Gadgetory

welcome back to harbor unboxed today we're opening up an old AMG wound by revisiting the FX 8350 and we will be comparing it to probably the most legendary cpu to be released in the past decade the core i7 2609 ago we did revisit the 2600 k and found that the performance of the now seven-year-old cpu was very respectable although it did travel the latest and greatest eighth generation 8700 k by a reasonable margin when using an extreme gaming GPU such as the GT x 1080 TI with the gtx 1070 or something even slower the margin was minimal therefore I concluded that if you currently own a 2600 Caine you are willing to overclock it chances are upgrading to an eighth gen core processor won't get you much in the way of extra performance at least for gaming but before we get too far into the comparison sponsoring today's video is high low and I greatly appreciate their support if you're after a respectable broker to trade with I suggest checking out halo registered since 2010 high/low is a trusted and regulated brand in fact it's one of the few Australian binary trading platforms to hold an Australian financial services license they offer industry high payouts with easy bonus terms as well as payouts up to 200% on initial investment quickly and securely with halo you can withdraw funds quickly and easily using a range of methods without any hidden terms you can also trade on the go with high-low iOS and Android apps halo also offers one of the easiest to access demo accounts I've ever seen simply click the demo button and you're away you don't have to deposit you don't even have to give them your email address just click the quick demo link and you'll get $10,000 worth of virtual money to play with moreover if you saw today high/low is offering the first 50 viewers an additional $50 so $100 on total and this software is available for the next four days so click the link in the video description and check them out right so when compared to the recent 2600 kaori visit the situation with the FX 8350 might be a little different despite packing more cos IPC performance is down and power consumption is way up this has always been the problem of AMD's bulldozer architecture and 2012 update codename for sheriff featuring piledriver cause didn't solve this I concluded my review back in October of 2012 by saying this with AMD's aggressive pricing the updated FX series isn't necessarily in an indefensible position against Ivybridge when purely comparing speed and price but it's not exactly an open-and-shut case either the FX 6300 might offer 22% more performance in the core i3 32:24 about the same price but our piledriver powered test rig also consumed 86% more power than the Ivy Bridge machine 227 Watts versus 116 watts the bottom line is that the piledriver FX series provides a quick affordable upgrade for folks still using loránd k10 hardware but there isn't a lot to see for those running high-end phenom ii x4 + x6 processors regardless of how cheap the new parts might be for those building a fresh rig from scratch ivory bridge will likely still be more attractive thanks to its superior single thread performance and efficiency so almost five and a half years later here we are games no longer used just two cores and in fact we're starting to see quad cores going out of fashion at the high end so while the core i3 3220 might have kept up with the 8 core FX 8350 back then it would likely struggle today still a 55 watt part is hardly sport for the mighty 125 watt FX processor so we're putting up against the much older 95 watt 2600 k the following benchmarks have been conducted with the core i7 2604 pointer key codes using ddr3 2133 memory then we have the FX 8350 clock slightly lower at 4.7 gigahertz but that's as high as I could get it and I don't expect that true percent difference in clock speed will amount to much the FX process has been paired with faster ddr3 2400 memory so that should well and truly make up for the clock speed deficit finally both the hyper transport link and Northbridge frequency you were overclocked to 2.6 kia hertz for testing we have almost a dozen games that were benchmarked at a range of settings and resolutions using the gtx 1080 TI so without wasting any more time let's get to the results like all good sessions we're starting with ashes of the benchmark and here we see the FX 8350 isn't able to get anywhere near the most out of the GTX 1082 a limiting performance to around 70 FPS on average this means best case the FX processor was 16% slower than 2600 K and this margin can be seen at 1440p when comparing the average frame rate that said though it was just 8% slower for the 1% low result under this GPU constraint s condition if we unleash the GTX 1080i with the medium quality settings at turn DP the FX 8350 is now almost 30% solar for both the 1% low and average frame rate or you could state it by saying that the 2600 K was 40% faster but that just sounds worse either way in this heavily threaded title a 2600 K appears to be vastly superior next up we have a game has been putting torrent sites out of business that was until it was recently cracked I am of course