WEBVTT

00:00:00.840 --> 00:00:06.640
so this video is meant to be a few things first of all it is meant to be a

00:00:04.319 --> 00:00:12.320
somewhat of a review of the PNY GeForce gts450 that I did unbox a little while

00:00:09.360 --> 00:00:20.359
back okay it is also meant to be a look at SLI performance with the PNY

00:00:16.160 --> 00:00:22.400
gts450 as well as the MSI Cyclone GTS

00:00:20.359 --> 00:00:25.960
450 now these are both very similar cards I'm going to stop holding up the

00:00:23.760 --> 00:00:30.320
box and instead I'm actually going to hold up the PNY card itself now they're

00:00:28.000 --> 00:00:35.280
quite similar cards you can actually see now that we have them side by side they

00:00:31.880 --> 00:00:36.760
do use the same reference PCB design

00:00:35.280 --> 00:00:40.960
okay so the biggest difference is the cooler we did a cooler comparison but

00:00:39.239 --> 00:00:44.680
remember these cards will work just fine in SLI mode so that's what we were able

00:00:42.920 --> 00:00:47.320
to Benchmark them in so I want you to have a look at my test bench here and

00:00:46.039 --> 00:00:53.399
I'm just going to show you what I was using for this particular little video

00:00:49.840 --> 00:00:57.199
review I've got a core i7 860 running at

00:00:53.399 --> 00:01:00.199
3.8 GHz down there and that is with

00:00:57.199 --> 00:01:02.920
hyperthreading on all right I've got uh

00:01:00.199 --> 00:01:06.640
4 gigs of Kingston hyperx H2O RAM don't even bother trying to look at that it's

00:01:04.280 --> 00:01:10.759
way in the back it looks like this I have a third stick here okay so that's

00:01:09.080 --> 00:01:15.360
what that looks like all right then we've got nus motherboard which is both

00:01:13.040 --> 00:01:20.280
SLI and Crossfire compatible which was necessary for the testing today and this

00:01:17.560 --> 00:01:24.520
is all running off an ocz vertex series SSD all right so let's have a look at

00:01:22.479 --> 00:01:29.439
all the cards that we ran with today you can see we've got the GTX 480 in there

00:01:27.240 --> 00:01:35.079
as a reference point okay we've got the 58 70 as a reference point as well so

00:01:32.200 --> 00:01:41.840
the whole point today was to see how does two mid-range or really value

00:01:38.399 --> 00:01:44.159
performance cards like the 450 or the

00:01:41.840 --> 00:01:50.280
5770 you can see we've tested with a single as well as with dual 57 70s as

00:01:47.079 --> 00:01:52.640
well how does two uh more mid-range

00:01:50.280 --> 00:01:57.600
cards compare to a single high-end card from each vendor and then how does the

00:01:54.719 --> 00:02:03.079
NVIDIA offering compare to the ATI offering I also threw in the GTX 460

00:02:01.000 --> 00:02:06.759
Cyclone Edition because I did have it on hand and it's such a popular card right

00:02:04.840 --> 00:02:11.800
now I figured okay well that'll be a good reference point for people as well

00:02:09.520 --> 00:02:15.120
so stay tuned I'm going to do boring Benchmark graphs and before then I'm

00:02:13.720 --> 00:02:19.280
going to show you all the settings I was using in the games that we did

00:02:16.920 --> 00:02:24.319
test one of the games we tested obviously crisis is still a benchmark

00:02:21.720 --> 00:02:28.560
for optimal system performance under incredible load if it can run crisis

00:02:26.480 --> 00:02:32.000
really well then typically it can run most games fairly well although that is

00:02:30.360 --> 00:02:35.760
changing there are some fairly demanding games coming out these days we're

00:02:33.519 --> 00:02:40.920
running at 1680 X 1050 which is a perfectly reasonable resolution for our

00:02:38.640 --> 00:02:45.959
mainstream level cards but what you'll discover when we show you the graphics

00:02:43.159 --> 00:02:50.080
benchmarks The Benchmark numbers is that some of the higher-end cards do start to

