Crysis 2 Tessellation Article

Page 21 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
The area that bothered me, that MoSox did raise, from a gamer stand-point was nVidia not supporting nVidia GPU's for a discrete GPU PhysX product with another vendor and rendering card.


http://physxinfo.com/news/330/official-nvidia-position-on-hybrid-ati-nv-physx-configurations/

PhysXInfo: Can you please explain to us what factors have led to such decision?

Nvidia: Today NVIDIA’s GPU and PhysX drivers are interconnected to optimize performance. In the future we expect this interdependence to deepen. This alone makes it difficult to support a third party GPU.

In order to make sure our customers have a great experience, we QA every release of our PhysX or Graphics drivers by testing approximately 14 NVIDIA GPUs for graphics processing with 8 GPUs for PhysX processing on 6 common platforms with 6 OS’s using 6 combinations of CPU and memory. This is over 24000 possible configurations. While we don’t test every possible combination, it should be clear that the work and cost to NVIDIA is substantial. AMD does not support PhysX for their customers. Adding AMD GPUs would significantly increase the necessary work and cost for NVIDIA. We prefer to invest in inventing new technologies that give our customers great new experiences.

Hats off for Zogrim for getting the answer but I really wish nVidia would rethink this though considering it is still a nVidia GPU and maybe get more adoption for GPU Physics in titles. One can still disagree with nVidia and be constructive about it.
 

Seero

Golden Member
Nov 4, 2009
1,456
0
0
The area that bothered me, that MoSox did raise, from a gamer stand-point was nVidia not supporting nVidia GPU's for a discrete GPU PhysX product with another vendor and rendering card.


http://physxinfo.com/news/330/official-nvidia-position-on-hybrid-ati-nv-physx-configurations/



Hats off for Zogrim for getting the answer but I really wish nVidia would rethink this though considering it is still a nVidia GPU and maybe get more adoption for GPU Physics in titles. One can still disagree with nVidia and be constructive about it.
Love to talk about that, but this really has nothing to do with this topic.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
The road barriers don't bother me because there is subtle detail and really enjoy the angle it was first represented -- gotta love it. Even the bricks are not in your face tessellation and subtle, and I believe the game wasn't designed from the ground up for tessellation and simply added. Most of their attention was for the console market so I wasn't expecting ideal in all circumstances.

The only nit-pick I have is the road curbing so far.

http://maldotex.blogspot.com/2011/09/tesselation-myth-in-crysis-2-el-mito-de.html

This article actually did more to help me than any other. Because obviously nVidia sabotaged the GTX 470, hehe, because I can't receive the magical 40 frames with everything set to ultra. There are so many settings and configurations for a gamer to achieve nice frame-rate and a good experience.

I think the bricks are POM (might be wrong, but I recall reading that).

The game not being made with Dx11 isn't the issue we're discussing. Although, I agree it makes proper use of tessellation more difficult. Not impossible, but more difficult, and it's obvious doing it right was put in the "too hard pile".

As far as "sabotaging the 470". Yes, they did. They actually sabotaged every card out there. All of them perform worse than the have to if the patch was done properly. In the end though, nVidia still looks better in benchmarks. So, it's a win for nVidia even if it isn't for their customers.
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
Love to talk about that, but this really has nothing to do with this topic.

Agreed! The subject is tessellation and, to some degree, did nVidia with nefarious intent try to sabotage the title to look good in performance over their competitor? I don't understand how they would get away with it, if even possible. What developer would desire to work with an institution that would sabotage with nefarious intent? All it would take is one developer on the team to air their dirty laundry and come clean about it -- to expose it -- yet, nothing.

This would not just represent nVidia developer relations for gaming but for the entire company, their Cuda teams, relationships with Google, with tegra, everything about nVidia and their pro-active nature. Why would anyone work with them?
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Agreed! The subject is tessellation and, to some degree, did nVidia with nefarious intent try to sabotage the title to look good in performance over their competitor? I don't understand how they would get away with it, if even possible. What developer would desire to work with an institution that would sabotage with nefarious intent? All it would take is one developer on the team to air their dirty laundry and come clean about it -- to expose it -- yet, nothing.

