Was the problem with Project Cars ever detailed?

[Azix]

If anyone has any resources on what the issue was with this game I would love to see it. Wondering if I missed something, I had stopped following the issue after a while.

Though it might take more than an outside would know about the game to get to the answer. I think it's a very interesting case.

The problem was Lower than expected performance on systems with Amd graphics cards installed. I only want to know why the game ran like that.

 

[tential]

You should be specific and specify which issue you're talking about. Many people had 0 issues with this game.

I think I know what you mean but detail in op.

Thanks

 

[Azix]

Lower than expected performance on systems with Amd graphics cards installed

 

[godihatework]

Didn't you just get a thread you necroed locked for the same thing?

 

[realibrad]

Didn't you just get a thread you necroed locked for the same thing?

No. The thread was locked because it was old, and the original topic was different from this one. If it was locked for asking the question that the OP asked in this thread, then it was wrong to lock that thread.

 

[RussianSensation]

If anyone has any resources on what the issue was with this game I would love to see it. Wondering if I missed something, I had stopped following the issue after a while.

Though it might take more than an outside would know about the game to get to the answer. I think it's a very interesting case.

The problem was Lower than expected performance on systems with Amd graphics cards installed. I only want to know why the game ran like that.

The problem is the Project CARS developer either has no clue to how make games OR intentionally continues to make games that 100% favour NV hardware by not optimizing the game engine for any other products. This developer was already suspect of not knowing how to code games for AMD products back then when they made NFS Shift 2:

http://www.pcgameshardware.com/aid,...-2-accelerates-Radeon-graphics-cards/Reviews/

Yet, despite knowing that NFS Shift 2 engine was outdated and unoptimized pile, they just continued to use it for Project CARS and ignored AMD user feedback during 2.5 years of crowd funding development on the game.

It's hard to fix a broken DX11 game engine that is reused from NFS Shift 2 and was already broken back then; or I suppose as NV supporters/loyalists would say it it's AMD's fault that they couldn't optimize the NFS Shift 2 game engine even though the developer somehow magically improved performance in NFS Shift 2 by 45% on AMD cards back in the days.

In 2015, it's popular to just blame AMD and ignore how AMD's cards perform very well in titles made by developers who actually make sure the games run well across both AMD and NV -- to some people a shock I suppose:

Dota 2, CS: GO, Diablo 3, Heroes of the Storm, League of Legends, Warframe, World of Tanks, WoW at 4K

Not surprisingly, Project CARS is a GW title so expecting it to run well on AMD products is like asking for a miracle.

 

[godihatework]

No. The thread was locked because it was old, and the original topic was different from this one. If it was locked for asking the question that the OP asked in this thread, then it was wrong to lock that thread.

[quote name="Makaveli" post=37437786]you doubt it why?[/QUOTE]<br />
<br />
because nvidia!!<br />
<br />
Was there any conclusion on why this game's performance was bad? Something was clearly wrong, but what?<br/>

Really? Are you sure about that?

 

[2is]

The problem is the Project CARS developer either has no clue to how make games OR intentionally continues to make games that 100% favour NV hardware by not optimizing the game engine for any other products. This developer was already suspect of not knowing how to code games for AMD products back then when they made NFS Shift 2:

http://www.pcgameshardware.com/aid,...-2-accelerates-Radeon-graphics-cards/Reviews/

Yet, despite knowing that NFS Shift 2 engine was outdated and unoptimized pile, they just continued to use it for Project CARS and ignored AMD user feedback during 2.5 years of crowd funding development on the game.

It's hard to fix a broken DX11 game engine that is reused from NFS Shift 2 and was already broken back then; or I suppose as NV supporters/loyalists would say it it's AMD's fault that they couldn't optimize the NFS Shift 2 game engine even though the developer somehow magically improved performance in NFS Shift 2 by 45% on AMD cards back in the days.

In 2015, it's popular to just blame AMD and ignore how AMD's cards perform very well in titles made by developers who actually make sure the games run well across both AMD and NV -- to some people a shock I suppose:

Dota 2, CS: GO, Diablo 3, Heroes of the Storm, League of Legends, Warframe, World of Tanks, WoW at 4K

Not surprisingly, Project CARS is a GW title so expecting it to run well on AMD products is like asking for a miracle.

By ignored AMD feedback you must actually mean, sent multiple emails to AMD that went unanswered until after the games release, since that's what happened. Not unlike TW3 where AMD approached them what? An entire 2 weeks before release or something retarded like that.

Are both developers lying?

