PDA

View Full Version : I don't get this post...


Bigus Dickus
11-10-02, 10:13 PM
Originally posted by pastorHR + AA + AF... those features are not so much interesting for me.OK, whatever floats your boat. But then this:

I think that the raw power of the new GC should focus more on stuff like :

* better lookin shaders wich include bump , shininess , and reflection mapping , anisotropic filtering, real time refraction etc

OK, so perhaps you were referring to an AF besides anisotropic filtering?

* better lightning with performance free shadows
* more polygons
* more IQ
* better physics
* and it could be much farther (Deph of field, soft shadows , radiosity, real time volumetric lightning w shadows, etc ... )

More IQ as in higher res and anti-aliasing? :confused:

I do agree with the rest though, more powerful shaders are nice. Now if companies would stop selling DX7 class hardware and move the common install base to the shader generation, so that developers actually implement these features in games, it would be really nice.

netviper13
11-10-02, 10:24 PM
dude, Sat shut that thread down for a reason, it's probably not the greatest idea to bring a part of it back that's likely to turn into the flame-war that other one turned into.

pastor
11-10-02, 10:29 PM
Originally posted by Bigus Dickus
Originally posted by pastor[b]OK, whatever floats your boat. But then this:

[b]

OK, so perhaps you were referring to an AF besides anisotropic filtering?


More IQ as in higher res and anti-aliasing? :confused:

I do agree with the rest though, more powerful shaders are nice. Now if companies would stop selling DX7 class hardware and move the common install base to the shader generation, so that developers actually implement these features in games, it would be really nice.

in fact i was referring to anisotropic shader , *hmpfff* not filtering :D ... i mean the ability of a material to have an orientation in his shininess ( helpfull for some realistic metal shaders )

for the IQ thing oh well .. i made a typo i guess, i was meaning better AI (though it was in a general sense cos Cpu are more dedicated to this kind of task)

Joe Cool
11-10-02, 10:36 PM
Originally posted by netviper13
dude, Sat shut that thread down for a reason, it's probably not the greatest idea to bring a part of it back that's likely to turn into the flame-war that other one turned into.

I'm curious as to why it was shut down? There was little or no flaming going on, and obviously there were people still interested in continuing that very on-topic discussion.

I'm sorry to say that I think it happened because the "Nvidia-side of things" was getting hammered, and the powers-that-be here didn't like it. This even though it was pelly himself who started things up again by his post on the main page. :(

Bigus Dickus
11-10-02, 10:36 PM
Originally posted by netviper13
dude, Sat shut that thread down for a reason, it's probably not the greatest idea to bring a part of it back that's likely to turn into the flame-war that other one turned into.

Dude, I didn't bring up any of the argument about Pelly's conclusion, only this part of pastor's post in which I didn't understand what he meant.

:)


p.s., Pelly knows he is wrong :D

Bigus Dickus
11-10-02, 10:37 PM
Originally posted by pastor
in fact i was referring to anisotropic shader , *hmpfff* not filtering :D ... i mean the ability of a material to have an orientation in his shininess ( helpfull for some realistic metal shaders )

for the IQ thing oh well .. i made a typo i guess, i was meaning better AI (though it was in a general sense cos Cpu are more dedicated to this kind of task)

That makes sense, and to be honest I'd never thought about that effect much before. I assume you're talking about materials such as a fine grooved metal surface, which has a completely different reflection and diffraction pattern at various viewing angles. Interesting, and yeah... that would be really cool to see in games. I hope it happens this decade! :D

Bigus Dickus
11-10-02, 10:39 PM
Originally posted by Joe Cool
I'm curious as to why it was shut down? There was little or no flaming going on, and obviously there were people still interested in continuing that very on-topic discussion.

I'm sorry to say that I think it happened because the "Nvidia-side of things" was getting hammered, and the powers-that-be here didn't like it. This even though it was pelly himself who started things up again by his post on the main page. :(

Agreed.

LORD-eX-Bu
11-10-02, 10:42 PM
God not this again.......:o

Bigus Dickus
11-10-02, 10:42 PM
Wow, I finally decided to read Pelly's post on the front page that started up this argument again (I think I missed it the first time around being on vacation and all).

lol, he used fanATIc on the front page of this site. :D Very professional.

pastor
11-10-02, 10:44 PM
Originally posted by Bigus Dickus
That makes sense, and to be honest I'd never thought about that effect much before. I assume you're talking about materials such as a fine grooved metal surface, which has a completely different reflection and diffraction pattern at various viewing angles. Interesting, and yeah... that would be really cool to see in games. I hope it happens this decade! :D

that's the point ;)

pastor
11-10-02, 10:49 PM
Originally posted by Bigus Dickus

I do agree with the rest though, more powerful shaders are nice. Now if companies would stop selling DX7 class hardware and move the common install base to the shader generation, so that developers actually implement these features in games, it would be really nice. [/B]

oh yes, anyway i don't think even DX 8 class GC with be very usefull for such effects ... for a demo maybe :p

