Is the X800 series simply a speed upgrade?

It´s a speed upgrade. 3dc gives the option for higher detailed characters but other than that pure speed upgrade.
 
3dc - compression for normal maps
ps 2.0b - higher pixel shader instruction count supported
hd - support of features at HD resolutions

other performance improvements

read our tech article linked above
 
Brent_Justice said:
3dc - compression for normal maps
ps 2.0b - higher pixel shader instruction count supported
hd - support of features at HD resolutions

other performance improvements

read our tech article linked above
Click to expand...

It is just a speed upgrade. But a serious one to be sure.
 
Sir-Fragalot said:
It is just a speed upgrade. But a serious one to be sure.
Click to expand...

as i just showed you it is not just a speed upgrade

3dc - which has the potential for a better image quality
ps2.0b - which has the potential for a better image quality

just like how

SM 3.0 on 6800U has a potential for a better image quality

whether we see any though will be up to the developers and if they use the features

only time will tell

so far we've seen no examples of SM 3.0 having an increase in IQ, but we have seen a few examples of how 3dc can improve IQ, HL2 and Serious Sam 2 will employ 3dc btw
 
Thanks, cause I'm spending $500.00+ on an xtpe and I don't want to sacrifice beauty for speed.
 
haroldmeyer said:
Thanks, cause I'm spending $500.00+ on an xtpe and I don't want to sacrifice beauty for speed.
Click to expand...

The niceest thing about the X800 series is the fact that you get less of a performance hit when enabling FSAA and AF at high resolutions than you would with the 6800Ultra. I don't know if that's just a driver thing that Nvidia can fix or not.
 
Brent_Justice said:
as i just showed you it is not just a speed upgrade

3dc - which has the potential for a better image quality
ps2.0b - which has the potential for a better image quality

just like how

SM 3.0 on 6800U has a potential for a better image quality

whether we see any though will be up to the developers and if they use the features

only time will tell

so far we've seen no examples of SM 3.0 having an increase in IQ, but we have seen a few examples of how 3dc can improve IQ, HL2 and Serious Sam 2 will employ 3dc btw
Click to expand...

I can´t see anything new there. The longer shader instruction length don´t mean the X800 can run longer shaders only running shaders without as much looping. So that is really onlya speed upgrade that too as with everything.

But what can you do with ATI:s PS 2.b shaders that you can´t do with PS 2.0?

The Crytek guys and many others say that the only benefit of PS 3.0 over PS 2.0 is longer shader instruction length??

And 3dc it don´t increase quality either really. It´s just a compression method which again tries to raise IQ through speed ;)

The only thing I can think of is Temporal AA that can raise the IQ higher than the previous radeons if you disregard the performance difference between them...
 
oqvist said:
I can´t see anything new there. The longer shader instruction length don´t mean the X800 can run longer shaders only running shaders without as much looping. So that is really onlya speed upgrade that too as with everything.

But what can you do with ATI:s PS 2.b shaders that you can´t do with PS 2.0?

The Crytek guys and many others say that the only benefit of PS 3.0 over PS 2.0 is longer shader instruction length??

And 3dc it don´t increase quality either really. It´s just a compression method which again tries to raise IQ through speed ;)

The only thing I can think of is Temporal AA that can raise the IQ higher than the previous radeons if you disregard the performance difference between them...
Click to expand...

I can´t see anything new there. The longer shader instruction length don´t mean the X800 can run longer shaders
Click to expand...

yes it does, it can run longer shaders than ps 2.0's limits

But what can you do with ATI:s PS 2.b shaders that you can´t do with PS 2.0?
Click to expand...

ps 2.0b allows longer instruction counts, so you can have shaders with loger instructions, refer to: http://www.hardocp.com/image.html?image=MTA4MzAzNjk0NGJ3UHFVM2NZQWJfNF82X2wuanBn also more temporary registers, 32 in ps2.0b versus ps 2.0's 12

The Crytek guys and many others say that the only benefit of PS 3.0 over PS 2.0 is longer shader instruction length??
Click to expand...

longer instruction length, dynamic branching, arbitrary swizzle, graident, and more stuff: refer to here: http://www.hardocp.com/image.html?image=MTA4MTMwNTg4MDRVNW0xSmRUZG1fNl8yX2wuanBn

And 3dc it don´t increase quality either really. It´s just a compression method which again tries to raise IQ through speed ;)
Click to expand...

very true, but my facts are correct, Half Life 2 and Serious Sam 2 are going to use 3dc to raise image quality

The only thing I can think of is Temporal AA that can raise the IQ higher than the previous radeons if you disregard the performance difference between them...
Click to expand...

