Support NeoGAF

[Always-honest]

For future reference:

TressFX on the PC version

I think they'll cut the strand count down a bit to improve performance. The problem with TressFX's appearance is also the way it doesn't drape over the surface of the character model that the textures are on. Not sure what's going on there.

is this an actual screenshot from the game? Cause then there are def. less and thicker strands on the console version.

 

[SmokyDave]

So, just re-downloaded Tomb Raider on Steam, no mods, just Vanilla. Cranked it up to Ultra and ran the benchmark @1080p.


i7 3770, 16GB Ram, 6950 2GB unlocked/OCed to 6970 spec. (which is approx 2.7TF compared to the PS4's 1.8)

and I get these results. Does it look fantastic? Yes. Mostly because of IQ.


To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

Man, that is interesting. Really puts the PS4 performance into perspective. Thanks for that.

 

[MaLDo]

So, just re-downloaded Tomb Raider on Steam, no mods, just Vanilla. Cranked it up to Ultra and ran the benchmark @1080p.


i7 3770, 16GB Ram, 6950 2GB unlocked/OCed to 6970 spec. (which is approx 2.7TF compared to the PS4's 1.8)

and I get these results. Does it look fantastic? Yes. Mostly because of IQ.


To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

You can test yourself again setting resolution to 1080p, AO to medium, aa to FXAA, disabling tessellation, shadows to high, and enabling multithreaded rendering

1. Open regedit (win key + R, type regedit) and navigate to HKEY_CURRENT_USER\Software\Crystal Dynamics\Tomb Raider\Graphics
2. Right Click anywhere in the empty space and select New > DWORD (32-bit) Value
3. Name the new value MultiCoreFlushMode then double click on it and change the value data to 1.

 

[Tagyhag]

is this an actual screenshot from the game? Cause then there are def. less and thicker strands on the console version.

That's because TRESSFX 2.0 is less of a resource hog, but not as impressive.

What will also matter though, is the actual animations.

 

[Van Owen]

So, just re-downloaded Tomb Raider on Steam, no mods, just Vanilla. Cranked it up to Ultra and ran the benchmark @1080p.


i7 3770, 16GB Ram, 6950 2GB unlocked/OCed to 6970 spec. (which is approx 2.7TF compared to the PS4's 1.8)

and I get these results. Does it look fantastic? Yes. Mostly because of IQ.


To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

lol, try changing the AA to fxaa and you'll likely get higher than 30fps average.

 

[Caayn]

So, just re-downloaded Tomb Raider on Steam, no mods, just Vanilla. Cranked it up to Ultra and ran the benchmark @1080p.


i7 3770, 16GB Ram, 6950 2GB unlocked/OCed to 6970 spec. (which is approx 2.7TF compared to the PS4's 1.8)

and I get these results. Does it look fantastic? Yes. Mostly because of IQ.


To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

Is that with tessellation and TressFX turned on? Disable tessellation and set the AA to the level of the definitive edition if that uses any AA at all.

Please post printscreens next time instead of taking a photo of your screen.

Edit: Is there a separate benchmark tool for Tomb Raider on the pc? Would like to see how my 6870 holds up.

 

[Ysiadmihi]

So, just re-downloaded Tomb Raider on Steam, no mods, just Vanilla. Cranked it up to Ultra and ran the benchmark @1080p.


i7 3770, 16GB Ram, 6950 2GB unlocked/OCed to 6970 spec. (which is approx 2.7TF compared to the PS4's 1.8)

http://img.photobucket.com/albums/v...B-EB86-4D09-B29A-5C0CD9F578C6_zpsnfduvbqm.jpg

http://img.photobucket.com/albums/v...0-2FE5-4348-A013-3CCE70942374_zpstla1mzof.jpg

and I get these results. Does it look fantastic? Yes. Mostly because of IQ.


To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

Um...are you using SSAA? I wouldn't do that.

I get 55fps in the benchmark with everything maxed and FXAA at 1080p with my 7970.

 

[SmokyDave]

Um...are you using SSAA? I wouldn't do that.

I get 55fps in the benchmark with everything maxed and FXAA at 1080p with my 7970.

Isn't that a £275 card?

What kind of hardware do you need for 1080p60?!

 

[UnrealEck]

So, just re-downloaded Tomb Raider on Steam, no mods, just Vanilla. Cranked it up to Ultra and ran the benchmark @1080p.

i7 3770, 16GB Ram, 6950 2GB unlocked/OCed to 6970 spec. (which is approx 2.7TF compared to the PS4's 1.8)

and I get these results. Does it look fantastic? Yes. Mostly because of IQ.

