Minecraft’s r-tx really surprised us over the past year apparently nvidia has made meaningful improvements to both the LSS moving out of Vaseline on camera blurry meme status and into potentially useful status and improve deterrent exp tricks in this benchmark we’ll be extensively testing at multiple r-tx GPUs against Minecraft RT X for Windows 10 including a 60 FPS normalized chunk render distance benchmark standard benchmarks with RT X on off DL SS on and off and will include visual analysis of many of the features in Minecraft snoo beta the benchmarks help set the bar for what level of performance is needed to Play Minecraft yeah that’s that’s where we are now we’re trying to play Minecraft before that this video is brought to you by Squarespace Squarespace is what we’ve been using for years to manage our own gamers Nexus store and we’ve been incredibly happy with the choice Squarespace makes ecommerce easy for those interested in Starting stores but it also has powerful tools to build all types of websites photo galleries for photographers resume and portfolio sites and small business sites are all easily done through Squarespace having built a lot of client websites the old way before running GN full-time we can easily recommend Squarespace as a powerful fast solution go to squarespace.com slash gamers Nexus to get 10% off your first purchase with Squarespace we had a pretty complicated test matrix for this one between the variations of our TX on and off and DL SS on and off we also had to deal with Things like the chunk render distance so this is a major component to Minecraft it effectively determines your your view distance in the game and given how the game is built and how it mechanically works that’s a large part of it so there’s a lot to test just to make sure We’re clear upfront on everything we’re going to be talking about a lot of graphical improvements that are specific to Minecraft but are not new to the industry so simple additions like roughness maps and normal maps are a commonplace in basically every game minecraft didn’t have native roughness Metalness normal Maps it’s lacking a lot of stuff so when we talk about these things please remember that these are not new technologies a lot of this stuff is not special RTX and video only features it’s just minecraft finally got this stuff because r-tx is rolling out with it and they Work together so all the textures right now it’s they more or less without the r-tx version are only a color map and then the game is rendered unlit to some extent chapters will be included in the description to help navigation particularly to make sure people see the Quality analysis and the charts like we have a lot of data in here and discussion towards the end on DL SS and this is also one of the games I’ve had the most fun benchmarking recently because I haven’t personally played minecraft since around got it in alpha And last played it when it was exiting beta so it’s been a while and this was a fun meta automation of a benchmark because we automated our benchmark within the game with the game parameters and so we basically use the mind cards to build the track to automate on a time Or what our benchmark was anyway it’s really difficult to benchmark accurately for a lot of reasons most of them are related to Minecraft optimization in general but we’ll talk about that more later let’s start off with an overview of the changes for minecraft RT X and some initial graphics analysis we’ll get Into the numbers and then we’ll go through the rest of the graphics analysis minecraft is almost an accidentally perfect candidate for RT X but it’s not for the obvious reasons traditionally NVIDIA has pushed for the feature to be included in major triple-a graphics titles like Battlefield 5 and That’s become more of a balancing act between the high poly highly complex games getting juggled with real time ray tracing add-ons the end result has typically been a trade-off of what graphics options in the game get turned off or sacrificed in order to turn on RT X and support the performance or how low of a resolution you’re willing to bring the game down to minecraft meanwhile has basically no native unmod at graphics quality to speak of the texture resolutions stock are abominable and the game itself is entirely comprised of simple geometry we’re not saying that in A mean or a critical way we’ve played a lot of Minecraft here so that’s the entire appeal of Minecraft it sort of works here because ray-tracing features aren’t competing with traditional triple-a graphics quality in this title it’s able to take almost all of the GPU horsepower to prioritize rendering ray-traced scenes In an amusing way this is also why D LSS if it works better in its new iteration which will inspect today is a good inclusion for Minecraft the LSS works on neural networks to build predictions of how a game will unfold frame by frame the end goal being that a player would Enable both DL SS and RT acts in order to negate one another in performance but end up with a net higher graphics quality that hasn’t really worked yet and DL SS was deservingly