StealthHawk
11-11-02, 05:56 AM
Originally posted by Joe Cool
I'm curious as to why it was shut down? There was little or no flaming going on, and obviously there were people still interested in continuing that very on-topic discussion.

it already degraded into a back and forth between Pelly and his attackers. look for the other thread that was started for feedback the first time around. do we really need to see that again? and of course, some of the same people were involved, again. it's not like they didn't make their points before, do we really need a repeat performance?

i'm glad the mods closed the thread, we don't need bickering about the same topic to start in another thread. the first time was bad enough.

saturnotaku
11-11-02, 07:34 AM
I've changed my mind. I'm going to reopen this thread for now. But if it turns into another back-and-forth, it and any subsequent threads will be closed.

pelly
11-11-02, 10:32 AM
My God...I've created a Jerry Springer episode online!!!

:D

Sorry for the added grief saturnotaku... :D

sbp
11-11-02, 10:34 AM
http://www.btinternet.com/~tonyrichardson/horse.gif

Joe Cool
11-11-02, 11:15 AM
saturn, could you please define what you mean by "back and forth," and why such a thing would warrant the locking of a thread?

I'm asking because it's been my understanding that the alternative to a "back and forth" discussion is just a one-sided one.... :(

pelly
11-11-02, 11:18 AM
I think Saturn is referring to "back and forth" as being a thread full of opinionated people who are unwilling to listen to each other. As such, it turns in to a thread full of rants where people keep saying the same things over and over...

I think sbp is right on the money with his comment...

:D

creedamd
11-11-02, 11:48 AM
Originally posted by pelly
I think Saturn is referring to "back and forth" as being a thread full of opinionated people who are unwilling to listen to each other. As such, it turns in to a thread full of rants where people keep saying the same things over and over...

I think sbp is right on the money with his comment...

:D

so if someone likes aniostopic filtering and has a 1600xp processor, would the 4200 and a 2400xp be better? Or the 1600xp and a 9700, Is there a link, for this?
I would much rather have af then speed as long as its playable.

Bigus Dickus
11-11-02, 01:21 PM
Is it possible to create and define anisotropic textures? Would that be an easier way to achieve such effects than using an anisotropic shader routine, or is it a requirement that some anisotropic data (like a texture layer) be used as an input to generate an anisotropic output? No, you should be able to do per-pixel lighting I guess completely as a shader function over an isotropic base texture.

Hmm... I wonder when such capabilities will crop up in video cards? NV40/R400 era? I wonder if ATi and nV are even thinking about such things yet?

pelly
11-11-02, 01:56 PM
creedamd,

I said in my review...that anyone looking to use FSAA or Anisotropic filtering would be best off with the 9700 Pro...regardless ( within reason folks ) of CPU speed...

pastor
11-11-02, 03:26 PM
Originally posted by Bigus Dickus
Is it possible to create and define anisotropic textures? Would that be an easier way to achieve such effects than using an anisotropic shader routine, or is it a requirement that some anisotropic data (like a texture layer) be used as an input to generate an anisotropic output? No, you should be able to do per-pixel lighting I guess completely as a shader function over an isotropic base texture.

Hmm... I wonder when such capabilities will crop up in video cards? NV40/R400 era? I wonder if ATi and nV are even thinking about such things yet?

Check it out :

source: http://www.technoa.co.kr/notes/content.asp?content_id=14249

apparently it's possible with nv30 (and maybe with any other programmable gpu ?)

i'm quite in a hurry to see this on my desktop cos i'm working on such complex shaders since many years and it's exciting to know that it will be possible to set them in real time :cool:

budd_wm
11-11-02, 06:16 PM
I think what pelly means to say is....he is wrong. ;)

Bigus Dickus
11-11-02, 07:30 PM
Woah, that's pretty cool pastor. I didn't realize that current programmable shaders could handle such effects. I guess that, looking in retrospect, a fully programmable shader should be, well... fully programmable, including fragment programs that are viewpoint dependent.

Really neat stuff. The R300 should be able to do the same effects, though I guess it's possible it would need to multipass if the shader program were really long (though I doubt it would be).

pastor
11-11-02, 09:45 PM
Originally posted by Bigus Dickus
Woah, that's pretty cool pastor. I didn't realize that current programmable shaders could handle such effects. I guess that, looking in retrospect, a fully programmable shader should be, well... fully programmable, including fragment programs that are viewpoint dependent.

Really neat stuff. The R300 should be able to do the same effects, though I guess it's possible it would need to multipass if the shader program were really long (though I doubt it would be).

yeah it's cool :)

and i can tell you (because i'm working on 3d stuff not real time) , there is much much room for a lot of other effects with a fully programmable gpu.

for instance take the 10 years old 3dstudio3 : the today graphics cards are able to handle in real time just a little part of what 3ds3 was able to do (obviously not in real time)

it's amazing to see all the material/shaders combinations you can do with modern 3d packages ... it go much much farther that what u can see on GC in real time

and i was speaking about shaders and materials .. but you can apply the same thinking for lightning*/shadows , rendering ,antialiasing modes, filtering modes , model deformation routines ....etc