barton 2415mhz (230FSB)

1 Light - 27.6m
8 Lights - 5.6m



GeForce 3 (220c, 460m)
Barton 1830mhz (230FSB)

1 Light - 27.5m
8 Lights - 5.6m



GeForce 3 (220c, 460m)
Barton 1150mhz (230FSB)

1 Light - 24.3m
8 Lights - 5.6m



GeForce 3 (220c, 460m)
Barton 900mhz (180FSB)

1 Light - 22.8m
8 Lights - 5.5m

Second Test, only adjusting the GPU speed.



Quote:

GeForce 3 (240c, 460m)
Barton 2415mhz (230FSB)

1 Light - 30.0m
8 Lights - 6.2m



GeForce 3 (220c, 460m)
Barton 2415mhz (230FSB)

1 Light - 27.6m
8 Lights - 5.6m



GeForce 3 (200c, 460m)
Barton 2415mhz (230FSB)

1 Light - 25.3m
8 Lights - 4.7m



GeForce 3 (180c, 460m)
Barton 2415mhz (230FSB)

1 Light - 22.9m
8 Lights - 4.7m

The proof is in the test. The polygon count decreases by 5 million with *only* a 40mhz reduction in GPU clock, while it takes a 1500mhz reduction of the CPU speed to get that same decrease of 5 million.



EDIT NOTE: Stuffed up the info from the last GPU test. Re-ran the test.
__________________
Snake's System:

[size=1][list][*]AthlonXP Mobile 2500+ (@2.5ghz, 1.850v)[*]Albatron KX18D PRO (NForce 2 Ultra @227FSB)[*]512MB's OCZ Platinum PC3200 EL (@DDR454, CAS 2.5, 6,3,3)[*]GeForce3 (@230c, 460m)[*]Fortissimo III, Gamesurround 7.1[*]POS Intel 56k Modem (soon to get high speed, though)[*]Maxtor DiamondPlus 9 120GB, ATA133 (8mb)[*]Samsung DVD/CD-RW Combo (52x32x52x16x)[*]Lite-On LTR 16102B (16x8x40x)

[/size][/list]

Last edited by NickSpolec; 03-24-04 at 07:55 PM.
NickSpolec is offline   Reply With Quote
Old 03-25-04, 02:08 AM   #29
EciDemon
Registered User
 
EciDemon's Avatar
 
Join Date: Nov 2002
Location: Sweden
Posts: 1,423
Send a message via ICQ to EciDemon
Default Re: Console graphics vs Computer TV-OUT

But still even tho, it's useless to compare with theoretical specs.
Those are selling arguments for those who want's to belive them.
You never see game pushing that amount of single lighted polygons.

If one must compare consoles and PC then look at multiplatform games in stead.
And what do we see ?
PC to Console ports = Lower quallity gfx on the consoles
Console to PC = Same quallity (IE: Same texture res etc) wrapped in a bad port.

But it's useless to compare consoles and Pc I think.
I have Metal Gear solid 2 for my PS2 and I have Substance for PC. My pc is fairly good (2400+ 512mb Audigy2 etc etc) Yet MGS2 always performs better on the ps2
Ive tried the game with both a Gf3 and an FX 5200u.
But then again I can play UT2003 with both of those cards and have good frame rates and the PS2 would never be able to process that game in the quallity used in PC
EciDemon is offline   Reply With Quote
Old 03-28-04, 07:30 AM   #30
OWA
...
 
OWA's Avatar
 
Join Date: Jul 2002
Location: Virginia
Posts: 9,481
Default Re: Console graphics vs Computer TV-OUT

Okay, I basically had my first experience with the Xbox. No, I haven't bought one yet but I went over a friends to test it out. We basically spent quite a while playing Halo in co-op mode which was another first for me. It was great. I do much better with someone helping me. We were using wireless controllers and that was very cool also. Overall, I really liked it but one problem was that it was really difficult playing the game with a controller. I had to practice a while before we could even start since I wasn't use to controlling things with two analog thumbsticks. Yeah, I use those in the sports games on my PC but the control is a lot different in those vs a FPS. It was difficult but my friend didn't have any problems with it so I guess it's like most things, with practice you get a lot better at it.
OWA is offline   Reply With Quote
Old 03-28-04, 02:04 PM   #31
Edge
3d animator for hire!
 
Edge's Avatar
 
Join Date: Oct 2002
Location: Minnesota, USA
Posts: 3,564
Default Re: Console graphics vs Computer TV-OUT

Yeah, it does take some getting used to controlling an FPS with a controller. For the first couple hours I was like "whoa, how do I even aim with this thing?", but after a while I became really good with it, and to be honest I even like it better than a mouse in some ways (note I say SOME). But strangely the controls in Halo with the pad must be slightly different than other games, because even though I'm great at aiming in Halo I'm really crappy in other FPS games for the Xbox like RTCW and RS3. But the pad really shines in games like Otogi or Ninja Gaiden. Speaking of which, Outpost.com (and Fry's electronics) has Otogi for $20. Very nice price for a good action game, hint hint
Edge is offline   Reply With Quote
Old 03-28-04, 03:33 PM   #32
Johnmcl7
Defiant
 
Johnmcl7's Avatar
 
Join Date: Apr 2003
Location: Edinburgh, Scotland
Posts: 247
Default Re: Console graphics vs Computer TV-OUT

Some interesting points here, for me it's the games, doesn't matter if the PC offers much better graphic quality, some of my favourite console games aren't available on PC and vice versa.

John
Johnmcl7 is offline   Reply With Quote

Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


Similar Threads
Thread Thread Starter Forum Replies Last Post
Thoughts from console owners on NVIDIA's GEFORCE GRID MikeC Console World 11 05-27-12 08:43 AM

All times are GMT -5. The time now is 04:34 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.