1 00:00:00,000 --> 00:00:06,000 AMD's RX 6950xt is almost a totally 2 00:00:03,720 --> 00:00:10,620 different class of GPU compared to when it first launched to tepid reviews 3 00:00:07,919 --> 00:00:14,519 earlier this year and with recent price adjustments it is such a good deal now 4 00:00:12,660 --> 00:00:19,199 that its green opponent no longer makes any sense and neither does its 5 00:00:16,740 --> 00:00:25,740 predecessor the 6900 XT but why am I talking about the 6000 6 00:00:21,119 --> 00:00:28,740 series because however the RX 7900 7 00:00:25,740 --> 00:00:31,439 series performs today and it's good I 8 00:00:28,740 --> 00:00:35,700 want you to also keep in mind AMD's track record for dramatically improving 9 00:00:33,660 --> 00:00:40,680 the performance of their gpus through new features and Driver optimizations 10 00:00:37,880 --> 00:00:45,120 personally I'm counting on it because now that I have publicly committed to 11 00:00:42,660 --> 00:00:49,260 skipping NVIDIA's latest 40 series cards AMD is my only hope for an upgrade this 12 00:00:47,640 --> 00:00:53,340 generation just like my only hope of paying our 13 00:00:51,000 --> 00:00:57,660 production team is our sponsor PDQ automate patch management and software 14 00:00:55,440 --> 00:01:02,399 updates with PDQ search your library of over 200 ready to deploy applications 15 00:00:59,640 --> 00:01:07,700 and install software zero touch from your disk start your free trial at 16 00:01:04,140 --> 00:01:07,700 pdq.com LTT 17 00:01:14,600 --> 00:01:20,580 right out of the gate the 7900 series is 18 00:01:17,820 --> 00:01:26,220 full of upgrades it features AMD's brand new rdna 3 architecture up to 24 19 00:01:23,340 --> 00:01:31,259 gigabytes of much higher bandwidth gddr6 memory substantially boosted core clocks 20 00:01:28,680 --> 00:01:37,259 across the board and this is great the step down 7900 XT comes with as many 21 00:01:34,740 --> 00:01:43,740 compute units as AMD's previous Flagship with the beefier XTX rocking even more 22 00:01:40,979 --> 00:01:46,979 despite having more and faster cores though one of AMD's boldest claims is 23 00:01:45,840 --> 00:01:52,979 that they've managed to improve efficiency by 54 this is thanks in large 24 00:01:51,000 --> 00:01:57,659 part to AMD's high bandwidth chiplet design the first of its kind in a modern 25 00:01:54,899 --> 00:02:00,960 GPU what's more they've beefed up their Ray accelerator cores for real-time 26 00:01:59,340 --> 00:02:06,060 raytraced Lighting in supported games and finally included AI accelerator 27 00:02:03,899 --> 00:02:10,739 cores to bring Radeon to feature parody with NVIDIA and Intel's offerings 28 00:02:08,759 --> 00:02:14,040 unlike those offerings however the Radeon RX 7900 series cards we're 29 00:02:12,780 --> 00:02:18,500 looking at today come equipped with DisplayPort 2.1 which enables support 30 00:02:16,560 --> 00:02:25,200 for up to 8K 165 Hertz displays is that even 31 00:02:21,720 --> 00:02:27,959 necessary who cares it's awesome 32 00:02:25,200 --> 00:02:33,720 and it's something that NVIDIA's top-end GPU the RTX 4090 cannot do in spite of 33 00:02:31,800 --> 00:02:37,860 its staggering sixteen hundred dollar sticker price 34 00:02:34,920 --> 00:02:42,780 all of this is while actually fitting in your case and costing 900 and a thousand 35 00:02:40,860 --> 00:02:49,980 dollars for the XT and the XTX respectively that is 200 to 300 less 36 00:02:46,860 --> 00:02:51,599 than the RTX 4080 that AMD says they 37 00:02:49,980 --> 00:02:56,840 should be able to compete with for Speed Of course we won't be taking their word 38 00:02:54,239 --> 00:03:02,040 for that AMD has been known to be optimistic at 4K it's clear out of the 39 00:03:00,480 --> 00:03:07,920 gate and Shadow of the Tomb Raider that whatever team red fans might have hoped 40 00:03:03,959 --> 00:03:10,800 for the RX 7900 XTX is not AMD's answer 41 00:03:07,920 --> 00:03:16,440 to the RTX 4090 it's got performance that is realistically more in line with 42 00:03:12,959 --> 00:03:19,200 the 4080 though that is what AMD said 43 00:03:16,440 --> 00:03:26,700 and this pattern continues in Hitman cyberpunk 2077 and Modern Warfare 2. 