Lost Planet 2 DX11 - noticeable tessellation

[evolucion8]

are you blind? the 2.8 i5 is matching and the 2.8 i7 is beating the 3.2 X6. not to mention the i5/i7 quads have a shit ton more oc headroom than that 3.2 X6.

Are you nuts? I see the Phenom II X6 1090T is having 20fps min and 65.1 max, the i7 860 is having 17fps min, but higher maximum of 70.8, you call that a beating?

Wreckage, like I said before, the GTX 460 1GB may only match the HD 5870 in Tessellation, but in general gameplay, its nonear of it, stop with the fud. Look at the HD 5870 beating even the GTX 480 in DX9 mode, or the fact that in most games like Battlefield Bad Company is in a spitting distance of the GTX 480, Tessellation doesn't tell you the whole story and by the time that Tessellation is widely used, neither the GTX 460 1GB or the HD 5870 will have the power to use it properly.

Doesn't worth to beat the dead horse again, plus none of the cards there was able to provide of decent playability, so doesn't matter if nVidia has 8X times more power in Tessellation than competition like they claim, I don't see such advantage here and neither makes a difference in playability, we need faster GPU's than we have currently to make Tessellation usable. So give up your fanboy antics, doesn't make you look well (Specially at your age )

 

[Keysplayr]

Are you nuts? I see the Phenom II X6 1090T is having 20fps min and 65.1 max, the i7 860 is having 17fps min, but higher maximum of 70.8, you call that a beating?

Wreckage, like I said before, the GTX 460 1GB may only match the HD 5870 in Tessellation, but in general gameplay, its nonear of it, stop with the fud. Look at the HD 5870 beating even the GTX 480 in DX9 mode, or the fact that in most games like Battlefield Bad Company is in a spitting distance of the GTX 480, Tessellation doesn't tell you the whole story and by the time that Tessellation is widely used, neither the GTX 460 1GB or the HD 5870 will have the power to use it properly.

Doesn't worth to beat the dead horse again, plus none of the cards there was able to provide of decent playability, so doesn't matter if nVidia has 8X times more power in Tessellation than competition like they claim, I don't see such advantage here and neither makes a difference in playability, we need faster GPU's than we have currently to make Tessellation usable. So give up your fanboy antics, doesn't make you look well (Specially at your age )

It sure is playable, on both ATI and NV. Why do you insist on driving home a point repeatedly that just isn't true?
With average framerates anywhere from 35-75 depending on the card, and occasional dips under 20, how is this not playable/usable? I also don't get why you're moving toward the insults again. Can't you refrain from this behavior?

EDIT: After running the bench again, That dip below 30 only comes for about 1.5 seconds at the very tail end of Test B when the creature dives into the River. The rest of the test averages well over 45 fps at the highest settings at 1680x1050 MSAA4x.

How is this not playable? This is on a single GTX480 at stock speeds, with a mildly overclocked i7 860 @3.4 GHz.

Here are my scores from the other thread:

1680x1050, Full Screen, Refresh: 59.88, VSync OFF, MSAA4x, Motion Blur: ON, Shadow Detail: High, Texture Detail: High, Rendering Level: High, DX11 Features: High.

Test A:
Lowest fps observed: 45fps
Scene1: 58.5, Scene2: 54.1, Scene3: 67.4, Overall Average: 58.1fps RANK "B"

Test B:
Lowest fps observed: 31fps, Scene1: 49.2, Overall Average: 47.2fps RANK "B"

Single GTX480 @ stock. 258.96 drivers
i7 860 @ 3.4GHz
Win7 64.
8GB DDR3

 

[toyota]

Are you nuts? I see the Phenom II X6 1090T is having 20fps min and 65.1 max, the i7 860 is having 17fps min, but higher maximum of 70.8, you call that a beating?

)

that min can be just one single place on the becnhmark for all we know. the POINT was the game may scale to 6 cores for AMD but clearly the slower clocked i5/i7 quads can match or beat even the fastest X6.

