WYP
News Guru
Between DX12 and Core Scaling, there is a lot to talk about here.
Read more about Shadow of the Tomb Raider's PC performance.
Read more about Shadow of the Tomb Raider's PC performance.
Last edited:
Shadow of the Tomb Raider PC Performance Review
- Thread starter WYP
- Start date
fadingfool
New member
Just use the 2080 FE chart - from what I've seen in all the reviews there's about a 1% difference between a 1080Ti and 2080 FE at 1440p 
NeverBackDown
AMD Enthusiast
Pretty impressive DX12 implementation. Nice review fella
kdawgmaster
New member
Can you guys seriously start including GTX 1080 ti's into your benchmarks. The more you dont the more you become an unreliable source more so considering new cards are out.
NeverBackDown
AMD Enthusiast
That's just a stupid statement. Facts are facts and results speak for themselves. Can't include everything. Besides that he doesn't have a 1080ti to test.
Anyway.. It's nigh on identitical to a 2080. Effectively just replace the name and boom you're done. Not that difficult.
kdawgmaster
New member
Your mearly assuming that for this game. There are some games where the 2080 does get a fair amount of performance more and vise versa. Right now since there are a lot of people looking at comparing the 1080 ti to the 2080 seeing which they should get right now and them not having it gives people a very good reason to not look at them for benchmarks. If he doesnt have one then he should have considered getting one long ago because its been atleast 6 months to a year that they havent had them.
NeverBackDown
AMD Enthusiast
Yeah you know he should have just bought one with his own money alongside all the other equipment he needs to test. What's another $800 right?
Nobody is comparing which to get. If you are buying new you get the 2080 it's not much more. Why would you buy old technology even if its new(not used) when it's already been replaced and cost almost the same? If you were looking used 1080ti then the 2080 is already out of the price for those people.
kdawgmaster
New member
Tons of people are comparing whether or not to get the 2080 or the 1080 ti because those are the most "reasonable" at this time and the 2080 hardly scratches that surface when you consider that its taking a 2080ti to run game demos with ray tracing at 1080p 60 fps and if thats the case then ray tracing as a tech is poinless for the 2080 and therefore shouldnt be considered in if you're going to buy it or not. The next thing is where I live the 2080 brand new is 1100 cad for the cheapest and the cheapest 1080 ti i can buy brand new is 870 cad so that right there is a huge chunk of change when considering as you said from what we can see they perform the same. Which basically leaves DLSS as the sole reason you would consider it and theres no guarantee that will kick off because it needs Devs to continue supporting it and as resolutions get higher the need for super sampling decreases.
As a review your job is to paint the CLEAREST possible picture and when you omit a card that people are still VERY much considering you arent helping consumers out and therefore become useless to consider for information.
I use to come to this site ALL the time because they had the largest amount of video cards tested HELL thats how i came to the conclusion of buying my hybrid 1080 ti, i use OC3D and another site that tested it, but now since they arent willing to test 1080 ti's what good are they to consumers looking at making a purchasing choice.
NeverBackDown
AMD Enthusiast
Have your opinion sure. He can't have everything on hand. He buys everything with his own money. It's not a big deal in DX11 and DX12 they aren't far off. Only in Vulkan is it such a vast difference between the two
His review is clear. You cannot test everything. Many people in fact most people buy lower end cards. How come you don't complain about that? They hardly ever get the spotlight. Or does it not matter because it doesn't fit your situation? Just cross reference benchmarks. You've already done that by looking at others. Why can't you just cross reference? You do that when buying things to. I'm sure you've looked at multiple sites and multiple different cards to see which you want to buy and then looked up reviews. The whole point is to get as much information as possible. Being picky about one thing is a waste of time especially for one game.
You already know in nearly every game they are very close and less than 5%. The 2080 is not worth the money and that's reflective of the various websites who reviewed it. They are just to expensive for much gain. And because of this Nvidias stock has slightly fallen to the mixed reaction this launch has received, even the all the analyst noticed how little of a difference they are. It's not that complicated honestly dude. Especially all over one game.
