» »

Aalchemy

Aalchemy

1
2
»

Senitel ::

Seveda se lahko brez skrbi lotita T&L-a vendar če si prebral jima ne rata ravno najbolje - 1.5M tps...

undefined ::

Pikachu (in mogoče še Senitel), razloži mi potem zakaj mi laufa Serious Sam na 1024x768x32 4xFSAA popolnoma gladko (1GHz Celeron2, V5 5.5k AGP)?? Kako to, če je igra tako T&L zahtevna?? I don't get it. >:D

Pikachu ::

Meni se ne da vec...

P.S. Senitel, ali se ti se da razlagati te stvari?

LP,
Pikachu.

Senitel ::

Eko jest sem se odloču:
I don't care kaj IceMan in MadManMato tuki prbijata. In če se kdo od teh dveh oseb hoče o čemer koli kregat z mano, mi naj najprej pripravi eno (presenečenje) lepo spodobno in predvsem POVSEM tehnično razlago zakaj in kako je Voodoo-jev FSAA boljši od GeForce-ovega. Do takrat pa ČAOO>:D...

Matek ::

Jasno in glasno. Klik

Matek ::

Če to ni dosti, še opis nVidiinega FSAA-ja:

NVIDIA's Direct3D FSAA

NVIDIA is probably the most flexible in terms of FSAA options under Direct3D, but unfortunately, only a couple of the settings are actually practical for use in normal gameplay. Let's take a look at the settings offered under Direct3D (these settings correspond to the slider positions on the D3D FSAA slider in the drivers):

1x horizontal resolution, 2x vertical resolution - 2 sample FSAA, unbiased mipmaps
2x horizontal resolution, 2x veritcal resolution - 4 sample FSAA, biased mipmaps
2x horizontal resolution, 2x veritcal resolution - 4 sample FSAA, unbiased mipmaps
2x horizontal resolution, 2x veritcal resolution - 4 sample FSAA, unbiased mipmaps, different filter
3x horizontal resolution, 3x veritcal resolution - 9 sample FSAA, biased mipmaps
3x horizontal resolution, 3x veritcal resolution - 9 sample FSAA, 9x unbiased mipmaps
4x horizontal resolution, 4x veritcal resolution - 16 sample FSAA, biased mipmaps*
4x horizontal resolution, 4x veritcal resolution - 16 sample FSAA, unbiased mipmaps*

*Note: These two settings require a 64MB card to run.

The first thing you have to realize is that unless you want to run your card at 1/9th or 1/16th of its current speed, the last four settings are completely useless. The GeForce2 GTS does not have the fill rate to allow for a 9 or a 16 sample FSAA algorithm to be implemented at a reasonable frame rate in most games.

Secondly, NVIDIA's lowest FSAA setting under Direct3D, the 2 sample setting, does not look as good as 3dfx's 2 sample FSAA since you're only effectively supersampling in one direction. In spite of this, the 2 sample setting still yields a relatively similar performance hit since you're forcing the chip to render twice as many pixels.

Še ni dosti? >:D
Bolje ispasti glup nego iz aviona.

undefined ::

Naj še jest udarim po nakovalu, dokler je železo še vroče. :D

16-bit mode in 22-bit post-filter usefulness:

GeForce2 Ultra (Q3A, 16-bit, 200%):


Voodoo5 5500 (Q3A, 16-bit, 200%):


Kompresija textur (S3TC, FXT1):

GeForce2 Ultra (Q3A, using S3TC):


Voodoo5 5500 (Q3A, using FXT1):


Full-Scene Anti-Aliasing:
Razlaga:

GeForce2 Ultra modes (Ordered Grid Super-Sampling - OGSS):

1.5x1.5 (2.25) FSAA:


NOTE: In antialiasing 1.5x1.5 the resolution increases by 50%. Since the native resolution is not divisible by the virtual, the number of sub-pixels varies from pixel to pixel (from 1 to 4). The quality offered by this type of antialiasing is also rather low.

2x2 (4x) FSAA:


Voodoo5 5500 modes (Rotated Grid Multi-Sampling - RGMS):

2x FSAA:


4x FSAA:


NOTE: In principle, Voodoo5 driver allows to change the arrangement of samples inside the pixel, but the 3dfx mask is very useful for obtaining proper image quality. A mask like this smoothes the vertical and horizontal edges of polygons better than the GeForce mask.

Primerjava:



NOTE: Za najboljšo vidljivost detajlov na tej sliki spustite resolucijo na 800x600 ali manj!!!
O primerjavi pa mislim, da mi ni treba komentirati, saj govori sama zase. :D

Uporabnost T&L na GeForce1/2 vs. Voodoo5 (no-T&L) (again, by Gary Tarolli :D):

So why would a Voodoo5 have a longer life span than a card with T&L? One reason is because it's feature-proof, as strange as that may seem. Hardware T&L cards like the GeForce have their feature set pretty much frozen. For example, the GeForce supports lerping between 2 matrices from what I understand, and no more. Developers basically cannot use this feature, and require 4 matrices at a minimum. When APIs support this, the GeForce will have to fall back to software T&L or not expose the feature. The Voodoo5 uses software T&L all along and thus can implement any feature that comes along. Basically hardware T&L will accelerate (and even that is questionable) today's features only. Once new features come along, these cards will either not support the feature or fall back to software t&l. Note: this is for gfx chips that are hardwired, ie. not microcoded, heavily microcoded chips can implement most new features in microcode. So I would argue that hardware T&L does not increase longevity and can actually shorten it.