 

[poofyhairguy]

Obviously some game developers are in over their heads, and Nvidia is the company willing to throw out a life raft. Just like them fixing Batman for WB.

 

[Azix]

[quote name="Makaveli" post=37437786]you doubt it why?<br />
<br />
because nvidia!!<br />
<br />
Was there any conclusion on why this game's performance was bad? Something was clearly wrong, but what?<br/>

Really? Are you sure about that?

I assumed it was because it was old. Though I thought it was better to post in that thread rather than start a new one. meh.

No reason to lock the thread for the actual post.

By ignored AMD feedback you must actually mean, sent multiple emails to AMD that went unanswered until after the games release, since that's what happened. Not unlike TW3 where AMD approached them what? An entire 2 weeks before release or something retarded like that.

Are both developers lying?

The thing is there was conflicting info from the devs about working with AMD. First they stated no communication and that seemed to have stuck. They more recently say they were in communication with AMD. AMD was also in contact with TW3 devs. They said they reached out to include tressfx later because the hairworks code came much later in development and ran badly.

But really... even if AMD was not in contact with them it's a pretty bad showing for a game and the problem was not in graphics output AFAICT. there is something fundamentally off about the game. The problem is beyond driver overhead unless they are doing some really strange things that cause the AMD drivers to do extra work.

 

[2is]

I don't see much in the way of conflicting story lines. Two different games, two different devs, same story.

 

[Hitman928]

The dev who originally said they hadn't heard from AMD for several months "checked the emails" and said that they had actually been in contact the whole time.

 

[PPB]

Why even bother with that game, just go for assetto corsa and be done with them.

 

[96Firebird]

To answer the original question, no one (here) knows. People can speculate all they want, which it looks like they have already started again, but anyone's guess is as good as theirs. No reason was ever given as to why AMD cards perform worse in this game.

 

[RussianSensation]

By ignored AMD feedback you must actually mean, sent multiple emails to AMD that went unanswered until after the games release, since that's what happened. Not unlike TW3 where AMD approached them what? An entire 2 weeks before release or something retarded like that.

Are both developers lying?

That's been debunked a long time ago. What has stood in the history books is NFS Shift 2 getting a magical 45% increase in performance months after release and yet this increase had nothing to do with AMD's drivers. That is the developer's resume track record when it comes to optimization -- or in this case lack thereof.

The interesting part is gamers keep ignoring many other glaring issues in Project CARS completely unrelated to AMD in their defense of this turd of a game:

1) Horrendously unoptimized CPU performance:

GTX980 OC pairing with:

i7 4770K 4.4Ghz = 78.6 fps

vs.

i7 4770K 2.5Ghz 4 cores HT disabled = 53.6 fps
i7 4770K 3.5Ghz 2 CPU cores HT enabled = 57.6 fps
i7 4770K 3.5Ghz 4 cores HT disabled = 62.9 fps (2 extra CPU cores add just 9% performance increase but....adding HT with 4 cores reduces performance:

i7 4770K 3.5Ghz 4 cores HT enabled = 62.2 fps.

SO let's get this straight going from 2.5Ghz to 3.5Ghz (+40% increase in CPU clock) provides only a 17% increase in performance on a 980 OC but going from 3.5Ghz to 4.4Ghz (26% increase) adds a 25% increase in performance (78.6 vs. 62.9 fps). FAIL. :sneaky:
http://www.computerbase.de/2015-05/project-cars-guide-grafikkarte-prozessor-vergleich/2/

2) Horrible Kepler performance and Kepler scaling

Let's not forget that this game absolutely bombed on Kepler GPUs too so the problem isn't entirely AMD related.

770 = 30.7
780 = 33.5
780Ti = 35.8 fps (aha just 5 frames more going from a 770 to 780Ti)

vs.

960 OC = 33. fps (almost as fast as a 780Ti)
http://www.computerbase.de/2015-05/project-cars-guide-grafikkarte-prozessor-vergleich/2/

But I guess it's convenient to just wipe all this data from the past to keep the AMD is at fault theme going, right?

It's amusing how people will die by the sword bashing AMD and ignore all of the other data points that prove how the entire game was broken on day 1 of release.

If the game is based on the inefficient and broken NFS Shift 2 engine, it only makes sense that any game built on this foundation will end up as a broken disaster.

Despite decent professional reviews, actual gamers didn't care much for this game:

http://www.metacritic.com/game/playstation-4/project-cars
http://www.metacritic.com/game/pc/project-cars
http://www.metacritic.com/game/xbox-one/project-cars

Total retail sales to date are an abysmal 600K units, of which 570K came from XB1/PS4 consoles. This game was riddled with bugs and glitches and optimization on the PC was basically non-existent.

