View Single Post
Old 07-04-08, 09:51 PM   #15
Xion X2
Registered User
Xion X2's Avatar
Join Date: May 2006
Location: U.S.
Posts: 6,701
Default Re: HD 4870 vs GTX 260

Originally Posted by SH64 View Post
There is no point of that FS bench since it dosen't show the min.fps
Since when does a minimum framerate not factor into an average?

its obvious & logical that the GTX260's min.fps would've been higher if it was in that benchmark i posted (due to its higher Video Memory).
No, it wouldn't, and here's why.

Frame-buffer and overall memory bandwidth have the same effect and are interrelated. In fact, you could think of memory bandwidth as composed of 3 key components:

1) Frame buffer
2) Bus width
3) VRAM speed

Each of these plays a role in boosting peak memory bandwidth to speed up texture loading/streaming.

As of 2007, advanced personal computers and graphics cards use even more combined buses than dual-channel, and combine four (e.g. Mac Pro), five (e.g. nVidia 8800GTS), six (e.g. nVidia 8800GTX) [b]or more sets of 64-bit memory modules and buses to reach 256-bit, 320-bit, 384-bit or greater total memory bandwidth.

The point of the entire exercise of using a large frame buffer is to increase bandwidth, but if a card already has it due to faster VRAM (such as 4870 with GDDR5 instead of GDDR3) then it doesn't need as large a frame buffer. See how the 4870 has just a little higher bandwidth than the 260 does despite the smaller frame buffer:

i7-2700k @ 5.0 GHz
Nvidia GeForce 570 2.5GB Tri-SLI
Asus P67 WS Revolution (Tri-SLI)
OCZ Vertex SSD x 4 (Raid 5)
G.Skill 8GB DDR3 @ 1600MHz
PC Power & Cooling 950W PSU
Xion X2 is offline