• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Working On An Entire Range of HBM GPUs To Follow Fiji And Fury Lineup – Has Priority To HBM2 Cap

The memory latency does not work like that. The chip read latency Is constant regardless of the interface clock frequency. Latency is measured in clock cycles, it is why ddr4 has higher latency measurements than ddr3. And the memory cells being used are no different to those in ddr5, just they are stacked etc.

Plus as I showed, the 1080p performance is fine when using mantle so the low 1080p performance is being caused elsewhere than the HBM.

I thought about it last night and I think the cause is 'Over aggressive memory management' in the driver. The memory managment is in the game for thief in mantle mode so will not be affected by amds driver memory management, causing lower res performance to tank.

If you look at game memory usage at lower resolutions, such as in gregs Pcars video, the FX is using half the ram of the TX yet there is plenty off memory to use.

You might be on to something here, a lot of buffering is being moved around to keep the VRam usage low, if its constantly flushing and uploading the pool it will cause a slowdown, if its moving objects between the GPU buffer and system RAM it will increase the latency massively.
 
With the whole is it the HBM or the FIJI architecture or just AMD's memory management, causing not so good 1080p discussion. Unfortunately we are not going to be able to prove things one way or the other until there is another HBM enabled card that is not FIJI to test against. I mean if the first NVidia HBM card also has issues with lower resolution performance then that would point to the HBM being a culprit, if it doesn't then that points to it being down to something AMD has done with FIJI.
 
With the whole is it the HBM or the FIJI architecture or just AMD's memory management, causing not so good 1080p discussion. Unfortunately we are not going to be able to prove things one way or the other until there is another HBM enabled card that is not FIJI to test against. I mean if the first NVidia HBM card also has issues with lower resolution performance then that would point to the HBM being a culprit, if it doesn't then that points to it being down to something AMD has done with FIJI.

No, if Pascal has enough Buffer (HBM2) then it has no need for Memory management, the same would be true for AMD's HBM2 card.
 
Current estimates are still 10-12 months aren't they? Now if they took 2 years to release something new and it didn't beat the competition that would be a big problem.

Big Pascal already taped out last month.

I expect Nvidia wont be in any rush to release pascal card given that they have no real competition right now, and if a price war did start they would win because their cards are simply much cheaper to produce than HBM/AIO solutions with poor yields.

If they wanted to they could aim at a March release I would think, that is what,some rumours suggested. I think they might just wait for summer 2016 and enjoy the current high sales.

The Pascal architecture is in some sense very similar to Maxwell v2. Maxwell is a very forward looking architecture wit plenty of life in it. Nvidia have talked about Volta (the successor to pascal) even being an evolution of Maxwell. that is why Nvida invested to heavily in it, why we see things like such a huge increase in performance per watt on the same node.
So the new architecture is not such a huge step forward, it is mostly about working with HBM and a the die shrink.

Nvidia have plenty of time to work out supply, manufacturing, testing and driver issues with HBM.
 
You might be on to something here, a lot of buffering is being moved around to keep the VRam usage low, if its constantly flushing and uploading the pool it will cause a slowdown, if its moving objects between the GPU buffer and system RAM it will increase the latency massively.

If this was true the performance problem would be @2160p where the memory was moving stuff around all the time, @1080p this is not needed.

The problem is the low clockspeed of the HBM.

It is the same problem you would get if your gaming PC had a massive multi core Xeon chip with a low clockspeed, massively powerful but still bottlenecking the GPUs.
 
If this was true the performance problem would be @2160p where the memory was moving stuff around all the time, @1080p this is not needed.

The problem is the low clockspeed of the HBM.

It is the same problem you would get if your gaming PC had a massive multi core Xeon chip with a low clockspeed, massively powerful but still bottlenecking the GPUs.

A good way to test this theory Kaap would be to test how much of a performance improvement overclocking HBM at 1080P offers.
 
A good way to test this theory Kaap would be to test how much of a performance improvement overclocking HBM at 1080P offers.

I think that's part of the reason I am making the argument in the first place.

In theory with the bandwidth available there should be no improvement from overclocking the memory yet people who have done it have got an improvement.