Odgovor na to temo - Tim Sweeney (Lead Programmer, Epic Games):

I agree with Gary on point #1, that the current DirectX7 / GeForce approach to "two matrix blending", which is shown off in NVidia's "bending cylinder" demo, isn't useful for skeletal animation. For skeletal animation, you need support for many matrices, with per-vertex *indexes* as well as blending factors.
I don't believe that simply having 4 matrices is enough; for realistic animated characters, you really need "lots" (i.e. 12 to 16 matrices), and up to 3 or 4 per-vertex matrix indices. In other words, any card that implements the DirectX7 spec is not going to be useful for accelerating skeletal animation in a realistic game scenario.

The new DirectX8 features add a huge amount of programmable flexibility that enable one to do implement skeletal animation with 100% hardware acceleration, and far more. Supporting this will require one of the upcoming 3D cards which support the new DirectX8 feature set.

PHEW! A long one!
Ok, I'm done... for now. :D

undefined ::

Še nekaj o multi-samplingu, da ne bo Senitel spet ratov slučajn preveč pameten. Se pravi feature, ki je GeForce2 NE podpira, Voodoo5 pa seveda jo. AJA, pa NV20 jo bo tud, za celih $600! :D



DirectX 8.0:

bod x3dfx (2/1/01 2:20:02 am)

The new drivers include T-buffer support for DX8

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

F.Y.I.



The new drivers are DX8. They contain multisample support. So everyone who was asking about T-buffer effects under D3D can start coding motion-blur, depth-of-field and other effects.



If you install the DX8 SDK and run the samples you can now use the dropdown box to select the multisample type used by the sample in fullscreen mode (none, 2x or 4x). Point sprites should also be hardware accelerated.


Še OpenGL extensions:

GL_3DFX_multisample

GL_3DFX_texture_compression_FXT1

GL_ARB_multitexture

GL_ARB_texture_compression

GL_EXT_abgr

GL_EXT_bgra

GL_EXT_blend_subtract

GL_EXT_clip_volume_hint

GL_EXT_compiled_vertex_array

GL_EXT_fog_coord

GL_EXT_packed_pixels

GL_EXT_point_parameters

GL_EXT_stencil_wrap

GL_EXT_texture_compression_S3TC

GL_EXT_texture_env_add

GL_EXT_texture_env_combine

GL_EXT_vertex_array

GL_S3_s3tc

GL_SGIS_multitexture

GL_SGIS_texture_edge_clamp

WGL_3DFX_gamma_control

WGL_ARB_extensions_string

WGL_EXT_extensions_string

WGL_EXT_swap_control


Eto, zdaj pa mislim, da bo dovolj dokazov za nekaj časa, or you hungry for more, Senitel? >:D

McHregec ::

Razprava je pravkar postala bolj zanimiva. :D

Senitel ::

No in zdej še povej za kakšen faktor x je RGSS boljši od ORGSS...



22bit post filer... Heh... A veš kaj 16bit je 16bit s post filterjem ali brez. Zato so pa 32bitne barve...



Texture compression smo že enkrat obdelal...



Eko kos dokumentacije iz DirectX 8.0 SDK...



Direct3D Feature Notes:



No significant API changes are contemplated beyond this point.

All implemented features have been tested individually

to some extent, but they have not been tested in exhaustive combinations

yet. Many features in DirectX 8.0 are awaiting new hardware, however

there is at least one implementation available for each of the features

shipped:




- Programmable vertex shaders are supported on the host CPU. On CPUs they

will operate with performance comparable to HW implementations and can be

used immediately for DirectX 8.0 prototyping and development.


- Programmable pixel shaders are supported on the host CPU using refrast

only. These can be used for prototyping DirectX 8.0 shaders for use when

hardware ships.

- Parallel DMA data input is supported on refrast and the SW geometry pipeline.

- Point Sprites are supported in refrast, software T&L pipeline, and should

be supported in a prototype DirectX 8.0 driver from at least one hardware

manufacturer.

- Volume Textures are supported in refrast.


- Multisample rendering is supported in refrast and future hardware.


- Higher-Order Primitives are supported in refrast and via D3DX utilities.




GL_3DFX_MULTISAMPLE ni enako GL_ARB_MULTISAMPLE (ki je pravi multi-sampling)...

