Gadgetory

welcome back to our brown box today I am finally at long last benchmarking a custom Vegas 64 graphics card and on hand for testing I have the Isuzu ROG Strix model which has now been GP peed into an Aries card anyway we have the Strix version on half a Testament please note it's now called the Aries after a sue spent two individuals will performance though should be the same ok so back when Vegas 64 was first released we compared it to the geforce gtx 1080 found edition and found it to be 5% slower on average at 1440p across 32 games since then we've mostly focused on Vega 56 as we've found that that's a much better value product and it stands out quite well to the competing gtx 1070 and I suppose near the 1070 TI anyway now that it's been almost a year since Vega 64 was released in its reference form I felt it was about time we revisited and why not do that with a custom Vega 64 graphics card you guys have only been screaming at me to test one for the better part of a year now so maybe we can do that for comparison I have the MSI GTX 1080 gaming x 8g and this is the standard 10 gigabits per second gddr5 X model although both models are pretty heavily overclocked out of the box we're still going to squeeze every last bit of performance out of them with a custom overclocked stock the Strix Vegas 64 OC gaming comes clocked at 1590 megahertz for the core and 950 megahertz for the HBM to memory we were able to increase the core clock to 1700 megahertz and mean 7% overclock and the memory to 1050 megahertz 11% overclock as for the GT x 1080 gaming x 8g it comes with a base clock of 1683 megahertz and 2502 megahertz for the gddr5 X memory we were able to hit eighteen hundred and seventeen megahertz for the core and this resulted in a bruise frequency of about 2.1 gigahertz of course that is depending on load the memory also reached 2.75 gigahertz so an 8% core overclock and a 10% memory of a clock now as mentioned earlier we have 27 games in total all of which have been tested at 1080p using both these stock and overclock configurations however I'm only going to discuss the results for eight of the games and then jump into our performance breakdown the rest of the graphs will be available for free on our patreon page so check for the link in the video description as usual our Corsair GPU test rebuilt inside the crystal series 570 X has been used and inside we have our core i7 8700 k clocked at five-year Goats with 32 gigabytes of ddr4 3,200 memory ok so I think that's about everything let's get to the good stuff first up we have f1 2017 here we find very competitive performance between the two GPUs the GTX 1080 was just four percent faster on average but three percent slower for the one percent low result overclocked the GTX 1080 does pull clearly into the lead having said that though we still are only seeing a single-digit percentage gain next up we have rec fest and the first time I tested with this game I use it to compare the RX 580 and GTX 1060 and this saw in video trailing by a 10% margin for that particular comparison however after multiple driver updates the GeForce performance in this title does a Peter we've improved somewhat now the GTX 1080 is able to pull ahead of Vegas 64 at least when looking at the average frame rate here we have a title that's pretty much always favored AMD and that is Tom Clancy's Rainbow six siege and we see that today that still remains true the GTX 1080 was eight percent slower out of the box and four percent slower once overclocked moving on to Far Cry 5 and here we see the Isuzu Vegas 64 card was a smidgen quicker than the MSI GTX 1080 then once overclocked we see basically the same performance although an AMD sponsored title far cry 5 is very well optimized so I do view this as a fair fight then next up we have fortnight using the Unreal Engine which is pretty much optimized for NVIDIA and Intel hardware and this gives the GTX 1080 a rather significant 19 percent performance advantage overclocking further extends this margin out at 23 percent of course in favor of the NVIDIA GPU Total War saga Thrones of Britannia is the newest game added here and we find very competitive results with Vegas 64 just edging ahead for the frame time performance overall though a very similar experience using either GPU ARMA 3 has now been optimized to the point where it plays very similar on AMD and NVIDIA hardware which is quite shocking because about a year ago the radio performance was a disaster so it's great to see such competitive results here like fortnight player owned owns battlegrounds is also built upon the Unreal Engine and as a result favors Hardware from the Green Team here we see the GTX 1080 enjoying a 14% performance advantage out of the box making it the preferred GPU for this title as for power consumption well as you might have expected at this point Vega does guzzle quite a bit more power pushing total system