 

[busydude]

Test B

Average FPS: 23.1
Dx11: High
AA: None
Quality Settings: Max
Resolution: 1680 X 1050
CPU: Phenom II X2 555
GPU: ATI 5750

 

[Xarick]

I would like to see benchs at 1920x1080. Cause looking at those numbers a 480 may be the only truly playable option with dx11 high at this resolution.

 

[edplayer]

I ran mine at 1920x1080


Test A: Sc1 - 36.8, Sc2 - 31.5, Sc3 - 40
Rank B, 36fps


Test B: Rank B, 31.4fps, Sc1 - 32.7


Q9550 at 3400Mhz, 4GB, GTX 460/768MB at stock. DX11, no AA, everything else at high or max

 

[evolucion8]

I also don't get why you're moving toward the insults again. Can't you refrain from this behavior?

EDIT: After running the bench again, That dip below 30 only comes for about 1.5 seconds at the very tail end of Test B when the creature dives into the River. The rest of the test averages well over 45 fps at the highest settings at 1680x1050 MSAA4x.

I didn't insult anyone really, toyota called me blind, and I called I called him nuts. Is easy to accuse someone without seeing both posts. What I can't really withstand is Wreckage's constant trolling and nvidia propaganda derailing the thread and even worse that he's getting away with it. I don't understand his obsession with nvidia. Who cares if the GTX 460 runs the same as the HD 5870 in heavy tessellation? Both runs incredibly slow in the benchmark results shown by Weakage's link anyways. In the end, the HD 5870 is the faster card, but the GTX 460 is the best value in its price point. Doesn't make any sense recommending a videocard based on one single scenario only. Now I can say that the HD 4890 is the better card because it outperformed the GTX 460 by a considerably margin in Lost Planet 2 in DX9 mode, please.....

that min can be just one single place on the becnhmark for all we know. the POINT was the game may scale to 6 cores for AMD but clearly the slower clocked i5/i7 quads can match or beat even the fastest X6.

Whatever, the results are there and proves you wrong, you are just speculating with the min fps results which showed that the Phenom II X6 1090T had higher minimum fps compared to the i7 870, even the slower 1055 ran 1fps higher at the minimum fps department. We all know that the i7 processors are better overall, but like Anandtech's review of the Phenom II X6 proofed that in very intensive multi threading scenario, the gap closes so much that in many tests, the X6 rivaled far more expensive processors from Intel.

Core i7 860
Min: 17
Max: 70.8

Phenom II X6 1090T
Min: 20
Max: 65.1

COre i5 760
Min: 14
Max: 64.6

Phenom OII X6 1055T
Min: 18
Max: 59.8

Phenom II X4 965
Min: 15
Max: 56.0

 

[Keysplayr]

I didn't insult anyone really, toyota called me blind, and I called I called him nuts. I can't withstand is Wreckage's constant trolling and nvidia propaganda derailing the thread and even worse that his getting away with it. I don't understand his obsession with nvidia. Who cares if the GTX 460 runs the same as the HD 5870 in heavy tessellation? Both runs incredibly slow in the benchmark results shown by Weakage's link anyways. In the end, the HD 5870 is the faster card, but the GTX 460 is the best value in its price point.

Well, you actually called him something else, then edited it. You do understand that I see you insisting that this game isn't playable as a bit of propaganda as well? Perfect opportunity. Big tesselation game is out, Nvidia can do it better due to their better tesselation performance, but it means nothing because of a low minimum framerate that only lasts a second. You're capitalizing on this, or trying to. It has been explained to you by a couple of people already that the low framerate is just a dip, not a constant by ANY means. So the bench is correct that the framerate can get that low, but it is also flawed as they represent the minimum framerate of only a few frames in the whole bench. So, Wreckage likes Nvidia, and you like AMD/ATI. So what? There isn't anything wrong with that. You guys should just accept that and not try to BS one another here. I mean, you're even battling each other with your sigs. Wreckage goes over the top, and so did you here. Nothing we haven't seen before from many including myself. We all have moments.