Če že kartica postavlja neke zastavice, še ni nujno, da to tudi res podpira. ATI Radeon tudi javi zastavice za multi-sampling. Torej ATI Radeon podpira multisampling??

undefined ::

Da, Senitel, tudi Radeon podpira Multi-Sampling. Sem pa mislil, da vsaj to veš. >:D



Evo ti snip:

Accumulation Buffer support



ATI supports an accumulation buffer to support multi-sample effects like Motion Blur and Depth-of-Field. Undoubtedly everybody already knows what these effects are, if not check out our 3dfx T-Buffer articles.



The accumulation buffer works by rendering the image to an off-screen buffer (kind of like rendering to a texture) and then you apply an accumulate step in which you copy and combine the off-screen buffer data with the data already in the framebuffer. By repeatedly rendering to the off screen buffer and combining this data into a final image inside the framebuffer we can create the multi-sample effects like Motion Blur. Imagine rendering a sphere to the off screen buffer and then blending it into the framebuffer, if you move the location of the sphere for the various steps you'll end up with multiple copies of the sphere at different locations blended into one final image… this is motion blur. Accumulation buffer based multi-sample effects are supported by OpenGL and DX8.


Senitel, saj ni GeForce bog in batina... če verjameš, da je, potem obstajata 2 možnosti:
1. nVIDIA te mastno plačuje
2. s tabo nekaj ne štima (ok, joke >:D)
Sam resno, Voodoo5, Radeon in še celo Kyro imajo več featuresov kot GeForce (nikar me ne pripravi do tega, da ti še o Radeonu in Kyru spišem romane), in bodo šele prišli z NV20 (za $600-$700 USD). Žal. :)

Senitel ::

Ej IceMan sej nevem kaj naj rečem... BUDALA!!!!!!
Sam resno, Voodoo5, Radeon in še celo Kyro imajo več featuresov kot GeForce

NOBENA DANAŠNJA KARTICA ZA OBIČAJNE PC-JE NE PODPIRA MULTI SAMPLINGA!!!!!!!!!!!!!!!! MULTI-SAMPLINH JE BIL TO, KAR NAJ BI 3DFX PREDSTAVIL ŠELE V RAMPAGE-U (M-BUFFER), IN NVIDIA V NV20. KAJ NAJ BI PA POL BIL M-BUFFER???? NEKI ZA NA KRUH NAMAZAT??? ČE KATERA KARTICA ŽE USPE KREIRATI MULTISAMPLE BUFFER ŠE VEDNO NI NUJNO, DA GA TUDI V RESNICI PODPIRA.
EKO IZ ATI RADEON SDK:
Multisample Antialiasing

The RADEON™ supports 2-, 3- or 4-sample buffers for full-scene antialiasing. Note that the RADEON™ exports the D3DPRASTERCAPS_STRETCHBLTMULTISAMPLE caps bit, which indicates that the multisample support is somewhat limited. In particular, this cap indicates that the D3DRS_MULTISAMPLEANTIALIAS render state cannot be turned on and off in the middle of a given frame and that multisample masking (D3DRS_MULTISAMPLEMASK) cannot be used.


GEFORCE DEFINITIVNO NI BOG IN BATINA. RADEON JE TRENUTNO DEFINITIVNO NAJBOLJŠI ČIP. ZELO BLIZU MU JE GEFORCE, KYRO PA VOODOO 5 STA PA DALEČ ZADAJ.
IN ČE TI NI JASNO ZAKAJ JE VSE TOLE NAPISANO Z VELIKIMI ČRKAMI. NAJ TI RAZLOŽIM: TVOJIH BEDARIJ IMAM POVHEN KUFER!!

THE END!!

undefined ::

Senitel, ravno toliko pomeni Multi-Sampling danes na Radeonu in Voodoo-ju 5, kot pomeni T&L na GeForce1/2. NOBENA od teh dveh stvari ni popolna (trenutni T&L je preveč "statičen"), tudi Anisotropic filtering na GeForcu ni (Radeon ima več passov). Vsi pa vemo, da se stvari izboljšujejo z vsako naslednjo generacijo kartic, in na NV20 bodo VSI ti featursi dosti boljši in bolj izpopolnjeni, ravno tako kot bi bili na Rampage-u, če bi izšel. Is that clear? Polhen kufer maš lahko le sam sebe, IMHO.

Senitel ::

A sem reku, da imam dovolj tvojih "bistroumnih" odgovorov??
1
2
»


Vredno ogleda ...

TemaSporočilaOglediZadnje sporočilo
TemaSporočilaOglediZadnje sporočilo
»

Radeon 8500 MAXX

Oddelek: Strojna oprema
28965 (568) HUDiR
»

Tema za vse oboževalce 3DFXa -----> dol z Nvidijo !!!!!!! (strani: 1 2 3 )

Oddelek: Strojna oprema
1375600 (3670) goranTNT
»

3dfx Rampage

Oddelek: Zvok in slika
371345 (938) undefined
»

voodoo 6000

Oddelek: Zvok in slika
301571 (1177) Senitel
»

DirectX 8.0

Oddelek: Novice / Grafične kartice
151943 (1943) undefined

Več podobnih tem