Updated List of Video Card GPU OVERALL Performance VP Ratings - TITAN update!

Toggle sidebar Toggle sidebar
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
Here is a gauge of overall graphics performance for as broad of a library of modern PC games as possible -- with Voodoopower being a unit similar to the horsepower rating for cars, yet much more comprehensive, like how a car would perform on many different racing tracks, put together for an average!

It is continually updated and refined, thanks to further sources from hardware review sites and forum enthusiasts alike.

(**For a quick search, just press Ctrl+F and then type in whatever brand/number you're interested in, and your browser will highlight it for you.**)


VP = Voodoopower

___Tier AAA Stupid High-End --(For somebody who cannot get enough out of it)
.........reserved for upcoming dual-GK110 card, and hopeful future release by AMD :biggrin:

___Tier AA Super High-End --(Plays like a wicked beast)
  1. 128%-- Radeon Ares II HD 7970 GHz Edition X2 6GB (DX11.1) -- 401 VP*
  2. 121%-- Geforce GTX 690 4GB (DX11) -- 380 VP
  3. 113%-- Radeon HD 7990 6GB (DX11.1) -- 353 VP* --- (* NEW ENTRY! *)
  4. 110%-- Geforce GTX TITAN 6GB (DX11) -- 343 VP*
  5. 100%-- Geforce GTX 780 3GB (DX11) -- 313 VP* --- (* NEW ENTRY! *)
    (times 9% over Tier A's 100%)

    Tier A Upper High-End --(Plays like a tamed beast)
  6. 100%-- Geforce Mars II GTX 580 x2 3GB (DX11) -- 287 VP
    (times 19% over Tier B's 100%)

    Tier B Mid High-End --(Plays most games at maxed-out settings @ 2560x1600 w/ 4x AA)
  7. 108%-- Geforce GTX 770 2GB (DX11) -- 257 VP* --- (* NEW ENTRY! *)
  8. 107%-- Radeon HD 7970 GHz Edition 3GB (DX11.1) -- 254 VP
  9. 104%-- Radeon R9 280X 3GB (DX11.1) -- 246 VP --- (* NEW ENTRY! *)
  10. 100%-- Geforce GTX 680 2GB (DX11) -- 237 VP
    (times 23% over Tier C's 100%)

    Tier C Lower High-End --(Plays most games at maxed-out settings @ 2560x1440)
  11. 120%-- Radeon HD 6990 4GB (DX11) -- 232 VP*
  12. 119%-- Geforce GTX 590 3GB (DX11) -- 230 VP
  13. 118%-- Radeon HD 7970 3GB (DX11.1) -- 228 VP
  14. 111%-- Geforce GTX 670 2GB (DX11) -- 214 VP
  15. 110%-- Radeon HD 7950 Boost Edition 3GB (DX11.1) -- 212 VP
  16. 109%-- Radeon Ares HD 5870X2 4GB (DX11) -- 210 VP*
  17. 107%-- Geforce GTX 560 Ti 2Win 2GB (DX11) -- 207 VP
  18. 106%-- Geforce GTX 760 2GB (DX11) -- 204 VP* --- (* NEW ENTRY! *)
  19. 100%-- Radeon HD 7950 3GB (DX11.1) -- 193 VP
    (times 12% over Tier D's 100%)

    Tier D Upper Mid-Range --(Plays most games at nearly maxed-out settings @ 2560x1440)
  20. 110%-- Radeon HD 7870 ("XT" / Tahiti LE / Boost Edition) 2GB (DX11.1) -- 190 VP --- (* NEW ENTRY! *)
  21. 109%-- Radeon HD 6870X2 2GB (DX11) -- 188 VP*
  22. 109%-- Geforce GTX 660 Ti 2GB (DX11) -- 187 VP
  23. 109%-- Radeon R9 270X 2GB (DX11.1) -- 187 VP --- (* NEW ENTRY! *)
  24. 106%-- Geforce GTX 580 3GB (DX11) -- 181 VP
  25. 105%-- Geforce GTX 580 1.5GB (DX11) -- 180 VP
  26. 105%-- Radeon HD 5970 4GB (DX11) -- 180 VP*
  27. 102%-- Radeon HD 5970 2GB (DX11) -- 175 VP*
  28. 100%-- Radeon HD 7870 GHz Edition 2GB (DX11.1) -- 172 VP
    (times 22% over Tier E's 100%)

    Tier E Mid-Upper Mid-Range --(Plays most games at nearly maxed-out settings @ 1920x1080 w/ 4x AA)
  29. 118%-- Geforce GTX 460 2Win 2GB (DX11) -- 167 VP
  30. 114%-- Geforce GTX 660 2GB (DX11) -- 161 VP --- (* NEW ENTRY! *)
  31. 111%-- Geforce GTX 570 2.5GB (DX11) -- 156 VP
  32. 110%-- Radeon HD 6970 2GB (DX11) -- 156 VP
  33. 109%-- Geforce GTX 570 1.25GB (DX11) -- 155 VP
  34. 109%-- Geforce GTX 480 1.5GB (DX11) -- 154 VP
  35. 103%-- Geforce GTX 560 Ti-448 1.25GB (DX11) -- 146 VP
  36. 102%-- Geforce GTX 650 Ti Boost 2GB (DX11) -- 144 VP --- (* NEW ENTRY! *)
  37. 100%-- Radeon HD 7850 2GB (DX11.1) -- 141 VP
    (times 37% over Tier F's 100%)