It suggests that for 1080p the clockspeed is holding HBM back.

I will test properly once mine are waterblocked up.
 
Not really been reading this thread but I did read Kaap saying that Fiji isn't ideal for 1080P and if you watch my vids, that was quite apparent. I would even say 1440P isn't suited either and it is only UHD that see's any benefit from HBM that I can tell.
 
If this was true the performance problem would be @2160p where the memory was moving stuff around all the time, @1080p this is not needed.

The problem is the low clockspeed of the HBM.

It is the same problem you would get if your gaming PC had a massive multi core Xeon chip with a low clockspeed, massively powerful but still bottlenecking the GPUs.

Memory IC's are not CPU's, you can have a thousand cores in your Xeon the performance in DX11 is exactly the same as it is if you only have 6 because only one of them is used for DrawCalls.
Run the same thing in DX12 and your DrawCalls are 6 times higher

A GPU Memory architecture does not use one 32Bit lane with the others idling, in the case of Hawaii it will use 16 of them to make up the 512Bit bus.

In the case of HBM its four 1024Bit lanes.

GDDR5: 16x 32Bit @ 1250Mhz = 320GB/s of memory performance

HBM: 4x 1024Bit @ 500Mhz = 512GB/s of memory performance

The end result is how much performance it has, that is measured in GB/s.

Overclocking HBM is not proof that the MHz are a bottleneck, its proof that overclocking something increases performance.
 
The problem is the low clockspeed of the HBM.

It is the same problem you would get if your gaming PC had a massive multi core Xeon chip with a low clockspeed, massively powerful but still bottlenecking the GPUs.

its nothing to do with the clock speed of hbm... the bandwidth is determined by line width AND memory clock together. And your analogy is wrong in a massive way... one is a processing information, the other is a memory system. the way the clocking works is completely different between the two.
 
its nothing to do with the clock speed of hbm... the bandwidth is determined by line width AND memory clock together. And your analogy is wrong in a massive way... one is a processing information, the other is a memory system. the way the clocking works is completely different between the two.

So why does overclocking the memory on a Fury X @1080p bring positive gains even though the card already has massive bandwidth compared to say a TitanX ?

Answer because the memory clockspeed is holding HBM back.

Perhaps you should get hold of a Fury card and try it yourself !!!

Here is the Firestrike 1080p standard bench you will notice the people who overclocked the memory on their Fury Xs have got the higher graphics scores.