This would not just represent nVidia developer relations for gaming but for the entire company, their Cuda teams, relationships with Google, with tegra, everything about nVidia and their pro-active nature. Why would anyone work with them?

People in business do it for the same reasons... money. I think you are stretching it a bit saying that this hurts nVidia's partner relations. If it did that, nVidia would have been in trouble long ago. All partners care about, for the most part, is if the relationship will make them money.
 

Seero

Golden Member
Nov 4, 2009
1,456
0
0
Agreed! The subject is tessellation and, to some degree, did nVidia with nefarious intent try to sabotage the title to look good in performance over their competitor? I don't understand how they would get away with it, if even possible. What developer would desire to work with an institution that would sabotage with nefarious intent? All it would take is one developer on the team to air their dirty laundry and come clean about it -- to expose it -- yet, nothing.

This would not just represent nVidia developer relations for gaming but for the entire company, their Cuda teams, relationships with Google, with tegra, everything about nVidia and their pro-active nature. Why would anyone work with them?

We may not be talking in the same channel but I am actually interested on the 2xbandwidth when tessellation used from AMD. Where is it? Knowing that the hardware itself should handle tessellation better, it clearly is not the case. As MutantGith pointed out, crysis is known to kill performance for fun and optimization is not a priority for crysis 2, seeing everything under perform is not to my surprise. Without the bias and conspiracy, with 580 the minimium FPS is only 2. If conspiracy is in place, I'll say Nvidia shot their own feet more than to AMD.

I say there exist bugs within the code path which prevents AMD's true power to be unleashed. I am guessing but I believe not even half of those tessellation unit is working. In fact, I do not even think the Dx9 code path is correct as AMD's lineup really should not get hit like that if it were for bugs. There are problems, but the invisible ocean is not one of them. At best, it hurts performance in general, not only to AMD, so the conspiracy theory does not hold, IMO.
 

SirPauly

Diamond Member
Apr 28, 2009
5,187
1
0
People in business do it for the same reasons... money. I think you are stretching it a bit saying that this hurts nVidia's partner relations. If it did that, nVidia would have been in trouble long ago. All partners care about, for the most part, is if the relationship will make them money.

I don't think this hurts nVidia's business at all. This is rumor mongering, conjecture, innuendo and conspiracy -- more entertainment than reality to me. If this had any substance, there would be clear poof and data, engineers and developers coming clean.
 

ocre

Golden Member
Dec 26, 2008
1,594
7
81
Nvidia sabotaged me and many others when I couldn't use my Nvidia card for PhysX together with my 5000 series ATI card. Nvidia sabotages those who want to run SLI on AMD rigs. Nvidia hobbles CPU PhysX.

Nvidia dumped XFX when they dared to make AMD cards.

Nvidia cuts off the HW review sites when they dare to breach their review guidelines. Nvidia (not Ubisoft) demanded that the HAWX 2 benchmark is used in the reviews.

Nvidia hates 3DMark (makes them look bad) so many HW sites including this one dumped it. Anand uses only Unigine and "compute performance" (mainly Civilization V) for the reviewed video cards.

Nvidia hates video quality benchmarks like HQV (those also make them look bad) so sites like Anand, Techpowerup, etc dumped that too even for the reviews of HTPC cards.


Here's from what I know the last video quality test done by Anand (only the anisotropic filtering)

http://www.anandtech.com/show/2977/...tx-470-6-months-late-was-it-worth-the-wait-/7


Here's an example of a review for a HTPC card not tested for image quality (HQV) but tested instead in...Metro 2033 and other demanding games

http://www.anandtech.com/show/4263/amds-radeon-hd-6450-uvd3-meets-htpc

Nvidia is a bully and this has to stop. If AMD was anything like them they would have made some DX11 only games two years ago. If AMD is anything like them they will create some games that work really bad on the Nvidia cards when they will release the 7000 series and split the gaming market for good, Nvidia started this.

