First REAL FX Benchmarks

[MuFu]

I bet the dustbuster amplifies the noise slightly... there must be a sort of resonating chamber/horn effect occuring.

MuFu.

 

[KimB]

I bet the dustbuster amplifies the noise slightly... there must be a sort of resonating chamber/horn effect occuring.

It's easy to prevent that by just having the dimensions of the chamber such that no frequency produced by the fan resonates.

 

[martrox]

Well, the BS is starting to bubble up to the top...... Quote from Kyle at HardOCP:

Solid apples-to-apples comparison on the way (in English) all done on an AMD system. While the TC "review" does have some merit to it, I dont think they really got into the meat of the GFFX.


__________________

Kyle "FrgMstr" Bennett
Webmonger @ [H]ard|OCP
Purveyor of Smoothness @ Ratpadz.Com

Let the spin doctoring begin......

 

[MuFu]

?

[H] reviews have been excellent of late, IMO. I don't see why it would be a bad thing if they reveal it isn't such a disappointment after all. :?

MuFu.

 

[MuFu]

Unless when he refers to "meat" he means smelly old dog food that's been left out for 3 months because poor Fido snuffed it due to food poisoning, having ingested the aforementioned foodstuff.

MuFu.

 

[Ante P]

1. ATI product is more quiet
2. ATI product offers better AA implementation
3. ATI product supports higher A.F. levels
4. ATI product is cheaper
5. ATI product historically has better 2D quality

Well, I have a number of reasons to want an FX over the R350:

1. Linux support. I need to work in Linux, and currently I can't on an ATI card.

Do you mean that you can't install the drivers from Ati.com or what?

 

[MuFu]

 

[Grall]

Doom,

I seriously doubt the FX fan draws 5-6W, it may spin fast but it isn't very large in diameter. Also it's a radial fan which means the fins are very small, hence it does less work than a propeller-like axial fan found on CPU coolers and such.

I'd be surprised if the fan draws more than half of what you state...


*G*

 

[KimB]

Do you mean that you can't install the drivers from Ati.com or what?

It's an agpgart issue. It's not entirely ATI's fault, but that doesn't help any.

 

[LittlePenny]

The ArtX team looks better and better everyday.

If that FXFlow were air tight, I would be fun to pour some water down there and watch it shoot out like a motor boat.

 

[MuFu]

To be honest I reckon about 45W of that 75W draw is the ASIC itself. I can't see how the fan, framebuffer and I/O are going to consume more than 30W.

MuFu.

 

[Doomtrooper]

?

[H] reviews have been excellent of late, IMO. I don't see why it would be a bad thing if they reveal it isn't such a disappointment after all. :?

MuFu.

Thats because of Brent doing the majority of the reviews.

 

[MuFu]

Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice! Justice!

MuFu.

P.S. Sorry to all the people hangin' on F5 expecting a constructive post.

 

[Sabastian]

Yeah I really do expect to see better reviews of the GeforceFX then this one. Of course I may be wrong but all we can do is wait and see. I would take this review with a grain of salt. There has to be more in it then what we’ve seen so far, but I am pretty sure that it won't be more then marginally faster over all.

 

[Doomtrooper]

Hmmmm a smile challenger I see

Lets get ready to rumble

 

[Doomtrooper]

Doom,

I seriously doubt the FX fan draws 5-6W, it may spin fast but it isn't very large in diameter. Also it's a radial fan which means the fins are very small, hence it does less work than a propeller-like axial fan found on CPU coolers and such.

I'd be surprised if the fan draws more than half of what you state...


*G*

Ya it may be a bit high, depending on RPM some small Delta fans draw as much as 7 watts...just a estimation

 

[Natoma]

I know it's a bit early, but does anyone have any clue when the R400 will be out? The reason I'm asking is that I currently have a GF2 and a P3-1Ghz.

Every 3 years I scrap all the current parts of my machine and I start from scratch. So in 2000 I built a new machine around a P3, GF2, 21" Hitachi CM814, and 512MB of PC133, and have upgraded incrementally since then.

This year I figure on getting a dual LCD or OLED 21", Clawhammer (maybe dual), and R350/R400, along with Serial ATA HD's in RAID array, which I want to last me for another 2-3 years as my base platform.

So again, any clues on R400 debut?

 

[Althornin]

Well, I have a number of reasons to want an FX over the R350:

1. Linux support. I need to work in Linux, and currently I can't on an ATI card.
2. Anisotropic implementation. I really do not like texture aliasing, of which I have noticed more on the 9700 Pro, and the blurring of text in some games is just as bad.
3. Various visual quality issues on the 9700, most likely related to drivers, in various games: z-buffer errors, color banding.

1) is not entirely ATI's fault
2) blurring of text WILL OCCUR if your FORCE ON aniso. There is no magic driver that knows "hey, this texture is text!"
3) The "z-buffer" error you speak of will be fixed in the next driver release, as per Sireric over at Rage3d. Its not actually a zbuffer error, its a driver issue with their T&L engine - the SW taht recodes fixed function as VS has a bug, irrc. Here is the thread : http://www.rage3d.com/board/showthr...adid=33661552&perpage=20&pagenumber=3
and relevant quote:

Regarding Morrowind: It's a false accusation in this case that it's a "Z-buffering" issue. It appears like that, but it has nothing to do with the Z buffering resolution. In fact, our resolution is oustanding, much better resolution than most products out there.

However, there was a SW glitch in our fixed function TCL unit (it's now just SW code in our vertex processors) when doing clipping that caused Z errors to be introduced, per polygon. This was identified and fixed in the next driver. It affected multiple programs, and caused "z-looking problems" and some texture issues as well. Beyond Morrowind, there is Mafia and a bunch of other apps, that had minor visual problems, which are now going to be fixed.

So, what are your reasons again

 

[Hellbinder]

2. Anisotropic implementation. I really do not like texture aliasing, of which I have noticed more on the 9700 Pro, and the blurring of text in some games is just as bad.
3. Various visual quality issues on the 9700, most likely related to drivers, in various games: z-buffer errors, color banding.

There is NO texture Aliasing with the Quality mode of the 9700pro. The only game i have seen texture aliasing is BF 1942 and it is the same on ALL hardware. Blurng text??? name the game..

as for your other point.. GIVE ME A BREAK.

You are digging, I mean seriously digging to the center of the earth to find something negative to complain about. And in some cases you are flat out MAKING IT UP. Then you are trying to play it as if they are Huge enough to not warrant buying the card. When you dont have the first clue as to what driver issues the GFFX may or may not have.

 

[Millennium]

MuFu wrote:
?

[H] reviews have been excellent of late, IMO. I don't see why it would be a bad thing if they reveal it isn't such a disappointment after all.

MuFu.


Thats because of Brent doing the majority of the reviews.

Exactly ...