    Tier F Mid-Lower Mid-Range --(Playable at up to 1920x1080 with high settings in most games)
  38. 136%-- Radeon HD 7850 1GB (DX11.1) -- 139 VP --- (* NEW ENTRY! *)
  39. 133%-- Radeon HD 6950 2GB (DX11) -- 136 VP
  40. 131%-- Radeon HD 5870 2GB (DX11) -- 134 VP
  41. 131%-- Radeon HD 6950 1GB (DX11) -- 134 VP
  42. 129%-- Radeon HD 5870 1GB (DX11) -- 132 VP
  43. 128%-- Geforce GTX 560 Ti 2GB (DX11) -- 131 VP
  44. 128%-- Geforce GTX 295 1.75GB (DX10) -- 131 VP
  45. 126%-- Geforce GTX 560 Ti 1GB (DX11) -- 129 VP
  46. 122%-- Radeon HD 6930 1GB (DX11) -- 125 VP
  47. 119%-- Geforce GTX 470 1.25GB (DX11) -- 123 VP
  48. 119%-- Radeon HD 7790 1GB (DX11.1) -- 122 VP --- (* NEW ENTRY! *)
  49. 119%-- Radeon HD 6870 2GB (DX11) -- 122 VP
  50. 119%-- Radeon HD 4870X2 2GB (DX10.1) -- 122 VP*
  51. 118%-- Geforce GTX 560 2GB (DX11) -- 121 VP
  52. 117%-- Radeon HD 6870 1GB (DX11) -- 120 VP
  53. 116%-- Geforce GTX 560 1GB (DX11) -- 119 VP
  54. 111%-- Geforce GTX 650 Ti 2GB (DX11) -- 114 VP --- (* NEW ENTRY! *)
  55. 109%-- Geforce GTX 650 Ti 1GB (DX11) -- 112 VP --- (* NEW ENTRY! *)
  56. 109%-- Radeon HD 5850 1GB (DX11) -- 112 VP
  57. 103%-- Radeon HD 6850 1GB (DX11) -- 106 VP
  58. 103%-- Geforce GTX 460 2GB (DX11) -- 106 VP
  59. 101%-- Geforce GTX 460 1GB (DX11) -- 104 VP
  60. 100%-- Geforce GTX 465 1GB (DX11) -- 103 VP
    (times 41% over Tier G's 100%)

    Tier G Lower Mid-Range --(Playable at up to 1920x1080 with medium-high settings in most games)
  61. 141%-- Geforce GTX 285 2GB (DX10) -- 103 VP
  62. 138%-- Geforce GTX 285 1GB (DX10) -- 101 VP
  63. 134%-- Geforce GTX 460 v2 1GB (192-bit) (DX11) -- 98 VP
  64. 132%-- Radeon HD 7770 GHz Edition 2GB (DX11.1) -- 96 VP
  65. 129%-- Radeon HD 7770 GHz Edition 1GB (DX11.1) -- 94 VP
  66. 128%-- Geforce GTX 275 1.75GB (DX10) -- 94 VP
  67. 126%-- Geforce GTX 460 768MB (DX11) -- 92 VP
  68. 125%-- Geforce GTX 280 1GB (DX10) -- 92 VP
  69. 125%-- Geforce GTX 275 896MB (DX10) -- 92 VP
  70. 124%-- Radeon HD 4850X2 2GB (DX10.1) -- 91 VP*
  71. 123%-- Geforce GTX 460 SE 1GB (DX11) -- 90 VP
  72. 121%-- Radeon HD 5830 1GB (DX11) -- 88 VP
  73. 121%-- Radeon HD 4890 2GB (DX10.1) -- 88 VP
  74. 119%-- Radeon HD 4890 1GB (DX10.1) -- 87 VP
  75. 118%-- Radeon HD 6790 1GB (DX11) -- 86 VP
  76. 116%-- Radeon HD 4850X2 1GB (DX10.1) -- 85 VP*
  77. 114%-- Geforce GTX 560 SE / GT 645 1GB (DX11) -- 83 VP
  78. 110%-- Geforce GTX 260-216 896MB (DX10) -- 81 VP
  79. 106%-- Geforce GTX 550 Ti 1GB (DX11) -- 78 VP
  80. 106%-- Radeon HD 4870 1GB (DX10.1) -- 78 VP
  81. 104%-- Radeon HD 5770 / 6770 1GB (DX11) -- 76 VP
  82. 103%-- Geforce GTX 650 1GB (DX11) -- 75 VP --- (* NEW ENTRY! *)
  83. 102%-- Geforce 9800GX2 1GB (DX10) -- 74 VP
  84. 100%-- Radeon HD 7750 1GB (DX11.1) -- 73 VP
  85. 100%-- Geforce GTX 260 896MB (DX10) -- 73 VP
  86. 100%-- Radeon HD 4870 512MB (DX10.1) -- 73 VP
    (times 46% over Tier H's 100%)

