1
00:00:00,000 --> 00:00:06,759
looked at his original tests last week on AMD's Dual Graphics or the ability to

2
00:00:04,839 --> 00:00:11,639
basically run your integrated GPU and your discret GPU that's like an addin

3
00:00:08,840 --> 00:00:11,639
card in

4
00:00:16,359 --> 00:00:23,199
crossfire have you got an unlocked Intel processor do you want access to

5
00:00:20,640 --> 00:00:27,920
exclusive offers click now to join Intel Club Enthusiast and get started so we

6
00:00:26,320 --> 00:00:35,000
showed you the numbers that we got in a couple of games with AMD's latest fa

7
00:00:30,560 --> 00:00:37,719
7850k Apu alone that Apu running with a

8
00:00:35,000 --> 00:00:41,879
discrete graphics card an r7250 running by itself and we showed you what it's

9
00:00:40,079 --> 00:00:47,480
like when we pair them together in dual Graphics mode well you guys asked for an

10
00:00:44,760 --> 00:00:54,559
expanded test so we've thrown two more discrete graphics cards into the mix an

11
00:00:50,039 --> 00:00:57,719
r7260x and a GeForce 750 Ti and we've

12
00:00:54,559 --> 00:00:59,640
thrown in more CPUs so we've got a 4670k

13
00:00:57,719 --> 00:01:06,400
that's a Core I5 from Intel that we' run with discreet and on board a 6800k so

14
00:01:03,079 --> 00:01:07,759
that's AMD's own last generation Apu and

15
00:01:06,400 --> 00:01:13,920
an Athlon 760k which is a very inexpensive CPU

16
00:01:10,600 --> 00:01:16,600
from AMD with no onboard Graphics at all

17
00:01:13,920 --> 00:01:21,040
but based on the same architecture as that last gen Apu now before we dive

18
00:01:19,200 --> 00:01:26,200
into this I want it to be clear that cavar has other benefits that are not

19
00:01:23,400 --> 00:01:31,200
being tested in this video HSA or heterogeneous system architecture allows

20
00:01:28,280 --> 00:01:36,680
the GPU cores on Andy's latest Apu to be utilized for General productivity tasks

21
00:01:33,799 --> 00:01:41,200
much more efficiently efficiently than previous architectures allowed it can

22
00:01:38,720 --> 00:01:46,520
share system memory rather than having data copied back and forth between the

23
00:01:43,079 --> 00:01:48,280
CPU and GPU RAM over the PCI Express Bus

24
00:01:46,520 --> 00:01:54,159
which opens up a whole new world of cooperation between your CPU cores and

25
00:01:50,600 --> 00:01:55,840
your graphics cores this enables today

26
00:01:54,159 --> 00:01:59,360
better performance with some image filters improved thumbnail Creation in

27
00:01:57,799 --> 00:02:03,719
Windows better performance in Libra office and some other things but HSA is

28
00:02:02,079 --> 00:02:08,239
mostly about the future with some very exciting developments such as opencl 2.0

29
00:02:06,840 --> 00:02:12,440
with a compliant driver expected in early 2015 and all that good stuff the

30
00:02:10,920 --> 00:02:16,280
bad news is that we'll have to wait for software development to catch up but the

31
00:02:14,599 --> 00:02:21,400
good news is that application developers and compiler writers should if AMD is to

32
00:02:19,040 --> 00:02:26,280
be believed find it easier to code for HSA systems which should help Drive

33
00:02:23,720 --> 00:02:31,080
adoption with all of that aside if you're a value gamer and you're looking

34
00:02:28,360 --> 00:02:35,040
for a system today what do you buy well we saw your comments on the previous

35
00:02:32,440 --> 00:02:39,519
video and on the Forum and science wins so I was let loose and I was able to

36
00:02:36,800 --> 00:02:43,200
test almost all of the things all right guys as these graphs are a lot more