And one other thing. You are comparing an AMD six core 3.2GHz (3.6 Turbo) to a 4 core i7 860 2.8GHz(3.4 Turbo). And also a six core 2.8GHz AMD to a i5 750 4 core at 2.66GHz. Fair much?
Why don't we throw the 980 six core in there. You're saying the AMD solution is better like it matters, because it shouldn't if you maintain that the min framerate renders all CPU's unplayable like you did with the video cards. You are being inconsistent in that respect.

 

[evolucion8]

Please, don't compare my behavior to Wreckage's. I'm neutral and I don't have any issues recommending nvidia, you can see my posts across the forums recommending nvidia with no issues. Believe me, toyota is a person with an angry character (like me) if I called him a name, in less than a second he would had report me or insult me and my family

About the fps, How can you know if the min fps lasted only a second? You should do some benchmarks yourself, otherwise, we have what we have and that's it. But in the end, current GPU's will run the game just fine with tessellation, specially with multi GPU configurations at least on DX9. Could someone do the DX11 tests with a multi GPU setup?

Yeah, we have moments, because we are humans. (For now )

 

[Keysplayr]

Please, don't compare my behavior to Wreckage's. I'm neutral and I don't have any issues recommending nvidia, you can see my posts across the forums recommending nvidia with no issues. Believe me, toyota is a person with an angry character (like me) if I called him a name, in less than a second he would had report me or insult me and my family

About the fps, How can you know if the min fps lasted only a second? You should do some benchmarks yourself, otherwise, we have what we have and that's it. But in the end, current GPU's will run the game just fine with tessellation, specially with multi GPU configurations at least on DX9. Could someone do the DX11 tests with a multi GPU setup?

Yeah, we have moments, because we are humans. (For now )

How can I know the min fps lasted that long? There is a beautiful line graph at the termination of the benchmarks, both A and B. The dip is at the very end for 1 or two seconds when the creature dives into the river. That's it. The rest of the bench is perfectly well above the playable threshold and then some. Don't just look at the numbers, look at the line graph. I'll get you a screenshot of my runs tomorrow morning (not running them again tonight, going to sleep now). But take a look yourself.


EDIT/UPDATE: First shot is a direct screenshot from the bench. Second is compiled in Excel from a Fraps FPS recording.

 

[Wreckage]

And as far as comparing your behavior to Wreckage's? Well my friend, just look at your sigs. Peas in a pod. Only two on the forums who are doing this (that I know of).

Don't compare me to Evo just because he copied me. I'm neutral compared to him. (I have many an AMD CPU yah know). Plus I don't hurl about the insults like he does. I also don't have his apparent anger issues.

Either way this is off topic, I just felt the need to defend my honor since others are sort of talking behind my back.

Keep me out of your little spat.

 

[evolucion8]

Don't compare me to Evo just because he copied me. I'm neutral compared to him. (I have many an AMD CPU yah know). Plus I don't hurl about the insults like he does. I also don't have his apparent anger issues.

Either way this is off topic, I just felt the need to defend my honor since others are sort of talking behind my back.

Keep me out of your little spat.

Honor? MWHAHAHAHAHahaha. I can't see the honor in a person that begs for attention from nVidia. I would prefer to kill myself rather than copy you, that's for sure. So please stay on topic or stay away from it.

 

[3DVagabond]

You are comparing an AMD six core 3.2GHz (3.6 Turbo) to a 4 core i7 860 2.8GHz(3.4 Turbo). And also a six core 2.8GHz AMD to a i5 750 4 core at 2.66GHz. Fair much?
Why don't we throw the 980 six core in there.