To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

Bit of a pointless exercise because PS4 doesn't run at Ultra. It runs at some new preset Crystal Dynamics have decided over to result in best frame rate to graphical quality ratio. Plus the TressFX is apparently different. The TressFX the Definitive Edition uses may have less strands and/or less of the various physics calculations to offer more performance.
Then there's things like Tesselation which don't seem to exist in the PS4 version. We don't even know what AA is used.

Without PS4 screencaps you can't accurately compare ultra performance to PS4 performance yet because we don't know what the graphics settings the game's running at will be on PS4. Judging by screenshots alone and footage, I'd say it looks better in some minor areas (skin for example and even then not by a lot) and in others it seems worse.

 

[BigTnaples]

Okay, so I followed your instructions Maldo.

Modded the Registry files, disabled tessellation, lowered the shadow detail, and AO, and ran the benchmark again.

I get 35fps.




(Also Keep in mind the benchmark is just Lara, standing on a cliff, no enemies, no particles, no explosions, etc)So, given that, I am running a vastly more powerful machine, and that the PS4 version will have enhanced physics, more particles, subsurface scattering, better lighting, LoD, touched up decals and mud/blood/water textures, as well as other enhancements, I think that a 5FPS difference on a $400, inferior machine, is a job well done.

 

[delta25]

To me, the Definitive Edition looks much better in almost every way aside from IQ, and runs at superior framerate, and on vastly inferior hardware.

This is quite interesting when you compare your PC alone to the PS4' hardware, it really puts things into perspective. By default your PC spanks the shit out of the PS4, yet the PS4 version runs at 1080p, a stable 30fps with reworked/new in-game assets, I guess good optimization can go along way.

 

[plagiarize]

I'm just chiming in here as a 7970 owner who beat the game on PC... The benchmark does not do a good job of representing how the game actually plays. Very regularly you will see much lower performance than you get in the benchmark. The benchmark is pretty useless in telling us how the game actually plays on a given machine.

 

[Mr. Fix]

This thread will be interesting if they finally announce a ps4 port of TLoU. I agree with the backlash though.

 

[Dahaka]

I'm just chiming in here as a 7970 owner who beat the game on PC... The benchmark does not do a good job of representing how the game actually plays. Very regularly you will see much lower performance than you get in the benchmark. The benchmark is pretty useless in telling us how the game actually plays on a given machine.

+1

That's what I am saying all the time. The 60fps claim is a huge lie by most that drive by in this thread and laugh at this release or the consoles in general. Let them be honest and see how shaky that claim is when the frames fall before their eyes.

I agree on the negative sentiment about the price though.

 

[delta25]

I'm just chiming in here as a 7970 owner who beat the game on PC... The benchmark does not do a good job of representing how the game actually plays. Very regularly you will see much lower performance than you get in the benchmark. The benchmark is pretty useless in telling us how the game actually plays on a given machine.

I think it has to do with the fact that his 6970 and your 7970 are vastly superior to what the PS4 has inside of it, yet after everything is said and done the PS4 still manages to match these high end rigs while still providing 1080p and a stable 30fps with re-worked/new in-game assets.

 

[Always-honest]

That's because TRESSFX 2.0 is less of a resource hog, but not as impressive.

What will also matter though, is the actual animations.

Ah right. thanx

Is it an actual screenshot though?

 

[C00kie M0nster]

Good news is you can still sell your Ps4/Xbone for full price & put the money towards a real gaming machine... a PC.
60fps last gen ports guaranteed on it!

I already have a great gaming PC along with Tomb Raider on it.

 

[Ysiadmihi]

That's what I am saying all the time. The 60fps claim is a huge lie by most that drive by in this thread and laugh at this release or the consoles in general. Let them be honest and see how shaky that claim is when the frames fall before their eyes.

What 60fps claim are you talking about? I assume you're talking about PC gamers, and I don't think I've seen one in here claim to have 60fps with TressFX enabled.

 

[UnrealEck]

http://www.techspot.com/review/645-tomb-raider-performance/page3.html

44 FPS 1920x1200 (higher than 1080p) at ultra settings (this includes tesselation) on a 6970.

 

[CoG]

How do we know the consoles will run 30fps solid as on the benchmark? I get 45fps average on my 4770K / GTX 760 setup but 60fps max (Ultra settings). The console version may dip in the teen on the benchmark for all we know.