mocked on our channel and others for its comical launch over a year ago now NVIDIA is Trying again with the LSS 2.0 and the company claims it’s really fixed at this time minecraft is probably the most predictable type of game of all time for that neural net to work with and you’re also working with a game where lower resolution doesn’t have as profound of An effect as it would in a game with a greater fine detail or nuanced texture to work with if the LSS doesn’t work here it really would have no chance to work anywhere get into minecraft content then it makes these key changes PBR compliance has been added for materials In the game physically based rendering is something we’ve covered in the past with the Crytek team but the crux of it is that a given material will now have metallic and roughness values which are two properties that define almost every surface in the real world the texture Maps added include roughness on items Like bricks in the game where a roughness map is used to influence reflectivity off the surface of the object the mortar meanwhile between the bricks is rougher and less reflective while the brick is smoother and shinier another addition was normal Maps which work by distorting how the light Interacts with the surface of the mesh also known as an object an example can be seen with blocks of iron where the normal Maps now make it seem as if there are five bars bulging off of the surface and the original graphics for that would look basically completely flat global Illumination was also added this is something we’ve defined fairly recently in great detail it’s the calculation of light bouncing off of multiple services multiple times RTX shadows make secluded areas significantly darker like in forests bloom is now enabled Authority exon pretty much everyone remembers what that Is from the many years ago when that was a big thing in every game and glass and water exhibit reflections and refractions are appropriate so none of these are brand new to the world of computer graphics but they’re new to Minecraft one quick note of fairness of analysis To some levels have resource packs which add emissivity from a block so you have light coming off of a block often the terracotta blocks but they don’t add it to the non r-tx variant remember that these light emissions could at least be added to the non r-tx block so you at Least have some emission of light like a torch would do some of the effects could also be faked you could make up for a lot of the r-tx features by working on the graphics with traditional techniques to fake the higher quality graphics but r-tx is a much easier way to execute Without wreak rafting the entire game and that’s where we get the meme phrase it just works if everyone wanted a fair demo they’d at least add a light emission to both the r-tx and non r-tx blocks and they might even consider faking some of the other effects for non r-tx We just wanted to make sure that you’re aware that it’s not always fair to simply show the r-tx on and off and toggle them to compare because some of the maps and resource packs with the artsy acts off version have had no additional work invested including a lack of basic parameters like light There’s more to it too but we should get into the initial benchmarks first talk through the numbers of how GBS perform and then we’ll look at DL SS we need to start with some initial research on a per map basis before we progress when testing minecraft we explored each of n Videos bundled in 6 RT acts Maps to better understand the environment because the game is almost completely player created there’s a wide range and performance potential our footage should illustrate this the aquatic map with its large glass dome under a waterfall has a lot of reflections and refractions that Strain the GPU in different ways from the jungle map which is more similar to standard survival mode minecraft and different still from the stressful dystopian city map that we standardized for testing the shadow and light map was good as a showcase of ray tracing but some of its indoor settings naturally Restricts viewing distance and restrict the amount of objects visible on the screen so the map itself is barren inside of each of the rooms the first chart is strictly research which we used to decide how we’re going to test a game prior to committing a few days of doing Testing this chart shows the results for a 2080 super tested with 8 chunks drawn at 1080p with RT axon and DL SS on well test other settings later in the video here you can see that the shadow and light map using the mirror room renders at 104 FPS average with the 2080 super On these settings the Crystal Palace map renders at 83 FPS average in our test area while the aquarium section of the shadow and light map renders at 76 FPS average the stress tests that GN used located in the dystopian City renders at about 71 FPS average so it’s more Similar to the aquarium area of the previous test this is like for like testing across different maps with the same GPU and the entire point of this is to illustrate that the vastly varied environments your framerate