1 GPU Scores.



  1. Score 19575, GPU 980 Ti @1510/2100, GFX Score 21409, Physics Score 23697, Combined Score 10285, CPU 5960X @4.873, Gibbo, Post No.283 - Link Drivers 353.30
  2. Score 19488, GPU TitanX @1554/1907, GFX Score 21992, Physics Score 21765, Combined Score 9692, CPU 5960X @4.5, tipes, Post No.172 - Link Drivers 350.05
  3. Score 19307, GPU TitanX @1530/1977, GFX Score 21841, Physics Score 20835, Combined Score 9751, CPU 5960X @4.625, ^JP^, Post No.120 - Link Drivers 350.05
  4. Score 19278, GPU TitanX @1441/2027, GFX Score 21317, Physics Score 22379, Combined Score 10015, CPU 5960X @4.6, Kaapstad, Post No.9 - Link Drivers 347.88
  5. Score 19090, GPU TitanX @1262/2000, GFX Score 21282, Physics Score 21347, Combined Score 9886, CPU 5960X @4.5, andrewohare, Post No.104 - Link Drivers 347.88
  6. Score 18760, GPU TitanX @1550/2009, GFX Score 22570, Physics Score 17905, Combined Score 8549, CPU 5820k @4.5, mac40, Post No.414 - Link Drivers 353.45
  7. Score 18026, GPU TitanX @1490/1950, GFX Score 21420, Physics Score 16627, Combined Score 8742, CPU 4930k @4.5, alex_123_fra, Post No.159 - Link Drivers 350.12
  8. Score 17757, GPU 980 Ti @1351/1808, GFX Score 21000, Physics Score 16671, Combined Score 8620, CPU 5930k @4.3, 0n1ons, Post No.321 - Link Drivers 353.30
  9. Score 17651, GPU 980 Ti @1450/2002, GFX Score 20394, Physics Score 17505, Combined Score 8843, CPU 5930k @4.375, elmarko, Post No.427 - Link Drivers 353.30
  10. Score 17637, GPU TitanX @1475/1750, GFX Score 20692, Physics Score 16510, Combined Score 8799, CPU 5820k @4.2, thefogo, Post No.146 - Link Drivers 350.12
  11. Score 17597, GPU 980 Ti @1310/2078, GFX Score 21438, Physics Score 13941, Combined Score 9024, CPU 4790k @4.9, whyscotty, Post No.417 - Link Drivers 353.49
  12. Score 17580, GPU 980 Ti @1595/2018, GFX Score 22325, Physics Score 13496, Combined Score 8216, CPU 4770k @4.9, amigafan2003, Post No.287 - Link Drivers 353.30
  13. Score 17326, GPU TitanX @1244/2003, GFX Score 21484, Physics Score 13371, Combined Score 8631, CPU 4790k @4.8, whyscotty, Post No.20 - Link Drivers 350.05
  14. Score 17240, GPU 980 Ti @1478/1950, GFX Score 21059, Physics Score 14078, Combined Score 8523, CPU 4790k @5.0, khemist, Post No.424 - Link Drivers 353.38
  15. Score 17032, GPU 980 Ti @1510/1930, GFX Score 21325, Physics Score 13035, Combined Score 8309, CPU 4790k @4.6, Razor Time, Post No.210 - Link Drivers 353.06
  16. Score 16988, GPU 980 Ti @1512/2000, GFX Score 21448, Physics Score 12663, Combined Score 8299, CPU 4770k @4.5, Robzere31, Post No.327 - Link Drivers 353.06
  17. Score 16964, GPU TitanX @1200/1917, GFX Score 20110, Physics Score 15785, Combined Score 8230, CPU 5930k @4.0, dessimpson, Post No.406 - Link Drivers 353.30
  18. Score 16915, GPU TitanX @1410/2007, GFX Score 20958, Physics Score 13087, Combined Score 8425, CPU 4790k @4.7, Twisted Kid, Post No.160 - Link Drivers 350.12
  19. Score 16816, GPU 980 Ti @1489/1852, GFX Score 20471, Physics Score 13506, Combined Score 8532, CPU 4790k @4.7, setter, Post No.402 - Link Drivers 353.38
  20. Score 16798, GPU TitanX @1524/2001, GFX Score 21731, Physics Score 12542, Combined Score 7659, CPU 2700k @5.0, mac40, Post No.93 - Link Drivers 350.12