37
00:02:41,239 --> 00:02:46,680
compact with data than our normal graphs I'll give you guys some tips on reading

38
00:02:44,599 --> 00:02:50,440
them fairly standard Fair the CPU and GPU are listed on the left hand side so

39
00:02:48,920 --> 00:02:54,040
you can see what row that correlates with to help you guys differentiate

40
00:02:52,319 --> 00:02:58,560
between the rows I've also color coordinated each row with based on its

41
00:02:56,080 --> 00:03:03,319
CPU not its GPU so you can see how the different CPU combinations stack up that

42
00:03:01,040 --> 00:03:07,200
way also there's a dollar value right beside the name of the combination and

43
00:03:05,360 --> 00:03:10,920
that dollar value is an estimated guess based on numbers that I pulled from

44
00:03:08,480 --> 00:03:14,840
newegg.com so these are American prices if you're an international viewer I

45
00:03:12,640 --> 00:03:19,200
would highly suggest when you get to the dollar per performance graphs later on

46
00:03:17,239 --> 00:03:23,480
that's the second set of two that you base them based on uh numbers in your

47
00:03:21,440 --> 00:03:28,120
own region also these dollar values do not include the cost of the motherboard

48
00:03:25,959 --> 00:03:31,519
as that can vary quite a bit and was not a super important variable in these test

49
00:03:30,239 --> 00:03:35,560
I'm sure you guys will come up with your own conclusions and let me know in the comments below and on the Forum and all

50
00:03:34,439 --> 00:03:40,799
that kind of stuff but here's some conclusions of my own that I gathered from looking at my graphs after I made

51
00:03:39,080 --> 00:03:45,560
them there seems to be a sweet spot for both of these games in Far Cry 3 that

52
00:03:42,920 --> 00:03:49,360
sweet spot is about $7 while in BioShock that sweet spot is about $5 that makes

53
00:03:47,480 --> 00:03:54,000
sense because bshock infinite is easier to run than Far Cry 3 Therefore your

54
00:03:51,599 --> 00:03:58,200
dollar per performance goes down which is a good thing um but you're going to

55
00:03:56,319 --> 00:04:01,439
need to be able to run certain games so it all kind of makes sense although be

56
00:04:00,040 --> 00:04:06,120
completely honest for most of these setups the price that you're paying

57
00:04:03,280 --> 00:04:10,640
you're getting 1080p on high at no problem so it makes a ton of sense uh

58
00:04:08,640 --> 00:04:16,320
especially one last thing personally I would not buy a solo 4670k for gaming by

59
00:04:14,000 --> 00:04:20,400
that I mean 4670k and no discrete graphics card it does not perform very

60
00:04:18,519 --> 00:04:28,680
well and personally I'd rather pay less than a 4670k and get a 760k and an

61
00:04:24,520 --> 00:04:31,520
r7260x or pay $10 more and get a 760k

62
00:04:28,680 --> 00:04:35,280
and a 750t and get even more performance on top of that that see makes a lot more

63
00:04:33,440 --> 00:04:40,120
sense to me but again just let me know what you guys think in the comments down below and on the Forum so a main

64
00:04:38,360 --> 00:04:43,919
takeaway I would grab from this is that going back a few generations and

65
00:04:41,639 --> 00:04:47,800
grabbing a processor from that time at a really low price and then buying a nicer

66
00:04:46,160 --> 00:04:51,440
more powerful graphics card and pairing those two together is an awesome

67
00:04:49,560 --> 00:04:56,360
combination and you get a lot of power for your price with that combination

68
00:04:53,880 --> 00:05:01,320
another thing is that 1080p gaming has been never this viable at a price point

69
00:04:59,080 --> 00:05:06,560
like this you can run games at high at 1080P and get epic frame rates that's

70
00:05:03,759 --> 00:05:11,039
really cool to see because value gaming has always kind of been a shove under

