Gadgetory

Raven Ridge ap use are interesting as products in a world where MSRP acted as an infallible decree handed down by Galactic overlords the GT 1030 would cost $70 the RX 560 would cost $100 and the G 45 60 would have always been $60 in this world the real one the GT 1030 has now used served both the GTX 1050 and rx 560 in price landed at 110 to 120 dollars for a really low end graphics card and the G 45 60 has actually fallen in price oddly down to 60 from an overpriced $80 previously and then the our 320 200 G and r5 2400 G entered the market priced at 100 and $170 respectively before that this video is brought to you by Thermaltake and the view 71 enclosure the view 71 is a full tower case that's capable of fitting three video cards and most configurations it's also one of the better cooling cases in our recent case testing bench lineup the view 71 has hinged a tempered glass doors on either side that make it easy to open and show off and it comes with at least one rain fan though you can get the RGB version if you prefer learn more at the link in the description below to recap our previous coverage of Raven Ridge basically we bought the APS and we immediately did some d lid and extensive thermal coverage if you want that go check the previous videos today we're focusing only on gaming benchmarks no power no thermals no nothing other than gaming because ultimately aps are meant to go in gaming machines that are low-end so ap use the argument is that it is for price conscious users this means that there's no room for the usual wishy-washy this part's fine except if you have $20 more you can buy this one or $20 less you buy this one there's no room for that you either buy it or you don't because when it's cut when it comes down to these cheap parts the alternatives are a G 45 60 and a GC 1030 for some price or an r3 and a GT 1030 or something like that so a GT 1030 should be $70 it's presently at time of filming a 110 to 115 which just really because that's what the 1050 in the 560 were priced originally but that's the world we live in the are the G 45 60 was $80 it's now 60 and the R 3 CPU is about $100 so ultimately you're looking at an r3 plus a GT 1030 they'll just pick that one for now because it's what we tested those come out to about $200 kind of insane a G 45 16 a 1030 thanks to the new in 4560 price reduction comes out to about 170 to 200 depending on where and when you buy it the R 3 and R 5 ap use are $100 and $170 which is what makes the argument sort of tempting and interesting especially for the our 320 200 G we think that's the rock star that you should be looking to because it's $70 cheaper than the r5 and it's pretty damn close to it in performance as you'll see today so for that previous coverage check the thermal content otherwise today we're talking about overclocking and gaming performance and we also previously talked about DRAM scalability where we tested memory scaling for frequency for cache latency and for channeling and you can find that content elsewhere but starting out we did a lot of overclocking on these AP years please note that the gaming k5 gigabyte motherboard we used exhibited some variability in the base clock so although we would set a multiply of 39 point five or 39 we would often see clocks that were a bit lower than 3900 because the base clock would fluctuate between basically 96 point something and 101 to pay on how it felt so what that means is that in the charts when you see three point nine five gigahertz it's set to that but it's possible that it was dithering between whatever 96 times thirty nine point five is versus a hundred times thirty nine point five let's just quickly put one of the memory tests on the screen from previously as well just to recap everyone we didn't do fire strike or time spy or something like that we saw some decent scalability on the APU as part of what we were testing was how do these perform with different kits of memory and we can cycle through a few of the previous charts to show the differences ultimately what we found was that our Trident the thirty two hundred megahertz seal 14 KITT seems to play better with the motherboards we used and we tested multiple of those and the AP use and so that's what we're using for the main testing and for the G forty five sixty we're using twenty four hundred megahertz Ram you can learn more about why we made these choices in the test methodology section linked in the description below please click that link before you ask questions it's probably answered there so with all that said articles below if you need more info on previous content otherwise let's get into the charted data for gaming today getting straight into the eSports titles we're starting on overwatch at 1080p medium as discussed in our previous bench Theory test duration content we run five-minute test passes for overwatch as its dynamic nature introduces a lot of variants the five-minute passes smooth this out you can learn more about our overwatch testing methodology over in the bench