44 00:03:23,040 --> 00:03:28,019 well great right I mean why did I spend 45 00:03:26,700 --> 00:03:34,800 all that time talking about the importance of AMD's ongoing driver developments keep that in mind well 46 00:03:32,280 --> 00:03:39,780 there's this for one thing Forza Horizon 5 is a clear loss for team red thanks to 47 00:03:37,920 --> 00:03:46,319 those subpar one percent and five percent low frame rates but also this 48 00:03:43,140 --> 00:03:48,780 for another reason that right there is a 49 00:03:46,319 --> 00:03:54,180 one thousand dollar GPU beating a sixteen hundred dollar GPU in F1 2022 50 00:03:52,200 --> 00:03:58,560 and if the results I'm looking at right now are any indication of the kind of 51 00:03:56,340 --> 00:04:04,860 raw power that AMD's driver team has on tap here there is some serious aging 52 00:04:01,799 --> 00:04:07,500 like fine wine potential on the table 53 00:04:04,860 --> 00:04:14,879 of course you should never buy a promise though and today AMD's best Trails 54 00:04:10,799 --> 00:04:17,760 NVIDIA's best by about 17 which sounds 55 00:04:14,879 --> 00:04:24,000 bad Until you realize that NVIDIA's card costs 60 percent more compared to the 56 00:04:21,060 --> 00:04:31,500 4080 the XTX is a no-brainer at nine percent faster and 20 cheaper and the 57 00:04:27,419 --> 00:04:34,020 7900 XT does even better I mean sure the 58 00:04:31,500 --> 00:04:38,340 4080 is seven percent faster but costs 33 percent more 59 00:04:36,080 --> 00:04:42,840 1440p doesn't always show us a difference compared to 4K but here we do 60 00:04:40,740 --> 00:04:48,840 see some strange behavior in Forza that we retested several times the 7000 61 00:04:45,900 --> 00:04:54,120 series is actually slower than AMD's older 6000 series cards we think this 62 00:04:52,199 --> 00:04:57,660 could be a problem with the game as it's the only one out of our suite that does 63 00:04:55,919 --> 00:05:03,120 it and when we went to test for it says Ray tracing performance we saw that it 64 00:04:59,460 --> 00:05:06,240 was in fact pretty broken we also see 65 00:05:03,120 --> 00:05:08,160 the RTX 4080 pull ahead of even the XTX 66 00:05:06,240 --> 00:05:11,820 in Shadow of the Tomb Raider and Red Dead Redemption which is especially 67 00:05:10,139 --> 00:05:16,440 surprising considering that virtually every other game shows AMD closing the 68 00:05:14,100 --> 00:05:20,220 Gap instead the biggest example is in Modern Warfare 2 where the minimum frame 69 00:05:18,419 --> 00:05:25,620 rates aren't quite up to the level of the RTX 4090 but the average frame rates 70 00:05:22,860 --> 00:05:30,600 are which again can set some potential for optimization in total these numbers 71 00:05:28,259 --> 00:05:34,500 put the 7900 Series in a very good position to compete with NVIDIA's 40 72 00:05:32,220 --> 00:05:39,199 series to the point where if traditional rasterized gaming was all that existed 73 00:05:36,600 --> 00:05:45,600 NVIDIA would need to lower their prices but it isn't all that exists and Ray 74 00:05:42,120 --> 00:05:48,060 tracing is sadly still not AMD's strong 75 00:05:45,600 --> 00:05:54,180 suit while substantially better in cyberpunk than the rx6000 series these 76 00:05:51,060 --> 00:05:58,080 7000 series gpus are behind NVIDIA by at 77 00:05:54,180 --> 00:06:00,180 least a factor of two at 4K at 1440p AMD 78 00:05:58,080 --> 00:06:04,680 at least manages to break the 30 FPS minimum Mark making the game playable 79 00:06:02,039 --> 00:06:08,820 and you can use FSR to render at a lower resolution and upscale for smooth 80 00:06:06,479 --> 00:06:13,680 gameplay but at that point we could also do the same thing for Team Green and the 81 00:06:11,340 --> 00:06:18,600 overall story would still be the same F1 2022 fares slightly better for AMD at 82 00:06:16,500 --> 00:06:23,400 least with the 7900 XTX pulling up to within 16 percent of the RTX 4080 in 83 00:06:21,419 --> 00:06:27,780 minimum frame rates and actually near parity