3dmark 2k1 no hardware support??

[Natedogg]

I'm getting on the Nature game and some directx 8 tests the error no hardware support.... What's up with that?? I have to 7089 drivers and directx8.. any suggestions??

[dustinyo]

are you sure the video card is in there? no hardware support could definatly come from not having hardware...

i would help, but my computer is out until tomorrow when my new cpu comes in

[Natedogg]

I know it is in there... do you mean securely or what??

[LordCrass]

Probably because the Radeon doesn't have a pixel shader, or it's not available in the current drivers? Not sure which of the above it is.

[Natedogg]

I'm gonna try using the ati 7072 drivers... A lot of other people seem to see the nature seen on the radeon.. I'm just left out :-(

[id_syne]

I don't thing the radeon has the capabilities to do that tests, seems like 3Dmark 2K1 is made for the upcomming Geforce 3, which i shall be getting.

[dustinyo]

Originally posted by Natedogg
I know it is in there... do you mean securely or what??

hehe i was kidding man, i'm just giddy about my new system coming tomorrow... i hope you didn't take offense

[Natedogg]

I know you were kidding.. But it's odd that everyone seems to be able to see this, but mine skips all of the dx8 tests eevn in software mode... Must this be a dx8 problem?? I reinstalled it, and still same problem.. I'm gonna try redownloading the dx8a setup and try it again in the morning.. Leave me some suggestions guys.. I would love to have them waiting for me in the mornin ;-) Thanx again.

[LordCrass]

Dude, that test doesn't run for me either and I'm using the 7089 drivers. Problems I encounter:

Lobby - low: One guy's head flickers next to him for a sec.
Lobby - High: Same as low, but also another guy's legs do the same thing.
Nature: Not supported by hardware
Pixel Shader: Not supported by hardware
Point Sprites: Black screen for 30sec or so, then on to next test. Results show N/A for this test. If I hit escape to abort it, it will abort at the end of the 30sec and I see a flash of a beige horse or something before it exists.

Otherwise, no problems other than frame rate on the high detail tests Of course, I only have a 500Mhz Athlon, so that could use an upgrading.

Does anyone know if the Radeon has a vertex shader or is that vertex test using software calculations?

[FullAction]

Except the skipping of the nature and pixel shader test, which is in it's current form not supported by the Radeon, all the other glitches, you mentioned, aren't present in software mode, which leads to the suspicion, that ATI has to rework the drivers a lot.

[APVM]

Benchmark rules! this 3dmark is a very good game.

[FullAction]

ALthough getting the highest benchmarks can be a quite amusing game, I'd rather find out, what of the promised features aren't working, by a free download, than after I have payed my bucks for a game and find out then.

[Fisu]

ATI h&w pixel shaders are based on dx8 beta, which is incompatible with dx8 final. what this means is, basically that we are not going to see Radeon pixels shaders ever used. too bad ..

[Paddy]

Here's what Jason from ATI posted about Pixel Shaders yesterday in the opengl.org forums !

"Unless you hack the registry, we will report 0.0 pixel shaders. The whole "0.5 pixel shader" thing turned into a real hairball, and you can get a lot of the RADEON™ functionality through DX8's enhancements to SetTextureStageState (stuff like LERP, MAD, TEMP and ARG0), so we decided to focus on other areas.

-Jason"

No comment ...

[FullAction]

What does that mean "it turned into a hairball"? Aren't these people getting paid, to sort those problems ?
I really would be interested, in what other areas they are going to focus. Probably drinking coffee and looking out of the windows. These people should get off their lazy a**es and get some work done!

[Furious]

Well I tried enabling pixel shaders in the registry but now I get an "Unable to initialize the video card" error.

Hmmmm.

I'll keep looking into this.

[j0harris]

"The whole "0.5 pixel shader" thing turned into a real hairball, and you can get a lot of the RADEON™ functionality through DX8's enhancements to SetTextureStageState (stuff like LERP, MAD, TEMP and ARG0), so we decided to focus on other areas."

