Gadgetory

so the original title of this video is gonna be something along the lines of I 580 600k slash 8400 verse r5 2600 X slash 2600 you guys get the point right a typical benchmark video but if you tech tubers already beat me to that I'm still gonna throw those benchmarks in this video but I want to focus more or less than something that has been covered very much yet and that is xfr 2.0 on the DL XFR stands for extended frequency range and it does just that allows the CPU to clock higher than the advertised turbo boost so it's almost like a turbo boost on top of a turbo boost which is a bit confusing but xfr 2.0 will usually kick in if you have an adequate cooling solution so if you have a beefy enough air cooler or a IO or custom loop then x4 2.0 will be able to take full advantage of that extra overclocking Headroom you might otherwise have on a risin two chip I just called it risin - and technically Zen + is not rising - at least AMD doesn't refer to their own architecture that way so I really don't think we should if you want to be consistent with what they're showing us so what I should have said was Zen plus or rise in second generation I don't know these naming schemes are getting really stupid anyway the topic today's video I'm gonna compare the r5 2600 x2 itself with xfr 2.0 enabled versus a manual overclock the 4.2 gigahertz yes on a be 350 board albeit with slightly crispy vrm temps but everything seems stable nothing was overheating during the 30 or so minutes of just straight-up benchmarking all of these games I hope that a dozen or so included in this video also some synthetics and I also decided to throw in another Adobe premier render so you guys can see how productive the CPU actually is in the real world when it comes to content creation something I want to mention though before jumping into these graphs the RFI of 2600 X in my benchmarks in particular this is a good reason for this I'll explain in a second sometimes outperforms the are 727 or text in games not in you know workload scenarios where you're either rendering a video or using blender or handbrake or what-have-you obviously the higher core countship will come out on top there all other things equal for the most part they were here's the distinction here's why in some cases for the games the r5 comes out on top with my benchmarks it's because the r7 was overclocked manually to 4.2 gigahertz across all course now the r5 was not manually clock save the RAM frequency which I had enabled XP profile for so 3200 Hertz on the RAM timings I think 16 18 18 and I want to say that is all I changed and then I left the CPU frequency V core base clock all of that unchanged again in a be 350 BIOS the 2600 x1 in a few cases because of x2 for 2.0 and what happens is this X of our 2.0 allow certain cores for very small amounts of time to turbo to a higher frequency than what I could manually get all of my cores on the 2700 X simultaneously - so 4.2 gigahertz was it across all course I didn't do per core overclocking just all eight at 4.2 and what that does is essentially offset any potential you know overclock you could have on a single core so maybe I could hit 4.4 gigahertz on core 1 but then 4.1 gigahertz on the other seven chords right I didn't do that because that's a very case-specific scenario well the next of our 2.0 enabled for any of those n+ cpus that aren't manually overclocked the turbo frequencies at any given time can be slightly higher than 4.2 gigahertz and because of this and also because of the fact that many games aren't optimized for six cores let alone eight cores and 16 threads the 2600 X occasionally comes out on top and that's why I'm willing to say in this video that if you play video games primarily with your Zen plus CPU then you might want to give manual overclocking a second thought is it really necessary I don't think so in fact I think you'll be better off in many cases when it comes to games with just letting X afar do its thing now today's test bench is exactly the same as it was three days ago when I benchmark the 2700 X so a 10-7 TTI from EVGA 16 gigs of DDR 4 o'clock to 3200 megahertz and depending on the platform either a Z 370 or X 470 motherboard with the exception of again the 2600 X which I used to be 350 motherboard for more or less just to prove a point right that you don't need the most expensive motherboard out there to hit the overclocks that you desire by the way a few of you tried to point out that my 1070 Ti and the test bench was somehow a massive bottleneck and that all my frame rates are being skewed as a result I wholeheartedly disagree with that first off the 1070 Ti is an insanely powerful car it's nothing to just look at man it's not bad at all in fact we compared it to last gen architecture right Maxwell stuff the 1070 TI is more powerful than the 980 TI so by that definition all of our benchmarks from three years ago were invalid because our graphics card was always the bottleneck there's always gonna be a bottleneck no matter what you do the idea is to eliminate the graphics card bottleneck as much as possible to leverage the CPUs the most but then if you scale the graphics card up all your frame rate should scale about the same assuming that your graphics card was the bottleneck to begin with so the logic there just doesn't add up to me and that's why I'm using the 1070 T I also it's just more representative because not everyone is a 1080 TI which is the best thing out there these are more real-world results so starting off first with Cinebench r15 I want you to pay very close attention to the x-axis where all the CPUs and their respective frequencies are listed as you can see the 1700 X from AMD reached 4 gigahertz that was the highest overclock I could attain with it 16:19 the multi-core side is not bad at all it was by far though the lowest scoring single core performer the 2700 X which is essentially the step-up right a generation above was able to reach 4.2 gigahertz with the x4 70 motherboard that jumped our single core performance up to 184 