on average which again gives me 84 00:06:26,160 --> 00:06:31,979 hope that some of AMD's woes with Ray tracing are a matter of optimization and 85 00:06:30,060 --> 00:06:36,660 that feeling only grows with shadow of the Tomb Raider where even at 4K it's 86 00:06:34,020 --> 00:06:43,560 within about 12 percent of the 4090 and turning out nearly 60 FPS it is an older 87 00:06:39,960 --> 00:06:46,139 game but that's not bad it's just that a 88 00:06:43,560 --> 00:06:52,860 good race ran when the other guy ran faster is still an l and ultimately the 89 00:06:50,039 --> 00:06:58,199 RX 7900 XTX just couldn't even come close to the RTX 4080s level of 90 00:06:55,440 --> 00:07:02,819 performance with Ray tracing enabled I suspect that most people myself included 91 00:07:00,539 --> 00:07:07,319 still don't use Ray tracing often and won't really be affected but given that 92 00:07:05,220 --> 00:07:12,360 these things do still cost a thousand flipping dollars it's definitely a skid 93 00:07:09,900 --> 00:07:16,680 mark on what has otherwise been a pretty good showing so far and I'm afraid 94 00:07:14,520 --> 00:07:20,580 there's more where that came from the professionals who have need of this kind 95 00:07:18,360 --> 00:07:24,660 of graphical horsepower will be well disappointed isn't quite the right word 96 00:07:22,500 --> 00:07:28,979 but looking at these numbers they're not going to be excited either yes blender 97 00:07:27,479 --> 00:07:34,800 runs a good eight to twenty percent faster on the 7000 versus the 6000 98 00:07:31,680 --> 00:07:37,440 series but NVIDIA walks all over AMD's 99 00:07:34,800 --> 00:07:42,180 entire stack at between three to four times the performance thankfully AMD's 100 00:07:40,740 --> 00:07:48,180 new encoding engine helps them push ahead of the RTX 4080 in DaVinci Resolve 101 00:07:44,759 --> 00:07:49,620 in both h.264 and especially h.265 which 102 00:07:48,180 --> 00:07:55,979 their encoding engine seems to favor that's a good 20 more speed for the XTX 103 00:07:52,560 --> 00:07:58,860 there not bad though the 49d does remain 104 00:07:55,979 --> 00:08:02,639 on top procyon video and photo are also pretty good for AMD with performance 105 00:08:00,240 --> 00:08:05,880 that I would call comparable it's within a few percentage points either way you 106 00:08:04,259 --> 00:08:10,979 look at it and rounding out the video editing Suite is topaz AI which despite 107 00:08:08,699 --> 00:08:19,500 the name is using GPU compute here rather than any AI course and wow even 108 00:08:15,599 --> 00:08:21,660 the 7900 XT outperforms the RTX 4090 109 00:08:19,500 --> 00:08:27,180 here this workload seems to heavily favor AMD's Infinity cash as the 6000 110 00:08:24,300 --> 00:08:32,399 series cards aren't too shabby either specvperf brings AMD about as many wins 111 00:08:29,639 --> 00:08:37,740 as losses but the degree to which the 7000 series wins over the competition is 112 00:08:35,099 --> 00:08:43,620 often much lower than the losses that they suffer with especially poor results 113 00:08:39,839 --> 00:08:45,420 in 3ds Max Maya and SolidWorks which 114 00:08:43,620 --> 00:08:50,880 even dip under the performance of the 6000 series that's pretty rough those 115 00:08:48,180 --> 00:08:56,700 are kind of important applications they pull off major wins in Catia Creo energy 116 00:08:53,519 --> 00:08:58,620 medical and especially Siemens NX though 117 00:08:56,700 --> 00:09:02,459 as we've said in the past this is because the version in specview perf 118 00:09:00,480 --> 00:09:06,660 doesn't support Hardware acceleration on NVIDIA and actually this is just going 119 00:09:04,800 --> 00:09:10,620 to be excluded from our averages when we calculate it later on spec workstation 120 00:09:09,000 --> 00:09:14,399 Paints the radons in a significantly better light though with product 121 00:09:12,060 --> 00:09:19,260 development life sciences and energy all pushing well past Team Green but media 122 00:09:16,980 --> 00:09:24,839 and entertainment as well as GPU compute still stay with NVIDIA likely thanks to 123 