R420 could be more future proof than NV40?

@

@trapine

Gawd
Joined
Mar 2, 2003
Messages
775
Read this bit from Xbit labs and it strikes me as pretty intresting.It will be a turn around for ATI if X800 is infact more future proof than 6800 :p

This is only part of the conclusion at xbitlabs

http://www.xbitlabs.com/articles/video/display/farcry20b.html

We will have more benchs with AA and Aniso active later.....

"...It is important to point out that ATI’s RADEON X800 XT and PRO graphics cards handle extreme geometry load better compared to NVIDIA’s GeForce 6800 Ultra and GT, which may mean that from this point ATI’s visual processing units have higher future-proof than NVIDIA’s latest graphics processing units do.

Additionally, ATI’s RADEON X800 XT traditionally calculates complex pixel shaders faster than everything else available today
Click to expand...
 
I don't know why they would say that...especially when they contradict themselves in the very next paragraph:
"Additionally, ATI’s RADEON X800 XT traditionally calculates complex pixel shaders faster than everything else available today. Unfortunately, the RADEON X800 PRO still lags behind the competing GeForce 6800 GT product, whereas the RADEON 9800 XT does not seem to be a really strong rival for the GeForce 6800, at least, in FarCry."

So they're saying the X800 which is basically a revamped 9700/9800 core has more potential than Nvidia's latest yet still lags behind in performance/price range? Makes no sense considering the features the 6800 has that the X800 does not..
 
Nice to see one good line could be salvaged for ATI stalwarts out of that review, even if it makes no sense. :)

You want future proof, you want SM3.0, period.
 
Although I agree the more features = future proof, let's take a loot at FX5200 vs. R9200.

Although the FX5200 supports DX9 it is completely useless and slower in real world use compared to the R9200.

I guess we just have to wait and see, don't we??
 
tranCendenZ said:
Nice to see one good line could be salvaged for ATI stalwarts out of that review, even if it makes no sense. :)

You want future proof, you want SM3.0, period.
Click to expand...

You mean SM3/2.1/2.015/2.0abcd etc.....
 
Ultra Wide said:
Although I agree the more features = future proof, let's take a loot at FX5200 vs. R9200.

Although the FX5200 supports DX9 it is completely useless and slower in real world use compared to the R9200.

I guess we just have to wait and see, don't we??
Click to expand...


LOL love the FX5200 compared to a R9200....classic

DASHlT
 
games in the future = more geometry = ati rendering faster then nv

was that really that hard to get from that quote by the thread starter ?
 
if the xt was going to be faster in future game why is it so slow in doom 3 or are you considering that an abberation? ati pulled a 3dfx and screwed themselves rehashing the same technology over and over and doing nothing but so minor changes.
 
games in the future = more geometry = ati rendering faster then nv

was that really that hard to get from that quote by the thread starter ?
Click to expand...

Looks like it is around hear at the moment with all this NV r00lz Doom3 Shit happening.Roll on X800XTPE to bring some sanity back :eek:

if the xt was going to be faster in future game why is it so slow in doom 3 or are you considering that an abberation? ati pulled a 3dfx and screwed themselves rehashing the same technology over and over and doing nothing but so minor changes.
Click to expand...

Do you really need a refresher of this again :rolleyes: 1word OpenGL is CURRENTLY better in Nvidia.NOT D3D!
And its 2.8FPS at that Not 20FPS like some people are Imagining?=Not imposable to bridge By a longshot.
 
Spank said:
games in the future = more geometry = ati rendering faster then nv

was that really that hard to get from that quote by the thread starter ?
Click to expand...

yeah, basically what it's just saying is it is more future proof (going to be faster in the future if all things stay the same on nvidias side), but as of right now it's lagging behind. - that's what i got out of it anyway.
 
NOTHING is really future-proof

I don't care who makes it or what it is. Eventually it will be replaced (in the future) with something more efficient, more powerful, or shinier.

So stop the pissing contest and buy me strawberry-flavored quik!!!!!!!!!!!!!!
 