what matters most for a benchmark is consistency so that we can fairly compare devices Head-to-head so as long as we pick one map and stick to it we can at least make a comparison for hierarchy of kart for this one we noticed that the city level had the lowest run to run deviation of all and was the most reliable and also The most stressful so we stuck with it as a representation of the extreme end everything should scale with relatively the same stack in the hierarchy so you won’t see much movement from where the cards are relative to each other just gonna be the absolute frame rate that Changes the city benchmark we used was sort of fun anyway because it was a meta automation that we were able to develop in this level where we built a simple minecart track with a known speed and on a known time each direction so we can consistently test using real-world Gameplay across a variety of graphics elements in the level one of the most important things for benchmarking minecraft is traversal of terrain such that the chunks are drawn and purged to and from the screen without this you lose the realistic and and most of the load intensive elements of the game are Lost with that our mind track helps us traverse far enough to influence trunk loading which also helps better understand performance for the chart that’s coming up in a minute we’re doing a simpler task that’s hopefully useful to more serious minecraft players who have experience with the games existing but limited graphics settings minecraft Renders its world in clusters of blocks and data known as chunks which is going to be an annoying name to say this video with our TX disabled we can see pretty clearly as each chunk draws into the world given our render distance settings without our TX those settings Include a 96 trunk default 128 trunk option 160 option and that goes down as well if you’re running on really low end hardware each draws increasingly more of the world before fading out of draw distance but also significantly increases system load for our TX your options start at 8 chunks rather than Something like 32 or 96 and then go to 12 16 20 and 24 it’s significantly more limited by nature of our TX we did a 60 FPS normalized benchmark to look at chunks draw distance while maintaining 60 FPS with our TX on and with the LSSI this means that we increase the draw Distance until we hit 60 FPS sustained plus or minus about 2 FPS we have some leeway there well have charts with more precise FPS numbers in a moment but this is more of a functional task for players who want to know how much of the world they can actually see here’s The chart the 20 ATT I draw 16 chunks at 1080p 60fps but couldn’t go higher at 60 at 1440p it does 12 chunks at 60fps and at 4k it is incapable of reaching 60 FPS in our test world even with 8 chunks drawn the r-tx 2080 super rendered 12 Chunks at 1080p 60fps 8 at 1440p 60 FPS and was below 60 FPS at 4k so unable to render the minimum 8 at 60fps it can still render it can still play and we’ll get to that later just not at 60 that said the 2080 super was within a couple Of frames per second of hitting 16 chunks drawn at 1080p so we’d count this as 16 and less intensive maps and workloads especially as things get optimized later by Nvidia and Microsoft the 2060 super is barely able to hold at 8 chunks for 60fps at 1080p it’s Technically below that it’s at 55 but it’s close enough that we’re counting it since this is a more intensive map the 2060 ko is incapable of reaching 60 with our TX and DL SS enabled in our test suite for a first set of comparative tests we’ll start with 1080p our TX on DLS has on and 8 chunks render distance tested in our custom course on the dystopian city map the RTR 2080 Ti Stream holds that about 90 FPS average with lows relatively consistent run to run at 59 FPS 1% and 50 for FPS 0.1% that puts the 20 atti About 26% ahead of the 20 80s super as wouldn’t make sense the 20 80 super is about 7.2 percent ahead of the 2017 / which maintains 60 fps and heavy load scenarios at 1080p and with a draw distance of again 8 this keeps the 2070 super and overall good classification in Terms of experience the two are so close though that we at first didn’t trust the numbers so we ran the tests a few more times 8 more passes on the devices and it didn’t change we think this is because the RT hardware is so similar That at 40 cores RT cores that is versus 48 RT cores they’re close enough that the more limiting factor is the RT Hardware for once rather than rasterization performance the 27 t super leads the 20 60 super by 21% which also leads the 20 60 KO by 21 percent the 20 60 KO struggles noticeably on this heavier map although do ok on less intensive worlds that are not as RTX feature focused if we look at the graphics options for the game there aren’t many that you can adjust to help with performance so the 2016 on super Cards will be the ones that are the most on edge of running adequately you can further reduce resolution you could try to stick to less arty XIV world you could turn off fancy rendering settings as they’re called in Minecraft but none of that’s