 

[Van Owen]

I'm just chiming in here as a 7970 owner who beat the game on PC... The benchmark does not do a good job of representing how the game actually plays. Very regularly you will see much lower performance than you get in the benchmark. The benchmark is pretty useless in telling us how the game actually plays on a given machine.

Yep, got much better framerates during gameplay than the actual benchmark. Only thing that annoys me is the fps hit when a cutscene zooms in on Lara's face and hair.

 

[nubbe]

Prices has started to reach €39 for the game even before launch...
I think i'll nab it

 

[Alastor3]

so does the PC have the new Lara model or it's only ps4/Xbone ?

 

[CadetMahoney]

How do we know the consoles will run 30fps solid. The console version may dip in the teens on the benchmark for all we know.

Weird how my 3 year old pc runs this game at 60.

I blame bone for the lack of 60 on PS4.

 

[nkarafo]

Wait, are you telling me that a PS3/360 game with a few graphical improvements here and there at an increased resolution can't run at 60fps on the mighty PS4?

Exactly how stronger are these new machines anyway? I still can't say.

 

[TheZjman]

We just got this video from S-E, showing the new effects in the next-gen version of the game. I'm sure it's been posted before, but the quality here is quite good
http://www.gamersyde.com/news_tomb_raider_next_gen_explanations-14964_en.html

Welp, I'm in. That looks beautiful. Didn't get to play it last year either.

 

[BigTnaples]

Wait, are you telling me that a PS3/360 game with a few graphical improvements here and there at an increased resolution can't run at 60fps on the mighty PS4?

Exactly how stronger are these new machines anyway? I still can't say.

It can't run at 60 on 90% of gaming rigs out there either.


Such a silly way to judge the consoles power.

 

[UnrealEck]

PS4 version will have enhanced physics, more particles, subsurface scattering, better lighting, LoD, touched up decals and mud/blood/water textures, as well as other enhancements, I think that a 5FPS difference on a $400, inferior machine, is a job well done.

Enhanced physics being that her radio, pistol arrows and axe swing a little on her belt.
More particles? I've not seen them. What I have seen particle wise in one video was snow blowing past on PC which was abscent on PS4.
I've yet to see better lighting too. Different maybe, but haven't seen better.
LoD? Level of detail? Haven't seen it demonstrated.
Touched up decals, mud, blood and water textures? I've seen better blood and stuff on her skin but that's it.
None of the above is going to make much difference at all to performance.
What are the "other enhancements"?

 

[UnrealEck]

I'm just chiming in here as a 7970 owner who beat the game on PC... The benchmark does not do a good job of representing how the game actually plays. Very regularly you will see much lower performance than you get in the benchmark. The benchmark is pretty useless in telling us how the game actually plays on a given machine.

Actually the minimum frame rate I get in the benchmark is very similar to what I get when playing.

 

[nkarafo]

It can't run at 60 on 90% of gaming rigs out there either.


Such a silly way to judge the consoles power.

I'm sorry, i got used to the previous generation jumps. You know, the time when each generation had a huge gap. The PS3 didn't have any problem running a late, taxing PS2 game at 60fps and HD graphics (Shadow of the Colossus).

Edit: wrong game

 

[RoboPlato]

I'm sorry, i got used to the previous generation jumps. You know, the time when each generation had a huge gap. The PS3 didn't have any problem running a late, taxing PS2 game at 60fps and HD graphics (Shadow of the Colossus).

SotC is 30fps on PS3

 

[Moofers]

I played halfway through the game on PS3 last year. I'm excited to get this on PS4 next week and start it over again. I think what I've seen in the videos and in the screens in this thread all look pretty great! And I appreciate the attempts at comparing benchmarks. Thanks for that effort, guys!

 

[nkarafo]

SotC is 30fps on PS3

My memory is fuzzy yes. Maybe i mixed it up with God of War collection.

 

[BigTnaples]

I'm sorry, i got used to the previous generation jumps. You know, the time when each generation had a huge gap. The PS3 didn't have any problem running a late, taxing PS2 game at 60fps and HD graphics (Shadow of the Colossus).

You're kidding yourself.

 

[Finaika]

SotC is 30fps on PS3

It was like 10fps on the original PS2, so it was a huge upgrade.

 

[nubbe]

PS2 games are quite primitive compared to PS360 games

 

[RoboPlato]

My memory is fuzzy yes. Maybe i mixed it up with God of War collection.

GoW collection is 60 on PS3 but it was also mostly 60 on PS2 as well. It's just steady on PS3.