Theory Series previously the GT 1030 results are close to margin of error of one another there's some overlap and so we can call the GT 1030 with a forty five sixty and twenty four hundred megahertz Ram as appreciably equal in performance to the overclocked are 329 33 megahertz memory and GT 10:30 the and the r5 2400 G overclocked to three point ninety five gigahertz and sixty hundred megahertz on IGP performed respectively close to the discrete devices the part lands at 60 FPS average with lows similarly scaled to its competition the apu ends up within margin of error of the GT 10:30 and 3.9 gigahertz r3 1200 with the g 45 60 and 1030 running a couple percentage points ahead stock the r5 2400 g with 32 hundred megahertz CL 14 Ram performs at 54 FPS average resulting in its overclocked granting a 12 percent jump over the stock apu stock the 2400 g doesn't impress in this game but it's overclocked changes the stack significantly the are 320 200 g operates at 50 FPS stock with a 2400 g leading by about 7% for a $70 price increase that's not a big jump in performance our 2200 g overclocked the IGP a bit higher than our 2400 g and ended up at 55 point 5 FPS average it seems that there's some inherent - these 2400 G over the 2200 e4 overwatch it's just a question of whether that's worth $70 to you for an extra couple percentage points of performance and that depends on how much you care about this one game rocket League has the discrete parts placing in the top of the chart plotting within margin of error of one another this is an instance where we've become bound by the GT 1030 and not by their partnered CPUs and illustrates why we eliminate bottlenecks and discrete component reviews this however is not one of those integrated GPU testing requires a more value driven method rather than a perfectly clean test method the are 524 G lands at around 61 FPS average when overclocked placing it's average FPS within our error margins for the top two discrete results these are functionally equivalent an average framerate they are not equivalent in lows however we noticed that the apu underperformed in 0.1 percent lo metrics something we can better show with this frame time plot the harder stutters are rare fortunately but they do occasionally pop up and get noticed overall the are 524 G is doing well to keep up with two dedicated processing components but the two discrete components are still doing a bit better back to the chart the our 320 200 G stock apu who places at about 54 FPS average led by the stock 2400 G by about 7.5% once again for a $70 difference this gap doesn't seem worth the jump even low-end frame time performance isn't all that disparate between the two AP years overclocked in the 2200 g pushed it to 60 FPS average impressively within standard deviation of discrete component performance and highlighting again that the 2200 g is really the part to look at here again the low-end frame time performance here isn't impressive not in this game but for $100 part we do have to remark again that this is hands-down very impressive AP is in the past things more similar to this one for example have not done too well it's basically always come down to really wishy-washy like well you could use it in this specific HTPC build I guess but ultimately a lot of the time you can buy a discrete component when the whole shortage thing didn't exist and as cheap CPU and you'd get better performance for gaming that's changed Eisen and Vega together have really done a lot for the APU discussion for Andy the 2400 G however really struggles to look valuable in the face of its own alternative the 2200 G its brother basically so in this testing so far and we haven't gone through all of it the $100 2200 looks really damn good because it's $100 a GT 1030 with a G 45 60 which is also incredibly impressive and performance for the price it doesn't come close in terms of price to performance even though the 2200 G is a bit behind in some scenarios you can always drop settings a little bit to recover some of those frame time differences so quite a good show so far but we do have a couple more charts to go through let's move on to Sniper Elite Four this game is sort of a best case for AMD hardware whereas something like Source engine games can be more of a worst case tonight for at leats asynchronous compute and proper dx12 implementation are well leveraged on the APU first note that this is 1080p high the settings are aggressive for an APU and you should drop to lower settings if you're actually going to play this game the point however is to use the higher settings because we already have a lot of GPUs tested with this game so we have some extra data we can show treat this more like a synthetic test though know that the scaling is linear as settings are reduced so all scales relatively equally anyway we just like this 4 d GPU data one more note the D GPUs if not otherwise listed we're tested on our 7700 K GPU test platform so you get an unrestrained to look at lower end GPU performance but not an equivalent