You don't think it's fair to compare a 4c/8t chip with a 6c/6t chip? That's the whole point between the 2 different designs. "What's better, cores or threads?" Intel says threads and AMD says cores.

 

[Keysplayr]

You don't think it's fair to compare a 4c/8t chip with a 6c/6t chip? That's the whole point between the 2 different designs. "What's better, cores or threads?" Intel says threads and AMD says cores.

Does this look like LP2 is utilizing the i7's 4 logical cores?

Now all we need is a Task Manager screenie of someone with an AMD x6. And while we're at it, a Gulftown TM screenie.

 

[3DVagabond]

How does that make it an unfair comparison? If cores works better than threads, then Intel's philosophy is wrong, and AMD's is right. Sounds like you want to discount the threads all together. You aren't trying to stack the deck to much in Intel's favor, are you?

 

[Keysplayr]

How does that make it an unfair comparison? If cores works better than threads, then Intel's philosophy is wrong, and AMD's is right. Sounds like you want to discount the threads all together. You aren't trying to stack the deck to much in Intel's favor, are you?

The comparison is unfair, because this game we are talking about here does not use the logical cores (Hyperthreaded logical cores) of my i7. It is only using the actual 4 cores as shown by the TM grid. That said, as Toyota had mentioned, a 4 core i5/i7 is damn near matching and most often exceeding the performance of a six core AMD CPU. There is no "deck stacking" when the data is in your face like this. I am not discounting threads, Lost Planet 2 is. It apparently does not recognize logical cores, which obviously means it's making due with the 4 actual cores. And as far as Intels' philosophy being wrong? What's Gulftown? Isn't that a six actual core CPU? HT is the bonus when the app uses them.

Stack the deck...... Oh brother. I laughed at that one.

 

[3DVagabond]

The game not wanting to use the logical threads of the processor is a good argument for "real" cores. Everyone knows clock for clock, core for core, Intel is currently faster. I'm still not seeing what's unfair about the comparison. Unless you want to bring price into the equation? And I sure don't see any reason for your last sentence, except to try and belittle.

 

[zebrax2]

The comparison is unfair, because this game we are talking about here does not use the logical cores (Hyperthreaded logical cores) of my i7. It is only using the actual 4 cores as shown by the TM grid. That said, as Toyota had mentioned, a 4 core i5/i7 is damn near matching and most often exceeding the performance of a six core AMD CPU. There is no "deck stacking" when the data is in your face like this. I am not discounting threads, Lost Planet 2 is. It apparently does not recognize logical cores, which obviously means it's making due with the 4 actual cores. And as far as Intels' philosophy being wrong? What's Gulftown? Isn't that a six actual core CPU? HT is the bonus when the app uses them.

Stack the deck...... Oh brother. I laughed at that one.

It could be that the game is already putting too much load to the 4 cores that it cannot process the threads anymore. You can see from your screen shot that when there is a spike of load to the threads there is also a drop of load in the cores.

As for the i5/i7 vs X6 discussion. We already know that Intel CPUs can handle more IPC than their AMD counter parts. What is important is how those CPUs stack up against each other on a price vs performance ratio rather just who is the faster. As i see it they are about equal in this game

 

[Keysplayr]

@ 3DVagabond:
If you cannot see the unfair comparison, that is your choice. Nothing I can do about it. And you know very very very well the reason for my last sentence. It was in response to your "stacking the deck" comment. Why pretend you didn't know that?

@Zebrax2:
It would be useful to have a Task Manager screenshot of someones AMD x6 while running Lost Planet 2. If we see only 4 of the 6 cores loaded, then I would agree with your theory. But if all 6 are loaded, the story is different.

 

[AtenRa]

In HT (Hyper Threading)CPUs like the Core i7 920, there are Logical Cores (four of them) and each Core can handle 2 Threads, but the OS(Operating System) sees the total number of Threads that the CPU can handle as 8 but we only have 4 Logical Cores, Phenom II X6 has 6 Logical Cores and the OS can handle maximum 6 Threads.