  21. Score 16623, GPU 980 Ti @1455/1900, GFX Score 20408, Physics Score 12983, Combined Score 8437, CPU 4790k @4.6, ciggybumly, Post No.191 - Link Drivers 353.06
  22. Score 16259, GPU 980 Ti @1497/1898, GFX Score 20819, Physics Score 12120, Combined Score 7632, CPU 3820k @4.75, lnoton, Post No.342 - Link Drivers 353.38
  23. Score 16258, GPU 980 Ti @1250/1850, GFX Score 20281, Physics Score 12679, Combined Score 7876, CPU 4770k @4.5, iwrox, Post No.301 - Link Drivers 353.30
  24. Score 16254, GPU TitanX @1330/1770, GFX Score 20519, Physics Score 12216, Combined Score 7879, CPU 4790k @4.4, Simon Belmont, Post No.179 - Link Drivers 350.12
  25. Score 16203, GPU TitanX @1326/1753, GFX Score 18946, Physics Score 15568, Combined Score 8005, CPU 3930k @4.625, Gregster, Post No.324 - Link Drivers 353.38
  26. Score 16113, GPU TitanX @1503/2000, GFX Score 21862, Physics Score 10395, Combined Score 7505, CPU 4870HQ @4.7, Meaker, Post No.195 - Link Drivers 353.06
  27. Score 16089, GPU 980 Ti @1252/1928, GFX Score 19638, Physics Score 12954, Combined Score 8077, CPU 4790k @4.6, Valenni, Post No.265 - Link Drivers 353.06
  28. Score 15939, GPU 980 Ti @1400/2000, GFX Score 20113, Physics Score 12308, Combined Score 7540, CPU 4790k @4.4, mufc802, Post No.292 - Link Drivers 353.06
  29. Score 15869, GPU 980 Ti @1260/1753, GFX Score 19738, Physics Score 12530, Combined Score 7665, CPU 4790k @4.6, savagesam117, Post No.368 - Link Drivers 353.30
  30. Score 15792, GPU 980 Ti @1567/1753, GFX Score 20446, Physics Score 11717, Combined Score 7227, CPU 2700k @4.6, Doco, Post No.351 - Link Drivers 353.30
  31. Score 15749, GPU 980 Ti @1255/1853, GFX Score 19383, Physics Score 13873, Combined Score 7148, CPU X5650 @4.0, Strife212, Post No.204 - Link Drivers 350.12
  32. Score 15637, GPU 980 Ti @1412/1980, GFX Score 19375, Physics Score 14370, Combined Score 6757, CPU 5930k @3.5, Clov!s, Post No.307 - Link Drivers 353.06
  33. Score 15505, GPU 980 Ti @1260/1753, GFX Score 19696, Physics Score 11935, Combined Score 7223, CPU 3770k @4.5, rapt0r, Post No.384 - Link Drivers 353.38
  34. Score 15446, GPU Fury X @1120/600, GFX Score 17633, Physics Score 18421, Combined Score 7112, CPU 5820k @4.7, uksoldierboy, Post No.435 - Link Drivers 15.7
  35. Score 15024, GPU 980 Ti @1457/1753, GFX Score 20284, Physics Score 8587, Combined Score 8254, CPU 4690k @4.3, 40Four, Post No.362 - Link Drivers 353.30
  36. Score 14799, GPU 980 @1400/2003, GFX Score 16468, Physics Score 23567, Combined Score 6384, CPU 5960X @4.874, topdog555, Post No.128 - Link Drivers 350.12
  37. Score 14732, GPU Fury X @1140/550, GFX Score 17597, Physics Score 13393, Combined Score 7115, CPU 4790k @4.7, whyscotty, Post No.365 - Link Drivers 15.15
  38. Score 14566, GPU TitanX @1329/2053, GFX Score 20249, Physics Score 8180, Combined Score 7533, CPU 2500k @4.5, sean, Post No.164 - Link Drivers 350.12
  39. Score 14540, GPU Fury X @1128/500, GFX Score 16922, Physics Score 15202, Combined Score 6856, CPU 3930k @4.4, Gregster, Post No.330 - Link Drivers 15.15
  40. Score 14505, GPU 980 @1472/2102, GFX Score 16032, Physics Score 21874, Combined Score 6536, CPU 5960X @4.5, Kaapstad, Post No.42 - Link Drivers 344.75
  41. Score 14402, GPU 980 @1540/2018, GFX Score 16318, Physics Score 18295, Combined Score 6548, CPU 5930k @4.625, JeffyB, Post No.130 - Link Drivers 350.12