I'm not impressed with Temporal AA, I'd rather have more physical samples being taken
 
Thanks okay I got the impression that you could just use looping to run as long shaders on the old generation but there would be a performance loss doing so so in that sense it don´t mean the R420 can do stuff the R300 can´t.

Have also heard only the dynamic branching and such come with shader model 3 and are not supported by ATI?
 
oqvist said:
Thanks okay I got the impression that you could just use looping to run as long shaders on the old generation but there would be a performance loss doing so so in that sense it don´t mean the R420 can do stuff the R300 can´t.

Have also heard only the dynamic branching and such come with shader model 3 and are not supported by ATI?
Click to expand...

You can multi-pass a pixel shader, but it would be slow in Direct3D because the F-Buffer only works in OpenGL. With the longer instruction count in PS 2.0b you have more room to run a long shader without having to multi-pass.

Example.

R300 - Can have a shader up to 160 (total) instructions, if you go over that it has to multi-pass.

R420 - Can have a shader up to 1,536 (total) instructions, after that it has to multipass.

So for example a pixel shader that is 500 instructions long would be multi-passing on the R300, whereas the R420 wouldn't have to.

The X800 series does not support dynamic branching in the pixel or vertex shader. It does support static branching in the vertex shader though.
 
Ah good so I hadn´t got it wrong then :)

What I am getting at there is no new real image quality boosting stuff for the new radeon series like displacement mapping for the 6800 for example. Though I don´t think we will ever see real displacement mapping have heard the performance hit is huge and Far Cry is using something called virtual displacement mapping and such.

So all these boosts in image quality is through speed only so in that sense it´s a speed upgrade but a damn big one lol. Me I am getting one :)
 
oqvist said:
Ah good so I hadn´t got it wrong then :)

What I am getting at there is no new real image quality boosting stuff for the new radeon series like displacement mapping for the 6800 for example. Though I don´t think we will ever see real displacement mapping have heard the performance hit is huge and Far Cry is using something called virtual displacement mapping and such.

So all these boosts in image quality is through speed only so in that sense it´s a speed upgrade but a damn big one lol. Me I am getting one :)
Click to expand...

The R300 and up support Displacement Mapping via N-Patches.

Displacement Mapping is not a SM 3.0 only feature.

Offset/Parrallax Mapping (often called Virtual Displacement Mapping) can be done in Pixel Shader 1.1, 2.0 and 3.0. It is simply a better bump mapping techique. It has nothing to do with Displacement Mapping.
 
oqvist said:
But true displacement mapping can the X800 do that?
Click to expand...

define true displacement mapping

the R300 and up can do Displacement Mapping via N-Pathces

the GeForce 6 series can do it too through Vertex Texture Lookups in VS 3.0

just a different way of getting to the same goal

it isn't a SM 3.0 only feature, it is simply part of DX9

its all moot anyways, there aren't any games that use it yet

when it starts getting used then we can argue about which way is better, but by that time ati is likely to have a SM 3.0 part as well
 
Quote:

And regarding displacement mapping: that's a Vertex Shader 3 feature. It makes a difference, but it is not used in any game, not even the Unreal Engine 3 demo. Real displacement mapping generates additional geometry (i.e. triangles) to be textured, filtered, etc. Even NVIDIA's new cards are only a first shot at it, because they lack a tesselation unit. That means, they can shift vertex positions dependent on texture lookups, but they cannot create vertices out of thin air. As far as I understand, on 6800 cards you may displace vertices of an already fine but flat mesh. So all the geometry must already be there
 
And regarding displacement mapping: that's a Vertex Shader 3 feature.
Click to expand...