Tenchi4U said:
NOTHING is really future-proof

I don't care who makes it or what it is. Eventually it will be replaced (in the future) with something more efficient, more powerful, or shinier.

So stop the pissing contest and buy me strawberry-flavored quik!!!!!!!!!!!!!!
Click to expand...

chocolate > * noob :mad:
 
Ultra Wide said:
Although I agree the more features = future proof, let's take a loot at FX5200 vs. R9200.

Although the FX5200 supports DX9 it is completely useless and slower in real world use compared to the R9200.

I guess we just have to wait and see, don't we??
Click to expand...

Except thats a totally invalid comparison since by all acounts the 6800 series and the X800 series are evenly matched (each wins some games and loses some games--but overall they are close to even in speed).

So since speed is essentially equal, the only thing left to compare is features.

SM3.0 and SLI for nvidia

Fast SM2.0 and lower power consumption for ATi.

And I think given the options, most gamers are going to go with Nvidia as the +'s on ATis side arent really futureproofing anything.
 
@trapine said:
Looks like it is around hear at the moment with all this NV r00lz Doom3 Shit happening.Roll on X800XTPE to bring some sanity back :eek:

Do you really need a refresher of this again :rolleyes: 1word OpenGL is CURRENTLY better in Nvidia.NOT D3D!
And its 2.8FPS at that Not 20FPS like some people are Imagining?=Not imposable to bridge By a longshot.
Click to expand...

Do your posts ever have a point other then trolling with useless drivel? lol

You call being an ATI fanboy forum sanity? Go to Rage3D if thats what your after.

Dependent Texture Limit
2.0b = 4
3.0 = No Limit

Position Register
2.0b = none
3.0 = Yes

Executed Instructions
2.0b = 512
3.0 = 65536

Interpolated Registers
2.0b = 2+8
3.0 = 10

Intstruction Predication
2.0b = none
3.0 = Yes

Indexed Input Registers
2.0b = none
3.0 = yes

Constant Registers
2.0b = 32
3.0 = 224

Arbitrary Swizzling
2.0b = none
3.0 = yes

Gradient Instructions
2.0b = none
3.0 = yes

Loop Count Register
2.0b = none
3.0 = yes

Face Register (2-sided lighting)
2.0b = none
3.0 = yes

Dynamic Flow Control Depth
2.0b = none
3.0 = 24

Minimum Required Full Precision Shader Capabilities
2.0b = FP24 (96-bit)
3.0 = FP32 (128-bit)
Click to expand...

Look at those specs and show me how PS 2.0b and FP24 is more futureproof than PS 3.0 and FP32.

Then look at these benchmarks using D3D Rightmark.

http://www.hexus.net/content/reviews/review.php?dXJsX3Jldmlld19JRD04MDQmdXJsX3BhZ2U9Nw==

The 6800u has better geometry processing performance than the X800XT PE.
 
I think the real problem for Ati at the moment is not really the performance of their cards but rather where they are positioned pricewise versus Nvidia's offerings. (as well as supply problems) Calling the X800pro the "pro" model you would expect it to perform just under the XT like in the previous generation yet it competes better with the 6800nu for a lower price. Nividia's mid-line GT not only outperforms the X800pro but competes well with the XTpe in early benches.

Ati simply needs to lower the price of the X800pro to around $300 to go head to head with the 6800nu and make a beefed up model to compete with the GT in the middle of the line. Bottom line is....any of these cards are going to be a step up from last generations fastest performers, they just need to adjust pricing (and maybe naming) for certain models.
 
Too bad you left out so much of the conclusion:
Even though ATI considerably improved performance of the RADEON X800-series graphics cards in FarCry’s “Pure Mode”, NVIDIA’s GeForce 6800 Ultra still can claim performance leadership here, as in quite a lot of cases speed delivered by the GeForce 6800 Ultra is a bit higher compared to the rivalling RADEON X800 XT.

...