  42. Score 14251, GPU 980 @1500/2000, GFX Score 16361, Physics Score 16941, Combined Score 6463, CPU 4930k @4.5, Spudley, Post No.29 - Link Drivers 350.12
  43. Score 14126, GPU 980 @1421/1923, GFX Score 16174, Physics Score 16787, Combined Score 6458, CPU 4930k @4.5, ubersonic, Post No.43 - Link Drivers 347.88
  44. Score 14028, GPU 980 Ti @1222/1828, GFX Score 19558, Physics Score 7830, Combined Score 7258, CPU 2500k @4.3, ZeroG, Post No.378 - Link Drivers 353.38
  45. Score 13776, GPU 980 Ti @1000/1753, GFX Score 16521, Physics Score 11617, Combined Score 7004, CPU 4790k @4.4, savagesam117, Post No.348 - Link Drivers 353.30
  46. Score 13642, GPU Fury X @1050/500, GFX Score 15917, Physics Score 14560, Combined Score 6297, CPU 4960X @4.0, dafadau, Post No.310 - Link Drivers 15.15
  47. Score 13546, GPU 980 @1622/2003, GFX Score 16397, Physics Score 12049, Combined Score 6397, CPU 4790k @4.4, passey89, Post No.10 - Link Drivers 347.88
  48. Score 13360, GPU 290X @1335/1675, GFX Score 15216, Physics Score 18310, Combined Score 5758, CPU 3930k @5.125, Geeman1979, Post No.18 - Link Drivers 14.12
  49. Score 13151, GPU 980 @1325/1953, GFX Score 15864, Physics Score 11895, Combined Score 6193, CPU 4790k @4.5, D3K, Post No.280 - Link Drivers 350.12
  50. Score 12713, GPU 980 Ti @1226/1903, GFX Score 20006, Physics Score 6186, Combined Score 5910, CPU 2500k @3.3, ZeroG, Post No.329 - Link Drivers 353.38
  51. Score 12620, GPU 290X8GB @1270/1260, GFX Score 15074, Physics Score 13208, Combined Score 5518, CPU 4790k @4.8, passey89, Post No.248 - Link Drivers 15.20
  52. Score 12578, GPU 290P @1400/1622, GFX Score 14935, Physics Score 14022, Combined Score 5381, CPU 4820k @5.204, Rangerjr1, Post No.286 - Link Drivers 14.12
  53. Score 12442, GPU 290X @1260/1625, GFX Score 14166, Physics Score 17546, Combined Score 5298, CPU 4930k @4.8, Kaapstad, Post No.54 - Link Drivers 14.6
  54. Score 12266, GPU 780ti @1250/1800, GFX Score 13714, Physics Score 18391, Combined Score 5353, CPU 5820k @4.7, smsmasters, Post No.140 - Link Drivers 350.12
  55. Score 12107, GPU 980 @1357/1753, GFX Score 14053, Physics Score 14443, Combined Score 5309, CPU 5930k @3.5, Clov!s, Post No.34 - Link Drivers 344.48
  56. Score 12099, GPU 290X @1250/1575, GFX Score 14107, Physics Score 15185, Combined Score 5100, CPU X5670 @4.4, weldon855, Post No.263 - Link Drivers 15.20
  57. Score 12003, GPU 290X @1250/1625, GFX Score 13451, Physics Score 17789, Combined Score 5230, CPU 3970X @4.9, Kaapstad, Post No.55 - Link Drivers 13.11
  58. Score 11811, GPU 780ti @1250/1750, GFX Score 13697, Physics Score 13467, Combined Score 5329, CPU 4790k @4.8, smsmasters, Post No.85 - Link Drivers 350.12
  59. Score 11720, GPU 290X @1195/1615, GFX Score 13903, Physics Score 12604, Combined Score 5135, CPU 4790k @4.4, savagesam117, Post No.200 - Link Drivers Win10 ES
  60. Score 11614, GPU 970 @1323/1940, GFX Score 13754, Physics Score 12712, Combined Score 5058, CPU 4790k @4.5, Nedb37, Post No.328 - Link Drivers 353.30
  61. Score 11601, GPU 970 @1515/1980, GFX Score 13734, Physics Score 12430, Combined Score 5123, CPU 4790k @4.4, madp0k, Post No.157 - Link Drivers 350.12