71
00:05:08,360 --> 00:05:15,199
the bus but now you can go with a very nice resolution very nice settings at a

72
00:05:13,639 --> 00:05:18,680
very nice price point and have a really good experience and the word console

73
00:05:17,039 --> 00:05:21,960
killer was brought around a lot when the consoles were first coming out and we

74
00:05:20,440 --> 00:05:25,759
all knew it would be easier to surpass consoles later on down the line I didn't

75
00:05:23,840 --> 00:05:30,280
realize it would be this fast with certain setups that we have in this

76
00:05:27,039 --> 00:05:33,759
video Even and with 750 Ti 750

77
00:05:30,280 --> 00:05:36,199
and 260x now taking up bat it's really

78
00:05:33,759 --> 00:05:39,440
easy to actually just beat the absolute crap out of a console a lot of console

79
00:05:37,680 --> 00:05:43,919
games aren't running at 1080P and they're not running that fast and with a

80
00:05:41,720 --> 00:05:48,240
setup like this you can run it at really good frame rates at 1080p at high

81
00:05:46,120 --> 00:05:52,199
settings with AA on and all that kind of stuff so it's a very interesting setup

82
00:05:50,400 --> 00:05:55,319
and I hope you guys are going to take advantage of this kind of stuff if

83
00:05:53,400 --> 00:05:59,960
you're building a console build in the future just so you guys know as well our

84
00:05:57,720 --> 00:06:02,280
Star team is in second place and I know most of the times I upload a video and

85
00:06:01,319 --> 00:06:07,319
call it what place we're in it's inaccurate but I think this might be the first time it's actually accurate so be

86
00:06:06,039 --> 00:06:13,479
sure to join up we're going to have a lot of fun and the dog fighting module should be coming around sometime around

87
00:06:10,880 --> 00:06:18,080
April 18th so Luke's testing reveals a bit of a problem with the way the term

88
00:06:15,720 --> 00:06:24,560
value gets thrown around to describe inexpensive products volume is not the

89
00:06:21,360 --> 00:06:26,360
same thing as cheap I understand that

90
00:06:24,560 --> 00:06:32,000
for some people they're going to have $300 to spend on a PC and even if they

91
00:06:29,599 --> 00:06:37,000
could have something 10 times better for another $10 that won't be feasible so

92
00:06:34,960 --> 00:06:42,080
they will need to settle for the reduced price to Performance ratio of a cheap

93
00:06:39,400 --> 00:06:46,479
computer for those folks who do have the regular income that would allow them to

94
00:06:43,680 --> 00:06:51,160
save a couple more allowances or skim some money off a couple more paychecks

95
00:06:48,479 --> 00:06:56,120
the advice I've always given wait until you can afford something decent

96
00:06:52,960 --> 00:06:58,680
continues to ring true grabbing a cavar

97
00:06:56,120 --> 00:07:03,639
today and running with on board then upgrading later with a discret graphics

98
00:07:00,879 --> 00:07:07,919
card does make a fair bit of sense if you're impatient and you don't want to

99
00:07:05,120 --> 00:07:13,680
wait to save up another $60 to $70 for the cheap CPU and mid-range GPU up front

100
00:07:11,080 --> 00:07:17,960
but make sure that when it comes time to upgrade to a mid-range GPU you are

101
00:07:15,919 --> 00:07:23,000
picking one based on its individual merits not just based on Dual Graphics

102
00:07:21,280 --> 00:07:28,840
compatibility the r7260x or the 750 Ti cannot take

103
00:07:26,440 --> 00:07:34,440
advantage of the onboard graphics for you know and ened gaming the way the

104
00:07:30,400 --> 00:07:37,199
r7250 can but because they're so much

105
00:07:34,440 --> 00:07:41,280
better it doesn't matter this may change in the future when the onboard can be

106
00:07:39,000 --> 00:07:45,759
used for AI or something as a co-processor but we're still a fair way