talking about Assassin's Creed origins and its DRM for days here we see the ultra quality settings were a bit taxing on the bulldozer processor and while it did keep frame rates above 30 FPS at all times it was noticeably laggy when compared to the 2600 K the higher quality settings did allow for much smoother performance though it was interesting to see that downgrading to the medium quality settings didn't really improve things basically when GPU limited 1440p the FX 8350 was still 14% slower on average and 33% slower for the 1% low result so a very comfortable win for the Sandy Bridge processor moving on we have some battlefield 1 single-player results and yes we are testing the single-player as it's impossible to accurately measure multiplayer performance we can get a rough idea but it's really not the most accurate test as I can't control what the other 60 plus players do and you won't believe how many times people from both sides completely dissipate by benchmark orders so the single-player results will have to do here the 2600 K was 30 to 35% faster for the high and medium quality tests and even at 1440p with the ultra quality settings the FX 8350 was still around 20% slower than intel's old quad-core the FX 8350 does quite well in Call of Duty World War 2 despite trailing their 2600 K by a reasonable margin and by this I mean the resulting performance was very playable interestingly even at 1440p the FX 8350 was still three percents are for the average frame rate that was only 13 percent slower for the 1% low result but I'm not quite sure why I'm saying only as that is still a decent margin moving on we have Dawn of War 3 and golly gosh what's going on here then and I'm not talking about that delightfully elegant expression previously when comparing the 2600 K to the 87 ok we did find that the older Sandy Bridge CPU was almost 30% slower in this tile and that was one of its worst losses despite that though the 2600 K was still 56 percent faster than the FX 8350 and that almost shouldn't be possible Dawn of War 3 clearly isn't optimized any shape or form for the bulldozer architecture if you're wondering why the frame rate doesn't change between the various quality sayings and resolutions even with the 2600 K the answer is simple we're extremely cpu limited in this title even at 4.8 gigahertz next up we have dirt 4 and this is a relatively harmless title it's been helping budget cpus and gpus feel good about themselves since June 2017 we see the demon at 1440p caresses the FX 8350 along allowing for a minimum of 60 FPS in fact as we continue to become more and more GPU bound the 1% low figures start to line the same note can't be said for the average frame rate looking at the average frame rate the FX 8350 was incredibly 34 percent slower at 1440p and 45 percent slower at 1080p so while the FX processor was able to deliver playable performance the 2600 K was just on another level this would probably be surprising if I hadn't actually tested the game with the FX series when it was first released last year still it is surprising in the sense that the Rison CPUs performs so well in this title though I probably shouldn't compare the effects of rising series as they're in no way interchangeable moving on from dirt 4 we have another racing title and again things move faster when powered by the 2600 K again although the FX 8350 was able to deliver playable performance it was still up to 24 percent slower at 1440p however at 1080p the margins were often around 30% as in the FX 8350 was 30% slower so again we see another title where the older 2600 K is just on another level the last racing title tested his project cars - as we found numerous times already the FX 8350 was up to 30% slower or you could say 1440p when comparing the average frame rate the 2600 K was 43% faster 43% that's clearly a very significant margin moving on things do look much more competitive when testing with Rainbow six siege in fact at 1440p things managed to almost come together yes I know we are heavily GPU limited here but by Jo I'll take it at this point at 1080p using the ultra quality settings the FX processor was up to 20 percent slower but still this is one of the better results we've seen capping the benchmark session off we have total war Warhammer 2 and again at 1440p the FX 8350 is afforded the ability to catch up here it's just 7% slower that said at 1080p using the ultra quality settings it's now 16% slower and 23 percent slower at high I'd say that went pretty much as most of you would have expected that seemed to be a one-sided beat down and the older 2600 K was swinging the hammer but to be sure let's quickly check out their results across the ten games tested to give us a clearer picture of how the battle unfolded well things honestly look very much how I imagined it the FX 8350 was 19% slower at 1440p using the ultra quality settings 16 percent slower if we focus on the 1% low result it might interest you to know that the 2600 K was just 10% slower than the 8th generation Core i7 87 era K for the average frame rate under the same test conditions and 17% slower for the 1% low result so this