This was supposed to be one of the best racers not just on the PC but across any gaming platform and it completely failed to live up to the hype. Using the old NFS Shift 2 engine was a big big mistake.

 

[brandonmatic]

I thought it was interesting that The Tech Report decided to exclude the Project Cars results from their Fury benchmarks "given how different they are from everything else." I don't recall ever seeing that before.

http://techreport.com/blog/28624/reconsidering-the-overall-index-in-our-radeon-r9-fury-review

 

[Silverforce11]

I thought it was interesting that The Tech Report decided to exclude the Project Cars results from their Fury benchmarks "given how different they are from everything else." I don't recall ever seeing that before.

http://techreport.com/blog/28624/reconsidering-the-overall-index-in-our-radeon-r9-fury-review

They discarded Dirt Showdown because DX11 compute usage in that game favors AMD GPUs. They even said that was the reason they won't keep on using that game in reviews due to the bias. Then they include Project Cars.. obviously their readers spot a HYPOCRITE and pointed it out to them.

The problem with Prj Cars isn't just poor AMD performance, its poor Kepler performance. Seen that somewhere before, it's a re-occurring theme with GameWorks.

For the latest interview from people in charge of GameWorks:
http://wccftech.com/exclusive-nvidias-amds-perspectives-gameworks-bottom-issue/

"positively influence both the performance and the visuals to offer users of its latest GeForce products a better experience"

If you don't keep up to date with their latest gen GPUs, you don't get to enjoy the "positive influence". It's working as intended.

Before the same false claims are made that Prj Cars does not use GameWorks, you should be informed that NV regards PhysX as part of its GW libraries (https://developer.nvidia.com/gameworks-physx-overview).

The Prj Cars lead engineer had this statement to say: "Physics running 600hz is causing problems for AMD because it's hammering the CPU". We don't know why it doesn't have a negative effect on Maxwell, better driver overhead maybe, some offloading maybe, but it affects Kepler so we can rule out driver overhead as an issue.

As to why anyone would run physics simulations 600 times per second .. well, it takes a special kind of developer to poll physics 10x beyond what most systems run the game at (60fps). Complete waste of CPU cycles.

 

[imaheadcase]

The problem is the Project CARS developer either has no clue to how make games OR intentionally continues to make games that 100% favour NV hardware by not optimizing the game engine for any other products. This developer was already suspect of not knowing how to code games for AMD products back then when they made NFS Shift 2:

http://www.pcgameshardware.com/aid,...-2-accelerates-Radeon-graphics-cards/Reviews/

Yet, despite knowing that NFS Shift 2 engine was outdated and unoptimized pile, they just continued to use it for Project CARS and ignored AMD user feedback during 2.5 years of crowd funding development on the game.

It's hard to fix a broken DX11 game engine that is reused from NFS Shift 2 and was already broken back then; or I suppose as NV supporters/loyalists would say it it's AMD's fault that they couldn't optimize the NFS Shift 2 game engine even though the developer somehow magically improved performance in NFS Shift 2 by 45% on AMD cards back in the days.

In 2015, it's popular to just blame AMD and ignore how AMD's cards perform very well in titles made by developers who actually make sure the games run well across both AMD and NV -- to some people a shock I suppose:

Dota 2, CS: GO, Diablo 3, Heroes of the Storm, League of Legends, Warframe, World of Tanks, WoW at 4K

Not surprisingly, Project CARS is a GW title so expecting it to run well on AMD products is like asking for a miracle.

Wait in another thread you said that those type of FPS difference was "Demolishing". lol

 

[Silverforce11]

Wait in another thread you said that those type of FPS difference was "Demolishing". lol

Did you have anything to say about how the same devs in the past, release a patch and AMD performance jumps 45%? Heck, do you have anything worth contributing to this thread at all or just trolling?

 

[tential]

Why is this game talked about so much on PC then? It's got 30k sales.... 30k? So why is everyone in a rush to talk about a game so few of us were willing to buy?

I may have had my gripes about AC Unity but at least people bought it.

Actually, this is what gets me with a lot of the PC community. It seems PC community gets caught up on these smaller titles and it's like wait a minute... barely ANY of us have this game. Why is this such a large part of reviews when this game is a game none of us play.

Games like Project Cars should get reviewed, maybe even follow up reviews on performance. But including them in GPU reviews is just senseless.