I see now nvidia is evil and Jen-Hsun Huang is the devil!

I do see a lot of hate from you too. Maybe you should stay far away from nvidia, i am sure you dont own a card from them. AMD is sooo much better and you should be very happy as an AMD card owner. Instead you spend your time full of hate for another company. this is sad
 

ocre

Golden Member
Dec 26, 2008
1,594
7
81
I don't think this hurts nVidia's business at all. This is rumor mongering, conjecture, innuendo and conspiracy -- more entertainment than reality to me. If this had any substance, there would be clear poof and data, engineers and developers coming clean.

seems like some ppl really really really want it to hurt their business. It is very important to them, obviously.

I dont see why ppl are so hateful. I have an nvidia card and i cant play it and i am no where near as upset. As a matter of fact, it doesnt bother me at all. I bet the next gen cards will be able to play this no problem, both AMD and Nvidias! Who wouldve thought that i couldnt play every game maxed out 1080 with my midgrade/budget 460! Stuff like this makes me want to upgrade soon. I actually get excited when its time to upgrade. But then again, i am not new and love PC gaming, always have!
 

WMD

Senior member
Apr 13, 2011
476
0
0
We may not be talking in the same channel but I am actually interested on the 2xbandwidth when tessellation used from AMD. Where is it? Knowing that the hardware itself should handle tessellation better, it clearly is not the case. As MutantGith pointed out, crysis is known to kill performance for fun and optimization is not a priority for crysis 2, seeing everything under perform is not to my surprise. Without the bias and conspiracy, with 580 the minimium FPS is only 2. If conspiracy is in place, I'll say Nvidia shot their own feet more than to AMD.

I say there exist bugs within the code path which prevents AMD's true power to be unleashed. I am guessing but I believe not even half of those tessellation unit is working. In fact, I do not even think the Dx9 code path is correct as AMD's lineup really should not get hit like that if it were for bugs. There are problems, but the invisible ocean is not one of them. At best, it hurts performance in general, not only to AMD, so the conspiracy theory does not hold, IMO.

Someone posted a link to a pcgh article sonewhere in this thread. Basically the 2x tessellation throughput of 6900 goes down the drain when tessellation factor is increased beyond a certain point. Crysis 2 uses extremely high tessellation factors. AMD users can manually clamp the tessellation factor in drivers to gain back some performance. But unfortunately this results such bad aliasing that it looks worse than no tessellation.
 
Last edited:

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Can someone tell me what extra effects (Tess, SSDO, POM etc) Ultra setting in the game have over the previous lower setting ??

If i recall, Tessellation is only on Ultra settings but thats not the only effect.

I believe the massive performance hit is not from the Tessellation but from an other effect.(perhaps SSDO or POM ??)
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Can someone tell me what extra effects (Tess, SSDO, POM etc) Ultra setting in the game have over the previous lower setting ??

If i recall, Tessellation is only on Ultra settings but thats not the only effect.

I believe the massive performance hit is not from the Tessellation but from an other effect.(perhaps SSDO or POM ??)

Ad google to this:
It also not just like the DX11 patch(es) were all about tessellation.
Besides tessellation the high ressolution textures and normal maps, the patch(es) add new effects:

Soft shadows with variable penumbra
Improved water rendering
Particle motion blur
Particle shadowing
Realtime Local Reflections
Screen Space Directional Occlusion
Sprite-based Bokeh depth of field effects

And as Seero's post shows it is really complex mathematics/code and hard to "101" boil down to laymen.
 

Red Hawk

Diamond Member
Jan 1, 2011
3,266
169
106
Judging by the effect of SSAO in Deus Ex (and Dragon Age II, for that matter...I actually should do some benches on that) I actually wouldn't be surprised if SSDO was causing more of a hit than tessellation. Or, at least, causing a major part of the performance hit.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
I don't think this hurts nVidia's business at all. This is rumor mongering, conjecture, innuendo and conspiracy -- more entertainment than reality to me. If this had any substance, there would be clear poof and data, engineers and developers coming clean.