00:09:21,899 --> 00:09:28,980 opencl's relative obscurity next to Cuda despite the blows the 7900 series car 124 00:09:27,120 --> 00:09:32,459 cards took though their beefed up media engines and GPU compute horsepower 125 00:09:30,779 --> 00:09:36,240 placed them with an average of two to ten percent of Team Green given the wide 126 00:09:34,740 --> 00:09:39,300 performance spread we saw depending on the workload potential prospective 127 00:09:37,980 --> 00:09:43,860 professional people will want to think long and hard about which capabilities 128 00:09:41,519 --> 00:09:47,940 are most important to them before making a final decision changing gears a little 129 00:09:46,500 --> 00:09:53,339 we found something interesting when we were testing resizable bar on both Intel 130 00:09:50,399 --> 00:09:56,880 and AMD platforms it seems like the performance improvements regardless of 131 00:09:55,140 --> 00:10:01,740 whether you have a team blue or team red CPU can be expected to be pretty similar 132 00:09:59,519 --> 00:10:07,260 despite AMD heavily marketing smart access memory as a selling point for the 133 00:10:04,140 --> 00:10:09,000 ryzen plus Radeon ecosystem bear in mind 134 00:10:07,260 --> 00:10:14,040 that this is a sample size of just two games but without resizable bar we're 135 00:10:12,360 --> 00:10:19,019 figuring you can expect anywhere from a 10 to 35 percent reduction in 136 00:10:16,019 --> 00:10:21,540 performance regardless of which side you 137 00:10:19,019 --> 00:10:25,920 swing for as for AMD's updated video encoder there's a lot to be excited 138 00:10:23,339 --> 00:10:29,820 about here av1 support and baked in machine learning to improve quality at 139 00:10:27,600 --> 00:10:33,480 the top of my list but despite receiving our cards early we actually only had a 140 00:10:31,860 --> 00:10:37,260 few days to test these things prior to embargo because we didn't have a driver 141 00:10:35,220 --> 00:10:40,200 so we were forced to push our image quality comparisons to a separate video 142 00:10:38,880 --> 00:10:46,200 that you can expect to see in the next couple of weeks one of AMD's other selling points is its 143 00:10:44,459 --> 00:10:49,860 smaller footprint compared to NVIDIA's 40 series gpus and I was concerned that 144 00:10:48,779 --> 00:10:55,079 this would come at the expense of thermals so I was extremely pleased to 145 00:10:52,500 --> 00:11:00,420 see that AMD's new cooler design has the 7000 series actually performing better 146 00:10:57,660 --> 00:11:04,320 than the 6000 series with that said though AMD does not have the luxury of 147 00:11:02,459 --> 00:11:09,300 only competing with themselves and NVIDIA is clearly in the better position 148 00:11:06,300 --> 00:11:10,740 here though with that said both of the 149 00:11:09,300 --> 00:11:16,200 NVIDIA cards we tested have substantially larger heatsinks with fan 150 00:11:13,440 --> 00:11:20,459 off modes while AMD's new cards always ran their fans it's probable that 151 00:11:18,720 --> 00:11:25,200 designs from the likes of ASUS or power color will run much cooler and hopefully 152 00:11:23,579 --> 00:11:29,160 quieter now we need to talk about core clocks 153 00:11:27,660 --> 00:11:33,540 because they're a little different this generation thanks to the decoupling of 154 00:11:31,200 --> 00:11:37,560 the front end and the Shader clocks the idea here is that the front end will 155 00:11:35,339 --> 00:11:41,459 benefit more from higher clocks while the more numerous Shader units crunch 156 00:11:39,420 --> 00:11:44,940 data in parallel at a lower clock speed resulting in a better balance of 157 00:11:42,899 --> 00:11:51,000 performance and efficiency and you can see that in action here where both the 158 00:11:47,100 --> 00:11:53,700 7900 XTX and XT have a consistent 200 159 00:11:51,000 --> 00:11:57,839 megahertz gap between the two clocks and even in a power virus like MSI combustor 160 00:11:55,980 --> 00:12:02,339 these clocks remained higher than the previous generation clock stability in 161 00:12:00,120 --> 00:12:06,839 game is similar to