    Tier H Upper Low-End --(Playable at 1680x1050 with default settings in most games)
  87. 140%-- Geforce GT 640 1GB (OEM, GDDR5) (DX11) -- 70 VP --- (* NEW ENTRY! *)
  88. 126%-- Geforce GTS 450 1GB (DX11) -- 63 VP
  89. 123%-- Radeon HD 5750 / 6750 1GB (DX11) -- 61 VP
  90. 121%-- Radeon HD 4860 1GB (DX10.1) -- 60 VP
  91. 118%-- Radeon HD 4850 1GB (DX10.1) -- 59 VP
  92. 118%-- Geforce GTS 250 1GB (DX10) -- 59 VP (beware of a plethora of underclocked versions)
  93. 114%-- Geforce 8800 Ultra 768MB (DX10) -- 57 VP
  94. 113%-- Geforce GT 545 1GB (GDDR5) (DX11) -- 56 VP
  95. 113%-- Radeon HD 4850 512MB (DX10.1) -- 56 VP
  96. 113%-- Geforce 9800GTX+ / GTS250 512MB (DX10) -- 56 VP
  97. 107%-- Geforce 8800GTX 768MB (DX10) -- 53 VP
  98. 106%-- Radeon HD 3870X2 1GB (DX10.1) -- 53 VP*
  99. 106%-- Geforce 9800GTX 512MB (DX10) -- 53 VP
  100. 104%-- Geforce 8800GTS 1GB (G92) (DX10) -- 52 VP
  101. 103%-- Radeon HD 4770 512MB (DX10.1) -- 51 VP
  102. 101%-- Geforce GTS 240 1GB (G92) (DX10) -- 50 VP
  103. 100%-- Geforce 8800GTS 512MB (G92) (DX10) -- 50 VP
    (times 25% over Tier I's 100%)

    Tier I Mid-Upper Low-End --(Playable at default settings, lower resolution with 8x AF in most games)
  104. 120%-- Radeon HD 6670 1GB (DX11) -- 48 VP
  105. 120%-- Geforce GT 545 / GT 640 1.5GB (DDR3 192-bit) (DX11) -- 48 VP
  106. 119%-- Radeon HD 4830 512MB (DX10.1) -- 48 VP
  107. 117%-- Radeon HD 4730 512MB (DX10.1) -- 47 VP
  108. 115%-- Geforce GT 640 2GB (DDR3) (DX11) -- 46 VP --- (* NEW ENTRY! *)
  109. 112%-- Geforce 8800GT / 9800GT 1GB (DX10) -- 45 VP
  110. 110%-- Radeon HD 7750 2GB, 4GB (DDR3) (DX11.1) -- 44 VP --- (* NEW ENTRY! *)
  111. 108%-- Geforce 8800GTS-112 640MB (DX10) -- 43 VP
  112. 107%-- Geforce 8800GT / 9800GT 512MB (DX10) -- 43 VP
  113. 100%-- Radeon HD 5670 1GB (GDDR5) (DX11) -- 40 VP
  114. 100%-- Radeon HD 6570 1GB (GDDR5) (DX11) -- 40 VP
    (times 29% over Tier J's 100%)

    Tier J Mid Low-End --(Playable at lower settings & resolution with 8x AF in most 2011 games)
  115. 126%-- Radeon HD 5670 512MB (GDDR5) (DX11) -- 39 VP
  116. 124%-- Geforce 9600GT 1GB (DX10) -- 38 VP
  117. 123%-- Radeon HD 6670 2GB (DDR3) (DX11) -- 38 VP
  118. 121%-- Radeon HD 2900XT 1GB (GDDR4) (DX10) -- 37 VP
  119. 121%-- Geforce 9600GT 512MB (DX10) -- 37 VP
  120. 118%-- Geforce 8800GTS 640MB (DX10) -- 36 VP
  121. 117%-- Radeon HD 2900XT 512MB (DDR3) (DX10) -- 36 VP
  122. 116%-- Radeon HD 3870 512MB (DX10.1) -- 36 VP
  123. 114%-- Geforce GT 440 / GT 630 1GB (GDDR5) (DX11) -- 35 VP
  124. 114%-- Geforce 8800GT 256MB (DX10) -- 35 VP
  125. 112%-- Geforce GT 240 1GB (GDDR5) (DX10.1) -- 35 VP
  126. 112%-- Radeon HD 6570 1GB (DDR3) (DX11) -- 35 VP
  127. 109%-- Geforce GT 240 512MB (GDDR5) (DX10.1) -- 34 VP
  128. 106%-- Radeon HD 4670 1GB (DX10.1) -- 33 VP
  129. 106%-- Radeon HD 3850 512MB (DX10.1) -- 33 VP
  130. 106%-- Geforce 8800GTS 320MB (DX10) -- 33 VP
  131. 105%-- Radeon HD 5570 1GB (GDDR5) (DX11) -- 33 VP
  132. 105%-- Geforce 8800GS / 9600GSO 384MB (DX10) -- 33 VP
  133. 104%-- Radeon HD 4670 512MB (DX10.1) -- 32 VP
  134. 102%-- Radeon HD 2900Pro 512MB (DX10) -- 32 VP
  135. 102%-- Geforce 9600GSO 768MB (DDR2) (DX10) -- 32 VP
  136. 101%-- AMD (Trinity APU) A10-series Radeon HD 7660D (dual channel DDR3-1866) (DX11) -- 31 VP --- (* NEW ENTRY! *)
  137. 100%-- Geforce GT 630 2GB (OEM, GK107-based) (DX11) -- 31 VP --- (* NEW ENTRY! *)
  138. 100%-- Geforce GT 440 / GT 630 1GB (DDR3) (DX11) -- 31 VP
    (times 27% over Tier K's 100%)

