Game does not run at 20 million pps thats just a sony marketing trick. Hardware just cant do that on the ps2 in theory it can but in real world it cant.doragasu wrote:DC is a great console, but I'm sure it will never be able to move near 20 million polys at 60 fps. I wrote what I'm going to write now again other times and a lot of people replied me saying I was mad and a lot other things like this, but I'll say it again: GT3 looks better than any other racing game I've never seen, it looks better than NFS Underground, better than MSR and even better than Project Gotham (it's a lot more realistic).Lord_Ball wrote:Ok let me ask why you think it couldn't handle them?doragasu wrote:Fast answer: Gran Turismo 3, Metal Gear Solid 2.Lord_Ball wrote:is there anything the PS2 can do that the DC can't (in regards to gaming and not in regards to things it can only BECAUSE it has a faster processor or more RAM).
Segatech Console Compare. Is it accurate?
-
- Soul Sold for DCEmu
- Posts: 4865
- https://www.artistsworkshop.eu/meble-kuchenne-na-wymiar-warszawa-gdzie-zamowic/
- Joined: Fri Jul 11, 2003 9:56 pm
- Has thanked: 2 times
- Been thanked: 4 times
Dreamcast forever!!!
- Segata Sanshiro
- Psychotic DCEmu
- Posts: 686
- Joined: Thu Oct 23, 2003 3:23 pm
- Location: the future
- Has thanked: 0
- Been thanked: 0
- Contact:
TXR2 may not have the realtime reflection or the trackside detail of GT3, but at simple view looks just as good. Yesterday i had a visit from a guy that works at sony as a game tester and when he saw TXR2 he went like: WOW!!!.... ehem, i mean cool game, it looks kinda like GT3! We had some healthy discussion- he gave me the press kit for VF4 and Tekken 4... He finished asking me to sell him one of my DCs with Sonic Adv 2 birthday pack. Draw your own conclusions!
Again i know that technically GT3 is far superior to any DC game, but people are equaly amazed by first looking at TXR 2, Zero Gunner 2, HeadHunter, Marvel vs Capcom 2... etc. I wont even mention monsters like Shenmue 1 and 2.
Again i know that technically GT3 is far superior to any DC game, but people are equaly amazed by first looking at TXR 2, Zero Gunner 2, HeadHunter, Marvel vs Capcom 2... etc. I wont even mention monsters like Shenmue 1 and 2.
-
- Soul Sold for DCEmu
- Posts: 4865
- Joined: Fri Jul 11, 2003 9:56 pm
- Has thanked: 2 times
- Been thanked: 4 times
-
- DCEmu Fast Newbie
- Posts: 19
- Joined: Mon Sep 15, 2003 7:47 pm
- Has thanked: 0
- Been thanked: 0
This topic was not meant to become a DC vs. PS2 topic. DC vs. PS2 is an argument potential vs. proven and those arguments never go good. This is simply a topic to discuss if this console comparison is valid. Yes all the consoles have good games, but lets just make sure this does not turn into a flame war.
-
- DCEmu Veteran
- Posts: 767
- Joined: Wed Mar 27, 2002 9:34 am
- Location: SoDak
- Has thanked: 0
- Been thanked: 0
- Contact:
I didn't mean for it to be a flame war, I just had an honest question as to whether or not the PS2 hardware supported any special features the DC doesn't support, or has limited support for. As the site that was making the comparison made little mention of the PS2, and what it mentioned made it's GPU sound like a piece of junk. I just wanted to know what it could actually do.Chuck D. Head wrote:This topic was not meant to become a DC vs. PS2 topic. DC vs. PS2 is an argument potential vs. proven and those arguments never go good. This is simply a topic to discuss if this console comparison is valid. Yes all the consoles have good games, but lets just make sure this does not turn into a flame war.
-
- DC Developer
- Posts: 9951
- Joined: Sun Dec 30, 2001 9:02 am
- Has thanked: 0
- Been thanked: 1 time
It does have a lot less features that the DC's PVR2. Speed wise, it approximately matches what the Dreamcast can do. However, it's connected to a more powerful processor, using a faster data bus, which is capable of shifting more data to the graphics system than the Dreamcast can. The limiting factor in DC games tends to be the CPU, although you can offload a bit of the work to the graphics system, while the limiting factor in PS2 games tends to be the rendering hardware.