 

[plagiarize]

I think it has to do with the fact that his 6970 and your 7970 are vastly superior to what the PS4 has inside of it, yet after everything is said and done the PS4 still manages to match these high end rigs while still providing 1080p and a stable 30fps with re-worked/new in-game assets.

I am planning on buying and enjoying this. I played the game on PC with Tress FX and spent a lot of time well below 60 fps and very much enjoyed it. If this is a stable 30 like AC4, I personally expect to find that a preferrable experience, at least until I have G-Sync at home and stuttering is a thing of the past.

I just wanted people to know that in game performance regularly drops well below the benchmark performance levels.

 

[Figboy79]

The only problem I have with this game is the price. I'd jump at $40, but at $60, I just can't shell out that money for a game I've already 100% completed.

I'll be waiting for a bit, and snag it when it drops in price on Amazon, which I'm sure it will do soon.

The framerate seems fine as long as it's locked at 30.

 

[Coolbgdog12]

Some of you guys need to bring your computer to Geek squad or something because I'm not understanding your performances.

Heres my performance benchmark at enthusiast settings.
My specs:
AMD FX 4100 OC'd to 4.2GHz
2GB Radeon HD 7870 oc'd to 1103mhz core/1210mhz memory
8GB RAM

This is hardly a top of the line rig.

Settings 1440p

FPS at 1440P.

Now If I want this I can play the game at these settings fine but I can get this close to 60fps if I cut down to 1080P and turn a few settings down to high or off such as Depth of field and level of detail. Turning off tressfx will net me 60fps.

I'll be renting Tomb Raider for PS4 just to get a glimpse of it and earn some trophies while I'm at it. If you haven't had the chance to play it on last gen consoles and you don't have a capable PC go ahead this is a great game. I'm still not too happy about it being $60 though.

 

[TyrantII]

Okay, so I followed your instructions Maldo.

Modded the Registry files, disabled tessellation, lowered the shadow detail, and AO, and ran the benchmark again.





I get 35fps.




(Also Keep in mind the benchmark is just Lara, standing on a cliff, no enemies, no particles, no explosions, etc)So, given that, I am running a vastly more powerful machine, and that the PS4 version will have enhanced physics, more particles, subsurface scattering, better lighting, LoD, touched up decals and mud/blood/water textures, as well as other enhancements, I think that a 5FPS difference on a $400, inferior machine, is a job well done.

Don't worry.

This is the thread where the GAF master race gets all salty over a console port.

Anyone mention the $5 steam sale yet?



What impresses me the most is this is a quick and dirty port to get some reps in for their next true next gen project.

As many have said, those claiming the consoles are just low end PC's are missing the forest for the trees.

 

[BruiserBear]

I think it has to do with the fact that his 6970 and your 7970 are vastly superior to what the PS4 has inside of it, yet after everything is said and done the PS4 still manages to match these high end rigs while still providing 1080p and a stable 30fps with re-worked/new in-game assets.

Yes, because PC's are inherently less efficient when it comes to games. The developer has to try and build a game that will run on all these different configurations, parts, etc. It means the game is far less optimized than it would be on a console.

So a PS4 at 1.8trflps is the equivalent to a much more powerful gaming PC.

 

[nubbe]

i7-3930k
Geforce 780
1080p

 

[Shancake]

I played a preview build today and unless they've changed it for final release, it was definitely 60FPS.

 

[BruiserBear]

I played a preview build today and unless they've changed it for final release, it was definitely 60FPS.

Which version did you play? Both?

 

[Finaika]

I played a preview build today and unless they've changed it for final release, it was definitely 60FPS.

Are you serious?

 

[mephixto]

Seems like the drivers have improved a lot, didn't touch the game since launch.

i7 3930k @ 4.5Ghz
2x 680 SLI
16GB RAM
Win 8.1

Ultra 4xSSAA - Tress FX on

ULTRA FXAA - Tress FX on

 

[ZeroAKA]

I played a preview build today and unless they've changed it for final release, it was definitely 60FPS.

Really? If someone else can confirm, I'll place a preorder ASAP.

 

[SniperHunter]

The game is very unoptimized on PC (or was when I played it). I beat it using a 7970 and the FPS was all over the place. I don't know what settings I used but I did have the hair tech enabled. The frame rate got very very low during the shanty town ambush area. While the game ran at 60 FPS in linear areas. The hair tech really kills the performance big time

 

[Spazznid]

.

Tomb Raider looks plenty good on PC.