  62. Score 11501, GPU 290P @1180/1600, GFX Score 13503, Physics Score 13512, Combined Score 4926, CPU 4790k @4.8, MadMatty, Post No.431 - Link Drivers 15.7
  63. Score 11447, GPU 290X @1152/1457, GFX Score 13096, Physics Score 16130, Combined Score 4810, CPU 5930k @4.1, 0n1ons, Post No.272 - Link Drivers 15.20
  64. Score 11427, GPU 980 @1347/1753, GFX Score 13363, Physics Score 13010, Combined Score 5036, CPU 4790k @4.6, chillidog, Post No.278 - Link Drivers 353.06
  65. Score 11347, GPU 970 @1501/1851, GFX Score 13413, Physics Score 12345, Combined Score 4986, CPU 4790k @4.4, CEUOTC, Post No.162 - Link Drivers 350.12
  66. Score 11236, GPU 290X @1180/1680, GFX Score 13450, Physics Score 11104, Combined Score 5070, CPU 2600k @4.4, new boy, Post No.24 - Link Drivers 14.12
  67. Score 11214, GPU 970 @1478/2000, GFX Score 13187, Physics Score 12245, Combined Score 4989, CPU 3770k @4.6, freddie64, Post No.215 - Link Drivers 350.12
  68. Score 11135, GPU 970 @1333/2002, GFX Score 13387, Physics Score 11033, Combined Score 4954, CPU 2600k @4.4, Andy7, Post No.123 - Link Drivers 350.12
  69. Score 11119, GPU nvTitan @1200/1802, GFX Score 12195, Physics Score 18406, Combined Score 4930, CPU 3930k @5.1, Kaapstad, Post No.56 - Link Drivers 320.00
  70. Score 11100, GPU 970 @1545/1977, GFX Score 13895, Physics Score 9048, Combined Score 5120, CPU 4670k @4.5, MrMD, Post No.89 - Link Drivers 347.88
  71. Score 10868, GPU 290X @1180/1640, GFX Score 13530, Physics Score 9084, Combined Score 4984, CPU 4690k @4.7, savagesam117, Post No.17 - Link Drivers 15.3
  72. Score 10573, GPU 970 @1501/1853, GFX Score 13396, Physics Score 8089, Combined Score 4988, CPU 4670k @4.0, CEUOTC, Post No.148 - Link Drivers 350.12
  73. Score 10402, GPU 7970 @1400/1700, GFX Score 11487, Physics Score 18579, Combined Score 4392, CPU 3930k @5.2, Rangerjr1, Post No.442 - Link Drivers 14.9
  74. Score 10400, GPU 980 @1404/1973, GFX Score 15934, Physics Score 8753, Combined Score 3131, CPU FX-8350 @4.4, EnjoyTroll, Post No.290 - Link Drivers 347.88
  75. Score 10294, GPU 780ti @1000/1750, GFX Score 12319, Physics Score 9759, Combined Score 4787, CPU 3770k @3.9, atheos, Post No.106 - Link Drivers 350.12
  76. Score 9864, GPU 780 @1143/1642, GFX Score 12207, Physics Score 7774, Combined Score 4845, CPU 2500k @4.3, RICKST20, Post No.13 - Link Drivers 347.88
  77. Score 9840, GPU 290P @1150/1450, GFX Score 11925, Physics Score 8594, Combined Score 4701, CPU 4690k @4.3, 40Four, Post No.289 - Link Drivers 15.20
  78. Score 9730, GPU 290X @1000/1250, GFX Score 11339, Physics Score 11340, Combined Score 4274, CPU 4790k @4.4, Anth4130, Post No.258 - Link Drivers 14.12
  79. Score 9527, GPU 780 @1120/1604, GFX Score 11583, Physics Score 8169, Combined Score 4578, CPU 4690k @4.2, easyrider, Post No.249 - Link Drivers 353.06
  80. Score 9411, GPU 780 @1031/1608, GFX Score 11338, Physics Score 8299, Combined Score 4539, CPU 2500k @4.6, Evian, Post No.156 - Link Drivers 350.12
  81. Score 9372, GPU 780 @1110/1502, GFX Score 10915, Physics Score 9930, Combined Score 4371, CPU 1230 V3 @3.6, CEUOTC, Post No.441 - Link Drivers 353.30
  82. Score 9344, GPU 970 @1317/1752, GFX Score 11455, Physics Score 8946, Combined Score 4037, CPU i7 960 @3.6, Strife212, Post No.118 - Link Drivers 350.12