However, PS2 games tend to have poorer texture quality than the more recent DC games. They tend to make up for this by simply throwing a lot of small polygons at the problem, which is why characters in PS2 games tend to look a little plasticy.
However, PS2 games tend to have poorer texture quality than the more recent DC games. They tend to make up for this by simply throwing a lot of small polygons at the problem, which is why characters in PS2 games tend to look a little plasticy.
- Quzar
- Dream Coder
- Posts: 7499
- Joined: Wed Jul 31, 2002 12:14 am
- Location: Miami, FL
- Has thanked: 4 times
- Been thanked: 10 times
- Contact:
it can. PCE, SNES and GEN were all in the same generation because of the graphics class even though they were a couple of years difference. Also, the N64 is grouped with the PSX and Saturn because it is like... 'first generation' 3D and all.Prophet][ wrote:buts is itsn't in the new generation, a generation doens't refer to how good it is, just when it came out
"When you post fewer lines of text than your signature, consider not posting at all." - A Wise Man
-
- DCEmu Cool Poster
- Posts: 1048
- Joined: Thu May 16, 2002 5:01 pm
- Location: Madrid, Spain
- Has thanked: 0
- Been thanked: 0
I don't know if GT3 really does 20 million, 10 million, 100.000, 10 or 1 pps, but when you look the cars, the backgrounds (those buildings you can see inside through the windows...) even the textures are pretty good, and all moving 60 fps. When I see the replays, sometimes I can't believe it's not real or at least it's not a CG animation. I haven't seen TXR2 (What game is that
?), but comparing GT3 to Project Gotham, PG has more effects but looks a lot less realistic to me. Of course I know XBox is a lot more powerful than PS2, but it looks impossible to me to move GT3 scenes with a DC.
I was recently playing Silent Hill 3 in my PS2 and it uses a pretty good effect I have not seen in the DC. I suppose it's not too difficult to do, but makes the game look like a film. It consists in blurring the background while keeping the front sharp or blurring the front while keeping the background sharp (like when you focus the lens of a camera). I think I saw it in some other games before, but dunno why I didn't put my attention in it. I don't know if I explained it properly, my english sometimes isn't enough to explain what I'm thinking about...
![Embarassed :oops:](./images/smilies/icon_redface.gif)
I was recently playing Silent Hill 3 in my PS2 and it uses a pretty good effect I have not seen in the DC. I suppose it's not too difficult to do, but makes the game look like a film. It consists in blurring the background while keeping the front sharp or blurring the front while keeping the background sharp (like when you focus the lens of a camera). I think I saw it in some other games before, but dunno why I didn't put my attention in it. I don't know if I explained it properly, my english sometimes isn't enough to explain what I'm thinking about...
-
- DC Developer
- Posts: 9951
- Joined: Sun Dec 30, 2001 9:02 am
- Has thanked: 0
- Been thanked: 1 time
Depth of field blur. Mario Sunshine and Wind Waker used it. The "cheap" way of implementing it is to use spare CPU time to read back pixels from the framebuffer, apply a blurring effect to it based on the value in the Z buffer, and send them back. That's how PS2 games do it, and that's also how most PS2 games do that terrible looking motion blur effect. I think that's how Gamecube games do it as well, but I don't know the internal design of the 'cube well enough.
The Dreamcast can't do that simply because we don't have a Z buffer.
On more modern graphics hardware, it's possible to do this and a whole lot more using vertex and pixel shaders. The Xbox has those, and most recent-ish PC graphics cards do too.
The Dreamcast can't do that simply because we don't have a Z buffer.
On more modern graphics hardware, it's possible to do this and a whole lot more using vertex and pixel shaders. The Xbox has those, and most recent-ish PC graphics cards do too.
