WEBVTT

00:00:05.839 --> 00:00:12.080
Adata XPG V2 memory kits are optimized for the latest Intel gaming platforms.

00:00:10.639 --> 00:00:16.080
Check the link in the video description for more details. Welcome to my

00:00:14.320 --> 00:00:20.880
performance overview, I guess you could call it a review of sorts, of AMD's

00:00:18.640 --> 00:00:26.320
Richland series of APUs. Now, we only got one of them. We got the

00:00:22.920 --> 00:00:28.560
A106800 K. And so we kind of went, okay,

00:00:26.320 --> 00:00:34.079
so this is going to be priced at around 145 bucks or so, uh, sort of in bulk

00:00:31.599 --> 00:00:38.440
pricing. So let's compare that against the obvious competition, which is, of

00:00:36.640 --> 00:00:42.960
course, a sea turtle. Well, no, but it felt that way a

00:00:40.960 --> 00:00:48.000
little bit because AMD's Richland is really in a class of its own when it

00:00:44.559 --> 00:00:49.800
comes to a lowcost value gaming

00:00:48.000 --> 00:00:56.079
platform. So AMD sort of released this Richland

00:00:53.520 --> 00:01:01.520
series of APUs. They're on the same FM2 socket as the older APUs. So you don't

00:00:59.760 --> 00:01:06.240
have to really upgrade your hardware. You can just get um a BIOS update for

00:01:04.559 --> 00:01:10.880
your board or whatever else and you can basically drop in a Richland. Also AMD

00:01:08.799 --> 00:01:15.280
has told me that we are going to be looking at FM2 for the next generation

00:01:13.280 --> 00:01:18.479
of APUs as well. So they haven't really changed the platform. In fact, they

00:01:16.960 --> 00:01:23.520
haven't really changed the architecture at all. So, it's mostly just a small

00:01:20.479 --> 00:01:26.799
speed bump over the last generation 5000

00:01:23.520 --> 00:01:28.479
series, like the 5800K APU. So, why did

00:01:26.799 --> 00:01:32.560
they do this? My guess is it probably has something to do with the release of

00:01:29.840 --> 00:01:36.720
Haswell. So, we took the lowest cost Haswell we could find and took a few

00:01:34.960 --> 00:01:41.360
mainstream games that you might want to play on a system that doesn't have a

00:01:39.040 --> 00:01:45.119
dedicated graphics card and we ran them head-to-head. Now, with that in mind, we

00:01:43.439 --> 00:01:49.920
didn't do any compute performance benchmarking because we already know the

00:01:47.280 --> 00:01:54.640
Intel chip is going to wipe the floor with the AMD one. We're talking about a

00:01:52.000 --> 00:01:59.439
very specific usage case here that is not focused on CPU. So, if you want

00:01:56.640 --> 00:02:05.320
better CPU performance, go with a Haswell. If you want better GPU

00:02:01.600 --> 00:02:10.080
performance, well, stay tuned because in

00:02:05.320 --> 00:02:12.160
Skyrim, our 4430, which is an i5 4430,

00:02:10.080 --> 00:02:17.440
Intel doesn't have an i3 yet. They're launching only the higher end of their

00:02:14.080 --> 00:02:22.480
Haswell CPUs, got beat by our 58 or

00:02:17.440 --> 00:02:25.120
rather 6800 K by a whopping 10 frames

00:02:22.480 --> 00:02:29.680
per second, 24 frames per second average on Richland versus 15. Now, with that

00:02:27.280 --> 00:02:34.080
said, the 5800K, the last gen one, was only one FPS behind this one. So, you

00:02:31.760 --> 00:02:38.400
can really see it's a small clock bump and not really anything revolutionary in

00:02:36.160 --> 00:02:45.319
terms of performance. Our next game up was Dirt 3. And in this one, Richland

00:02:41.040 --> 00:02:48.400
outperformed Intel Solution by about

00:02:45.319 --> 00:02:52.879
75%. That is fantastic. All the way from

00:02:48.400 --> 00:02:54.800
22 FPS or so to 37 FPS going for a

00:02:52.879 --> 00:03:00.160
different platform. Bearing in mind this one costs about $30 less than the Intel

00:02:57.680 --> 00:03:05.680
solution. Finally, in Starcraft 2, which is a very sort of mainstreamy type

00:03:02.480 --> 00:03:07.440
title, Richland wins by about 25% versus

00:03:05.680 --> 00:03:11.680
the Haswell GPU, and we see it once again very close to the 5800K. So, was

00:03:10.000 --> 00:03:16.239
this the most exhaustive test of benchmarks that we could have possibly

00:03:13.280 --> 00:03:21.360
run? No. Was this pretty much all you need to know about Richland? I would say

00:03:18.319 --> 00:03:23.040
yes. AMD's delivered a new chip that

00:03:21.360 --> 00:03:26.239
goes on their existing platform that you don't need a new board for. They've

00:03:24.640 --> 00:03:31.920
delivered outstanding graphics performance in a CPU or APU or whatever

00:03:29.680 --> 00:03:35.120
the heck you want to call it, which really isn't any different from what

00:03:33.200 --> 00:03:38.159
they did last time, but that fine because what they did last time was

00:03:36.319 --> 00:03:42.000
great, too. I've done a fair number of videos about their last generation APU

00:03:40.239 --> 00:03:45.200
and it's awesome. And so, I think that pretty much wraps it up. If you want to

00:03:43.680 --> 00:03:48.560
do entry-level gaming and you don't want to buy a graphics card, buy an APU.

00:03:47.200 --> 00:03:54.350
Thanks for checking out this episode of Linus Tech Tips. Don't forget to subscribe for more unboxings, reviews,

00:03:51.599 --> 00:04:05.570
and other computer videos.