particularly helpful aside from the resolution reduction if you’re Playing in survival mode and you’re unlikely to build a metropolis that is replete with shiny objects that are TX likes then the 2060 would fare better and remain more playable for reference standard deviation and average FPS was 0.4 to 0.6 FPS so it’s actually incredibly consistent with our TX on our TX office a different story but on its pretty good standard deviation in 0.1 percent lows is always much higher by the nature of 0.1 percent lows but it was still fairly controlled for a beta of a game which is notoriously unoptimized already we measured that at about three points 1440p comparative Benchmarks show similar scaling just with overall lower frame rates the twenty atti drops to 72 FPS average from ones 1080p result of my FPS average which is a reduction of 20% from the resolution increase keep in mind that DL SS changes properties based on resolution so it’s not fully like for Like as a comparison between the resolutions the 2080 super is still around the 60 FPS marker at 56 FPS average the 2070 super is falling behind but we’d still consider it playable on most survival worlds with less rate raise the intensity and the 2060 ko falls off at this point and feels slow On input as a result of its lower frame rate 4k is rough even with the performance optimized settings applied for DL SS at this resolution it doesn’t really look that great for the 2080 TI the 2080 TI has now fallen another 33 percent from its 1440p results to 48 fps Average the can it run minecraft mean that we’ve seen in our comments lately has actually become a reality so good job internet commentaries you got what you wished for it now our only hope is Crysis remastered the 2080 Super and the 2070 super don’t hold up well either and At this point it’d be best to drop resolution and keep our TX fortunately with Minecraft that decision is easy to make you’re not really missing out on a ton with a lower resolution in most instances especially if you’re talking 4k down to 1440 and using DL SS even with the heavier Texture packs and RT X in this instance adds disproportionately more value than it would in other titles where you have to juggle between whether you want higher shadow quality or more geometry detail and this one everything’s a cube so it’s pretty easy toss up for some Perspective as to how things work in the game this chart shows more of our research with just the RT X 2080 super we have a complex test matrix here there’s our TX on our TX on with the LSS on our TX off our TX on with the LSS off And then there’s also a chunk distance and our TX off and chunk distance and so forth obviously disabling our TX worked the best at 1080p with the 2080 Super we were bound by the CPU so we’d have an even higher frame rate of the CPU were capable That’s expected of a game where the well of this caliber and it means that the framerate hit from enabling our DX is greater than what we can demonstrate in this example that’s especially true because we can’t provide a like-for-like drop in frame rate from enabling RT acts Because we can’t in the game set the chunk draw distance equal in a way that it’s within the bounds of what the CPU can do with our TX off and within the bounds of what the GPU can do with our TX on so it’s not even comparison with a 96 chunk draw distance which is significantly and noticeably farther than the a chunk our TX default option we saw 180 FPS average with pretty far behind lows due to high variability when our TX is off which is either a Minecraft issue in general or a minecraft r-tx issue with Windows 10 we Should note that we had another test run at 300 FPS + so our TX disabled with this chunk count is almost impossible to test consistently since it draws sort of differently each time either way that’s our baseline here our next best was 40% of the framerate of our TX off in this Chart at 71 FPS for the our TX on DL SS on render which helped improve performance with DL SS but what’s still a big drop this is a heavy hitter to performance even in this game but that’s what makes RT acts such a natural candidate for Minecraft there’s not much Else going on in the scene so RT X can get full focus of the GPU expanding our draw distance to 12 reduced FPS by 6% expanding it to 16 reduced FPS by a further 12 percent on top of that and it’s a nonlinear drop in framerate as the draw distance increases Our next our TX on jump goes to 20 chunks the LSS enabled at 47 FPS average or a 19 percent drop from the 16 junk option at 59 FPS average this is about the same performance as disabling RT X and rendering 160 chunks which one you prefer will depend mostly on preference But there’s no doubt that RT X looks significantly better for what’s drawn finally the maxed out our TX option is at 23 FPS average on the 2080 super so not quite ready for this hardware or with this iteration of Minecraft RT X and finally here’s a chart of DL SS on Versus off this will precede our DSS on versus off visual inspection that we’ll do right after this chart if you wanted to ditch the LSS you’d also be giving up about a 60% performing