His review is clear. You cannot test everything. Many people in fact most people buy lower end cards. How come you don't complain about that? They hardly ever get the spotlight. Or does it not matter because it doesn't fit your situation? Just cross reference benchmarks. You've already done that by looking at others. Why can't you just cross reference? You do that when buying things to. I'm sure you've looked at multiple sites and multiple different cards to see which you want to buy and then looked up reviews. The whole point is to get as much information as possible. Being picky about one thing is a waste of time especially for one game.
You already know in nearly every game they are very close and less than 5%. The 2080 is not worth the money and that's reflective of the various websites who reviewed it. They are just to expensive for much gain. And because of this Nvidias stock has slightly fallen to the mixed reaction this launch has received, even the all the analyst noticed how little of a difference they are. It's not that complicated honestly dude. Especially all over one game.
Zoot
Active member
This is a very good and informative piece, the best I've read regarding this game - Thanks for all your work.
I just got this and downloaded it overnight since I enjoyed the last two (particuarly the 2013 game). I ran the benchmark before heading to work this morning, and I got averages in the 70's at 1080p, the minimum being about 64 if I remember correctly. This is with a 6700k and a 1070.
I just got this and downloaded it overnight since I enjoyed the last two (particuarly the 2013 game). I ran the benchmark before heading to work this morning, and I got averages in the 70's at 1080p, the minimum being about 64 if I remember correctly. This is with a 6700k and a 1070.
WYP
News Guru
To address the "where is the GTX 1080 Ti" question. I can only test what I have in the office. I try my best to have a good range of hardware to test in these pieces.
In this case, our RTX 2080/RTX 2080 Ti results come as the result of a collaboration between myself and Tom, who both work from different locations (roughly 330 miles apart as the crow flies).
I don't have a GTX 1080 Ti graphics card here, and even if I did there is Denuvo to contend with, which has a habit of locking testers out of games (albeit temporarily, usually 24 hours) if hardware configurations are changed too often. This limits the amount of hardware that can be put on our gaming test suite.
In this case, our RTX 2080/RTX 2080 Ti results come as the result of a collaboration between myself and Tom, who both work from different locations (roughly 330 miles apart as the crow flies).
I don't have a GTX 1080 Ti graphics card here, and even if I did there is Denuvo to contend with, which has a habit of locking testers out of games (albeit temporarily, usually 24 hours) if hardware configurations are changed too often. This limits the amount of hardware that can be put on our gaming test suite.
WYP
News Guru
Haha. You assume that I have the money to spare for a GTX 1080 Ti. Beyond that, as RTX 2080 would be a much better buy for me, as I'd love to go in-depth with DLSS testing.
Cmon, you're irish. You guys are soon to get rich after Brexit plummets and crashes the country into disarray.
(That was a joke btw, dont need the brexiteers jumping in for a debate)im actually keeping mine. Both have clocked very nicely together. With my prev gen, I couldnt keep the same clocks in SLI as I did for each individual card. These ones are handling this wonderfully.
I just looked forward to the SLI benchmarks or 2nd version of the RTX next year.
WYP
News Guru
Cmon, you're irish. You guys are soon to get rich after Brexit plummets and crashes the country into disarray.
im actually keeping mine. Both have clocked very nicely together. With my prev gen, I couldnt keep the same clocks in SLI as I did for each individual card. These ones are handling this wonderfully.
I just looked forward to the SLI benchmarks or 2nd version of the RTX next year.
It will be interesting when Nvidia jumps to 7nm.
Also, Northern Irish, which is not a great position Brexit-wise. I'm in that UK bit of Ireland that is usually forgotten about. No need to discuss Brexit any further here though, that is waaayyyyy off-topic.