    Tier K Lower Low-End ----(Playable at lower settings & resolution with 8x AF in most 2010 games)
  139. 125%-- Geforce 9600GSO 512MB (G94) (DX10) -- 30 VP
  140. 123%-- Geforce GT 240 1GB (DDR3) (DX10.1) -- 30 VP
  141. 122%-- Radeon HD 5570 1GB (DDR3) (DX11) -- 30 VP
  142. 122%-- Geforce 7900GX2 1GB (DX9.0c) -- 30 VP
  143. 122%-- Geforce 7950GX2 1GB (DX9.0c) -- 30 VP
  144. 117%-- Radeon HD 3850 256MB (DX10.1) -- 29 VP
  145. 114%-- Geforce GT 430 / GT 530 1GB (DX11) -- 28 VP
  146. 112%-- Radeon X1950XTX 512MB (DX9.0c) -- 27 VP
  147. 110%-- Radeon X1900XTX 512MB (DX9.0c) -- 27 VP
  148. 109%-- Radeon HD 5550 1GB (GDDR5) (DX11) -- 27 VP
  149. 106%-- Radeon X1900XT 512MB (DX9.0c) -- 26 VP
  150. 105%-- Radeon HD 2900GT 256MB (DX10) -- 26 VP
  151. 103%-- Radeon X1950XT 256MB (DX9.0c) -- 25 VP
  152. 102%-- AMD (Llano APU) A8-series Radeon HD 6550D (dual channel DDR3-1866) (DX11) -- 25 VP
  153. 100%-- Radeon X1900XT 256MB (DX9.0c) -- 24.3 VP
    (times 62% over Tier L's 100%)

    Tier L Really Low-End --(Barely able to enjoy 4x AF with lowest settings and resolution in some 2010 games)
  154. 160%-- Geforce GT 520 / GT 610 / GT 620 (OEM) 1GB (DX11) -- 23.9 VP
  155. 154%-- Geforce GT 220 1GB (DX10.1) -- 23.1 VP
  156. 154%-- AMD (Trinity APU) A8-series Radeon HD 7560D (dual channel DDR3-1866) (DX11) -- 23.0 VP --- (* NEW ENTRY! *)
  157. 153%-- Geforce GT 220 / GT 315 512MB (DX10.1) -- 22.9 VP
  158. 152%-- Geforce 7900GTX 512MB (DX9.0c) -- 22.8 VP
  159. 146%-- Geforce 7900GTO 512MB (DX9.0c) -- 21.9 VP
  160. 145%-- Radeon HD 5550 512MB (DDR3) (DX11) -- 21.8 VP
  161. 138%-- Geforce 7800GTX 512MB (DX9.0c) -- 20.7 VP
  162. 136%-- Radeon X1950Pro 512MB (DX9.0c) -- 20.4 VP
  163. 135%-- Geforce 8600GTS 512MB (DX10) -- 20.3 VP
  164. 133%-- Geforce 7950GT 512MB (DX9.0c) -- 19.9 VP
  165. 132%-- Radeon X1950Pro 256MB (DX9.0c) -- 19.8 VP
  166. 131%-- Geforce 8600GTS 256MB (DX10) -- 19.7 VP
  167. 126%-- Geforce 9500GT 1GB (DX10) -- 18.9 VP
  168. 125%-- Geforce 9500GT 512MB (DX10) -- 18.8 VP
  169. 123%-- AMD (Trinity APU) A6-series Radeon HD 7540D (dual channel DDR3-1866) (DX11) -- 18.4 VP --- (* NEW ENTRY! *)
  170. 122%-- Radeon X1900 AIW 256MB (DX9.0c) -- 18.3 VP
  171. 120%-- AMD (Llano APU) A6-series Radeon HD 6530D (dual channel DDR3-1866) (DX11) -- 18.0 VP
  172. 119%-- Radeon HD 4650 512MB (DX10.1) -- 17.8 VP
  173. 115%-- Geforce 510 / 605 1GB (DX11) -- 17.3 VP
  174. 112%-- Geforce 7900GT 512MB (DX9.0c) -- 16.8 VP
  175. 111%-- Intel (Ivy Bridge APU) HD Graphics 4000 (dual channel DDR3-1866) (DX11) -- 16.7 VP --- (* NEW ENTRY! *)
  176. 111%-- Radeon X1800XT 512MB (DX9.0c) -- 16.7 VP
  177. 110%-- Radeon HD 5550 1GB (DDR2) (DX11) -- 16.5 VP
  178. 108%-- Radeon X1800XT 256MB (DX9.0c) -- 16.2 VP
  179. 107%-- Radeon HD 6450 1GB (DDR3) (DX11) -- 16.1 VP
  180. 107%-- Geforce 7900GT 256MB (DX9.0c) -- 16.1 VP
  181. 106%-- Geforce 7800GS+ 512MB (AGP) (DX9.0c) -- 15.9 VP
  182. 104%-- Geforce 8600GT 512MB (DX10) -- 15.6 VP
  183. 103%-- Geforce 7800GTX 256MB (DX9.0c) -- 15.4 VP
  184. 101%-- Radeon X1900GT 256MB (DX9.0c) -- 15.1 VP
  185. 100%-- Geforce 8600GT 256MB (DX10) -- 15.0 VP
    (times 50% over Tier M's 100%)