and the 1900 and 5 on the multi-core side which is very impressive the 8700 K from Intel which is by far the best gamer in this lot scored 16 84 at 5 gigahertz again not bad and is also the highest scoring single core performer the r5 2600 X at stock scored a 13 97 which isn't bad 186 in the multi-core side and the r5 2600 X with a manual overclocked to 4.2 gigahertz scored 1489 so we can see how exif are is really only benefiting a few cores which is why when running its stock it scored lower on the multi-core side of things and we can see our lonely i5 8400 on the far right of course running at stock because it has a lock multiplier which is a shame only 942 on the multi-core side 176 for the single core peak bench 4 tells a similar story not gonna spend too much time here the r7 2,700 is by far the best multi-core performer and with a score 4890 is also the best single core performer although the i5 8400 at stock isn't far off at 48 40 on the far right now for Adobe Premiere Pro more or less a real-world render scenario a five-minute 1080p 60 FPS video file run through the h.264 YouTube 1080p preset resulted in the following times the I 970 900 X at 4.6 gigahertz was the fastest render but again not by far and the CPU is it's just really not one you should consider because it runs extremely hot it's very expensive the platform in general is very expensive and look the i7 8700 K is not far off at 5 gigahertz it runs cooler the platform is cheaper and yeah you're just gonna have a better time with this one but then look to the left again at the r7 2,700 X is only 10 seconds slower at 4.2 gigahertz this is a manual overclock we do expect the manual overclock to benefit us more in these work oriented scenarios the course with games things will change you'll see that shortly the r5 2600 X add stock resulted in a 264 second render time not too bad especially when compared to the similarly priced I 580 400 it's stock and if we manually overclock to 4.2 gigahertz across all our five cores 260 is our time in seconds that's just 4 seconds shaved off going from stock to a manual 4.2 overclocked this kind of show us how Adobe Premiere Pro slightly favors frequency a bit more than core count to an extent now these gaming benchmarks or where I think the r5 2600 X definitely shines not only because it's again around $200 which is an insanely good price for what you're getting 6 cores 12 threads overclockable and on cheap platforms - that is everything you'd ever want in an affordable content creation and gaming oriented platform so all of these CPUs you see from here on out these benchmarks were tested at the following frequencies the 1700 X at 4 gigahertz the 2700 X at 4.2 gigahertz the 8700 K 5 gigahertz and the 2600 X left in stock 4 xf r 2.0 to do its thing and check out grand theft auto 5 in the 1080p resolution the 2600 X definitely holds its own only a for fps Delta between averages with it in the 2700 X and it actually performs better with 0.1% lows which means a stuttering was less frequent with the cheaper 6 core counterpart we see the opposite trend with player unknowns battlegrounds in 1080p the 2700 X holds a 124 average framerate 70 on 1% lows and 36 the lowest point 1 percent of frames and the 2600 X basically performs neck-and-neck with 700 X down to the lowest point one percent of frames where it falls short by 12 fps so bit more stuttering with the 2600 X I would say the extra cores in this scenario do aid the 2700 X despite the frequency difference now fortnight was a bit weird the 2700 X performed better than at 1700 X counterpart and the i7 was by far the best performer we also experienced the highest point 1 and 1% lowest with the i7 but the 2600 X performed very well in the average side actually outperforming the 2700 X and also in the 1% lows but then check out the 21% lowest the most stuttering occurred of all for these CPUs with the 2600 X so I would say that in this case both the frequency range as well as the limited core count even though 6 is plenty are holding the r5 back and deal with a super weird one I almost threw this benchmark out because I don't really believe what I'm seeing here the game did not update again everything with these benchmarks is exactly the same but the r5 actually performed better on average and then the i7 8700 K in OpenGL these CPUs by the way have the same core counts and they both have multi-threading enabled and even down to the 1% in point one percent Louis the r5 kept up with the 8700 K which is clocked much higher mind you so again an OpenGL kind of confusing the 2700 did not do very well all I saw a lot of stuttering with that CPU so I'm kind of curious as to why the r5 did so much better this is more or less just an observation I wouldn't take this one face value take it instead with a grain of salt battlefield 1 DirectX 11 1080p ultra preset this one's also super interesting the r5 and the r7 from Zen plus are both performing basically neck-and-neck with each other here the 2600 acts actually pulls ahead when it comes to 1% and 0.1% lows and almost matches the i7 8700 K at 5 gigahertz being this one's more CPU core oriented right having more cores is definitely better because i5 s that are quad core i5s 10 to max out see if utilization with battlefield one I expected this one to be a bit different than what we're seeing here but this is great news for anyone looking for a budget system right the r5 is doing very well right now rise of the Tomb Raider and DirectX 12 1080p high settings no anti-aliasing resulted in the r7 2700 XR losing in the first title this is the peak it's the first one you see when you run the benchmark and the r5 2600 ex actually basically performing on par with the 1700 X and the i7 8700 K coming out on top something else to note the r5 2600 X is basically outperformed in the second scenario at least the 80s 700 K the 2700 X and the 1700 X so this one was the top dog for the second scenario the third one it was a few frames off on the 8700 K and still managed to outperform the 2,700 X so things