means the 2600 K is closer to the 8700 K in terms of gaming performance than the FX 8350 is to the 2600 K now it's just mind-blowing getting back to these results though we see when using the ultra quality type settings at 1080p the FX 8350 is 26% slower on average and 21% slower for the 1% low result these margins continue to grow as we reduce the quality settings by imagine you've probably heard enough at this point okay so I'm sure you get at this point even after all these years the FX 8350 is much slower than the core i7 2670 probably not a real surprise that said let's address a few things that might be upsetting some FX owners who view this video is just another chance to rag on the FX series although much slower in the vast majority of tests the FX 8350 did almost always deliver perfectly playable performance so there is that I often upset FX owners when I express my honest opinion about the FX series that being that I believe it's complete garbage I always give my honest opinion though about everything whether that be Intel's hot and overpriced I like X range or Nvidia stupidly named tight next little P if you're a regular viewer you'll know that the list just goes on anyway I appreciate that the FX series was cheap or at least it was discounted to become very cheap the fx-8150 that landed at $250 us and most of all I bet a year later the FX 8350 landed at $195 us and most of all liked but eventually they started going on sale for 150 to 160 dollars us and in early 2017 we're often selling as low as 130 maybe even a hundred and ten dollars us at that price the eight core CPUs tempted some though I recommended viewers who were predominantly gaming pick up something like a core i3 6100 instead and upgrade later on or alternatively as the evidence suggests you'd be better off buying a second-hand Sandy Bridge ivory bridge or even a Haswell system that said I would just like to note that for certain productivity workloads the FX 8350 does offer a reasonably good bang for your buck but it is very application dependent and power consumption is still going to be that insane something we'll look at in a moment of course now with rising on the scene you'd be absolutely mad to ever buy an FX processor for productivity work a risin is genuinely good and a great value choice for gaming and productivity tasks getting back to gaming many will argue that when paired with a mid-range or low-end GPU the FX 8350 or any of its bulldozer relatives are perfect as they are cheap and provide playable performance it's my opinion though that they're not in fact they're terrible pairing with a low-end graphics card because while your rx 560 is just sip power the FX process is guzzling it like a drunken viking this ultimately is the real problem with the FX 8350 and the entire FX series if it offered the same performance per watt as the 2600 k or was even in the same ballpark but I might turn a blind eye to the lackluster performance the problem though is this playing ashes of the singularity at 1080p using the crazy preset the FX 8350 was 22% slower and that's no small margin however in order to deliver that lackluster performance it gathered considerably more power pushing total system usage 25% higher so that's 22 percent less performance while consuming 25% more power and that right there is why the FX series was the biggest disaster in AMD's history and remember the 2600 K was released almost two years before the FX 8350 not only that but this is total system consumption with an extreme 250 watt TDP graphics card with an Rx 560 for example that percentage margin will grow considerably so what about application performance is the FX series more efficient there in a word no running the Cinebench r15 multi-threaded tests we saw the FX 8350 push system consumption almost 60% higher 60% higher you might be thinking but with those eight core that probably scored higher right well if you thought that you'd be wrong the 2600 K produced a score of 832 points making it 15% faster honestly the Cinebench r15 tests combined with a power readout really tells you all you need to know about this comparison those power consumption figures were taken with both CPUs overclocked but don't think the FX series is far less efficient once overclocked stock you'll still see total system consumption around 60% higher with the FX processor so to recap my problem with the FX series isn't just that it's slower it's that it's slower and consumes significantly more power and let's be honest that's certainly not a good thing and ultimately it led to the FX series being a complete and utter failure for AMD on that cheerful note that's gonna do it for this one again I'd like to thank high/low for supporting our work it's been great to be a part that them on this one don't forget to check out high-low by using the link in the video description remember the first 50 viewers to sign up and contact high-low support with the code high-low 50 extra for within the first 4 days of this video going live will get an additional $50 so what are you waiting for go sign up and finally thanks for watching I'm your host Steve see you soon you