    Tier M Really, Really Low-End --(For most modern games other than PC-only games, use a gaming console!)
  186. 139%-- Radeon X1950GT 512MB (DX9.0c) -- 13.9 VP
  187. 137%-- Radeon X1900GT Rev2 256MB (DX9.0c) -- 13.7 VP
  188. 136%-- AMD (Trinity APU) A4-series Radeon HD 7480D (dual channel DDR3-1600) (DX11) -- 13.6 VP --- (* NEW ENTRY! *)
  189. 136%-- AMD (Llano APU) A4-series Radeon HD 6410D (dual channel DDR3-1600) (DX11) -- 13.6 VP
  190. 136%-- Geforce 7900GS 256MB (DX9.0c) -- 13.6 VP
  191. 135%-- Geforce 9500GT 512MB (DDR2) (DX10) -- 13.5 VP
  192. 123%-- Geforce 7800GT 256MB (DX9.0c) -- 12.3 VP
  193. 119%-- Radeon X1800XL 256MB (DX9.0c) -- 11.9 VP
  194. 117%-- Radeon HD 2600XT 256MB (DX10) -- 11.7 VP
  195. 113%-- Intel (Sandy Bridge APU) HD Graphics 3000 (dual channel DDR3-1866) (DX10.1) -- 11.3 VP
  196. 107%-- Radeon HD 3650 256MB (DX10.1) -- 10.7 VP
  197. 106%-- Radeon X1650XT 256MB (DX9.0c) -- 10.6 VP
  198. 105%-- Geforce 7600GT / 7600GTS 256MB (DX9.0c) -- 10.5 VP
  199. 104%-- AMD (Llano APU) E2-series Radeon HD 6370D (dual channel DDR3-1600) (DX11) -- 10.4 VP
  200. 102%-- Geforce 7800GS 256MB (DX9.0c) -- 10.2 VP
  201. 101%-- Radeon X1800GTO 256MB (DX9.0c) -- 10.1 VP
  202. 100%-- Geforce 6800 Ultra 256MB (DX9.0c) -- 10.0 VP
    (excluding the rest of older cards for now)

    Tier N Modern GPUs That Does Not Play Modern Games --(Worse than some integrated graphics--only to be for HTPC audio functions, etc..)
  203. 99%-- Intel (Ivy Bridge APU) HD Graphics 2500 (dual channel DDR3-1866) (DX11) -- 9.9 VP --- (* NEW ENTRY! *)
  204. 87%-- Radeon HD 2600Pro 256MB (DX10) -- 8.7 VP
  205. 75%-- Radeon HD 5450 512MB (DDR3) (DX11) -- 7.5 VP
  206. 73%-- Intel (Sandy Bridge APU) HD Graphics 2000 (dual channel DDR3-1600) (DX10.1) -- 7.3 VP
  207. 69%-- Radeon HD 4550 512MB (GDDR3) (DX10.1) -- 6.9 VP
  208. 52%-- Radeon HD 4350 512MB (DX10.1) -- 5.2 VP
  209. 51%-- Radeon HD 5450 512MB (DDR2) (DX11) -- 5.1 VP
  210. 50%-- Radeon HD 2400XT 256MB (DX10) -- 5.0 VP
  211. 49%-- Geforce 210 512MB (DDR3) (DX10.1) -- 4.9 VP
  212. 48%-- Intel (Clarkdale APU) HD Graphics (dual channel DDR3-1333) (DX10) -- 4.8 VP
  213. 46%-- Geforce 9400GT 512MB (DX10) -- 4.6 VP
  214. 41%-- Radeon HD 3450 512MB (DX10.1) -- 4.1 VP
  215. 33%-- Radeon HD 2350 / 2400Pro 256MB (DX10) -- 3.3 VP
  216. 31%-- Geforce 205 512MB (DX10.1) -- 3.1 VP


* - Note for Crossfire users, these dual-GPU Radeon cards still have frame pacing (extreme microstuttering with runt frames) issues due to the lack of frame metering except for some DX10+ games on single-monitor resolutions only. Currently, only a handful of reviews covered frame pacing game fixes introduced by Catalyst 13.8 beta drivers, with a few game benchmarks. In addition, Catalyst 13.10beta "Includes a number of frame pacing improvements for the following titles: Tomb Raider, Metro Last Light, Sniper Elite, World of Warcraft, Max Payne 3, Hitman Absolution, and for Cat13.10beta2, "frame pacing improvements for CPU-bound applications." So far, not enough data exists for a more ideally accurate representation of these dual-GPU Radeon cards, so these ratings should only be considered valid for DX10+ games with resolved frame-pacing issues for now. AMD seems to be working on it, so more remains to be seen, and hopefully more reviews cover more than just a handful of games using FCAT, for a clearer picture as to just how severe the penalty should be applied overall to the ratings.


* - Note for non-overclockers (who do not want to adjust the Temperature Target slider), these stock GPU Boost 2.0 cards can perform several percent faster on default settings during "cold starts" especially if benchmarks runs are less than 2 minutes long, let alone 30 seconds long (for the reasons to such, see: http://forums.anandtech.com/showpost.php?p=34664254&postcount=54 ), while the rating only reflects an average of ALL reviews put together - despite most reviewers not ensuring that the cards are warmed up in advance for more accurate measurement of normal gameplay that lasts for more than 15-30 minutes.

Using Titan as an example, if we look at just the review sites that compare warmed-up results against "cold starts" across their benchmarks and apply the difference to the overall average, we get this:
111%-- Geforce GTX TITAN (warmed up, default*) 6GB (DX11) -- 320 VP
instead of:
120%-- Geforce GTX TITAN 6GB (DX11) -- 343 VP
 
Last edited:
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
NEW ADDITION TO THE VOODOOPOWER RATINGS: MOBILE NOTEBOOK GPUs

VP = Voodoopower


___Tier A Super High-End --(Plays like a lapdog beast)
  1. 108%-- Radeon HD 7970M 2GB (DX11.1) -- 154 VP
  2. 100%-- Geforce GTX 680M 4GB (DX11) -- 143 VP
    (times 44% over Tier B's 100%)

    Tier B Mid High-End --(Plays like a lap-puppy beast)
  3. 117%-- Geforce GTX 675MX 4GB (DX11) -- 109 VP --- (* NEW ENTRY! *)
  4. 107%-- Geforce GTX 580M / GTX 675M 2GB (DX11) -- 99 VP
  5. 105%-- Radeon HD 6990M 2GB (DX11) -- 98 VP
  6. 100%-- Geforce GTX 670MX 3GB (DX11) -- 93 VP --- (* NEW ENTRY! *)
  7. 100%-- Geforce GTX 485M 2GB (DX11) -- 93 VP
    (times 26% over Tier C's 100%)