- Segata Sanshiro
- Psychotic DCEmu
- Posts: 686
- Joined: Thu Oct 23, 2003 3:23 pm
- Location: the future
- Has thanked: 0
- Been thanked: 0
- Contact:
@ doragasu:
Somewhere i read that GT3 cars are made out of 5000 polys, that's having in mind the LOD. That sounds a lot to me, the cars that i make (i'm a modeller for RT3D) are around 1500, 1900 and dont look much more blockier than GT3 cars (dont get me wrong, my modelling skills can never be compared to Polyphony's modellers, i'm just an amateur) just a polycount comparison. Dont get me wrong, i like GT3 (i have GT3 and Concept), but i LOVE TXR2. here are some shots
![Image](http://www.the-laser.com/TXR2_4.jpg)
![Image](http://www.the-laser.com/TXR2_1.jpg)
![Image](http://www.the-laser.com/TXR2_3.jpg)
![Image](http://www.segalife.com/images/screens/txr21.jpg)
BTW:
- TXR2 was released in 2000
- it looks very crisp and runs at 60fps
- The PS2 version osf TXR2 looks much worse than the DC
And this is a shot of a car a made some time ago (dont laugh ath the texture) which is 1400polys
![Image](http://www.geocities.com/kkw_racing/MS3D/cobra/cobra_LV41.jpg.txt)
and without texture
![Image](http://www.geocities.com/kkw_racing/MS3D/cobra/cobra_uv2.jpg.txt)
@ BlackAura:
Even the latter generation PS1 games and N64 have Z-buffer. I dont understand
??
Somewhere i read that GT3 cars are made out of 5000 polys, that's having in mind the LOD. That sounds a lot to me, the cars that i make (i'm a modeller for RT3D) are around 1500, 1900 and dont look much more blockier than GT3 cars (dont get me wrong, my modelling skills can never be compared to Polyphony's modellers, i'm just an amateur) just a polycount comparison. Dont get me wrong, i like GT3 (i have GT3 and Concept), but i LOVE TXR2. here are some shots
![Image](http://www.the-laser.com/TXR2_4.jpg)
![Image](http://www.the-laser.com/TXR2_1.jpg)
![Image](http://www.the-laser.com/TXR2_3.jpg)
![Image](http://www.segalife.com/images/screens/txr21.jpg)
BTW:
- TXR2 was released in 2000
- it looks very crisp and runs at 60fps
- The PS2 version osf TXR2 looks much worse than the DC
And this is a shot of a car a made some time ago (dont laugh ath the texture) which is 1400polys
![Image](http://www.geocities.com/kkw_racing/MS3D/cobra/cobra_LV41.jpg.txt)
and without texture
![Image](http://www.geocities.com/kkw_racing/MS3D/cobra/cobra_uv2.jpg.txt)
@ BlackAura:
Excuse my ignorance, but i dont understand very well. As far as i know Z-buffer is this:The Dreamcast can't do that simply because we don't have a Z buffer.
Code: Select all
Definition:
An array used to store the maximum Z coordinate of any feature plotted at a given (X, Y) location on the screen, used for hidden line removal in a 2D rendering of a 3D scene. The Z axis is perpendicular to the screen with values increasing towards the viewer so that any point whose Z coordinate is less than the corresponding Z-buffer value will be hidden behind some feature which has already been plotted.
![Confused :?](./images/smilies/icon_e_confused.gif)
-
- Mental DCEmu
- Posts: 483
- Joined: Tue Oct 21, 2003 2:41 am
- Location: Minneapolis MN
- Has thanked: 0
- Been thanked: 0
@ Segata Sanshiro: I'm not positive, but, I believe that it has something to do with a display list renderer not requiring a Z-buffer, as opposed to an immediate-mode renderer. With the DC's PowerVR graphics system, there's no need for a Z-buffer because there are no hidden lines to remove, because they're not rendered in the first place.
That's interesting to read about the "depth of field blur," though. After playing Devil May Cry, I was thinking that it was some special PS2 hardware capability... Remember "blast-processing" back in the Genesis/Super Nintendo days?
