Gadgetory

welcome back to harbor unboxed now it always surprises me how often we see a request for crossfire and sli benchmarks particularly given that we flat-out tell you guys not to invest in either technology we were doing that for years now but there still seems to be a lot of interest it's it perhaps even more surprising given the fact that AMD and in particular Nvidia have made no secret about the fact that they are pulling back on investment into multi GPU technology I've been pretty stubborn when it comes to multi-gpu benchmarks and for at least a year now I've basically refused to check out the crossfire or SLI but recently I did have two rx 590 cards land at my feet and I thought why not seems like there are enough of you that want to see some update and testing and well I've probably made you hold out long enough now probably not long enough for it to be any good I suspect but I've made you wait long enough for the update today's video sponsor is PCB way if you're after a custom PCB for your next project look no further they make high quality PCBs at affordable prices right now new members get a $5 bonus which covers their first order of ten one to two layer PCBs and they also offer an assembly service furthermore right now all PCB assembly orders enjoy free shipping worldwide so check them out link is in the video description last time we visited in-depth crossfire testing was back when Matt was still hosting the channel and I was locked away benchmarking and never to be seen or at least that was the plan of time so you can blame Matt for allowing my mug on the channel anyway back in June of 2016 we grabbed two R X 480 graphics cards threw them in our test rig and enabled crossfire for the most part the results were quite good but as we've come to expect for a multi GPU technology the experience was far from flawless of the 23 games tested six didn't work at all with crossfire enabled and the problem here is for the games that didn't work with crossfire the RX for ID cards were on average 47% slower from the GTX 1080 but just 9% slower when working sorry for today's test we have 20 much newer games and we're going to see how well the two RX 5 90s compare to just a single card in all 20 titles at 1080p and 1440p for comparison we have nine other graphics cards including high-end models such as Vegas 64 and the r-tx 2070 I'll be discussing the results for a dozen of the titles tested with the rest of the graphs available for free on our patreon page and the link will be in the video description for this benchmark I used my core i7 87 okay GPU test rig which is overclocked to 5 gigahertz and is featured inside the Corsair crystal 570 X packing 16 gigabytes of ddr4 3400 memory then for the AMD driver we use Radeon adrenaline 18 point 11.2 ok I think that's about everything let's get into the results first up we have battlefield 5 and unfortunately we're not seeing any support for crossfire in this title so no extra performance from the second card at 1080p and the same was also true at 1440p that's really quite surprising as crossfire did work in battlefield 1 here we see a mold 33% gain for the average frame rate at tele P so not great but a big improvement over nothing and scaling did improve at 1440p here the RX 5 nodes in crossfire boost the performance by 46% again not amazing but it is world's better than what we're currently seeing in battlefield 5 the best example of crossfire scaling that I've come across in this new batch of games was seen in strange Brigade here the crossfire RX 5 nineties boosted the average frame rate by almost 90 percent and better yet frame time performance was still very good then at 1440p we see over 90% scaling hitting 94% so this is an exceptional result for the crossfire 5 90s AMD's Radeon GPUs perform quite poorly in Assassin's Creed Odyssey so unsurprisingly crossfire support is not existent in this title so whereas the 5 90s were faster than the RTR 2070 in strange Brigade here they're 44% slower moving on the hitman 2 and again another time lacks crossfire support and therefore running with the technology enabled actually slightly reduces performance this is also seen at 1440p so we have another example of why multi GPU technology really isn't a good investment still it's not all bad and another example of that is shut off the Tomb Raider here we see a 48% performance boost at 1080p and this place the RX 5 90s alongside GTX 1080 and Vega 64 furthermore scaling is drastically improved at 1440p and now we're seeing a 61 percent performance boost and this place the 592 on par with the r-tx 2070 so pretty solid result they're mixed up we have Forza horizon 4 and this is another title where crossfire isn't supported and therefore we saw no gains at turn DP or 1440p in fact we saw a slight performance regression to my surprise monster hunter world does support crossfire though the frame time performance at 1080p was a little sketchy whereas we see a 32% boost for the average frame rate we only see a 14% improvement for the frame time performance jumping a 1440p does help to iron out this issue but even so scaling is below 40 percent which is pretty weak and certainly doesn't justify the investment of a second graphics card here we have an example of a title looks good when focusing on the average frame rate but the experience was actually pretty horrible despite averaging 100 fps and Star Wars Battlefront 2 at 1080p the average frame time performance was pretty shocking dropping down to well below the result of a single 590 the game was basically unplayable for example a constant 30 FPS would provide a much nicer experience unfortunately the frame time issue did persist at 1440p I'm not sure if there is a workaround for this title but