I am not encouraged by this statement. It does not put much confidence in ATI's driver team. i.e. "We can't get it to work right now, so we give up. You should use other features instead."

[Ogreboy]

I don't know about you guys, but I refuse to stand for these goddamned excuses from ATI. I say we write up a large petition and demand that ATI explain to use why the DX8 features that we were promised are not working, and apparently, they are not trying to get them to work.

I know now that I'm never ever buying an ATI product again, unless things change dramatically. And I know that that's not going to happen.

[Natedogg]

I agree about possibly not getting another ati product. At the time I purchased it, I thought I was going to get something reliable and would last a long time. Instead, I have ran into endless problems, some possibly because of my poor motherboard... I'm thinking of getting a different one along with a new processor.. Anyone have any suggestions?? Let me know because I would liek to have something affordable and would make the best of my Radeon until I move on.

[FullAction]

I, for one, will keep my eyes at the Kyro2. Quite affordable and the Overdraw features could make up for the missing T&L engine. At least with a powerful cpu, you will probably smoke the Radeon, like the first tests showed. Well, let's see how it does, when it really comes out, driverwise and in real games performance.

[cveale]

Originally posted by FullAction
I, for one, will keep my eyes at the Kyro2. Quite affordable and the Overdraw features could make up for the missing T&L engine. At least with a powerful cpu, you will probably smoke the Radeon, like the first tests showed. Well, let's see how it does, when it really comes out, driverwise and in real games performance.

I bet the Kyro II will suck on 3DMark 2001. There are too many high poly tests and current processors can only render about 2 mil max. It should do well in the high resolution tests though where bandwidth restricts the card.

[Daneel]

Yeah i must say that i'm dissapointed by ATI. By spending over 800 DutchGilders (over $300) for a Full-proof DX8 grafix-card i was thinking that it would last long before showing any lack of hardware support. Now i've got the card only shortly and already it is showing big shortages...
If this does not change, i don't think i will have much faith left to ever buy ATI again. No matter what reviews state or ATI claims.

It's a sad day

[scr0t]

Cheer up guys! I'm sorry to hear that you are disapointed eith your Radeons. You say that you are disapointed with the lack of support for DX8 stuff. But you had a choice of that or none in the way of the GF2 series or V5's etc..

I mean so some of the stuff wont work, so what. You can have everything. If you had bought a GF2 GTS then you would have been saying for the last 6 months, hahhaha look at my faster card than yours. Now we have the faster card, well sorta. We also have better visuals. I think we got a good deal outta it.

[euan]

For the PointSprite test I get a slideshow of 6 or so frames of a horse being rotated. Is the horser supposed to be made out of dots or sumfing?

[Ogreboy]

Yep, about 50,000 or 50 million particles or something like that, I forget now There's another post in the forums that has the details on what exactly it does, etc etc. but I don't remember where

[Furious]

Originally posted by Ogreboy
I don't know about you guys, but I refuse to stand for these goddamned excuses from ATI. I say we write up a large petition and demand that ATI explain to use why the DX8 features that we were promised are not working, and apparently, they are not trying to get them to work.

I know now that I'm never ever buying an ATI product again, unless things change dramatically. And I know that that's not going to happen.

They don't work on the GeForce either. At least our EMBM works.

[DuSTman]

Remember that directX8 was only recently added to the drivers? DX8 support only first appeared in the 3100 drivers. Now, using a tweaker to enable the pixel shader, I am then able to play about with the pixel shader instructions using an application that came with the DX8 sdk. It should be noted, that in Direct3D8 there are two versions of the pixel shaders defined: 1.0 and 1.1. At the moment the drivers only support 1.0. Big deal, 1.1 aint much better than 1.0, but if 3dmark uses 1.1 that would explain it's reporting the non-presence of the pixel shaders.

It isn't easy writing something as complex as driver sets for modern video cards, especially when there are elements as complex as programmable pixel shaders to worry about. I am absolutely confident we'll see support for version 1.1 pixelshaders in a future driver revision, after all, 1.1 and 1.0 are not very different, the main difference being that in 1.1 texture registers are read/write and read only in 1.0.