Typically with nVidia it takes class action for them to come clean. Even then, when they are replacing equipment because of it, they admit no wrong.
 
Last edited:

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Judging by the effect of SSAO in Deus Ex (and Dragon Age II, for that matter...I actually should do some benches on that) I actually wouldn't be surprised if SSDO was causing more of a hit than tessellation. Or, at least, causing a major part of the performance hit.

According to benches from BEHardware it's " Ultra Objects" that incurs the biggest single hit. Objects is tessellation and POM.

http://www.behardware.com/articles/...ser-look-at-performance-and-tessellation.html
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
So, HD5870 has the same performance in Tessellation with HD6950 ???
HD5870 lost the same amount with Tessellation as HD6950 and HD6970 ??

no way


HD6870 should have better performance in tessellation than HD5870 and yet HD5870 is faster.

Well, I was just answering the question of what caused the biggest performance hit. Now, as far as performance of the AMD cards relative to each other goes, I would have to think it's driver optimization. Or, more accurately, the lack of. I doubt AMD had a chance to optimize drivers before release since the patch was sponsored by nVidia.

This doesn't change the fact of stupid tessellation of invisible objects and flat surfaces, though.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Thats what i wanted to show, that Tess could not be the only thing hammering the performance in AMD cards, perhaps in NV cards too, although NV cards scale better in Ultra Setting.
 

Arkadrel

Diamond Member
Oct 19, 2010
3,681
2
0
Can you show some of these objects?

You didnt read the artical that this thread is about did you?


Water? where?




Oh here it is!




Lets use lots of tessellation on a invisible ocean that cant be seen

nvidia user: But but... it hurts my performance during game play, for no reason?
nvidia: the reason is it hurts AMD card users even more!
(20% ouch for us, 40% ouch for them)




No tess:



MAX TESSELATION!:






Because useing insane amounts of triangles on a silly rock, without any visual differnce shown between the 2, makes alot of sense.


I honestly think Nvidia payed Crytek to make Crysis2 run better on their cards, and to do it, they cranked up tessellation, without any need too do so.
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
I'm playing the game, so i think you guys speaking of this:








I'm not taltking about the water mesh. That's a problem of the engine. And has nothing to do with tessellation.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Can you show some of these objects?

You need to catch up, mate. It's what the whole thread is about. Check it out, because we don't really need to hash everything out from the beginning again. We've been going around in circles enough.
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
You need to catch up, mate. It's what the whole thread is about. Check it out, because we don't really need to hash everything out from the beginning again. We've been going around in circles enough.

I read the thread, but i only find words. One site complaining about the barriers (and was wrong), the other one about the curbs. So that's all? Two objects in a game which has more than 20+ tessellated objects?
 

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
I read the thread, but i only find words. One site complaining about the barriers (and was wrong), the other one about the curbs. So that's all? Two objects in a game which has more than 20+ tessellated objects?

No. It's about accusing Nvidia of collusion with Crytek to create super high amounts of tesselation to make AMD performance tank. Really, if Nvidia didn't have such good tesselation performance there probably wouldn't be any issue. Tesselated invisible ocean or not.
On a REAL note, the tesselation of the invisible ocean, which I believe is a side effect of the game engine, should be addressed in a patch by Crytek. Definitely a waste of GPU resources. As far as the K-rail barriers, I don't see a problem with that at all. It's not invisible and there is noticeable differences between no tess and tess.
You will see complaints of the LEVEL of tesselation due to the fact that AMD CAN handle lighter amounts of tess without too much trouble. As soon as the tess reaches a level that starts to make AMD GPU's struggle, then it was done by Nvidia on purpose. This has been quite the track record the past few years. Everything Nvidia does over and above, is unfair. Unless it benefits AMD users as well.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |