Gadgetory

welcome back to hard run box today it's time to take a look at rate racing on videos Pasco GPUs the feature that opened up to all last week through the release of driver version 4 25.30 one prior to this you needed an Nvidia chewing r-tx graphics card to get all the visual benefits of ray-tracing but for a variety of reasons I'll get to later you can now access rate racing in today's games on Pascal GPUs from the gtx 1066 gigabyte and up along with nuturing GTX GPUs like the gtx 1660 TI i'll be honesty a bit a little sick in the past week so i haven't been able to do as much testing on this topic as i'd normally like but i have managed to take the most powerful pascal GPU i have the NVIDIA Titan X and put it against invidious RT X lineup in all three of the games available today that support rate racing now we all know ray tracing on Pasco GPUs is going to suck this isn't some big surprise it's just a simple fact of having a GPU that doesn't have any specific acceleration built in for a very performance intensive graphical effect but there are some interesting questions about ray tracing on Pascal that I did want to answer the first is to what degree does Pascal suck is a card like the Titan X faster than the slowest RT X GPU v RT X 2060 NVIDIA has given out a few benchmarks they give us some indication but of course we want to verify this with our own data and in particular look at not just average frame rates but 1% lows and general performance swings within a gaming session the other question is whether you can still get some form of acceptable performance from a Pascal card while ray-tracing despite knowing full well the performance in general is not going to be great for example can this card run a game at above 30 FPS at 1080p with acceptable ray-tracing quality this will allow some gamers to genuinely try out ray tracing without being forced to view a slideshow so let's get into the data and as always I've used my Korra $9.99 hundred K test rig for these benchmarks with 16 gigabytes of ddr4 memory not that these titles are CPU or memory bottleneck to our ray tracing all of the data you see here is from the latest versions of the games with the latest drivers installed I know that slight battlefield 5 are continuing to optimize their ray-tracing performance so it's crucial with these sorts of tests to use up-to-date benchmark data you'll see the areas we've tested in each game as we get to them let's kick things off here with shadow of the Tomb Raider the latest game to include ray tracing through the form of shadows when I tested this game previously I found that only one of the three ray tracing modes makes any sense to use and that's the ultra mode the high mode produces worst visual quality than disabling ray tracing in my opinion while medium has a very limited scope for ratio shadows so for the testing here I've used the ultra mode we'll start here with a 1080p data and as expected the Titan X is falling between the artex 2070 and RT x 2080 when ray tracing is disabled and all of the settings are set to their maximum level however when enabling ultra ray tracing the Titan X immediately drops to a level of performance below the r-tx 2060 in fact it's 10% slower when looking at average frame rates but a huge 34 percent slower when viewing 1% lows and this begins to illustrate one of the key problems with ray tracing on Pascal specifically the experience is extremely inconsistent this is because there is such a large difference between the capabilities of a card like the Titan X without ray tracing and with ray tracing so as you move around an environment with varying ray accounts interactions and degrees of ray tracing the performance of the Titan X fluctuates massively in areas with little ray tracing performance is decent but when you're in an area with lots of shadows your frame rate can absolutely tanked as you can see in this chart one percent low performance so 21 FPS is unplayable and that's just at 1080p but if you had gone by just the average perform it's 47 FPS sounds all right you know it sounds decent the actual experience of playing the game though is far from that now of course you do also get a fluctuating framerate with our TX GPUs including the RT X 2060 but the issue is less pronounced the GPU is simply not as fast in areas without ray tracing and it can keep up better when ray tracing is enabled the one percent low frame rate for the our Tex 2060 was a touch over 30 FPS which isn't amazing but it's an experience that is somewhat ok move you to 1440p and it gets even worse for the Titan X we're down to a 30 FPS average and just a 14 fps 1% low which is completely unplayable the margins between the our TX 2 sixteen tight necks are a little narrow here as the 2060 also struggles at 1440p but with the titan clocking in more than 30 percent slower in the most intensive areas Pascal's simply can't keep up let's take a look at a more positive game for Pascal and that's battlefield 5 here are recommended setting to use is low reflections it doesn't have as many effects as the high or ultra modes but it's a good starting point and these days the performance hit isn't nearly as bad as it once was at 1080p we're looking in a reasonably significant drop in performance for the Titan X coming from well over 120 fps with the ray tracing disabled down to just 70 FPS on average with ray tracing however once again 1% loads get hammered having here from just low ray tracing perhaps the most interesting thing is that unlike with shadow of the Tomb Raider in battlefield 5 Pascal is more competitive with the RT X 2016 on average it's a little faster and in the most intensive areas it's a little slower we aren't anywhere near the performance of VRTX 2080 which comes closest to the Titan X when ray tracing is disabled but you do a fairly light implementation of ray tracing with the low mode the Titan X isn't as overwhelmed and it performs all right here in fact with the 1% low of 46 FPS the game is playable here performance still does fluctuate a lot but at least it's not dipping to slideshow levels on the regular so at 1080p with low ray tracing and a cut as powerful as a Titan X or 1080 Ti you could conceivably play the game without tearing your hair out would anyone sacrifice over a hundred FPS at this resolution for this performance well probably not but at least it is possible to try it out at 1440p the situation isn't as promising with 1% lows closer to 30 FPS for the Titan X which in a multiplayer shooter is really not acceptable we're also starting to see a divergence between the Titan X and RT X 2060 on average of the Titan is faster but it's much slow in the most intensive areas of our benchmark run this makes sense as more Ray's need to be cast at a higher resolution and any increase in ray tracing will punish Pascal more than Turing RT X the final game we're looking at here is Metro Exodus which uses ray-traced global illumination when I first tested the game I used the benchmark tool but I've switched to an in-game run here focusing on the high mode which I feel is the best to use in the game based on my previous Explorer I didn't even bother testing 1440p in this toilet because at 1080p were already at an unplayable level with the time next a 30 FPS average with the 1% low of 23 fps is simply not good enough and in this game there is no rate racing level below high so you can basically rule out Pascal entirely if you want to use rate racing in this title the Titan X is well behind the ITX 2060 here more than 30 percent behind in fact which again makes sense as global elimination is one of the most intensive ray tracing effects and one where acceleration is very useful and again comparing the tight next to something like the ITX 2080 shows that the RT cores are providing double to triple the performance with this effect enabled we can expect a similar difference between pascal and accelerated tutoring in games that use multiple ray tracing effects the more effects are added the more useful the RT course become so what does this investigation tell us overall well the first thing is that I really didn't need to test any slower cards that the Titan X 2 of the three games are already unplayable at 1080p with the Titans so you can guess how fast a card like a gtx 1070 would be the only test condition that was remotely usable was battlefield 5 at 1080p with low rate racing but even then I would expect a cheat X 1080 to barely hit 30fps in the most intensive areas and performance to fall away further from there so this sense is one of the key questions I had going in no Pascal really is not fast enough to give game as usable rate racing experience even if they're just you know investigating or playing around unless you have one of the top and Pascal cards and the ray-traced effects aren't too intensive as more rate race games come out Pascal it's only going to fall further behind so I don't see much of a future for any last generation card when it comes to rate racing in the absolute best case scenario the Titan X matches the r-tx 2064 rate racing capabilities but often Falls 30% behind or more especially when looking at crucial 1% load data the Titan also delivers a less consistent experience which will annoy gamers that hate frame rate fluctuations normally the Titan X is at least 25% faster than the ITX between 60 and more up around RTX 2080 territory but without those RT calls it just can't keep up with invidious latest mid-range GPU and I guess this brings up the question many people have been asking which is how much do the arty cause and other bits and bobs in the chewing architecture help accelerate rate racing compared to pascal this is more of just an interest sake type question but we do not have some rough data of it can give some insights the main comparison would have to be between the Titan X and the RT X 2080 which are reasonably matched outside of ray tracing the architects 2080 is faster but not by all that much but when ray tracing is factored in the RT x 2080 is anywhere from 26 percent faster with low reflections in battlefield 5 to more than 50% faster in shadow of the tomb raider 2 over twice as fast in metro Exodus considering tu-104 the GPU used in the art x 2080 packs in only 13% more transistors than GP 102 which was used for the titan x and GT x 1080 TI I'd say this level of acceleration is actually reasonably impressive and justifies the extra RT caused at least for these games and effects Nvidia really couldn't have just brute force ray tracing through cramming in more cuda cores the data does show definitively that ray tracing with RT core acceleration is more efficient I don't think the level of acceleration is disappointing either more than a 2x improvement when ray tracing is heavily used is a decent start from a first gen design and one that only packs in that 13% more transistors of course spending die space on specialized cores for ray tracing doesn't help performance in the vast majority of games that don't support the feature but I've already talked endlessly about the value proposition of ray tracing in RTX cards so really don't want to rehash that here also you've already heard us say that we still recommend playing with ray tracing disabled even on our TX cards that hasn't changed with this investigation but bashing ray tracing on our TX cards again isn't really the point of this video either however after doing all of this testing I'm sort of left with a question of what is the point why hasn't video bother testing and enabling ray tracing on Pascal you know it doesn't run well even on high-end GPUs it's unlikely to improve with future games and it just seems like something people wouldn't use even if it was available aside from allowing people like us to test ray tracing on Pascal and show how much acceleration the arti cores provide another technical insights from average consumer the benefits don't seem to outweigh the cost of Nvidia developing this support so I have two theories as to why the first is for developers let's say you have a development studio that invested heavily into Pascal and has loads of cards like this very tight and EXO use for this testing in their dev machines rather than forcing developers into upgrading to cheering to develop games with ray-tracing allowing Pascal cards to rate race albian slowly could improve the adoption of rate racing in games developers don't need 60 or even 30 FPS to test rate racing in their games so it could be handy for them this has a variety of benefits from Nvidia as they you know her interests in improving adoption to sell more r-tx cards the other theory is basically pure marketing by enabling Pascal owners to run rate racing they're getting a first-hand look at how terrible it runs on their existing GPU which might you incentivize an upgrade to an art xgp you in our opinion rate racing alone isn't enough to justify an upgrade but I'm sure this will work for a number of Pascal gamers I don't think this is a bad move having features unlocked for more GPU owners is generally a good thing and down the line there might be a ray traced game that actually does run well on Pascal but like most of our videos on ray tracing so far we're still in the very early stages of the technology and will only become a significant factor in buying decisions in a few generations time that's it for this one time to rest up and edit this video give it a thumbs up if you liked it subscribe for more analysis and investigations like this in the future because this is supporting us on patreon to get access to some cool perks I'll catch you in the next one