Gadgetory

who knew well hello there you're watching Jays two cents some people call me Jay and some people call me well we won't get into what some people call me but today we're going to talk about mantle that is that new API that everyone's talking about that came out for battlefield 4 and AMD cards and all that sort of junk and I went ahead and tested a couple of different cards and we're going to see if mantle really lives up to all the hype that everyone keeps talking about let's go ahead and talk about what mantle is mantle is an API that was developed on the AMD platform specifically for GCN architecture cards but not all GCN architecture cards see it debuted just recently and it's being used on the r9 series on AMD now that's the 290x the 290 the 280x the 280 the 270x the ii said it you kind of see where we're going with this but it is going to be available on many 7000 series cards in fact from what I understand it's already usable on the 7000 series now when it comes to mantle is strictly an AMD API or can anybody use it well it is open source but it's going to be up to the developers of both games and the card manufacturers aka a and B and NVIDIA to actually make use of this new API but more importantly how is mental interface with the graphics card well as you guys know or may not know when games are developed they have to use some sort of architecture for the engine on the game to operate well here we are today at eleven point two DirectX 11 point two and we've been using DirectX for many many years so a new API was developed and that is mantle but instead of being developed by a third party like it was with Microsoft it was developed directly by one of the card manufacturers AMD or ATI which is a subsidiary of AMD AMD body ATI in their technologies now why should you even care about mantle especially if you have an AMD card today because as of the time of making this video mantle is not available on anything other than GCN architecture AMD cards well you should care because quite honestly it has everything to do with how your CPU and your GPU interact with each other you've always heard of the term bottleneck and bottlenecking is one of those things that has scared a lot of first time and even second third or even some novice builders who are concerned about trying to eliminate that bottleneck in their system typically the bottleneck happens to be the CPU because it can't handle all of the processes being sent to it by the graphics processing unit the GPU aka the video card well basically mantle alleviates a lot of that load which means lower end CPU based gaming rigs saw an immediate boost in performance now the same goes for lower end graphics card that may have been paired with a lower end processor now let's go ahead and talk about the test bench behind me because this is what we used I'm using an Intel i5 46 70 K at stock clocks now if we were to overclock the CPU we kind of alleviate some of that bottleneck which means that the differences in mantle would be a lot less apparent now to make sure that we can further create a bottleneck both of the graphics cards that I tested here were overclocked to their max so for clarification stock speed I 546 70 K and an overclocked GTX I thought we were talking about mantle here J there's no GTX in mantle the r9 270 from MSI and the r9 290 both from MSI they are the boxes as you can see both are the gaming Edition cards with their aftermarket coolers so that meant that we could bump up the overclock even farther than if they had been referenced models further creating a bottleneck or that's kind of what we're going for or we're looking to see if maybe there is no bottleneck hmm now let's go ahead and talk about my results and how I came to those results the results of that is how I came to the results this is a very clean show we're going to keep it that way I use the fraps frames per second benchmark utility which will record the frames per second over a 60 second duration giving me a minimum and maximum and an average frames per second I then did that test five times during the parcel storm map 64-player ultimate preset then took the average of the five averages so that's how I came up to these results now first up we have the r9 270 from MSI that is the overclocked version that I overclocked even farther to eleven hundred and fifty megahertz it's a pretty good overclocking card and with the thirteen point one two drivers which is not mantle I got fifty four point three nine frames per second average at 1080p now by doing nothing more changing no settings the same map in fact the exact same server I loaded up the beta fourteen point one mantle drivers and I got an average of using the same methodology I just described are you ready for this sixty two point five three average frames per second for a 15 percent frames per second increase on the r9 270 that is pretty amazing considering that was fifteen percent extra performance by doing nothing but installing a free driver now how did it do on the r9 290 well if you guys follow on Twitter you know I was having a hell of a time getting the driver to work right behind Lee got catalyst to cooperate and stop trying to run catalyst with thirteen point one two and fourteen point one on top of it it was just being really weird with the r7 it was perfect but the r9 it wanted me to completely uninstall catalyst completely wipe the driver about three times bring up the basic VGA driver install the catalyst fourteen point one four which then was showing up as thirteen point one two having me go into the device manager delete the driver again reinstall catalyst fourteen point one two more times and it finally then showed up as fourteen point one I have no idea why the 290 required me to go through all that but it did I figured it out and we were able to get our results so for the r9 290 gaming overclocked edition overclocked to one thousand 85 megahertz we got ourselves a 13.1 - which is a mantle or non mantle driver an average of 94 frames per second which is great that is absolutely great so how did it do with the mantle drivers we got ninety five point five that is one and a half frames per second more with mantle or a little bit over one and a half percent well guys remember if you look at mantle and what it's intended to do it is intended to remove some of the load off of your CPU allowing your GPU to perform to its full potential through better efficiency of the API which is what mantle is doing it allows your GPU to do what it's meant to do and allows your CPU to do what it's meant to do and it further widens basically the virtual bandwidth between the two and it all has to do with virtual optimization of that API now we can hope that maybe we'll see with refined drivers better improvement on the r9 290 now I know you guys are saying what about the 280 guys I don't have a 280 to test but I can tell you this if you are looking at buying a high-end r9 graphics card at 280 X 2 980 a 290x do not expect mantle to be giving you any sort of an amazing increase now I know there's other benchmarks out there that have shown as high as 10% improvement on the 290 for me that was absolutely not the case now that might be the case if I ran higher resolution this actually is a 1200 P monitor behind me but I didn't think that that would be fair to the 270 since the 270 is a 1080p graphics card so it may be a little bit better for the r9 running mantle at higher resolutions maybe we'll test that in the future but these are my results and take this as this is what GA experienced with mantle this is not the one benchmark and on the planet telling you all there is to do with mandala just shows you what my experience was with it now if you're on a budget this means that you can really expect to get better performance out of your graphics cards or your lower end AMD cards that you may have already purchased without worrying about busting the budget and still getting more performance to get 15% improvement in fps by buying new hardware could be a very costly move however the new mantel API allowed that to happen for absolutely no cost whatsoever so tell me what you guys think are you using mantle what card are you using with sound off down in the comments and let everyone who may be searching for mantle and mantle capable cards and help them decide what they should buy let me know about your experiences with mantle and whether it did or did not increase your performance in battlefield 4 as always guys I appreciate you watching this video I hope it's opened up your eyes for what mantle really is and what you can expect from it if you want to see more videos like this head on over to my channel and mash that subscribe button if you're not already a subscriber and if you are you're awesome and you know it let me get on out of here guys and as always I will see you in my next video