    Tier C Lower High-End --(Plays most games at high settings @ 1920x1080)
  8. 116%-- Radeon HD 6970M 2GB (DX11) -- 86 VP
  9. 111%-- Geforce GTX 670M 3GB (DX11) -- 82 VP
  10. 107%-- Geforce GTX 570M 1.5GB (DX11) -- 79 VP
  11. 106%-- Radeon HD 7870M 2GB (DX11.1) -- 78 VP --- (* NEW ENTRY! *)
  12. 103%-- Radeon HD 6950M 2GB (DX11) -- 76 VP
  13. 100%-- Geforce GTX 480M 2GB (DX11) -- 74 VP
    (times 21% over Tier D's 100%)

    Tier D Upper Mid-Range --(Plays most games at high settings @ 1600x900)
  14. 114%-- Radeon HD 7850M 2GB (DX11.1) -- 69 VP --- (* NEW ENTRY! *)
  15. 110%-- Geforce GTX 470M 1.5GB (DX11) -- 67 VP
  16. 110%-- Geforce GTX 560M 1.5GB (DX11) -- 67 VP
  17. 108%-- Geforce GTX 660M 2GB (DX11) -- 66 VP --- (* NEW ENTRY! *)
  18. 103%-- Radeon Mobility HD 5870 1GB (DX11) -- 62 VP
  19. 101%-- Geforce GTX 460M 1.5GB (DX11) -- 61 VP
  20. 100%-- Radeon HD 6870M 1GB (DX11) -- 61 VP
    (times 35% over Tier E's 100%)

    Tier E Middle Mid-Range --(Plays most games at high settings @ 1366x768)
  21. 131%-- Radeon HD 7770M 2GB (DX11.1) -- 59 VP --- (* NEW ENTRY! *)
  22. 129%-- Geforce GT 650M 2GB (GDDR5) (DX11) -- 58 VP --- (* NEW ENTRY! *)
  23. 129%-- Radeon Mobility HD 4870 1GB (GDDR5) (DX10.1) -- 58 VP
  24. 127%-- Radeon Mobility HD 5850 1GB (GDDR5, 625MHz core) (DX11) -- 57 VP
  25. 124%-- Geforce GT 645M 2GB (GDDR5) (DX11) -- 56 VP --- (* NEW ENTRY! *)
  26. 122%-- Radeon HD 8790M 2GB (DX11.1) -- 55 VP --- (* NEW ENTRY! *)
  27. 122%-- Geforce GTX 560M 2GB (DX11) -- 55 VP
  28. 120%-- Radeon Mobility HD 4850 1GB (GDDR5) (DX10.1) -- 54 VP
  29. 115%-- Radeon HD 7750M 2GB (DX11.1) -- 52 VP --- (* NEW ENTRY! *)
  30. 115%-- Radeon HD 6850M 1GB (DX11) -- 52 VP
  31. 113%-- Radeon HD 8770M 2GB (DX11.1) -- 51 VP --- (* NEW ENTRY! *)
  32. 113%-- Geforce GT 640M 2GB (GDDR5) (DX11) -- 51 VP --- (* NEW ENTRY! *)
  33. 109%-- Radeon Mobility HD 5850 1GB (GDDR5, 500MHz core) (DX11) -- 49 VP
  34. 106%-- Geforce GTX 285M 1GB (DX10) -- 48 VP
  35. 104%-- Geforce GTX 280M 1GB (DX10) -- 47 VP
  36. 100%-- Geforce GT 445M 1.5GB (GDDR5) (DX11) -- 45 VP
(excluding the rest of older cards for now, including much slower DDR3 versions of the above cards that are explicitly stated as GDDR5 variants)


(This website was of tremendous help for the ratings: http://www.notebookcheck.net - with a cool comparison chart:
http://www.notebookcheck.net/Mobile-Graphics-Cards-Benchmark-List.844.0.html )

It was very very difficult to do this, with there being so few benchmark reviews out there for the notebook GPUs. I had to use my understanding of how each particular architecture (Juniper, Barts, GT200, GF106, etc..) perform according to their bigger discrete video card brothers (for the desktop PC), by trying to as accurately evaluate the bottlenecking factors (bandwidth, ROPs, and TMUs) with the given specs and clocks. For example, HD 7750 has roughly half the shader compute power of HD 7770 but the 7770 is only about 29% faster overall mainly because it was more bottlenecked by the same amount of bandwidth - this is how important it was to really understand how each of the different specifications for each GPU affected its overall performance.

I plan on adding more cards, but would like to know if you see any deviations that stand out for now. If anything, it would be appreciated, so please feel free to help with pointing anything out.
 
Last edited:
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
Hope you enjoyed the internet's most accurately categorized list of top ranked video cards with relative performance ratings, for newcomers and enthusiasts alike!

Somewhere close to 1,000 online video card review articles were studied for the culmination of this data, whew.

For the beginners, keep in mind that all the Radeon 7/6/5xxx and Geforce 7/6/5/4xx cards can play DX11 games. All of the Radeon 2xxx to 4xxx and Geforce 8xxx to 2xx cards can play DX10 games, but not DX11. As to the Radeons with numbers X19xx or below and the Geforces with numbers 7xxx or below, these cards are limited to DX9.0c.