Is brent going to have to repeat himself? It is not a VS3.0 feature, it is part of DX9 and can be done in multiple ways, if you want to learn about it do a search on microsoft.com
 
Brent_Justice said:
define true displacement mapping

the R300 and up can do Displacement Mapping via N-Pathces
Click to expand...

i thought they could only do virtual displacement mapping, i thought the 6800 was the only one who could do displacement mapping.
 
virtual displacement mapping and displacement mapping is two total different things from what I understand. And at least nVidia have made a good job making it sound like it´s only available through VS 3.0 hardware...

I am not saying I care a bit about that because from what I have understood the performance hit is huuuge...
 
In reality BOTH cards are just attemping to speed up the past generation, ATi did it mostly through things which reduce bandwidth and passes while largely raising Core speed and memory bandwidth. Nvidia used PS3.0 as a way to use even larger PS instructions to avoid multipassing even more then ATi did, without addressing other issues like texture compression and memory bandwidth on as large a scale as ATi did.

So in reality, both companies are using "cheap tricks" to go faster, the only thing differentiating the two cards at the moment are both in nvidias favour but those really only apply to professionals: Higher color depth (64bit) and sepperate bridge chip for rendering video and real time effects and passing them back to system memory on the 6800 series, this won't effect 90% of the population as no game is using the added colorspace yet, and hardly any gamer does enough video processing to need such a chip.

The end result leaving the ATi card as, basically, the faster, safer (lower heat = less power needed), and more usefull for current games option, with the Nvidia being the more feature rich option, and the best choice for anyone that uses professional graphics and video applications.
 
Bad_Boy said:
i thought they could only do virtual displacement mapping, i thought the 6800 was the only one who could do displacement mapping.
Click to expand...
That is because Nvidia marketing wants to make you believe this........
 
oqvist said:
Quote:

And regarding displacement mapping: that's a Vertex Shader 3 feature. It makes a difference, but it is not used in any game, not even the Unreal Engine 3 demo. Real displacement mapping generates additional geometry (i.e. triangles) to be textured, filtered, etc. Even NVIDIA's new cards are only a first shot at it, because they lack a tesselation unit. That means, they can shift vertex positions dependent on texture lookups, but they cannot create vertices out of thin air. As far as I understand, on 6800 cards you may displace vertices of an already fine but flat mesh. So all the geometry must already be there
Click to expand...

and where did this quote come from?
 
Bad_Boy said:
i thought they could only do virtual displacement mapping, i thought the 6800 was the only one who could do displacement mapping.
Click to expand...

Offset/Parrallax Mapping can be done on any card that supports Pixel Shader 1.1 and up.

One of the Radeon 9700Pro's new features that was introduced was Displacement Mapping capability.

Read these two PDF's that were created by ATI back in GDC of 2003:

http://www.ati.com/developer/gdc/GDC2003-DisplacementMapping.pdf

and

http://www.ati.com/developer/gdc/GDC2003-DisplacementMappingNotes.pdf

You will see that it is NOT a SM 3.0 only feature, it is simply a feature of DX9 that has been in there since December 2002 when DX9 was released.

In fact, look what's in our 9700Pro tech article, bottom of the page: http://hardocp.com/article.html?art=MzI3LDI=

and look at Beyond3D's: http://www.beyond3d.com/reviews/ati/radeon9700pro/index.php?page=page2.inc
 
Azaroth1 if you are familiar with him. And from where he got it I don´t know from beyond3d I would guess. But he is into 3d modelling from what I could remember he seem quite knowledgable to me :)

But the point is from what I see virtual displacement mapping and parallax mapping isn´t the same as displacement mapping. It´s just other techniques for creating the same effect.

Kind of semi displacement mapping that is more performance efficient.
 
oqvist said:
Azaroth1 if you are familiar with him. And from where he got it I don´t know from beyond3d I would guess. But he is into 3d modelling from what I could remember he seem quite knowledgable to me :)

But the point is from what I see virtual displacement mapping and parallax mapping isn´t the same as displacement mapping. It´s just other techniques for creating the same effect.

Kind of semi displacement mapping that is more performance efficient.
Click to expand...