consumption 18% higher when comparing the out of the box figures overclocked that margin is extended to 23% and that is the most voltage optimized overclock we could achieve with Vega 64 okay so from that small 8 game sample we saw quite a bit of back and forth between these two though that's probably be expected and now though let's get the full picture by looking at the results across all 27 games tested first up we have these stock out-of-the-box results comparing the MSI GTX 1080 gaming xhg to the asou Strix Vega 64 OC gaming here we see that the GeForce graphics card was just 5% faster on average the outliers include frost Punk where the GTX 1080 was 33% faster and then dirt 4 where it was 27% slower that said Vega 64 also struggled quite poorly in GTA 5 Warhammer 2 fortnight project cars to Sniper Elite 4 when using DirectX 11 and Assassin's Creed origins overclocking each graphics card extended the GTX sally's lead out to 8% there overall we did see very similar performance trends across the board overall the GeForce graphics card was faster by a margin gray than 5% in 16 of the titles tested and slower by a 5 percent or greater margin in just two of them so a pretty clear advantage going away of the green team here as I noted at the start of this video almost a year ago now in Vega 64 was first released we compared it the GeForce GTX 1080 founders ition and found it to be 5% slower on average at 1440p across 32 games but whatever reason that testing really seemed to stir things up extremists in either camp appeared quite upset with our findings though I do typically deem our content a success if I've managed to upset fanboys on both sides Nvidia fans are upset because I used an Fe model despite the fact that I wanted to make a reference to reference comparison since Vega 64 cards or custom Vega 64 cards I weren't yet available then AMD fans are upset because hell I can't even remember why they're upset now probably a whole heap of things that weren't really warranted or made any sense that's typically how those things seemed to go our point is there was a big hoo-ha about the results being misleading and for whatever reason one team was going to come out well on top at some point in time for some reason I actually said at the time that I expected custom cards just like this one to actually be quite a bit faster avoid throttling handle higher frequencies and just in general deliver better performance turns out I was pretty much wrong about that one almost a year later pretty much the same story different set of games but still a lot of games and what we found was that 5% difference going in favor of the GTX 1080 at 1440p these findings aren't unusual either at this point should be pretty common knowledge that overall the GTX 1080 is going to come out ahead more often than not I could have very easily handpicked six to nine titles to test with that would heavily favor one company more than the other but with well over a 20 game sample you get a pretty clear picture of where things stand when it came time to overclock it was pretty straightforward with the GTX 980 Vegas 64 though was a different kettle of fish and I tried pretty much everything to unlock that next level performance that under vaulting is said to achieve but no matter what I tried the end result was well it was pretty underwhelming every time I also failed to find any credible sources that have shown significant performance gains from overclocking the Zeus rx and Vegas 64 graphics card so at the other day it seems quite clear that AMD's come out well short at the and sure at this point that doesn't surprise a lot of you but we still see arguments in the comments saying that Vegas 64 is now better than the GTX time clearly not true today the cheapest of Vegas 64 graphics card costs around 580 dollars u.s. that's around $100 u.s. more than the cheapest GT X 1080 models therefore we don't see sense in paying around 20% more for about 5% less performance on average that said right now I suggest you buy neither of these products as we're knocking pretty much on the door of the next generation release from Nvidia so really an absolutely terrible time to buy a new graphics card I highly recommend you do not do that also as a side note this video did include a lot of decks 11 and X 12 testing and all the tiles that supported both api's and I will be creating a separate video very shortly that focuses on comparing those results so that should be quite interesting and that is going to do it for this one if you did enjoy the video be sure to hit the like button please subscribe for more content if you're interested in more content and if you appreciate what we do in our box then maybe support us on patreon there's some pretty cool perks there with discord and monthly live streams and yeah it's a whole lot of fun anyway thank you for watching I'm your host Steve I'll see you again next time you