uplift on the 20 atti at 1080p a 63% uplift on the 2080 super from off to on 66% on the 2070 super 70% on the r-tx 2060 super and about 64 percent only 2060 k out for the most part other than the KO where it’s just choking on hardware in general the uplift is nonlinear and increases as the cards get lower and the LSS does a lot in this Game for performance now we just need to look at what it does from a visual point of view finally after all this testing we sort of made the assumption that most people would be playing with mostly arti axon and the LSS enable as the game becomes too difficult to run without the LSS this also allows players to maximize render distance which is arguably more important than a crisper image still we need to do a DL SS comparison to really know minecraft with DL s s can be nitpicked for differences which is what we’re about to do but it’s important not To lose the forest for the trees although a lot of small details can be seen in side-by-side comparisons actually playing the game which will shift player focus more toward actions and less toward pure observation means that the ability to notice small details will go down compounding this all the Movement will decrease the player’s ability to spot those differences in realtime anyway purely to help show the technical differences we’ll start first with an image of the aquatic palace in this image going between the LSS on and off at 4k shows some difference in color within the keyhole passage mostly and That’s not necessarily better or worse but just different there are also differences in lighting on blocks near the rightmost parapet or the leftmost towers we think this is because there’s less pixel data to work with when DLS s is enabled and the resolution is natively 1080p rather than 4k and super Sampled up so the approximation is less accurate to the source the bloom around the floating points of light up in the air is also much stronger bloom and some 3d applications like blender is done by Counting pixels to determine the radius so starting from a smaller base image Could scale up the bloom larger than the intended native size the differences are in the same spots at 1080p but are more pronounced since there’s less data to work with that the native 720p render resolution the next image is a castle within a jungle for this one we Noticed that the candle light in the bottom center entrance is visible in both 4k with and without the LSS but is not visible at all with 1080p that 4k draws this with D LSS on is the interesting takeaway here we also notice that the textures on the left side of The image in the trees is significantly different this may be because it has more pixels to show the leaf density but we’re not positive it could also be something to do with the randomization of the texture if that is a thing but we don’t work with Minecraft enough anymore To remember on the right side we also see lighting changes in the forest similar to the previous image in our Crystal Palace example next we noticed more highlights on the edges of objects like the red bridge on the left and also noticed slightly brighter lighting toward the upper right of the castle we Think this is related to reduced geometric information at 720p native resolution finally in the shadow and light demo we see that DLS s still has some of its blurry roots the water reflections at the bottom of the screen are blurrier with TLS s on there’s reduced ambient occlusion or less Profound shadows in the corners and that’s toward the back wall the sharpness of edges is reduced for the green text we also noticed slight aliasing on the edges of the green text shown as thin black line artefacts overall we classified 1080p with TLS s off as visually similar or slightly Superior to 4k with the LS s on picking between them doesn’t matter too much just go with whichever one has more performance after talking with Andrew our video editor who does a lot of work in 3d modeling applications and game engines we decided that most of these differences are only really noticeable When you’re comparing the images side-by-side and looking for them we’d say that forgoing DL s s is better in a technical sense of visual quality but that the performance gained in Minecraft r-tx specifically far outweighs the less noticeable quality impact from DL s s being on and videos implementation is Much better in this game then its attempts last year of course it does have its imperfections to make sure everyone’s on the same page we do find ourselves toggling our TX off somewhat regularly for either specific bugs or for navigation of the map a list of issues and imperfections we encountered Is as follows the noise our artifice particularly at lower fps can become very visible we demonstrated this in the ceiling of the aquarium where rapid movement showed denoising at the edge of the view frustum we also noticed this especially heavily in darker settings are at lower frame rates when near Clouds it’s not ever prevalent but it’s there and again especially there’s lower frame rates speaking of clouds seem bugged with our TX on or done in a