the 6000 series cards but NVIDIA's 40 series manages flatter 162 00:12:05,040 --> 00:12:10,800 lines throughout the Run suggesting that AMD is trying to be pretty aggressive 163 00:12:08,880 --> 00:12:15,120 with managing power consumption which makes sense because AMD talked a 164 00:12:13,560 --> 00:12:18,540 lot about performance per watt in their marketing and looking at these in-game 165 00:12:16,800 --> 00:12:23,100 paragraphs no that seems to suggest that their 166 00:12:20,820 --> 00:12:29,040 self-congratulations are well Justified the Radeon RX 7900 XTX Drew comparable 167 00:12:26,579 --> 00:12:33,720 power to the 4080 and this held even when we hit it with MSI combustor the XT 168 00:12:31,320 --> 00:12:37,680 was even better still staying in line with its predecessor but what's 169 00:12:35,640 --> 00:12:42,660 interesting is when we take a magnifying glass to how many watts the 7000 series 170 00:12:39,660 --> 00:12:45,180 poles per frame in this example while 171 00:12:42,660 --> 00:12:50,519 the raw wattage drops much closer to the RTX 4080 when we capped the frame rate 172 00:12:47,880 --> 00:12:55,320 to 100 the total watts per frame increases substantially this suggests 173 00:12:53,220 --> 00:12:59,639 that while the RTX 4080 is able to throttle itself effectively AMD is 174 00:12:58,079 --> 00:13:04,139 actually having some trouble with power management and AMD's idle power 175 00:13:02,160 --> 00:13:11,820 consumption numbers seem to corroborate this where the RTX 4080 idles at around 176 00:13:07,260 --> 00:13:14,940 15 to 20 watts the 7900 XTX regularly 177 00:13:11,820 --> 00:13:18,120 sucks Back 40 and sometimes over 50 178 00:13:14,940 --> 00:13:20,100 Watts while doing nothing at all what's 179 00:13:18,120 --> 00:13:24,480 even more baffling is that we noticed what seems like a major bug during our 180 00:13:22,139 --> 00:13:29,339 testing depending on the monitor that we plugged into the GPU we saw a wildly 181 00:13:27,120 --> 00:13:34,639 different total system power at the wall from a respectable 52 Watts with nothing 182 00:13:32,100 --> 00:13:41,040 plugged in to a whopping 171 Watts with our 27 inch Rog Swift 183 00:13:37,860 --> 00:13:44,040 bench display using some other monitors 184 00:13:41,040 --> 00:13:46,740 even caused performance to drop now 185 00:13:44,040 --> 00:13:51,180 given that I have a high-end display and I live in Vancouver Canada where energy 186 00:13:48,480 --> 00:13:54,540 is the one thing that is still cheap none of this is a deal breaker for me 187 00:13:52,860 --> 00:14:00,660 but if I was in Europe where Energy prices had surged pun intended it would 188 00:13:58,019 --> 00:14:05,639 give me pause before I was willing to look at team red adding insult to injury 189 00:14:03,300 --> 00:14:09,839 this high idle power draw means that the fans are spinning at all times kicking 190 00:14:07,980 --> 00:14:14,100 that heat out into your room even if you're just scrolling Reddit meaning 191 00:14:11,820 --> 00:14:17,820 that the 7900 series is going to be a little less comfortable to live with 192 00:14:15,959 --> 00:14:22,740 maybe if you go team red you could pick up a blank T rather than a retro polar 193 00:14:20,100 --> 00:14:26,459 fleece from ltdstore.com space heater one possible explanation for this high 194 00:14:24,779 --> 00:14:32,639 idle power consumption could be that this is a similar situation to the ryzen 195 00:14:28,500 --> 00:14:34,620 7 5800 X 3D that CPU's use of 3D V cash 196 00:14:32,639 --> 00:14:39,240 limited its stability if voltages weren't dialed in just right and we 197 00:14:37,620 --> 00:14:43,139 could be witnessing similar teething issues with this new chiplet 198 00:14:40,680 --> 00:14:47,279 architecture and supporting that theory is the fact that the driver straight up 199 00:14:45,060 --> 00:14:51,720 locks out power adjustments below 90 percent of total board power when we 200 00:14:49,500 --> 00:14:56,579 asked them about this AMD played pretty Coy with us which tells us either we're 201 00:14:53,820 --> 00:15:00,720 right or