SweViver: RTX 2080 Ti vs GTX 1080 Ti on Pimax 5K+ VR Game Benchmarks


You do not need to assume, @SweViver explains it in the video.


No, he does not say anything about the overclocking of the 2080. But again I assume it’s factory. :slight_smile:


It’s an MSI Gaming Trio X which is already factory overclocked and has a custom PCB and cooling solution.


They were right if you take all announcements of Pimax in the KS campaign literally, including that Brainwarp will halve the performance requirements just like that, and the 1070 will be all you need.

But I must say that I automatically dismissed those statements as marketing talk just as I don’t believe in many other marketing statements of various products - my detergent will not make everything look like new, my car won’t run with 5 liters per 100 km, etc. etc.

To me it was always clear that a headset with such increased resolution would be very demanding and that there would not be a magic wand to make all the downsides evaporate to thin air. I accepted that I would possibly see some games have higher resolution but only run acceptable with lower setttings, which would perhaps be an unfortunate combination.

Still I hoped for a substantially improved overall experience in many games, with the potential to get even better when next generations of GPUs arrive (while being conscious that I will jump to the next best headset which betters the Pimax headsets).

In that sense, it appears as if the Pimax headsets will deliver - and this is where I think that TomsHardware was wrong, because they emphasized the shortcomings on certain promises and certain flaws, making it sound as if the VR enthusiast is better off with a Vive Pro or Samsung, which in view of the statements of many other voices we have heard seems an unsustainable position.

Perhaps the VR professional is better off with the Vive Pro, because it is proven tech, likely will have faster, better business customer support (well, don’t count on that but in principle it should be the case), and you need a sturdy work horse, not a fancy, sometimes erratic race horse, in your business application. But TomsHardware did not make this point, but made it sound as if I as an enthusiast should also prefer the Vive Pro, which doesn’t make any sense to me unless most of the beta tester and backer reviews were completely off track.


He states both are OC.


That’s not true.

“both GPUS are overclocked to it’s [sic] maximum”

1080Ti ~2000
2080Ti ~2100


He says both are overclocked to “max”.


EDIT AFTERTHOUGHT: 4) These are however only 10 games, and not the games I’d have chosen. Clearly the easy ones are fine, and clearly the hard ones are too hard to bother with anyway (eg hellblade, fo4vr). But there is still a difference and in some cases that difference means going from PiTools 1.0 to 1.25 or converting something unplayable to barely playable. In other games this may be a crucial little nudge therefore to help make a game playable or decent quality. What I’m saying is 20% or so may not be much but it may be just that little bit that is needed to make a game go from Pi 1.0 to Pi 1.5 and hence look decent. Or push the framrate up from an unplayable 45 to a playable 60. Sometimes small differences make all the difference.

So I dunno it may be something I will consider…


Well said, I agree.

The TH review was BS. Especially as these are the guys who told us to “just buy” the 2080Ti. Biased f***wits.

Having said that I can see why many would be disappointed with a HMD that even a 2080Ti can’t run in many cases, but that’s just the general masses, who should be happy with an OG Vive. For the rest of us it’s a fantastic bit of kit and TH just screwed up, I completely agree. EDIT: In fact it’s a revolutionary piece of kit and blows the competition out of the water. Even the fruit-ninja-playing masses will be amazed by it.


I would expect that before arguing you would at least watch the video. He says it at the beginning.


Yes he does. He says both cards are heavily overclocked.


I did watch the video


I have a liquid cooled overclocked 1080ti and was hoping that the 2080ti would be a big enough leap to justify buying it. Looks like SweViver helped make my mind up.:+1: I wander if Nvidia will release drivers that could help make a big difference between the two? Is there likely to be software enhancements from Pimax that could help optimize performance for the 5k/8k or is this it?


He did? I still think it’s a difficult decision man. Even 10-20% performance increase is very welcome when even that doesn’t yield yet the fps you want. Every percent is welcome in such scenario in my opinion. Although I can surely understand it when people want to hang on to their 1080TI’s because of the steep price for those few extra percents … Maybe when new games will take full advantage of the RTX series that we’ll start to see more difference


Hi guys, haven’t been online until now :slight_smile:

Yes both GPUs are overclocked:
GTX 1080Ti is on 2012Mhz
RTX 2080Ti is around 2100Mhz (pending between 2050-2140)

None of the GPUs has VRAM overclocked as I need to do more OC tests…


The extra outlay might be ok if you just play in 2d, but its doesn’t seem to help that much when it comes to sims in VR and that’s my main interest. I am in the middle of building a new X299 rig and I have the money, but am loath to spend it on a new GPU with out seeing a decent hike in performance.


Another great video thank you.


I am beginning to agree, which is why I added point 4 above to the list of my comments.

Yeah on the face of it you can say that 10-20% is a stupid advantage but yeah that’s totally true when you’re running a game at say 100fps and only get another 20fps on top. But when you’re at some low fps levels every single fps is crucial and can be the difference between a game being playable and not being playable at all. Like I can’t remember but was it hellblade where it’s something like 45fps then goes up to 60fps. Yeah it seems a silly improvement but that might just be the little extra required to make a game playable. Forget about absolute values and percentage increases - think about real world effect on your ability to play games or increase settings. It’s the straw that breaks the camel’s back in many cases.

I may well be tempted to get it. Either that or wait for 7nM next year.

The other thing though is that I am still interested in large FOV mode and for that I’m sure there will be many more games where just that extra bit of FPS will be great. I would love to SweViver to do some large FOV comparisons but I won’t tag and ask him as I know he will be inundated with lots of requests.


Yes as the minority & not the masses. Bur as it said it’s a pre-release review with software not finished. Buu the 2080 ti has nothing to do with individual reporters but TH itself. You should check out tge Real o Virtial review.

I suspect once consumer release nears basic software will function properly with Steam’s Dynamic Resolution & BW & other Refresh rates will be implemented.


I completely agree, and I am suprised sweviver said he doesn’t recomend it, yeah, sure, its a rip-off, but that additional small gain makes a large differenence in the games that were hard to run, I would say the 2080ti is optimal performance for the 5K+, price/performance ratio is terrible on it given its 2x the price for a 1080ti for 20-40% increase, but fact is the 1080ti is exactly 20-30% underpowered for the 5K+ in many -but not all- games.

focusing on the games that were hard to run on a 1080ti.

hellblade at PiTool 1.25, 45fps isn’t acceptable on a 1080ti, 62 is though, and its a 39% jump.

Arizona sunshine at PiTool 1.25, just a 18% jump but it makes it doable, I’m sure the image quality is much better for it.

Skyrim at PiTool 1.5, gives a 32% jump and I’m sure it will look noticably better than 1.25 (although in this case he said it looked good on 1.25)

Fallout/The Forest, sadly the boost wasn’t good enough to get to 1.25, but at 1.0 we have more FPS so maybe higher graphics can be one instead, there is scope for better FPS later with nvidia drivers and PiTool updates, and you only really need another 15% on top of this which may be reachable on this.