Special Note: The overall average performance of multi-GPU cards from review site benchmarks receive a somewhat modest 15% penalty for having two major impediments: microstuttering that is still not yet fully resolved with the current generation, and multi-GPU scaling that does not work for at least 15% of games out there - actually performing worse than a single-GPU configuration in many cases. Another impediment is that triple buffering cannot be forced as true triple buffering with Alternate Frame Rendering, so if one wishes to play with Vsync enabled, one would experience the fractioning of the frame rates whenever it drops below the monitor refresh rate; however, the lack of true triple buffering does not count as a penalty. (For multi-GPU setups, generally multiply the single card score by 1.55 to 1.6, depending on how new the cards are - it scales a little better with each new generation for newer games, with better driver support.)

The percentages are for comparing cards within the same class (they could be multiplied with the next class by using 1.23 for 23% for example.. anyways), while the patented (just kidding) trademark, GPUpower[sup]TM[/sup] is a linear overall performance scale of a card in an average out of several dozen popular modern PC games released in the last few years that scales normally well without being CPU-bound or dependent on unusual or exclusive features like PhysX or CUDA for example.

EDIT: GPUpower is changed to Voodoopower[sup]TM[/sup] (just kidding on the TM part again) for the following reason:
That way, homage is paid to the last Voodoo card ever made, the mythical Voodoo 5 6000, since it has approximately ONE Voodoopower (GPUpower) unit rating. Actually, it's the closest card to exactly 1 GPUpower that I could think of.

There! We have a unit similar to the horsepower rating for cars, yet much more comprehensive, like how a car would perform on many different racing tracks, put together for an average! A Geforce GTX 580 has 180 Voodoopower--180 times as much power as 3Dfx's last GPU. Also, it has 18 times as much power as a 6800 Ultra, the first DX9.0c flagship card. A Radeon 9700 Pro, the first DX9 card, would have nearly 5 Voodoopower. Radeon 8500, the first DX8.1 card would have just over 2 Voodoopower. Geforce 2 GTS would have about 1.3 Voodoopower, while Geforce 2 Ultra, the last mighty DX7 card, possesses 1.6 of these mysterious units.

This Voodoopower is a magical unit captured from the last incarnation of 3DFx's glory--it was the mightiest unit that 3dfx could barely impose on us gamers before taking its Voodoo5 6000 with it to the grave. Word has it that only 150 engineering samples escaped this horrendous fate and that only 30 of them actually worked with slightly tweaked drivers (although with a workable PCI hardware bug yet literally suicidal incompatibility with many motherboards). (L) Loving tribute--for more on the 6000, see: http://www.x86-secret.com/articles/divers/v5-6000/v56kgb-6.htm



*DISCLAIMER - (whoa, yes, there's a serious, eerie disclaimer lurking in the dark corner)-- The ranking list tries to take into account the overall performance in as many games as possible, not just a selected handful of games. The performance under consideration does not include PhsyX. It is just to let you know where your video card would rank in overall performance using the popular de-facto standard of today (1080p for modern video cards, and including 1440p and higher for high-end cards that came out in the last couple years). By the way, AA is sometimes automatically enabled if the game pre-configures the settings according to the video card, and is highly recommended even if it means upgrading your video card.

With anything lower than the lowest-ranked card, forget about playing even Bioshock from 5 years ago at 1680x1050 the way the developer intended the game to be played! (The 6800 Ultra that cost $500 7 years ago cannot even handle critical areas of Bioshock on default settings at over 15fps average on today's low-end resolution of 1680x1050)! ;)


Out of nostalgia, here are some of the historical notables: --(Approximated for now - just my own quick guesses):