that they're not sure either now it's time to talk about what this 202 00:14:58,560 --> 00:15:05,040 data means for the GPU Market first of all anyone who was hoping that the 7900 203 00:15:03,000 --> 00:15:11,100 series would come in and compete head to head against the RTX 4090 will be sorely 204 00:15:08,279 --> 00:15:15,959 disappointed at least for today we saw a brief glimpses of Glory but the RTX 4090 205 00:15:14,040 --> 00:15:21,720 remains on its Throne making it tough to dislodge its price from that TI Titan 206 00:15:19,740 --> 00:15:26,639 perch that it currently occupies as a Halo product but while AMD doesn't have 207 00:15:24,540 --> 00:15:33,540 a direct competitor to it I still think the 7900 XTX is a much better value 208 00:15:30,300 --> 00:15:36,120 as for the RTX 4080 well the comparison 209 00:15:33,540 --> 00:15:39,480 gets a little more complicated even the XT model comes really close in 210 00:15:37,980 --> 00:15:44,880 traditional rendering and the value proposition of either 7900 GPU is a 211 00:15:43,199 --> 00:15:49,980 damning indictment of NVIDIA's pricing but AMD does remain on the back foot in 212 00:15:48,300 --> 00:15:54,060 Ray tracing performance and that's something that NVIDIA can point to and 213 00:15:51,959 --> 00:16:00,000 claim makes up the difference even if nobody seriously cares it'll take people 214 00:15:56,639 --> 00:16:02,040 like me running out and buying 7900 215 00:16:00,000 --> 00:16:04,440 series cards instead of 40 series to make NVIDIA realize that they're out to 216 00:16:03,660 --> 00:16:09,720 lunch but wait why are we saying this now you 217 00:16:07,500 --> 00:16:14,880 might think we took a lot of flack for covering for NVIDIA's pricing in the RTX 218 00:16:12,000 --> 00:16:18,839 4080 review didn't we well not so much first off we compared it to cards that 219 00:16:17,279 --> 00:16:23,820 were available at the time it released and we ended the video with a message to 220 00:16:20,940 --> 00:16:29,040 wait for our dna3 and we're reviewing the 7900 Series in much the same way 221 00:16:26,820 --> 00:16:34,800 second it had already been made known that the RX 7900 XTX was not going to 222 00:16:32,339 --> 00:16:40,620 compete with the 4090 which depending on how you look at it means that either the 223 00:16:37,139 --> 00:16:43,139 RTX 4080 is actually a 90 class card and 224 00:16:40,620 --> 00:16:49,620 priced accordingly with the 490 being a TI or something or the 7900 XT and XTX 225 00:16:47,459 --> 00:16:55,920 are the current gen equivalent of the 6800 and 6800 XT and are themselves 226 00:16:52,860 --> 00:16:57,300 overpriced basically it's kind of a game 227 00:16:55,920 --> 00:17:01,980 of pick your Villain at this point because the reality is that 600 top of 228 00:17:00,060 --> 00:17:06,500 the line gpus are a thing of the past and that was an assumption that our 229 00:17:04,079 --> 00:17:10,679 review made but did not fully explain it's worth pointing out by the way since 230 00:17:08,819 --> 00:17:16,679 I'm being a downer right now anyway that while AMD supports DisplayPort 2.1 and 231 00:17:14,040 --> 00:17:22,439 NVIDIA's 40 series does not AMD went with a half measure by supporting only 232 00:17:18,600 --> 00:17:24,419 the uhbr 13.5 spec what this means is 233 00:17:22,439 --> 00:17:28,559 that it can only achieve its lofty resolution and frame rate support 234 00:17:25,740 --> 00:17:32,880 through display stream compression DSC allows up to three times more bandwidth 235 00:17:30,360 --> 00:17:37,460 than is both near zero latency and Visually lossless though that doesn't 236 00:17:34,980 --> 00:17:41,100 mean that there is no quality loss practically speaking this won't be a 237 00:17:39,419 --> 00:17:44,280 problem even if it were noticeable given that these gpus aren't really capable of 238 00:17:42,960 --> 00:17:48,360 pushing those resolutions and frame rates anyway but we brought it up with 239 00:17:46,380 --> 00:17:52,440 the RTX 40 series so we're bringing it up here for the sake of completeness it 240 00:17:50,340 --> 00:17:56,700 