look in terms of price obviously we still have the GT 1030 and the cheap CPU for that the are 524 G experience is at 10% uplift when an overclock from baseline the GT 10 30 and 45 60 using the slower memory mind you at 2400 Hertz end up with a stock r5 leading by a noteworthy 17% the overclocked are five leads the GT seven thirty and forty five sixty by 29% with again no overclocking on the 1030 and none can be done with 45 60 the 2400 G does well in this game and would clearly be the better option to the GT 10 30 and 45 60 we don't have data on this one for the 2200 G though our civilization 6 benchmarks use turn time as a metric for performance rather than frames per second this is for a few reasons one of which is that FP analysis terrain sieve is largely irrelevant and the other is that it can often be invalid the longer a processor takes to complete a turn the higher its framerate in other words slower processors perform better in fps value depending on how you test which is because they spend more time staring at an unchanging screen to process the turn because this doesn't rely upon GB is retesting we can show all other CPUs we've tested lately any CPU with unspecified memory speeds is running at 3,200 megahertz in our testing we found that over clocks are 524 G to perform roughly equivalent lead to the r5 1600 X and turn time processing planning at about 19 to 20 seconds per turn for five players it would take about one point six minutes to complete a full round until your next turn stock the 2400 G completes its turns approximately 2 seconds slower than overclocked granting the overclocks 2400 G at 9% turn time reduction the overclocked 2200 G completes its turns in about twenty four point seven seconds with a stock 2200 G distant from the pack down at twenty seven point five seconds the stock 2400 G does hold a meaningful time reduction of 22 percent over the stock 2200 G and the overclocked two variants put the 2400 G as 21 percent reduced we haven't yet run the G 45 60 in his bench but the R 3 1300 X runs between the stock and overclocked r5 2400 G CPU entries csgo was one of the more sensitive games to memory changes as we discussed previously we're still using the tried NZ kit as our baseline as it worked the best with the gaming k5 and the AP use the r5 2400 G stock CPU performs at 95 FPS average with lows reasonably in lockstep the our 320 200 G operates at 86 FPS average permitting the 24 energy a lead of 10.5% the discrete components land between 111 and 120 FPS average which is a substantial lead over the overclocked and similarly priced between 0 and $20 different our 520 400 G the ro 322 energy is the real winner here at $100 it's providing 83% of maximum performance as rated relatively versus the chart-topping combo to achieve 83% of performance at 58% of price plus or minus the impact of $10 is pretty damn impressive granted this is after we overclocked the r3 APU but overclocking is not particularly hard on these CPUs anyway rise and master makes it more accessible than before despite a preference for bios overclocking for this title specifically we'd recommend either the 2200 g4 something ultra cheap or the discrete component combination the r5 2400 G has limited usefulness in its price bracket for this particular title dota 2 will likely mirror csgo given the same roots but let's take a look we've become GPU bound and dota 2 with the G 45 60 and overclocked r3 1200 both maintain in 63 FPS average either would be a suitable choice for our settings the our 320 200 G with a 16 50 megahertz GPU overclocked roughly ties the overclocked r5 2400 GE both at 52 51 FPS average joining 4G it does manage to maintain advantaged 1% low frame times though stock the 24 of G holds a 45 FPS average with a stock 2200 G not meaningfully different it would seem that we are bound elsewhere in the stack total war hammer at 1080p medium at places the overclocked AP use and commanding positions managing to take the top two slots with the IGP overclock there are three 1200 and GT 1030 follow in third with the are 320 200 G outperforming this discrete combo by roughly 6 percent the r5 leads the r3 by about 10% thanks to the game's actual utilization of additional resources as for stock the AP has run behind the discrete components our r5 24 energy stock apu operated at 35 FPS average with overclock in the 24 g granting a significant 20 ish percent uplift we also have some 1080p high results in Ghost Recon results for scaling performance in the written review linked below so that recaps the performance a couple things here as always if you're new with any kind of game benchmarking we can only speak to two things the game as we tested and the games related to the games we tested that's why we choose a smaller suite of games that uses known engines ideally something a little bit different for each one while still being popularly played titles so with the games we've tested and ignoring all other factors focusing strictly on gaming performance what we're in is a situation