I explained what Parallax Mapping is at the bottom of this page: http://www.hardocp.com/article.html?art=NjEwLDY=

It has nothing to do with Displacement Mapping. It is better than bump maps, but not as good as displacement mappping.

It goes in this order from worse to best:

Bump Mapping < Parallax/Offset Mapping < Displacement Mapping

Bump Mapping and Parallax Mapping do not physical transoform the Polygons, the image is still 2D flat.

But Displacement Mapping does physically transform the polygons, making it truely 3D.

Bump Mapping and Parrallax Mapping are a Trick of the eye.

Displacement Mapping is actual, physical.
 
Yup exactly.

But here is the full reply about displacement mapping

Oh my goodness

PS3 and VS3 (aka Shader Model 3, aka SM3) allow for loops, thus you may execute potentially infinite shader instructions. Right. The problem is, your card won't be able to run long shaders in realtime It won't be FPS but SPF. X800 will run any shader of performance-wise reasonable length. nV's 65k instruction length limit is utterly pointless for many years to come, even X800's current limits are hopelessly overfeatured. So, instruction count is not it.

What is it then? Well, loops are implemented by branching conditionally. A SM3 shader may first calculate if it is reasonable to calculate the pixel value at all (it may be hidden or in deep shadow) and, based on that, either do the whole complex calculation or simply use e.g. black. Otoh branches don't come for free. They have the potential to flush the whole pipeline and, if not used carefully, SM3 shaders may severely decrease performance.

SM3 is SM2 plus some instructions and capabilities plus some limits relaxed. There is no visual difference between PS2 and PS3. You may create any effect in both models, and if you know what you are doing very well, you may make it running faster in SM3, at least in some cases. From a programmer's point of view SM3 is certainly more elegant, though.

And regarding displacement mapping: that's a Vertex Shader 3 feature. It makes a difference, but it is not used in any game, not even the Unreal Engine 3 demo. Real displacement mapping generates additional geometry (i.e. triangles) to be textured, filtered, etc. Even NVIDIA's new cards are only a first shot at it, because they lack a tesselation unit. That means, they can shift vertex positions dependent on texture lookups, but they cannot create vertices out of thin air. As far as I understand, on 6800 cards you may displace vertices of an already fine but flat mesh. So all the geometry must already be there.

What FarCry and UnrealE3 do is called Virtual Displacement Mapping. That's perfectly possible on any SM2 machine, and it doesn't need even many PS instructions. In fact that's the only reason why it can be done in realtime at all. The illusion it creates is quite good as long as you don't look at the surface from a very flat angle. If you do so, you notice that it's really flat but "painted".

The only reason why we probably wouldn't see Virtual Displacement Mapping on Radeons could be, that Crytek implement VDM via SM3 instructions only because they get payed to do so. Otoh, they want to sell their engine, so why should they purposefully p*** off ATI customers? Does not make sense to me.

Hope things are clearer now


and here is another

You're welcome, fishlore

Sorry for the "oh my goodness", but, while following the technical discussions on www.beyond3d.com and www.3dcenter.de vividly, I see that NVIDIA's propaganda already works all too well.

As regards instruction count, you will certainly have seen pictures of ATI's Ruby demo. Now, that is code that won't run on R300/R350 any more, at least not without a wrapper that replaces some shaders with slightly simpler ones. Same with nV's Mermaid. Still, those demos use shaders of not much over 100 instructions. With about 1000 instructions you can make "Finding Nemo" and "Final Fantasy" (the movie), and that takes hours per frame on a rendering farm. That's for 65k instructions being pointless in realtime

Some more thoughts about real SM3 displacement mapping:

Have you ever tried to model something in Maya or 3DSmax? Say, a rock? I have. Do you know how many triangles it takes to let something look round? Well, let's suppose we need 100 triangles for a rock like those in the foreground of jakup73's second shot. How many of them do you have on a beach? 10000? Now multipy. That's what real VS3 displacement mapping would have to generate on the fly. That's what would have to be textured, filtered, lit, etc. That's what with NVDIA's current implementation would already have to be there, because the hardware can't create new vertices!!! Impossible.