(Source: http://pvr-extremist.com/pvrsgbg.asp )PVR-extremist.com wrote:In a conventional 3D system each polygon is preceded through the 3D pipeline in tern, in the order that they are sent to the hardware. This is referred to as "immediate mode" 3D processing and means that the hardware cannot know whenever any part of the scene is complete until the last polygon for that scene has been processed.
... in PowerVR systems the hidden surface removal is completed in the first phase of the pipeline before the texturing, lighting and shading steps. This has the benefit that only the visible pixels to be finally drawn in the display memory are texture, shaded and lit elimiating redundant work formed in conventional 3D systems.
That's interesting to read about the "depth of field blur," though. After playing Devil May Cry, I was thinking that it was some special PS2 hardware capability... Remember "blast-processing" back in the Genesis/Super Nintendo days?
« ˆˆ -ˆ-ˆ ˆ » [ Let's help! Trizeal DC ]
-
- Soul Sold for DCEmu
- Posts: 4865
- Joined: Fri Jul 11, 2003 9:56 pm
- Has thanked: 2 times
- Been thanked: 4 times
We will really never know the, Dreamcast life span was way to short. Take a old machine like the c64 it took 10 years almost to get "sms" and "nes" looking grafix from it . When the first games started coming out they,looking only slightly better then Atari2600 games. Im sure With time Games on the dreamcast would of got better and better. How good or what new flaws in the hardware or tricks that could of been achived no one but homebrew developers are going to find.
There were hardware flaws in the c64 hardware, that allowed when found not even known by commodore that alowed more then 16 colours on screen This is how many demos look so nice since only 8 colours were the norm. Im sure there are a few bugs that are in the dreamcast hardware that might allow things it's not suposed to do this is the case on all computers and consoles it's finding them.
Ps2 does nothing much more then the dreamcast. I cant really say the dreamcast could not make games as good as it can. It was killed before people could really get the grips with it. Ive had a chat with a sega developer He told me while working on a game they used manly the examples and had no idea about many parts of the dreamcast hardware and no time to learn it we get rushed ps1 ports with a res changed only many times.
There were hardware flaws in the c64 hardware, that allowed when found not even known by commodore that alowed more then 16 colours on screen This is how many demos look so nice since only 8 colours were the norm. Im sure there are a few bugs that are in the dreamcast hardware that might allow things it's not suposed to do this is the case on all computers and consoles it's finding them.
Ps2 does nothing much more then the dreamcast. I cant really say the dreamcast could not make games as good as it can. It was killed before people could really get the grips with it. Ive had a chat with a sega developer He told me while working on a game they used manly the examples and had no idea about many parts of the dreamcast hardware and no time to learn it we get rushed ps1 ports with a res changed only many times.
Dreamcast forever!!!
-
- DC Developer
- Posts: 9951
- Joined: Sun Dec 30, 2001 9:02 am
- Has thanked: 0
- Been thanked: 1 time
The PVR renders the image in small tiles, which are something like 32x32 pixels, into a small on-chip buffer. For opaque geometry, it renders each pixel once, rendering only the pixel which is closest to the camera. For transparent stuff, it does some other stuff that I don't fully understand. Effectively, the internal process in the PVR rendering hardware simulates a Z buffer, but there is no actual Z buffer.
That's possible because the PVR hardware isn't like conventional 3D hardware. Instead of rendering polygons as soon as it recieves them from the CPU, it stores all the data required for a scene in display lists. It essentially sorts the display lists, and renders what's visible, without requiring a depth buffer.
That's because the PVR was designed to be comparatively cheap, and as efficient as possible, especially with memory bandwidth. Very nice design. It's just a shame that it's not caught on with any other graphics card developers yet.
That's possible because the PVR hardware isn't like conventional 3D hardware. Instead of rendering polygons as soon as it recieves them from the CPU, it stores all the data required for a scene in display lists. It essentially sorts the display lists, and renders what's visible, without requiring a depth buffer.
That's because the PVR was designed to be comparatively cheap, and as efficient as possible, especially with memory bandwidth. Very nice design. It's just a shame that it's not caught on with any other graphics card developers yet.