View Single Post
Old 01-07-09, 04:39 AM   #5
walterman's Avatar
Join Date: Oct 2003
Posts: 1,525
Talking Re: ok Whats better on a graphics card Texture fill rate or memory bandwith

Bandwidth !!!

Texture Fill Rate = #TMUs x Core Clock

All your TMUs could become useless if they are not feeded with the texels from memory that they need.

Why do you think that we use texture compression ? Because you save 1:4 or 1:8 of the needed bandwidth (& space of course), and your TMUs can reach their maximum theoretical load quicker.

Also, the TMUs have access to a very efficient L1 & L2 cache (fully associative), precisely to save bandwidth.

The problem with the modern games is the heavy use of shading power. But, in older titles, with more simple shading programs, the situation is inverse, and the bandwidth is what matters.

Sucks, because bandwidth isn't growing as fast as the shading power, and we still cannot use SSAA with the older titles.

Hope this helps
ASUS Rampage Formula X48 | Xeon 3350 @ 3.6 GHz (450x8/1.26v) | 4x1GB OCZ DDR2 PC2-6400 Reaper CL3 @ 900 MHz 3-4-4-15 | 1 x eVGA GTX 285 SSC | 1 x ASUS EN8800GTX (PhysX/CUDA -> Burnt by nVidia 196.75 driver) | X-Fi Titanium Fatal1ty PCIe | 1 x Intel X25-M G2 80GB | 2 x 750GB WD RE2 7500AYYS SATA2 16MB | Samsung SH-B083L SATA | Enermax Revolution 1250W | Samsung SyncMaster 275T 27" 1920x1200 | Thermaltake Black Armor | BloodRayne 2 FSAA Patch
walterman is offline   Reply With Quote