means that those new 8K ultrawide displays being shown up at CES will just 241 00:17:54,840 --> 00:18:00,900 barely be able to run at 85 Hertz without DSC but I would never expect 242 00:17:59,100 --> 00:18:05,100 that to be a problem for an actual user one slightly problematic thing that I do 243 00:18:03,120 --> 00:18:09,240 need to address is that we've taken a very glass half full approach today and 244 00:18:07,860 --> 00:18:14,280 I can understand why some of you might be frustrated by it we've gone and spun 245 00:18:12,059 --> 00:18:19,679 AMD's potential future driver improvements as a good thing when 246 00:18:17,100 --> 00:18:23,580 actually they're kind of bad another perspective is that it means that AMD is 247 00:18:21,840 --> 00:18:27,840 releasing these products even though they haven't really finished optimizing 248 00:18:25,500 --> 00:18:31,919 them and if you've kind of felt that way up to this point I totally get where 249 00:18:29,580 --> 00:18:37,200 you're coming from we ran into some kind of bizarre no really bizarre crashes and 250 00:18:35,700 --> 00:18:43,260 there are certainly features that could be added or refined but In fairness to 251 00:18:40,320 --> 00:18:47,760 our charitable approach today AMD is pricing these gpus according to what 252 00:18:45,120 --> 00:18:51,660 they are now and their track record for fixing these things is actually pretty 253 00:18:49,679 --> 00:18:57,660 good even if sometimes it takes a while as they focus on new features like FSR 254 00:18:54,200 --> 00:19:00,419 3.0 and the hyper RX performance 255 00:18:57,660 --> 00:19:04,799 Optimizer that have yet to Launch and if this is where performance is now with 256 00:19:02,760 --> 00:19:10,140 some of what broken certainly unoptimized drivers AMD clearly has some 257 00:19:08,280 --> 00:19:13,980 great Hardware on their hands here they even brought back the USBC Port 258 00:19:11,940 --> 00:19:19,020 something that is great for hooking up a monitor via a single cable and a godsend 259 00:19:16,679 --> 00:19:21,600 for enthusiasts like myself I'm still mad that NVIDIA removed it after the 20 260 00:19:20,700 --> 00:19:27,059 series so the RX 7900 XTX is going into my 261 00:19:25,140 --> 00:19:33,120 gaming rig and I'm officially switching to team red for the first time since the 262 00:19:30,419 --> 00:19:37,620 3870. goodness gracious 263 00:19:35,160 --> 00:19:42,059 the performance is good enough the price is good enough well at least compared to 264 00:19:39,660 --> 00:19:46,200 the competition and the downsides are small enough that I am comfortable 265 00:19:43,620 --> 00:19:50,580 making the leap just like I'm gonna comfortably leap to this Segway from our 266 00:19:48,240 --> 00:19:55,799 sponsor MSI and their recently released mag z790 Tomahawk Wi-Fi motherboard the 267 00:19:53,460 --> 00:20:00,000 mag 790 Tomahawk is a gaming oriented ATX motherboard that balances value with 268 00:19:58,080 --> 00:20:04,919 premium Hardware it's built around the latest Intel z790 chipset which supports 269 00:20:02,580 --> 00:20:09,179 12th and 13th gen Intel core processors with dual 8-pin power connectors and 270 00:20:06,900 --> 00:20:13,559 core boost technology is able to sustain heavy CPU power loads to better support 271 00:20:11,400 --> 00:20:17,940 demanding game settings it also features m.2 Shield frozer MSI's Advanced thermal 272 00:20:16,500 --> 00:20:21,900 solution which offers great protection for maximum SSD transfer speed 273 00:20:19,860 --> 00:20:25,740 performance with integrated options for multiple cooling methods the z790 274 00:20:24,000 --> 00:20:28,919 tomahawk ensures Optimal Performance when you need it the most make sure to 275 00:20:27,720 --> 00:20:31,740 check it out at the link in the video description if you enjoyed this and 276 00:20:30,780 --> 00:20:36,720 you're looking for something else to watch maybe go check out that Arc 30 day challenge video I mentioned Intel's 277 00:20:35,520 --> 00:20:40,940 doing some really good things over there but they also have a lot of work to do