out of the box the performance is pretty horrible using crossfire and you're certainly much better off with a single rx 590 frame time performance was also a little sketchy and Deus Ex mankind divided though nowhere near as bad as what we just saw in Battlefront 2 the issue was a little more noticeable at 1440p and although the average frame rate is much improved with a second 590 the overall experience really wasn't and due to the disparity between the average and 1% low result when using crossfire I'd rather play this title on a single 590 I also saw a little bit of stuttering going on a far cry 5 and this was present at both 1080p and 1440p project cars 2 was another title where we did see performance gains but the frame time performance wasn't great and much worse than that of a higher-end single GPU graphics card now please note all graphic card configurations with the exception of the crossfire cards were tested in Crysis 3 to measure power consumption however crossfire isn't work in crisis 3 so we used f1 2018 as this was one of the better games for scaling typically you're looking at a total system increase of around 60% when using a second rx 590 that leads to a pretty brutal power consumption figure here the crossfire 5 90s push total system draught a 576 watts and I did observe just over 600 watts in strange brigade this is quite a bit more than even a Vega 64 liquid graphics card and almost twice that of a single RT X 2070 so pretty horrible stuff when it comes to power consumption okay so when compared to a single 590 we saw a 37% boost on average to the frame rate at 1440p though that figure alone is a little bit misleading frame time performance in Star Wars Battlefront 2 and Davis X mankind Nevada was very bad I'd much rather use a single card in those titles while stuttering was a bit of an issue in Far Cry 5 project cars - and The Witcher 3 normally I also test with dirt 4 but that title suffered serious graphical glitches with crossfire enabled so I had to drop it from the batch of games that I tested with so are you better off buying a higher end single GPU graphics card or two cheaper graphics cards well if you hadn't already worked that one out here is your answer the r-tx 2070 costs $500 us while the RX 5 90s cost two hundred and eighty dollars each and even if you compared the r-tx GPU to a pair of $200 RX 5 80s the outcome would be much the same get the more expensive higher and graphics card when everything's going to the multi-gpu plan the RX 5 90s killed it beating the our TX 2070 by a whopping 25% margin but out of the 20 games tested we saw that kind of margin exactly once next-best was 40% in f1 2018 9% and prey 5% in project cars but the frame time performance was quite poor 4% in Deus Ex mankind divided but again the frame time performance was quite poor and 3% in The Witcher 3 and again frame time performance was suboptimal so in true titles the crossfire 590 showed the r-tx 2070 what for and in everything else they got their PCI Express connectors handed to them two years later we find that once again - GPU technology it seems like a good idea on paper but in practice it is a bit of a fail it's really only ever made sense for those with money to burn for example right now our TX 28 ECI SLO graphics cards I'm about the only multi-gpu configuration that makes sense but in almost every way they make no sense so I guess the point I'm trying to make is that multi GPU technology you really only make sense if you're at the end of the road you've got the fastest possible single GPU solution and yet you can't get any more performance without adding a second one so if you've got money to burn on an RT X 2000 UCI and you want another one just because then I guess SLI is okay it doesn't really make sense if you want to use SLI for say RT X xx 80s in that case I would just get a single RT X xx atti as you'll receive smoother performance in the vast majority of titles as for the RX 5 90s in crossfire well I'd much rather have a single Vega 64 graphics card for example that's extremely rare that - five 9s will provide higher frame rates than a single Vega 64 card while also offering smoother stutter free gaming if you're only ever going to play games like f1 2018 that support crossfire really well they're getting to say rx 570s for example costing around three hundred dollars would be a hard combo to beat in fact and I think you can beat it but who buys graphics cards to only ever play one or maybe two games of course there are other drawbacks that I'm yet to discuss such as heat and power consumption those two five 90s were dumping so much heat into the corsair crystal 570 x case and they spent more money would need to be invested in case fans and even then you're still running hotter due to the way the cards are stacked you'll also lose out on the power supply the RT x 2070 system works without an issue with a 500 watt unit and 600 watts would be more than enough the crossfire 5 90s though they'll need an 800 watt unit 750 watts would be the absolute minimum here and although we're only talking about a twenty to thirty dollar increase in PSU cost that's still a factor that needs to be considered ultimately though it is the poor support that kills these multi-gpu setups and that's why I feel no one should burden themselves with SLI or crossfire technology especially if they're trying to gain higher end performance with mid-range cards just by a faster single GPU much less stressful anyway I think we will end the video there if you did enjoy the video hit the like button subscribe for more content and if you appreciate the work during our box then can sit up supporting us on patreon thank you for watching I'm your host Steve and I will see you next time you