way that is mechanically worse with our TX off they’re translucent as they probably should be but our TX on makes them solid White this means that they become annoying from a gameplay perspective and they don’t seem particularly realistic anyway it seems that backface Coleen might be disabled with our TX on as well something that we were able to postulate based on what happens when you go into a Cloud the player is drawn at the same time as the rest of the world with our TX enabled and so the players hand and the objects will clip through the walls with our TX on but not with our TX disabled that’s because it has to be drawn at the same time as everything Else to ensure proper lighting of the players hand and the object whereas they’re drawn separately with our TX off this is just a small mechanical item it’s way lower draw distance with our TX on of course so even when you’re at lighter weight levels your best bet is Going to be 24 chunks and that’s taxing on the more complex levels you get a pretty standard 96 chunks with our TX off and can even do more and the bottom of that is also higher than 24 anyway we found ourselves toggling our TX off in order to quickly find places or to Evaluate the map so it can become a mechanical hindrance in some scenarios and you might find yourself toggling it on and off just to navigate at least in the early stages of the game some areas also become extremely dark which becomes a subjective debate it could provide more gameplay Opportunities for a creator or a mater who could light paths to better guide players in their creations but it can also slow down the mechanics of the game if you’re just trying to navigate quickly in that older style of Minecraft so it really this comes down completely To what you or the map creator is trying to do there are a lot of avenues here with the way our TX works to create new mechanical ways to interact with the game and there’s also ways to annoy players just depends on how it’s used finally bloom sometimes follows the Screen similar to the paper over the water in battlefield 5 where you get some ghosting effect of objects as you move the camera around so that’s it for our minecraft r-tx video and video has played this somewhat brilliantly by partnering with Minecraft and video had a really rocky launch to RTX It was something like 50 55 days before they even had a title out that would work with RT x part of that was because microsoft delayed its windows update past the RT x launch window for the 20 series and so NVIDIA was left high and Dry with no DXR or anything to work with and launched therefore was was pretty rough because there was literally zero value in RT x at launch and there remains to be value that has extremely gained dependent but it’s growing and the reason that this particular step is A brilliant play is that minecraft is a large enough community that it’s got games within games so you’re talking about an ecosystem of players who will find ways to mechanically leverage RT X in addition to visually leverage it to work with new viable mechanics that might not have really been viable before Again having really dark forests is something that either like or don’t like subjectively but it can be leveraged mechanically to help better guide a player to where they’re supposed to go and just basic level design so value goes up disproportionately from adding this one title to the roster as opposed To the average triple-a title where it’s pretty locked down and developer controlled additionally NVIDIA has shifted the conversation towards ray tracing in general real-time retrace and specifically ray tracing is not new they didn’t invent it it’s been around for ages and it’s not special but real time Ray tracing is what NVIDIA has been pushing for and now you see that suddenly consoles are including it suddenly am these getting involved so the conversation has shifted and it’s taken a while and the value is still questionable depending upon if games that you like have RT x support in a way That’s competent but minecraft is a massive addition to that it’s one of the best selling games if not maybe the best to get selling game aside from things like peggle or Tetris so it’s up there but anyway there’s still plenty of issues framerate issues stuff like that this is not a Trivial game to run it’s just it’s interesting because of how it’s so graphically non intensive at its core RTX can be can dominate the entire GPU and take all those resources for itself so very interesting stuff but looks good overall and we’d say compared to the previous attempts at the LSS job well Done on video on this version at least with this game it may change we’ll see so that’s it for this one subscribe for more always go to store documents access net or patreon.com slash gamers Nexus helps out directly and we’ll see you all next time Video Information
This video, titled ‘Minecraft RTX GPU Benchmark & DLSS On vs. Off Comparison: Significant Improvement’, was uploaded by Gamers Nexus on 2020-04-21 10:28:35. It has garnered 250004 views and 6563 likes. The duration of the video is 00:29:52 or 1792 seconds.