OK, you could do some LOD magic, i.e. only do it in the foreground, but I guess you would have to do that in the CPU, thereby making an already CPU-bound game even more so. And you wouldn't even be able to do it that way on current NVIDIA hardware, because, you guess, "no new vertices"!!!

From this point of view nV's VS3 seem pretty pointless also, because they don't save triangles, not even in the non-displaced case. That's not entirely true though. While I can't think of things that can't be done in normal VS2 programs, it may be much easier for the developer. I mean, it's a difference whether you have to specify a relief as mathematical opertions or as an image you can paint. The latter may be even done by an artist. Again I think the main advantage of SM3 (at least in nV's restricted implementation) is convenience for the developer, not that it won't run as well otherwise.

All in all it boils down to the question whether developers will use what is most convenient for them or what reaches the biggest group of customers. And that's where money gets in. I can't imagine a publisher not pressing the developers to use SM2 when that's what the masses have and that's where the money is.

In other words: I strongly expect to see SM3-only code in some nV demos, but not in any game before, say, 2006, maybe later. And by that time games will crawl on even the most advanced current hardware anyway.

Again Aztaroth11
 
Far Cry general discussion forum

forums.ubi.com

search for displacement mapping and choose the first post from with Astaroth11...

He also referrs to 3dcenter and beyond3d :)
 
kinda confused by oqvist's quote he found. so the new cards can do Displacement Mapping but they cant?

And regarding displacement mapping: that's a Vertex Shader 3 feature.
Click to expand...
uhhh....the x800 doesnt have VS.3.0, no? i thought it just had PS.3.0 + VS.2.0

virtual displacemenet mapping can make an object look raised or 3d from a angle, but when you look at on eye level its just flat. like shown in the far cry 3.0 patch vid. correct?

but true virutal displacement mapping makes the object look 3d in all angles including eye level, correct?

--------------
this is from the [H] x800 review:
The X800 series is Vertex Shader 2.0 compliant and does support point sampled Displacement Mapping with N-Patches ala TruForm 2.0. What it does not support is filtered Displacement Mapping via vertex texture lookups.
Click to expand...

ok so it supports displacement mapping. but which kind? whats the difference between

point sampled Displacement Mapping with N-Patches
and
filtered Displacement Mapping via vertex texture lookups

would one happened to be nick named virtual displacement mapping and the other true virtual displacement mapping?

thx in advanced for clearin this up for me brent + oqvist

Remark removed. Consider this a warning that this is not acceptable - Kyle
 
Bad_Boy said:
kinda confused by oqvist's quote he found. so the new cards can do Displacement Mapping but they cant?


uhhh....the x800 doesnt have VS.3.0, no? i thought it just had PS.3.0 + VS.2.0

virtual displacemenet mapping can make an object look raised or 3d from a angle, but when you look at on eye level its just flat. like shown in the far cry 3.0 patch vid. correct?

but true virutal displacement mapping makes the object look 3d in all angles including eye level, correct?



this is from the [H] x800 review:


ok so it supports displacement mapping. but which kind? whats the difference between

point sampled Displacement Mapping with N-Patches
and
filtered Displacement Mapping via vertex texture lookups

would one happened to be nick named virtual displacement mapping and the other true virtual displacement mapping?
Click to expand...

the X800 series does not have Pixel or Vertex Shader 3.0 support

but they can do Displacement Mapping via N-Patches, which is NOT a SM 3.0 only feature

Parrallax Mapping/Offset Mapping/Virtual Displacement Mapping = all the same thing, 3 terms that describe one 3D technique, which is better than Bump Mapping, but yes, still flat 2D, if you look at it on its side its flat

Displacement Mapping = true, actual, physical polygon changes, if you look at it from the side it really is raised
 
thanks for clearing it up for me Brent :)

but i have one more thing....

whats the difference between:
point sampled Displacement Mapping with N-Patches
and
filtered Displacement Mapping via vertex texture lookups
 
Bad_Boy said:
thanks for clearing it up for me Brent :)

but i have one more thing....

whats the difference between:
point sampled Displacement Mapping with N-Patches
and
filtered Displacement Mapping via vertex texture lookups
Click to expand...

read the second PDF i linked above, it will explain what the differences are
 
You must log in or register to reply here.
Back
Top