where if you're buying a 100 to $170 product you probably don't care about things like necessarily consumption or thermals although we covered that extensively and they're fine or things like that we care about is game fronts for the dollar for the dollar the APU that hands down gets our recommendation if you're really struggling to piece something together would be the our 320 200 G you have a lot fewer expansion options going forward with an r3 2200 G it's more limited on the CPU side if your plan is to fill what I would assume is a more limited use case but one that is still somewhat common if your plan is to fill the system with a DGP later a discrete card and upgrade and keep the other thing as a CPU then you're gonna be more limited with a 2200 that's just how it's gonna be so that's your use case you can or the recommendation but if it's not if you're just trying to build something cheap that games pretty well we like the 2200 G a lot at $100 it's really good value it's within single-digit percentage points in most cases of the 2400 G and it's behind yes the discrete components but it's $100 and the 4560 with the GT 1030 which we also really like by the way those two parts are closer to 170 on a good day plus or minus a bit so either way you look at it you're at least seventy dollars cheaper with the 2200 G now of course you would want to go with faster memory we did some memory frequency and kit scaling and a previous content piece if you're curious about the impact of 2429 33 3200 whatever check the content will tell you there how much it matters to buy the kits I believe our conclusion was something along the lines of 29 33 is perfectly fine it's not a huge performance disparity if you can save another $10 there or what it depends on your region then do it because ultimately if you're buying a $100 part saying well but going up to 3200 from 29:33 is only $13 more no stop it no it's like bad because what you're doing that game when you're saying it's only $13 more that's a 13% increase on what you're spending on the APU when you're spending that much more money you're getting close to just doing either the higher-end APU and then still spending $13 more or doing discrete components because if you buy a 2400 G at $170 or whatever it is and you're spending an extra couple bucks to go from 29:33 to 3200 depending on your region and where you live then you're entering territory where the price is now at equivalents with the discrete components and 2400 megahertz of RAM so consider that keep that in mind that's why the 2400 G doesn't impress enough frequently enough in these titles mind you to really get an outright recommendation it's fine but there's nothing special about it the 2200 G is quite special in that it's very affordable it's surprising how well it performs with reasonably high settings and you can stick 29:33 megahertz of memory in there and be quite happy with it in the US the memory prices aren't as disparate as in other regions we've looked sometimes going from 29 33 or 2666 to 3200 can be a big jump so you don't necessarily lose a lot by going with slightly slower memory it's not as crazy needed as just going with two sticks of it that's important but the couple hundred megahertz jump not quite as much and if you're value driven then keep the value in mind spend the money on something else that's also important for your system like cooling or whatever it may be so yeah recapping this briefly from a games we've tested we quite like the 2200 G the 2400 G we don't really feel like there's a good spot for it that's as clear of a standout like screaming by me spot in the market it's one of those things you look at and more or less way the options at the time what is the discreet combination cost what is the 24 energy cost it might be better to buy the discreet combination in a lot of these tests the GT 1030 and an R 3 or a G 45 60 and the the latter being more comparable in price were decently better than the 24 energy for not withstanding and games like sniper doom would likely be factored in Wolfenstein would likely be factored in so those games notwithstanding if you're looking at eSports titles and you're trying to spend about two hundred bucks the discrete combination still looks better but if you're trying to spend less we do like the 2200 G is worth buying so those would be my two choices based on the data we've collected it's 2,200 G and cheap discrete combo not a lot of space for the 2400 G until it's price drops and it will it'll eventually come down once it's 150 that'll be a really compelling argument but at 170 not quite so much so that's it for now if you want to see more information check the article linked in the description below where we'll have additional charts and information on all of this subscribe for more as always go to store that gamers Nexus net to pick up at one of our mod mats they will be shipping the end of this month or early next month most likely and that's our next round so thanks for watching I'll see you all next time