107
00:07:43,479 --> 00:07:50,319
away from that and if you're buying a graphics card today for games you can

108
00:07:47,960 --> 00:07:55,560
play now and you want the most for your money then you're going to want to get a

109
00:07:52,879 --> 00:08:00,800
proper discrete graphics card just like has been the case up till now sorry dual

110
00:07:58,199 --> 00:08:05,199
Graphics uh you're just not the most for my money at this point in time speaking

111
00:08:03,919 --> 00:08:09,599
of getting the most for your money we've got an exciting new sponsor now I

112
00:08:07,599 --> 00:08:13,720
switched to an electric shaver 3 years ago because I was too cheap and more

113
00:08:11,919 --> 00:08:16,159
importantly too lazy to deal with the [ __ ] of buying replacement blades

114
00:08:15,319 --> 00:08:21,159
for my Razor it does a decidedly worse job of

115
00:08:18,840 --> 00:08:25,400
removing hair from my face and to my dismay it turns out I'm still supposed

116
00:08:23,280 --> 00:08:29,720
to buy replacement blades for the thing I haven't replaced the blade on it once

117
00:08:27,599 --> 00:08:34,039
so I guess it's time for a new strategy Dollar Shave Club you know the our

118
00:08:32,000 --> 00:08:37,880
blades are [ __ ] great guys their whole concept is that for a few bucks a

119
00:08:36,279 --> 00:08:41,880
month less than the price of buying brand name razors and cartridges with

120
00:08:39,399 --> 00:08:46,240
unnecessary like vibrating handles and like built-in speakers and stuff you can

121
00:08:43,919 --> 00:08:50,720
get fantastic blades shipped straight to your door by a highly skilled and

122
00:08:48,399 --> 00:08:54,279
diverse team of people and bears the cost savings is great but for me the

123
00:08:52,480 --> 00:08:59,440
appeal is really the convenience of having fresh bathroom stuff delivered

124
00:08:56,760 --> 00:09:03,920
every month no more like you know maybe maybe I can get like one more shave out

125
00:09:01,320 --> 00:09:07,760
of this Elvish replica blade bam there goes the jugular they've also got Dr

126
00:09:06,160 --> 00:09:11,519
Carver's easy shave butter which is clear instead of foamy so you can

127
00:09:09,320 --> 00:09:15,160
actually see what you're doing and one wipe charlies's which are basically butt

128
00:09:13,200 --> 00:09:19,120
wipes for men with a peppermint infusion I tried to say that with a straight face

129
00:09:16,720 --> 00:09:23,079
but I couldn't do it so start shaving time and shaving money join now at

130
00:09:21,200 --> 00:09:27,920
dollarshaveclub.com Linus and see what it's all about it's available in the US

131
00:09:24,920 --> 00:09:29,760
Canada and in Australia guys like and

132
00:09:27,920 --> 00:09:33,160
share this video if you liked it dislike it if you disliked it and leave a

133
00:09:31,160 --> 00:09:36,440
comment on the lineus tech tips Forum Linked In the video description if you

134
00:09:34,600 --> 00:09:39,680
want to discuss this product or technology we didn't really feature a

135
00:09:37,880 --> 00:09:43,200
product in this video or if you have constructive criticism for me and my

136
00:09:41,320 --> 00:09:46,120
team don't worry we can handle it also linked in the video description is our

137
00:09:44,680 --> 00:09:50,240
support link with options to buy T-shirts give us a monthly contribution

138
00:09:48,160 --> 00:09:53,920
to help us keep making videos or give us a kickback whenever you buy random junk

139
00:09:51,920 --> 00:09:57,959
on Amazon so check that out if you enjoy our videos it helps us out a whole bunch

140
00:09:55,600 --> 00:10:04,409
oh and always thanks for watching guys and don't forget to subscribe

141
00:10:07,880 --> 00:10:10,880
he