Just remember that dx8 support in the drivers is new and it's really suprising that they work this well. Stay calm, there's nothing really to worry about..

[Mael]

Well, i loved 3DMark2k, and love 2001 even better. But the lack of being able to see the nature shots which are in the demo, just kinda suck. I will be building a new machine here soon, and how ATI handles this fiasco will decide weather the machine i build had an ATI product in it or an nVidia card in it.

[Atarion]

I can see the nature parts of the demo (they look incredible), but not in the benchmark. Go figure!?!

[JasonM [ATI]]

Paddy, would you mind dropping me a line when you're going to quote me like that, mate? By quoting me out of context like that, it feels like you'd rather not get the straight story. When I make a post to the Advanced Developer forum on OpenGL.org, I assume a certain level of knowledge among the readers of that forum. When you randomly spray sub-quotes into general end-user forums without the surrounding context, it's really easy for people to misinterpret what was originally intended.

FullAction wrote:

What does that mean "it turned into a hairball"?

What this means is that prior to the final release of DirectX 8.0, Microsoft decided that it was better to expose the RADEON™'s and GeForce{2}'s extended multitexture capabilities via the extensions to SetTextureStageState() instead of via the pixel shader interface. There are various practical technical reasons for this. Much of the same math that can be done with pixel shaders can be done via SetTextureStageState(), especially with the enhancements to SetTextureStageState() in DirectX 8.0. At the end of the day, this means that DirectX 8.0 exposes 99% of what the RADEON™ can do in its pixel pipe without adding the complexity of a "0.5" pixel shader interface.

Additionally, you have to understand that the phrase "shader" is an incredibly ambiguous graphics term. Basically, we hardware manufacturers started using the word "shader" a lot once we were able to do per-pixel dot products (i.e. the RADEON / GF generation of chips). Even earlier than that, "ATI_shader_op" was our multitexture OpenGL extension on Rage128 (which was replaced by the multivendor EXT_texture_env_combine extension). QuakeIII has ".shader" files it uses to describe how materials are lit. These are just a few examples of the use of the word shader in the game industry (nevermind the movie production industry which uses many different types of shaders, including those used by Pixar's RenderMan).

With the final release of DirectX 8.0, the term "shader" has become more crystalized in that it is actually used in the interface that developers use to write their programs rather than just general "industry lingo." In DirectX 8.0, there are two versions of pixel shaders: 1.0 and 1.1. (Future releases of DirectX will have 2.0 shaders, 3.0 shaders and so on.) Because of what I stated earlier, RADEON™ doesn't support either of the pixel shader versions in DirectX 8.0. Some of you have tweaked the registry and gotten the driver to export a 1.0 pixel shader version number to 3DMark2001. This causes 3DMark2001 to think it can run certain tests. Surely, we shouldn't crash when you do this, but you are forcing the (leaked and/or unsupported) driver down a path it isn't intended to ever go. The chip doesn't support 1.0 or 1.1 pixel shaders, therefore you won't see correct rendering even if we don't crash. The fact that that registry key exists indicates that we did some experiments in the driver, not that we are half way done implementing pixel shaders on RADEON™. DirectX 8.0's 1.0 and 1.1 pixel shaders are not supported by RADEON™ and never will be. The silicon just can't do what is required to support 1.0 or 1.1 shaders. This is also true of GeForce and GeForce2.

There are no shipping products that support pixel shaders from any vendor. You'll have to wait for those.

j0harris wrote:

I am not encouraged by this statement. It does not put much confidence in ATI's driver team. i.e. "We can't get it to work right now, so we give up. You should use other features instead."

This is a misinterpretation of what I meant in my post to the OpenGL list. I hope it is clarified by my answers above. This has nothing to do with us not being able to get something to work, it has to do with the fact that a Radeon-specific "shader" interface is simply not in DirectX 8.0 and the 1.0 and 1.1 shader interfaces are not supported by the RADEON™ silicon.

-Jason