To sum everything up, it looks like both ATI and NVIDIA have some performance reserves with their latest families of graphics chips. In the following 6-9 months until the new breed of graphics processors emerges, we are not only going to see a fierce battle between ATI’s RADEON X800 and NVIDIA’s GeForce 6 architecture, but probably are also going to witness the competition between Shader Model 2.0b and Shader Model 3.0 render paths in terms of performance and possibly quality.
Click to expand...

The x800 will be more future proof if MS cancels DX9.0c/SM3.0. :p
 
@trapine said:
Read this bit from Xbit labs and it strikes me as pretty intresting.It will be a turn around for ATI if X800 is infact more future proof than 6800 :p
Click to expand...

I noticed this comment when I read the review...and I can imagine ATi fans (nothing wrong with being a fan) would latch onto it...however...remember the part about how ATi couldn't execute enough shader instructions per pass to do 4 lights per pass? I think that negates the fact that thier geometry processing in farcry alone on beta everything has any indication of making the X800 more "future-proof"...if they are already hitting the shader limits of the X800 series within a couple months of its release...I want nothing to do with the card...
 
Whoa, dont you even dare say that Microsoft might cancel DX9.0c.

That would have Nvidia and every Nvidia fan fuming.

Although I'm sure when Microsoft planned SM3.0 32bitFP with partial precision support (AKA 16bitFP) they never ever intended for both to be used at the same time.
 
ZenOps said:
Whoa, dont you even dare say that Microsoft might cancel DX9.0c.

That would have Nvidia and every Nvidia fan fuming.

Although I'm sure when Microsoft planned SM3.0 32bitFP with partial precision support (AKA 16bitFP) they never ever intended for both to be used at the same time.
Click to expand...

How do you know that?

P.S. FP32 = 128bit, FP16 = 64bit
 
coldfusion71 said:
if the xt was going to be faster in future game why is it so slow in doom 3 or are you considering that an abberation? ati pulled a 3dfx and screwed themselves rehashing the same technology over and over and doing nothing but so minor changes.
Click to expand...

The Geforce 4 was a rehash of the geforce 3 and it was successful
 
The going rumor is that Nvidia actually planned a triple precision mixture FP32, FP16 and INT12, all in one scene (so you could double the performance when doing FP16, and maybe have 5 or 6x the performance when doing INT12 calcs)

The original plans for DirectX never mentioned anything about mixing modes while rendering. Sort of like how your desktop 2D supports 16 and 32-bit colour but not both at the same time. I think the first initial "issues" with mixed precision is starting to show too, at least in the Farcry 1.2 patch.
 
bboynitrous said:
The Geforce 4 was a rehash of the geforce 3 and it was successful
Click to expand...

The geforce 4 series was supposed be a 6 month refresh because the fx line was delayed for more then a year.
 
I think they're referring to raw fillrate, which the ati parts have plenty of, significantly more than the nvidia parts.

Problem is harnessing that fillrate, where ATI seems to be dropping the ball. Their d3d benchmarks should be significantly better than nvidia---if we only went by fillrate, instead of marginally better.

It'll be interesting to see if/can/how they optimize their drivers for greater efficiency without sacrificing picture quality. You know who pointedly mentioned that colored mipmap trickery wasn't the way to go, lol.
 
bboynitrous said:
The Geforce 4 was a rehash of the geforce 3 and it was successful
Click to expand...
The GeForce 4 was as much of a rehash of the GeForce 3 as the X800 is a rehash of the 9800. Both have more pipelines, faster core and memory speed, more shaders units, upgraded shader support, etc. It's not a simple renaming/minor updates like "rehash" sounds like. I wouldn't call either a "rehash" anyways... "based-on" would more accurately reflect that it relied on an earlier design, but doesn't imply the upgrades are minimal.

3dfx did a simple rehash with the Banshee, but even the Voodoo3 had significant updates from the Voodoo2. (if you were responding to the other guy's comment)
 
bboynitrous said:
I could have sworn the Geforce 3 ti series was the refresh
Click to expand...
Just renaming and binning of the same NV20 core... but I think you're right. The Ti-200 and Ti-500 were the fall refreshes in 2001.