Radeon X850XT 256MB (DX9) -- 10.9 VP
Radeon X800XT 256MB (DX9) -- 10.4 VP
Radeon X800XT All-In-Wonder 256MB (DX9) -- 10.4 VP
Geforce 6800GT 256MB (DX9.0c) -- 8.8 VP
Geforce 6800 (DX9.0c) -- 7.2 VP
Radeon X800 (DX9) -- 7.2 VP
Geforce 6600GT (DX9.0c) -- 6.3 VP
Radeon 9800 XT (DX9) -- 5.8 VP
Geforce FX 5950 Ultra (DX9) -- 5.5 VP
Radeon 9800 Pro (DX9) -- 5.4 VP
Geforce FX 5900 Ultra (DX9) -- 5.1 VP
Radeon 9700 Pro (DX9) -- 4.9 VP
Geforce FX 5800 Ultra (DX9) -- 4.6 VP
Geforce 4 Ti 4600 (DX8.1) -- 2.9 VP
Geforce 3 Ti 500 (DX 8.0) -- 2.2 VP
Radeon 8500 (DX8.1) -- 2.1 VP
Geforce 3 (DX 8.0) -- 2.0 VP
Geforce 2 Ultra (DX 7.1) -- 1.6 VP
Geforce 2 GTS (DX 7.1) -- 1.3 VP
~~~Voodoo 5 6000 (DX 7.0) -- 1 VP~~~
 
Last edited:
M

masian

Junior Member
Jan 10, 2013
4
0
0
well, i'm tier C and i only play at 1920x1080... which means i'm golden at max settings everything! if i overclock, i think i can imitate tier B!?
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
masian said:
well, i'm tier C and i only play at 1920x1080... which means i'm golden at max settings everything! if i overclock, i think i can imitate tier B!?
Click to expand...
Not everything! Overall, yes pretty much! Also, it depends on the card, and on how high the overclock is.
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
blackened23 said:
Tier AAA soon to be occupied by Titan.
Click to expand...
That'd be Dual-Titan if it comes out SOON enough, as a single big K still wouldn't beat GTX 690... we'll see! If not soon enough, it will just push GTX 690 down a class! Go, Nvidia, go Nvidia!! (And go, AMD too - hurry up with the Sea Islands Tenerife!)
 
F

f1sherman

Platinum Member
Apr 5, 2011
2,243
1
0
Uhh... I thought VoodooPower incorporates microstuttering and game/profile issues by penalizing dual cards with 10-15%
 
hyrule4927

hyrule4927

Senior member
Feb 9, 2012
359
1
76
@f1sherman

BoFox said:
Special Note: The overall average performance of multi-GPU cards from review site benchmarks receive a somewhat modest 15% penalty for having two major impediments: microstuttering that is still not yet fully resolved with the current generation, and multi-GPU scaling that does not work for at least 15% of games out there - actually performing worse than a single-GPU configuration in many cases. Another impediment is that triple buffering cannot be forced as true triple buffering with Alternate Frame Rendering, so if one wishes to play with Vsync enabled, one would experience the fractioning of the frame rates whenever it drops below the monitor refresh rate; however, the lack of true triple buffering does not count as a penalty. (For multi-GPU setups, generally multiply the single card score by 1.55 to 1.6, depending on how new the cards are - it scales a little better with each new generation for newer games, with better driver support.) ** NEW EXCEPTION: GTX 690 is penalized by only 7.5% instead of the usual 15% due to microstuttering finally being fixed to a good extent!
Click to expand...
 
U

UzairH

Senior member
Dec 12, 2004
315
0
0
It must have taken you a fair bit of effort to get all these numbers. They also look legit to me, based on the comparison I have been doing myself among recent generations.
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
Thank you UzairH!

UPDATE: Added HD 7870 ("XT" / Tahiti LE / Boost Edition) 2GB , rated at 190 VP.

reviews:

http://www.kitguru.net/components/graphic-cards/zardon/sapphire-hd7870-w-boost-tahiti-le-review/
http://www.rwlabs.com/article.php?id=760
http://www.tweaktown.com/reviews/51...hiti_le_2gb_myst_video_card_review/index.html
http://www.tweaktown.com/reviews/51...oost_overclocked_video_card_review/index.html
http://www.hardwareheaven.com/revie...iti-le-graphics-card-review-introduction.html
http://www.hardocp.com/article/2013...eon_hd_7870_myst_edition_review/#.URVl72dZMWA
http://hexus.net/tech/reviews/graphics/49569-powercolor-radeon-hd-7870-myst/
http://www.hardwareluxx.com/index.p...est-2x-tahiti-le-by-powercolor-and-vtx3d.html
( ^^^ SUPERB ARTICLE)
http://www.overclockersclub.com/reviews/powercolor_hd7870_myst/
http://www.techpowerup.com/reviews/Club_3D/HD_7870_jokerCard_Tahiti_LE/
http://www.techpowerup.com/reviews/VTX3D/Radeon_HD_7870_XT_Black/
http://www.overclock3d.net/reviews/gpu_displays/club_3d_7870_joker_xt_edition_review/1
http://www.computerbase.de/artikel/grafikkarten/2012/test-vtx3d-hd-7870-black-tahiti-le/
( ^^^ SUPERB ARTICLE)
http://www.pcgameshardware.de/AMD-R...D-Radeon-HD-7870-Black-Boost-Edition-1039564/
http://www.hkepc.com/8796
http://ht4u.net/reviews/2012/vtx3d_radeon_hd_7870_black_tahiti_le_benchmarks/
( ^^^ SUPERB ARTICLE)
 
Last edited:
Jaydip

Jaydip

Diamond Member
Mar 29, 2010
3,691
21
81
Bo unless the top tier is named "STFU" this should be un-stickied :p
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
Darn, I was gonna name it in your honor,
"Tier AAA: Stupid High-End --(Causing Jaydip to go into an uncontrollable drooling fit!)
:biggrin:
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
What is going to be really hairy is that Nvidia has so many different versions of GT 640M, 650M, etc.. that the consumer would easily be confused about which is which. If I state GDDR5 version, it could still be misleading to the consumer if the DDR3 version is not also explicitly and clearly noted in the ratings. The performance difference can be extreme between the two, like for example (see how all the DDR3 GPUs are completely bottlenecked:

IMG0039916.gif
(the Batman:AA chart using "Tri" view - the image cannot be linked here)
http://www.hardware.fr/articles/885-5/performances-battlefield-3-far-cry-3.html

Whereas the GDDR5 versions are far better off.
 
railven

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Dammit Bofox you're still missing the Ares II. What are we paying you for? :p

I want to see if it beats Devil 13 (not that I'd ever own one, but my money is on Ares II).
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
Problem is, you didn't pay me yet. $10 and I'll go for it. :D But hey, your first order is on the house! Hehe.. :p

So, put your seatbelts on and stay braced!
 
Last edited:
railven

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Nice, I knew the Ares II would beat the Devil 13.

I really dig the design with the included water cooler, too. I'd love to own one...someday...naaaah haha.
 
BoFox

BoFox

Senior member
May 10, 2008
689
0
0
It actually runs pretty cool for its temperament, and can be overclocked like MAD!!! Truly the world's fastest!! $1500 is about as high as it gets now. Will we be seeing an overclocked "Mars III" dual-Titan card for $2000 on Newegg?

I fine-tuned some of the ratings, (mostly notebook GPUs) after checking all of their correlation with each other. It's really difficult to do, yet critical so that things do not ever get messed up badly as I sometimes use certain cards as reference (with most reviews only using a handful of cards for comparison).

EDIT - Looks like some more fine-tuning still needs to be done. What is a pain in the neck is how some review sites use older drivers (with the exact same results) for comparing the older cards against the newer cards on newer drivers, without even admitting it. Oh well.. enough complaining!
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom