yeah totally right there, but with the AthlonXP Barton 3000+ setup it was kinda of a reversed world for the NV chips, there were thoughts that a even an AthlonXP  Barton 3000+ isn't even powerfull enough for a V6K to render with.

I really hope that the AGP x8 Dual Opteron boards will support a Voodoo3 like you asked earlier, then the Voodoo 5 6000 could show it's upper sides.

hehe I know we all went offtopic as I did too oh well doesn't matter

I really wonder how ATi is gonna cool the Mobile R600 aka R600M, I mean 534 Mil transistor's is a heck of a lot to cool some even mentioned 617 Mil transistors

Or will they reduce the amount of Transistors like they did with the R420? aka M18 Mobillity Radeon 9800?

a X1800 XT Platinum Edition had an R420 with 160 Mil Transistors my M18 R420 has only 110 Mil Transistors, would the R600M get the same type of degrade?

@ elfuego

Read this before you complain
http://www.bjorn3d.com/read.php?cID=264

and stop polluting this topic, if the R600 isn't your thing, Please do not intefere

Quote:

Chill out a bit I am not complaining; R9600 is indeed better then ti4200, but please do take a look at this: http://users.vtkom.net/~benchmark/

in fact, the only interesting part is this:

http://users.vtkom.net/~benchmark/html/gf4ti4200.htm //overclocked ti4200 on a AthlonXP Tbred B 2.5Ghz (16605 marks)
http://users.vtkom.net/~benchmark/html/radeon9600.htm //overclocked R9600 XT on a Athlon 64, Newcastle, 2.7Ghz (16760 marks)

Of course, in newer 3Dmark 03, with the lack of shader model 2.0 support, ti4200 is giving away:
http://users.vtkom.net/~benchmark/html/gf4ti4200.03.htm // 2132 marks
http://users.vtkom.net/~benchmark/html/radeon9600.03.htm //5355 marks

But it's still one hell of a card and you know it. The difference between the cards varies from "miserable" (oldies) to "not that great" in actuall games. Before you dig out more links check this out too: http://www.digital-daily.com/video/vga-roundup/

And I wouldn't call a little NV-ATI-3DFX chatter pollution. But it is a bit offtopic. If you think that it should not be here, then let this (and all other offtopics) be deleted or moved. I will do it myself if you only but ask. Cheers friend, no harm done 

BTW here ATi's new logo 8)
http://www.3dfxzone.it/news/puntatore.php?uid=4046

The old one:


The new one


The new one has more style, though the old has the nice chrome boarder around it

the naming would be like the following:

Enthousiast:

X2800 XTX
X2800XT

High End:

X2800 XL
X2800 Pro

Mainstream:

X2600XT
X2600 Pro

Budget stream:

X2300 XT
X2300 LE

The refresh of the AMD.ATI R600 would most likely be the R650/R680

which would create the following groupings with product names:


Enthousiast:

X2900 XTX
X2900XT

High End:

X2900 Pro
X2900 GT

Mainstream:

X2700XT
X2700 Pro

Budget stream:

X2350 XT
X2350 LE

well got this info from the dutch ATI topic

Quote:


it seemed to be one of the latest stories about the R600 and it's memory support, layout and memory capacity.

Well since my computerclub play's F.E.A.R. as main game my X1900 XTX + X1950 XTX CFE setup is fast enough, no need for a R600 for that game hehe.


Only I wonder how it's DX9.0c performance will be with FLight Simulator X Which runs very sweet on my R580/R580+ CFE set but if a single R600 can do it alot better, I wonder how much that will differ to my R580+R580+ CFE setup, since that single R600 would consume alot lesser energy than the dual R580+R580+ CFE setup I use now also.

And also R600 is DX10.1 aka DX10.0b compliant instead of DX10.0a which the G80 only is. PWNed 

ATi always had the upper hand in designing Ultra high end chips, thier R580 has about 9x the physical rendering power as that of a single G71 GTX and thier RV560 [X1600XT] has about 4 x the physical rendering power of a G71 GTX, so on the Physics side ATi has a greater upper hand than NV,, just imagine what a R600 would do, dual R580 or R580+ would do about 16 to 18 times the Physical rendering power of a G71 GTX, that's rendering Pure Physics ony hehe.


Also interesting

    * 64 4-Way SIMD Unified Shaders, 128 Shader Operations/Cycle
    * 32 TMUs, 16 ROPs
    * 512 bit Memory Controller, full 32 bit per chip connection
    * GDDR3 at 900 MHz clock speed (January)
    * GDDR4 at 1.1 GHz clock speed (March, revised edition)
   * Total bandwidth 115 GB/s on GDDR3 aka X2800 XT ???
    * Total bandwidth 140 GB/s on GDDR4 aka X2800 XTX ???
    * Consumer memory support 1024 MB
    * DX10 full compatibility with draft DX10.1 vendor-specific cap removal (unified programming)
    * 32FP [sic] internal processing
    * Hardware support for GPU clustering (any x^2 [sic] number, not limited to Dual or Quad-GPU)
    * Hardware DVI-HDCP support (High Definition Copy Protocol)
    * Hardware Quad-DVI output support (Limited to workstation editions)
    * 230W TDP PCI-SIG compliant

there will be a GDDR3 model with 115 Gb per sec equiped with 512Bit GDDR3 and a 140 GB per sec equiped  512Bit GDDR4 model, I suppose that the GDDR3 model is the X2800XT and the GDDR4 model the X2800 XTX, just a rough guess