AlienALX
Well-known member
No offence meant here but Tom really needs to send him a 1080Ti. It was the top end card from last gen and many want to know if it is worth spending up to £1200 so it's kinda important it's in there. Even more so that the 1080 IMO. I'm not having a go btw, I am just speaking constructively.
Wraith
Bettyswollocks
Given the general consensus surrounding all reviewers.
1080Ti (£650) Vs 2080 (£800)
Trading blows across the board, no brainer! buy a 1080Ti
1080Ti (£650) Vs 2080Ti (£1100)
2080 Ti gives roughly 25 - 30% Performance increase at a 65 - 70% higher price (Titan money) you'd need your head examined, given NVIDIA are claiming that the 2080Ti is now the new Titan the price makes sense.
RTX is exciting I'm not going to lie but it's in its infancy right now, for gaming it's a ways off yet, same goes for Blender as the Kernal is not ready to support it and APIs have to be updated too. As Jayz put it "you can't sell a product on a promise"
Zoot
Active member
I find it hard to be excited about RTX to be truthful, at least at the moment. I actually find it hard to get excited about new things in the PC space these days. Maybe I'm getting too cynical.
Each time a new technology is announced, it always takes a long time before it's common place. I remember back in the early days of DDR3, most people were just sticking with DDR2 since the difference was minimial between the two initially. A more recent example is DirectX 12, it hasn't been the earth shattering change we were initally made to believe it would be, at least not yet.
Anyway, here's my result at 1080p with a 6700k at stock and a 1070.
CPU thread scaling
" was able to utilise all sixteen threads within the processor, showcasing how parallelised Shadow of the Tomb Raider can be."
Uh, how EXACTLY did you determine this? I assume you are aware that code can JUMP between cores/threads so even a game like Starcraft 2 that's designed to run on only two threads can give the ILLUSION of running on all threads simultaneously but in reality segments of code start, finish then jump to another thread with the overall result appearing like it actually benefits from all cores.
The only difference is the MINIMUM drops to 61FPS from 66FPS for that CPU but minimums can be tricky and that's a very small difference for something that rarely happens.
Finally, the average FPS for that CPU at 4C/8T was 94% that of 8C/16T so doubling the core/thread count adds only 6%? Yes that's better but I think saying it utilizes 16 threads is stretching it a bit.
Probably the extra threads are helping mitigate drops from background applications more so than actually running game code. That doesn't mean it's not helping but again it's not by very much from 6 to 8 cores (minimum only).
Anyway, just to be clear if you run TASK MANAGER and see 16 threads utilized it does NOT mean that code is running on all sixteen threads AT THE SAME TIME and even if it did it doesn't mean it can benefit much over a lower core count as some branching threads of code may use very little of the CPU thread it's running on.
" was able to utilise all sixteen threads within the processor, showcasing how parallelised Shadow of the Tomb Raider can be."
Uh, how EXACTLY did you determine this? I assume you are aware that code can JUMP between cores/threads so even a game like Starcraft 2 that's designed to run on only two threads can give the ILLUSION of running on all threads simultaneously but in reality segments of code start, finish then jump to another thread with the overall result appearing like it actually benefits from all cores.
The only difference is the MINIMUM drops to 61FPS from 66FPS for that CPU but minimums can be tricky and that's a very small difference for something that rarely happens.
Finally, the average FPS for that CPU at 4C/8T was 94% that of 8C/16T so doubling the core/thread count adds only 6%? Yes that's better but I think saying it utilizes 16 threads is stretching it a bit.
Probably the extra threads are helping mitigate drops from background applications more so than actually running game code. That doesn't mean it's not helping but again it's not by very much from 6 to 8 cores (minimum only).
Anyway, just to be clear if you run TASK MANAGER and see 16 threads utilized it does NOT mean that code is running on all sixteen threads AT THE SAME TIME and even if it did it doesn't mean it can benefit much over a lower core count as some branching threads of code may use very little of the CPU thread it's running on.
Similar threads