What we see from the task manager, is that Lost Planet 2 is not optimized for HT and that’s why only four(4) Threads are being utilized. I’m sure Phenom II X6 will utilize all 6 Threads.

 

[Petey!]

Fermi only performs better by price. If you take the overall performance without comparing price and it's far worse.

Are you seriously saying that if you don't compare price.. the SINGLE most, arguably, important number attached to a Video Card, then the 460 doesn't match up to a 5870? Are you kidding me?

This is probably the most retarded statement I've read anywhere on the net. Oh no, that $200 card is slower then that $400 card. Really??

Performance comes down SQUARELY in regards to how it's priced. Who would want to buy a 460 if it was priced the same as a 5870? Nobody.

The fact that a card can beat a card double its price in DX11 it's pretty significant, especially when both are part of the current line-up offered right now.

 

[Scali]

It could be that the game is already putting too much load to the 4 cores that it cannot process the threads anymore. You can see from your screen shot that when there is a spike of load to the threads there is also a drop of load in the cores.

You cannot tell that from Task Manager.
If code is running on a core, it will show load (usually 100%). Windows doesn't know anything about any kind of internal load between the two logical cores in a physical core.
Even if one logical core were to claim all resources (which is pretty much impossible given the way Intel allocates the resources), it would still show the second, stalled, logical core as running 100%.

0% load means there's no thread running.

 

[Madcatatlas]

Are you seriously saying that if you don't compare price.. the SINGLE most, arguably, important number attached to a Video Card, then the 460 doesn't match up to a 5870? Are you kidding me?

This is probably the most retarded statement I've read anywhere on the net. Oh no, that $200 card is slower then that $400 card. Really??

Performance comes down SQUARELY in regards to how it's priced. Who would want to buy a 460 if it was priced the same as a 5870? Nobody.

The fact that a card can beat a card double its price in DX11 it's pretty significant, especially when both are part of the current line-up offered right now.

Alot of people would buy the gtx460 if it was priced the same as the hd5870. Ask Wreckage

Consider also that according to a HardoCP review, the GTX460 in SLI beats the Radeon HD 5870 in CrossfireX.

disclaimer: im not sure if your also being sarcastic somewhere in your post. Its complicated

 

[Scali]

In HT (Hyper Threading)CPUs like the Core i7 920, there are Logical Cores (four of them) and each Core can handle 2 Threads, but the OS(Operating System) sees the total number of Threads that the CPU can handle as 8 but we only have 4 Logical Cores, Phenom II X6 has 6 Logical Cores and the OS can handle maximum 6 Threads.

No, a Core i7 920 has 8 logical cores (how often do we have to keep explaining this?).

What we see from the task manager, is that Lost Planet 2 is not optimized for HT and that’s why only four(4) Threads are being utilized. I’m sure Phenom II X6 will utilize all 6 Threads.

Nonsense... if it didn't have any HT-specific code, it would just see it as an 8-core processor, and run 8 threads.
If it only runs 4 threads, then it has SPECIFICALLY checked that it is:
1) A quadcore CPU
2) With HT

And concluded that it wants to run only 4 threads.

 

[Petey!]

Alot of people would buy the gtx460 if it was priced the same as the hd5870. Ask Wreckage

Consider also that according to a HardoCP review, the GTX460 in SLI beats the Radeon HD 5870 in CrossfireX.

disclaimer: im not sure if your also being sarcastic somewhere in your post. Its complicated

Anyone who would buy a 460 if it was 400$ would be a fool, obviously, but I suppose some people still might. And as far as SLI vs Crossfire, well, Fermis SLI scaling has been pretty amazing on 2 cards, and Crossfire has never really stacked up as far as efficiency goes.

I was more responding to the fact that he was trying to say that if price wasnt taken into consideration (when it's what sets all cards apart), then the 460 performance wasn't on par with the 5870.