00:02:47.840 --> 00:02:54.680
get CPU bottlenecked at that resolution and we don't see scaling even with more

00:02:52.159 --> 00:02:57.599
impressive graphics cards again once again this is mainstream cards that

00:02:56.200 --> 00:03:03.519
we're mostly having a look at so I'm using all of the medium presets for ch

00:03:00.920 --> 00:03:09.400
PR here's the settings we used for Mafia 2 but I do want you to note one thing

00:03:06.440 --> 00:03:13.280
the Apex physics engine here is off because I do have a Radeon card on the

00:03:11.120 --> 00:03:18.120
platform right now but I did run it two separate ways for all the NVIDIA cards I

00:03:15.360 --> 00:03:22.319
ran it once with Apex physics on high and then once with it off so the only

00:03:20.599 --> 00:03:27.680
time I'm going to compare against Radeon cards is with physx off but it should be

00:03:25.360 --> 00:03:33.519
noted that this game does have substantial uh gameplay innov ations

00:03:30.560 --> 00:03:38.239
that rely on physx that really you're not getting the ideal experience from

00:03:36.319 --> 00:03:43.120
unless you have physx on Batman Arkham Asylum yeah I know this

00:03:41.200 --> 00:03:46.720
game's not that new but frankly I don't care I really enjoyed it so I'm going to

00:03:44.920 --> 00:03:52.720
go ahead and Benchmark it also it serves as a reference point as uh an unreal

00:03:49.439 --> 00:03:54.280
three engine game so that allows us to

00:03:52.720 --> 00:03:57.439
find out how these cards will compare against each other in a pretty wide

00:03:56.000 --> 00:04:00.879
variety of games so these are the settings that we did use and you can see

00:03:59.400 --> 00:04:08.079
once again here I have Hardware accelerated physics off but I had it set

00:04:04.959 --> 00:04:09.519
to high with the supported NVIDIA cards

00:04:08.079 --> 00:04:13.439
and so what I'll do is I'll show you again two points of comparison one with

00:04:11.480 --> 00:04:19.479
physx off with all of the cards in the mix and then one with high with only the

00:04:15.959 --> 00:04:19.479
NVIDIA cards

00:04:38.280 --> 00:04:44.440
so in conclusion uh that was my second video review and I had so many people

00:04:42.800 --> 00:04:48.160
requesting it I couldn't help but do more I guess there's not really too much

00:04:46.560 --> 00:04:52.120
to say I mean you've seen the numbers already and really the prices haven't

00:04:50.479 --> 00:04:57.960
really haven't settled enough for the GTS 450 series cards to really make a

00:04:55.199 --> 00:05:01.919
clear recommendation in terms of FPS per dollar or some kind of a Hardline

00:04:59.680 --> 00:05:05.560
measurement but what we can see is that the cards do scale quite well in SLI

00:05:03.880 --> 00:05:09.759
they do perform quite well for their price point and if you want to check it

00:05:07.520 --> 00:05:17.080
out here I just hope you've all enjoyed this review of the PNY gts450 as well as

00:05:13.560 --> 00:05:19.000
the SLI running review I mean I guess

00:05:17.080 --> 00:05:23.199
it's uh it's more of a review of both of these cards than only one but I hope

00:05:21.240 --> 00:05:26.360
you've enjoyed it and I also wanted to make a special note here in the

00:05:24.560 --> 00:05:31.160
conclusion that I totally forgot about before I am running the absolute latest

00:05:28.880 --> 00:05:35.360
driver revision for all of these cards so that means the latest 260 download

00:05:33.080 --> 00:05:39.880
from the NVIDIA website as well as Catalyst 10.9 for all of the at cards so

00:05:38.400 --> 00:05:43.199
these are pretty much the most upto-date numbers that you can get for any of

00:05:41.479 --> 00:05:48.600
these cards don't forget to subscribe to lineus Tech tips and I hope you've

00:05:44.520 --> 00:05:48.600
enjoyed our video review