  83. Score 9312, GPU 780 @1006/1502, GFX Score 10587, Physics Score 11496, Combined Score 4256, CPU 4770k @4.4, MeatLoaf, Post No.71 - Link Drivers 347.52
  84. Score 8986, GPU 7970 @1320/1875, GFX Score 9921, Physics Score 17609, Combined Score 3681, CPU 3960X @4.9, Geeman1979, Post No.64 - Link Drivers 13.10
  85. Score 8799, GPU 7970 @1330/1900, GFX Score 9684, Physics Score 17550, Combined Score 3617, CPU 3970X @4.9, Kaapstad, Post No.60 - Link Drivers 13.2
  86. Score 8242, GPU 7970 @1152/1752, GFX Score 9695, Physics Score 9127, Combined Score 3633, CPU 3570k @4.8, Lokken86, Post No.421 - Link Drivers 15.7
  87. Score 7825, GPU 7950 @1270/1600, GFX Score 9215, Physics Score 9340, Combined Score 3296, CPU FX-8350 @5.0, fx63007850, Post No.16 - Link Drivers 14.12
  88. Score 7362, GPU 680 @1180/1800, GFX Score 7917, Physics Score 17576, Combined Score 3071, CPU 3960X @4.9, Kaapstad, Post No.62 - Link Drivers 313.96
  89. Score 6661, GPU 960 @1190/1753, GFX Score 7769, Physics Score 9299, Combined Score 2671, CPU E3-1230V2 @3.59, CAT-THE-FIFTH, Post No.418 - Link Drivers 353.
  90. Score 6592, GPU 280X @1100/1500, GFX Score 8167, Physics Score 7484, Combined Score 2512, CPU FX-8320E @4.0, PapaLazaru, Post No.113 - Link Drivers 14.12
  91. Score 6384, GPU 7950 @1000/1500, GFX Score 7851, Physics Score 5311, Combined Score 3044, CPU X4 940 @3.6, TheMorningStar, Post No.447 - Link Drivers 15.7
  92. Score 5495, GPU 270X @1180/1400, GFX Score 6433, Physics Score 7458, Combined Score 2209, CPU FX-8320E @4.0, PapaLazaru, Post No.109 - Link Drivers 14.12
  93. Score 4498, GPU 570 @850/1050, GFX Score 4857, Physics Score 12532, Combined Score 1789, CPU 4790k @4.4, Aretak, Post No.167 - Link Drivers 350.12
  94. Score 4487, GPU 750ti @1320/1350, GFX Score 4898, Physics Score 9838, Combined Score 1836, CPU E3-1230 v3 @3.5, CEUOTC, Post No.138 - Link Drivers 350.12
  95. Score 4299, GPU 750ti @1272/1500, GFX Score 5095, Physics Score 4397, Combined Score 1950, CPU i3-2120 @3.3, bigjimmyauk, Post No.304 - Link Drivers 353.30
  96. Score 3773, GPU 860M @1097/1252, GFX Score 4128, Physics Score 7013, Combined Score 1615, CPU 4710MQ @2.5, CEUOTC, Post No.144 - Link Drivers 350.12
  97. Score 3748, GPU 560ti @1005/1100, GFX Score 4031, Physics Score 8496, Combined Score 1585, CPU 2500k @4.7, Azza, Post No.23 - Link Drivers 347.88
  98. Score 3154, GPU 6850 @1030/1150, GFX Score 3318, Physics Score 9150, Combined Score 1341, CPU 4690k @4.5, megatron, Post No.211 - Link Drivers 15.4
  99. Score 2694, GPU 6950 @800/1250, GFX Score 2873, Physics Score 4936, Combined Score 1256, CPU X4 965 @3.4, MadMatty, Post No.90 - Link Drivers 15.4
  100. Score 2465, GPU 6850 @850/1100, GFX Score 2559, Physics Score 9114, Combined Score 1041, CPU 4690k @4.5, megatron, Post No.136 - Link Drivers 15.4
  101. Score 1764, GPU 550ti @900/950, GFX Score 1877, Physics Score 4403, Combined Score 751, CPU i3 2120 @3.3, bigjimmyauk, Post No.77 - Link Drivers 350.12
  102. Score 1811, GPU 250 @1100/1200, GFX Score 2162, Physics Score 1844, Combined Score 807, CPU OP 180 @2.9, MadMatty, Post No.431 - Link Drivers 15.7
  103. Score 1460, GPU 5870 @850/1200, GFX Score 1466, Physics Score 7038, Combined Score 659, CPU i7 920 @2.66, hegemon, Post No.75 - Link Drivers 14.12
 