We’ve exhaustively benchmarked Minecraft RTX on Windows 10, including DLSS on vs. off, RTX on vs. off, chunk render distances and FPS, and more. Sponsor: Get 10% off Squarespace purchases (https://geni.us/BqEpf)
If you’re wondering what the best video card is for #Minecraft RTX, our benchmarks will answer that question. NVIDIA’s RTX and DLSS 2.0 came to Minecraft recently. Our GPU benchmarks include the RTX 2080 Ti, RTX 2080 Super, RTX 2070 Super, RTX 2060 Super, and RTX 2060 KO. These tests look at framerate (FPS) and frametimes at various resolutions, including 1080p, 1440p, and 4K, and with DLSS and RTX toggles. We also tested for 60FPS-normalized chunk render distance (how many chunks can be rendered at a fixed framerate) and tested performance on multiple maps, as it is variable. Additional graphics analysis includes explainers of new graphics arriving to the game, side-by-side comparisons of RTX on vs. off, comparisons of DLSS on vs. off, and discussion of glitches, imperfections, and differences. If you’re lagging in Minecraft RTX, there’s good reason, but there are a few ways to optimize (albeit not many). We’re using our standard GPU test bench, including a 5GHz 8086K. More details in our GPU review methodology posts.
Gigabyte RTX 2080 Ti Xtreme on Amazon: https://geni.us/CpdoO Gigabyte RTX 2080 Super on Amazon: https://geni.us/DQSpF6 RTX 2080 Super on Newegg: https://geni.us/M1mN EVGA RTX 2070 Super on Amazon: https://geni.us/4uvATWW MSI RTX 2060 Super on Amazon: https://geni.us/pwNg EVGA RTX 2060 KO on Amazon: https://geni.us/qerUgl
Thanks for watching! We said it a few times in the video, but lots of people skip down to the comments. To really reiterate a point: Minecraft is a game that is made by its players, and so performance is mostly a representation of an expected hierarchy of cards and of an expected delta between cards. The absolute FPS numbers are largely meaningless, as they will change from map-to-map and creation-to-creation. We show that here, so you can check out the video if you want to know more.
The best way to support our work is through our store: https://store.gamersnexus.net/
Like our content? Please consider becoming our Patron to support us: http://www.patreon.com/gamersnexus
ALTHOUGH we provide these timestamps, there is important information each step of the way. If you’re going to comment, you should watch the whole thing so you understand it. Skipping steps will leave out info that will help in the next section.
00:00 – Important Background Information 03:25 – Technical Vocabulary & Changes 05:26 – Physically Based Rendering, Normal Maps, & More 08:00 – Benchmarking Research & Initial Results 11:45 – Chunk Draw Distance at 60FPS Normalized 13:10 – 1080p Comparative GPU Benchmark 15:25 – 1440p Comparative GPU Benchmark 16:05 – 4K Comparative GPU Benchmark 17:30 – RTX Draw Distance vs. RTX Off, DLSS Off 19:48 – Minecraft DLSS On vs. Off Benchmark 20:30 – DLSS 2.0 Visual Comparison On vs. Off 24:20 – Imperfections, Ghosting, Noise, & Bugs 26:50 – Conclusions
** Please like, comment, and subscribe for more! **
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video (“this video is brought to you by”) and above the fold in the description. We do not ever produce paid content or “sponsored content” (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
Follow us in these locations for more gaming and hardware updates:
t: http://www.twitter.com/gamersnexus f: http://www.facebook.com/gamersnexus w: http://www.gamersnexus.net/
Editorial, Testing: Steve Burke Video: Keegan Gallick Video, Graphics Analysis: Andrew Coleman