 

[Silverforce11]

Good point, on steamcharts, its not even in the top 200 list, its gone.

http://steamcharts.com/top

Arma 3 has consistently been in the top 20 for YEARS yet so few review sites use it as a benchmark. Very popular game, crushes GPUs/CPU, good game too.

 

[Azix]

Why is this game talked about so much on PC then? It's got 30k sales.... 30k? So why is everyone in a rush to talk about a game so few of us were willing to buy?

I may have had my gripes about AC Unity but at least people bought it.

Actually, this is what gets me with a lot of the PC community. It seems PC community gets caught up on these smaller titles and it's like wait a minute... barely ANY of us have this game. Why is this such a large part of reviews when this game is a game none of us play.

Games like Project Cars should get reviewed, maybe even follow up reviews on performance. But including them in GPU reviews is just senseless.

I just think its an interesting case. Would have been nice to learn what they messed up but I guess that's unlikely for people outside of the loop to ever find out.

As for review sites using it, good point. They should not be. That's something else worth investigating.

"positively influence both the performance and the visuals to offer users of its latest GeForce products a better experience"

Don't want this to be about gameworks or just nvidia bashing but I did find that interesting when I saw it. If any nvidia person wants to know where they stand with nvidia, thats it right there.

 

[tential]

I mean that's why the performance is bad imo. It's a small fame reaching a small number of people. It'll get optimized later probably but really they focused on ensuring the review sites had builds that would work on their rigs and everything else comes later if ever. Pc isn't that big of a deal with 30k sales compared to the 570k other sales on console. They probably just don't really care about pc as it didn't give them much revenue

 

[RussianSensation]

Wait in another thread you said that those type of FPS difference was "Demolishing". lol

What? I am completely confused by your comment. Did you actually click the link on those popular Steam titles tested at 4K? Notice how most of them run well across either NV or AMD hardware? Don't you find it odd that the most popular PC games made by the best development houses not only run well across both NV and AMD but they scale very well on lower to higher end AMD cards and similarly on lower end NV to higher end NV cards?

It's interesting how you completely ignored the 2 points I made above regarding Kepler performance and Project CARS's atrocious CPU scaling performance.

Did you have anything to say about how the same devs in the past, release a patch and AMD performance jumps 45%? Heck, do you have anything worth contributing to this thread at all or just trolling?

It's better to keep the theme going that it's AMD's fault in all GW titles and that AMD spends little money on their drivers and that their drivers are crappy than to accept that 95% of all brand agnostic non-GW titles "magically" run amazingly well on ATI's/AMD's videocards and historically AMD's cards actually age better! But no, let's pretend 95% of all PC games in the last 30 years do not exist and start singing the HardOCP tune by focusing strictly on GW titles.

I would love to see actual Revenues vs. Unit sales of GW PC games because after Batman AK, Project CARS and Dying Light performance issues, I bet many gamers are staying well clear of GW until they drop to $30 and below. Right now for me a GW games is almost automatically a sub-$10 purchase because I know there is nearly a 95% chance it will be broken and have horrendous CF support for months. It's kinda hard to blame AMD here because AMD's card perform well in 95% of all other brand agnostic titles. There is clearly a theme. :sneaky:

Why is this game talked about so much on PC then? It's got 30k sales.... 30k? So why is everyone in a rush to talk about a game so few of us were willing to buy?

I may have had my gripes about AC Unity but at least people bought it.

Actually, this is what gets me with a lot of the PC community. It seems PC community gets caught up on these smaller titles and it's like wait a minute... barely ANY of us have this game. Why is this such a large part of reviews when this game is a game none of us play.

Games like Project Cars should get reviewed, maybe even follow up reviews on performance. But including them in GPU reviews is just senseless.

I should point out that 30K were retail figures, not including online sales but I am 100% onboard with the key theme of your post.

That's why I love reviewers that think outside the box. Look at Computerbase that actually tested 8 of some of THE most popular PC games on 4K across various GPU configs. This is pretty relevant since someone with a 1080P or 1200P or even 1440P monitor might enable DSR/VSR for these uber popular games:

http://www.computerbase.de/2015-07/...t-wow-ultra-hd/4/#abschnitt_performancerating

Good point, on steamcharts, its not even in the top 200 list, its gone.

http://steamcharts.com/top

Arma 3 has consistently been in the top 20 for YEARS yet so few review sites use it as a benchmark. Very popular game, crushes GPUs/CPU, good game too.

:thumbsup:

 

[3DVagabond]

Obviously some game developers are in over their heads, and Nvidia is the company willing to throw out a life raft. Just like them fixing Batman for WB.

Never mind. Off topic.