Last edited:
So why does overclocking the memory on a Fury X @1080p bring positive gains even though the card already has massive bandwidth compared to say a TitanX ?

Answer because the memory clockspeed is holding HBM back.

Perhaps you should get hold of a Fury card and try it yourself !!!

Agree. The bandwidth on the fury x is already huge, overclocking should bring zero gains as overclocking ram on TX and 980ti does nothing, if it was an increase in bandwidth providing the boost it would work on both.
 
So why does overclocking the memory on a Fury X @1080p bring positive gains even though the card already has massive bandwidth compared to say a TitanX ?

Answer because the memory clockspeed is holding HBM back.

Perhaps you should get hold of a Fury card and try it yourself !!!

Here is the Firestrike 1080p standard bench you will notice the people who overclocked the memory on their Fury Xs have got the higher graphics scores.

You should also notice that all three people are using different processors... the one at the top having the newest/highest clocked.

And all you have shown as what kaap stated is that overclocking can improve performance. Not that the memory is holding back the system, if anything it shows the cpu choice making the largest difference. but you need to show it on the same system, its not comparable when using different systems.

with the mantle/dx comparison on gregs system its showing the Driver/api being the problem. not the memory.
 
Last edited:
Gfx scores aren't affected by cpu, nice try though.
Overall score and physics score yes, gfx score no, which is what kaap is talking about.

You've got a lower clocked furyx beating one clocked higher, because it also has another +50 on the memory
 
Last edited:
So why does overclocking the memory on a Fury X @1080p bring positive gains even though the card already has massive bandwidth compared to say a TitanX ?

Answer because the memory clockspeed is holding HBM back.

Perhaps you should get hold of a Fury card and try it yourself !!!

Here is the Firestrike 1080p standard bench you will notice the people who overclocked the memory on their Fury Xs have got the higher graphics scores.


Because your overclocking it, by overclocking you are increasing the memory bandwidth, 550Mhz is a 10% overclock, your increasing the memory performance from 512GB/s to 563GB/s.
In the same way that if you overclock GDDR5 you are also increasing the memory performance, from 1250Mhz (320GB/s) to 1375Mhz (352GB/s)

Its overclocking, thats what it does, your too hung up on the 500Mhz thing, you think because its a low number it must be slow, its the same mentality that makes people think Nvidia's 1750Mhz memory must be faster than AMD's 1250Mhz, its very simplistic logic which ignores the 512Bit bus on one vs the 256Bit bus on the other.
512Bit @ 1250Mhz = 320GB/s
256Bit @ 1750Mhz = 225GB/s

Its nothing to do with the HBM architecture and the IC's Mhz. it is faster. much faster than GDDR5, doesn't mean its fast enough for what it needs to do on Fiji.
 
Last edited:
Gfx scores aren't affected by cpu, nice try though.
Overall score and physics score yes, gfx score no, which is what kaap is talking about.

of course it is affected by CPU... because DX11 is affected by the driver which is affected by the CPU... if it was not affected by the CPU then every card at the same frequency should have the same scores within a small margin of error.

The above is also true when you compare driver versions, newer ones improving performance by lowering driver overhead.

yet there are lower clocked cards with more powerful cups getting higher scores than higher clocked cards with less powerful cpus. you only have to look at the chart to see that.
 
Last edited:
CPU Physics is very different to DrawCalls, CPU Physics can be run across multiple cores, so the more cores you have the higher performance you will get.

DrawCalls in DX11 are one thread and one thread only, so it makes no difference how many cores you have, only by increasing the CPU clock rate can you increase DrawCall performance.
 
of course it is affected by CPU... because DX11 is affected by the driver which is affected by the CPU... if it was not affected by the CPU then every card at the same frequency should have the same scores within a small margin of error.

The above is also true when you compare driver versions, newer ones improving performance by lowering driver overhead.

yet there are lower clocked cards with more powerful cups getting higher scores than higher clocked cards with less powerful cpus. you only have to look at the chart to see that.

Firestrike graphics scores are not effected much if all the people are using top notch i7s which is the case here.

Also if overclocking the memory did not work on the Fury X why have the guys above done it. They are still using the same CPU before and after lol.
 
Back
Top Bottom