again looking really good here for the 2600 X now planet coaster in 1080p max settings is more less what I'd expect from the CPUs basically neck-and-neck the 2700 X did have a 2 frame rate jump over the 2600 X but I would call that within the margin of error and everything else was basically within the same range 2 FPS 2 1 FPS with the 0.1% lows you're seeing a side-by-side comparison here the two and they look practically the same and that's why one percent in point one percent lows are so important because they account more or less for the micro stutters you would otherwise not see if you just saw the average minimum frame rates f1 20:17 was another weird one just by the way that these two stacked up against each other the r7 2,700 X outperformed the r5 by a long shot on average but the r5 still managed to somehow maintain a 5 FPS lead on the minimum side of things so take this with a grain of salt because these aren't one percent pointers and Louis but still an interesting observation nonetheless universe sandbox 2 resulted in just 60 FPS on average for the r7 2700 axises with the earth and sphere of moon simulation the one I always run in the high setting preset the r5 2600 ex scored 11 fps higher on average this I think is where acts of our 2.0 definitely shines in games that are leveraging two to three cores heavily you're gonna want to keep your CPU just at stock settings and let X afar do its thing because if you manually overclock you're hindering your cpu from hitting any higher frequency in a few cores where it matters and lastly with witcher 3 i was able to breathe again so the 2700 x and the 2600 x scored the exact same score and this might juice with me manually benchmarking there's no built-in benchmark with winter 3 so I'm kind of just trying to do the same thing in game and that's why these fluctuate just a little bit but 112 fps on average is great to see for both CPUs because this is so GPU intensive we don't expect our CB to make much of a difference up to a certain extent so there are three things that I ultimately want you to take away from this video the first is they're interested in content creation or doing some CPU intensive tasks that require multiple threads multiple cores obviously the r7 CPU is the better vet but the second point if you're gonna play primarily games you're not gonna see much of a difference at all especially when you don't manually overclock between the r5 2600 or 2600 X and the r7 2700 or the 2700 X and third time back again into X afar and manually overclocking if you primarily play video games you don't need to manually overclock just let X 4 do its thing because most games we can assume will usually leverage between four and six cores efficiently leaning more towards four cores still but eventually we expect six cores to be the mainstream thing right and then manually of core like I might benefit a bit more before right now I think that letting xfr do its thing is the better bet for gamers and even if you content create look you saw with the 2600 X scenarios at 4.2 gigahertz across the board and XFR 2.0 enabled we know one or the other they basically render the video file in exactly the same amount of time and actually if you're scrubbing or throwing work stabilizer do anything that's more or less like single core oriented then EXIF are 2.0 is going to benefit you more in those scenarios so even on the content creation side I don't necessarily see the benefit of overclocking unless again you're gonna max out all your threads in those scenarios so I hope this video at least sheds a bit of light into the true value of horizon 5 CPUs especially with Zen Plus this is why I've said for the longest time that the r5 1600 is the best bang for the buck CPU because 6 cores 12 threads 200 bucks and it's overclockable boom it it plays games fine it content creates fine I'll let you stream fine it's the best of everything and it's not very expensive and you don't have to buy an expensive board to overclock the thing this same goes for the 2600 X the benefit here though is that in some cases you might have slightly higher slightly lower power draw so it really didn't change much even though we got a higher boost frequency out of that but at the same time you don't even need to manually overclock to get the best out of your trip especially when it comes to most games out there and that is just a win-win for everybody right ease of use come on who doesn't want that and that's why I'm willing to say that the are five twenty six hundred and twenty six hundred X are the best value CPUs I've ever been chucked on this channel hands down you just cannot get a better combination of both value and performance from a cpu out there unless you buy something from your buddy for 10 bucks it's worth $100 obviously great value there but on the in the new market right MSRP prices these two CPUs are just phenomenal for what they are now if you're concerned about the choice between the 2600 and the 2600 X I I really can't say too much on it yet because I haven't tested the 2600 but I'm willing to say that if it's anything like the 1600 was with respect to the 1600 X then the 2600 is going to be the better buy because it's gonna be cheaper it's not gonna come with as beefy a cooler so if you wanting to stock cooler that might influence your decision and rightfully so but if you have an aftermarket CPU cooler like an AIO or you know a bigger air cooler then I think the 2600 is gonna be the better bet I'll definitely get my hands on one eventually and we'll confirm that but for now I think you're gonna be perfectly fine with the regular 2600 save the extra 30 bucks buy yourself a nicer CPU cooler and let that thing just cruise at its turbo frequency with that folks if you like this video let me know if I give this one a thumbs up I appreciate that thumbs down for the opposite click to subscribe button if you haven't already and that Bella notification icon so you're notified with videos like these go alive this is science studio thanks for benchmarking with us