text
stringlengths
11
654
[928.60 --> 930.74] that hopefully looks somewhat good on a TV.
[930.86 --> 932.46] You're going to have to have an interface for Chromecast
[932.46 --> 933.56] that looks somewhat good on a TV.
[933.78 --> 936.42] You're going to have to design an interface for Android TV
[936.42 --> 937.74] that looks good on a TV.
[937.96 --> 942.66] And it just seems like an awful lot of hassle.
[943.58 --> 945.96] But maybe they have a better plan this time.
[946.30 --> 948.94] And maybe all that I'm saying right now
[948.94 --> 951.68] is going to have nothing to do with anything.
[951.78 --> 954.04] I see a lot of people complaining about mods in the Twitch chat.
[954.14 --> 955.98] Sorry, there's not a whole lot I can do about it right now.
[955.98 --> 959.62] But we do have plans to work out a different chat system
[959.62 --> 961.30] that's a little bit easier for us to moderate.
[961.54 --> 963.02] So stay tuned for that.
[963.06 --> 963.46] All right, guys?
[964.40 --> 964.88] This.
[965.90 --> 968.82] This is something that I'm getting pretty tired of,
[968.84 --> 969.80] this next topic here.
[969.92 --> 972.26] Now, NVIDIA released a new driver.
[972.72 --> 976.94] And they came out with all kinds of claims
[976.94 --> 978.14] with respect to performance.
[978.22 --> 979.14] So they're basically saying,
[979.26 --> 981.66] look, we're going to have DirectX 12
[981.66 --> 985.38] on all of our GPUs that supported DirectX 11 AMD.
[985.38 --> 989.08] And we're going to release a DirectX 11 driver
[989.08 --> 991.02] that stomps all over Mantle anyway.
[991.40 --> 992.38] So here, check this out.
[992.70 --> 994.42] So screen sharing here.
[994.50 --> 996.16] This was posted originally on the forum
[996.16 --> 998.22] by actually a couple of different folks.
[998.28 --> 998.96] I'll credit them later.
[999.02 --> 1001.94] But the 337.50 beta driver
[1001.94 --> 1003.82] was claiming that it would deliver
[1003.82 --> 1007.52] up to 64% faster single GPU performance
[1007.52 --> 1010.60] and up to 71% faster SLI performance
[1010.60 --> 1015.20] just through optimizing the DirectX 11 driver
[1015.20 --> 1016.46] that's available now.
[1016.46 --> 1021.12] Now, I have heard members of the press
[1021.12 --> 1022.86] muttering under their breath,
[1023.28 --> 1024.24] talking to each other,
[1024.34 --> 1025.48] not going to name any names,
[1025.48 --> 1029.38] about every time AMD or NVIDIA
[1029.38 --> 1031.28] releases a big driver update
[1031.28 --> 1035.54] that claims to completely turn the world inside out
[1035.54 --> 1038.10] by delivering amazing new performance
[1038.10 --> 1040.40] to the existing GPUs that they produce.
[1040.98 --> 1043.52] And the comments go a little something like this.
[1044.64 --> 1048.24] Maybe, maybe they'll actually have changed something
[1048.24 --> 1050.34] a lot this time, but I doubt it.
[1050.34 --> 1052.84] And I even heard one of them talking directly
[1052.84 --> 1054.30] to either AMD or NVIDIA,
[1054.36 --> 1055.72] not going to name any names here,
[1055.82 --> 1058.42] saying, look, if I spend the time
[1058.42 --> 1059.54] to benchmark this thing,
[1059.74 --> 1061.86] am I actually going to find this number you're claiming
[1061.86 --> 1063.94] or is it just a waste of my time again?
[1064.62 --> 1068.42] So the press in general is a little bit sick
[1068.42 --> 1071.36] of the whole claims that don't really seem to add up thing.
[1071.82 --> 1076.78] And unfortunately, it looks like this is maybe more of the same.
[1076.80 --> 1077.82] Okay, there are improvements,
[1077.82 --> 1080.94] but there might be a little bit more of the same going on here.
[1081.04 --> 1084.56] So we referenced articles from two of my favorite sites.
[1084.68 --> 1086.10] We actually didn't have the time,
[1086.46 --> 1087.34] just didn't have the bandwidth
[1087.34 --> 1088.58] to run the benchmarks ourselves,
[1088.70 --> 1090.32] but these are both very trustworthy sources.
[1090.60 --> 1092.68] Trust me that they're good for it, I guess.
[1093.10 --> 1095.16] So Anantec and PCPer.
[1095.28 --> 1097.94] So they both did pretty darn good articles
[1097.94 --> 1099.44] on the new NVIDIA driver
[1099.44 --> 1102.04] and the conclusions from both of them
[1102.04 --> 1103.68] basically amounted to this.
[1103.68 --> 1108.68] Anantec's test indicated about a 10% improvement
[1108.68 --> 1110.24] in performance in Bioshock Infinite
[1110.24 --> 1111.18] and Metro Last Light.
[1111.76 --> 1114.76] SLI performance increased 10% for Grid 2
[1114.76 --> 1117.88] and 76% for Total War Rome 2,
[1118.06 --> 1121.56] although Total War Rome 2,
[1122.06 --> 1124.68] you know, was pretty far back
[1124.68 --> 1125.56] in terms of optimization.
[1125.56 --> 1127.56] So anyway, a 76%,
[1127.56 --> 1128.68] that means they did achieve it.
[1128.74 --> 1130.16] So 76% performance improvement
[1130.16 --> 1132.18] in a particular game is good.
[1132.78 --> 1135.08] NVIDIA boasts huge performance gains
[1135.08 --> 1136.10] in older games.
[1136.42 --> 1136.96] So, you know,
[1137.12 --> 1139.04] Call of Duty Black Ops is getting 20%,
[1139.04 --> 1139.68] I guess.
[1140.02 --> 1141.52] And, you know,
[1142.26 --> 1146.28] SLI has seen slight performance increases.
[1146.28 --> 1148.92] And NVIDIA has been working on,
[1149.04 --> 1149.24] you know,
[1149.30 --> 1150.62] some mantle-optimized games
[1150.62 --> 1152.16] and all this kind of stuff.
[1152.24 --> 1154.28] But the basic conclusion was that
[1154.28 --> 1156.40] performance was only really increased
[1156.40 --> 1157.76] in some pretty specific games.
[1158.06 --> 1159.36] And for the most part,
[1159.52 --> 1162.04] the sensationalist headline performance improvements