(rancor) And it was Xbox design work that delayed the NV25 (GF4 Ti), not the FX series. That started the pattern where NVIDIA began missing 6 month deadlines.
 
tornadotsunamilife said:
but the xbox uses a modified geforce 3 ti-500 no? so surely not too much hard work was put into it
Click to expand...

No just like the guy you quoted said:

The Xbox uses an NV25 (ie Geforce4 Ti) at 200MHz (50Mhz slower than a GF4 ti4200)

Considering it came out about 6 months before the GF4ti launch, alot of work was put into it--but they couldnt hit high yields with it yet. You have to realize that 1) GPUs are on the drawing board and planing cycle for about 18 months. This means Nvidia was designing and building the Geforce4 Titanium back in October-November of 2000 (the time that the Geforce1 was being launched).

Think about it, the Xbox2 is going to be launched soon and it has an R500 in it (a chip not due for almost another year).
 
tornadotsunamilife said:
but the xbox uses a modified geforce 3 ti-500 no? so surely not too much hard work was put into it
Click to expand...
The NV2A was built in parallel with the NV20 and even has some features that appeared later in the NV25. Normally NVIDIA (and I assume ATI) build the current and next generation chips in parallel.

The Xbox project did delay the NV25, which is why the fall 2001 refresh was so weak (same exact NV20 core). I can't think of another time where the 6 month refresh was the *exact same chip*: The Riva128->Riva128ZX was a newer core (8MB support, fixed precision problem), TNT->TNT2 had a changed core, Geforce256->GF2, NV11->NV17->NV18 (GF2MX/GF4MX & refresh for AGP4x/8x), ****GF3->GF3****, GF3 Ti ->GF4 Ti, GF4 Ti->GF4 Ti (NV28), NV28->FX (FX was also late), NV30->NV35/NV38, NV35/38->6800. The NV28 was a minor update to the NV25 (FX was late), but at least it wasn't the exact same core like the GF3 to GF3 Ti (all models were NV20). It was a whole year before the new core came out (NV25), excluding the NV2A for the Xbox.
 
joemama said:
It's actually O GNOES now so both of you are n00bs :p
Click to expand...

bu...bu..the website said....

anyway, i never knew the xenon was going to use the r500 although i did know it was going to be some form of an ati card (any chance of it being modified too)?
 
ZenOps said:
The going rumor is that Nvidia actually planned a triple precision mixture FP32, FP16 and INT12, all in one scene (so you could double the performance when doing FP16, and maybe have 5 or 6x the performance when doing INT12 calcs)

The original plans for DirectX never mentioned anything about mixing modes while rendering. Sort of like how your desktop 2D supports 16 and 32-bit colour but not both at the same time. I think the first initial "issues" with mixed precision is starting to show too, at least in the Farcry 1.2 patch.
Click to expand...

...what issues?
 
@trapine said:
Looks like it is around hear at the moment with all this NV r00lz Doom3 Shit happening.Roll on X800XTPE to bring some sanity back :eek:



Do you really need a refresher of this again :rolleyes: 1word OpenGL is CURRENTLY better in Nvidia.NOT D3D!
And its 2.8FPS at that Not 20FPS like some people are Imagining?=Not imposable to bridge By a longshot.
Click to expand...

Uhm, you're not talking about this, are you?
image.html

image.html
 
hordaktheman said:
Uhm, you're not talking about this, are you?
image.html

image.html
Click to expand...

rofl

I guess that dude didn't see the part where the GT outperformed the XTPE...cuz that was the important part IMHO...
 
@trapine said:
Read this bit from Xbit labs and it strikes me as pretty intresting.It will be a turn around for ATI if X800 is infact more future proof than 6800 :p
Click to expand...

LOL.. let the excuses and justification begin. I also fell a whole load of "yeah but this card does this and that....". :rolleyes:
 
You must log in or register to reply here.
Back
Top