Far Cry 3 Performance, Benchmarked

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
I too would like to have seen "High" settings benchmarked. I'll just have to wait for HardOCP's "Highest Playable Settings" review; that can be a useful approach and one I hope you will consider in the future.
 
I do not understand the results made ​​with medium settings even I do not drop below 50 FPS with phenom 955 BE @ 3.8 + HD 6950.
I think you get the benefits of manufacturers to participate in planned obsolescence...
 


The only difference between the G860 & i3-2100 is the clock multiplier (30 vs 31) and the i3-2100 has HT. Other than Hyper Threading and 100mhz their is no difference between the two CPU's. I don't understand how 100mhz effected the playability that much. Their must be something else.

Pentium G860 & i3-2100 tech specs
 
Still looks to me as if those heavy GPUs should do better than they do at the higher settings, think it could have been optimised way better. Things like textures and meshes also still look like they haven't received a giant upgrade from the consoles.

I know most people say this is a very good-looking title, but I don't really see why it's such a demanding game... regardless of which game looks better I thought BF3 or Crysis used way more intensive effects.
 
still wondering when people will realise that the bottleneck they keep talking about is null. Hypertransport and quickpath have the same bandwith, quickpath is almost a direct rip off of hypertransport.
 
For those mentioning how the FX-8350 being better at heavily-threaded productivity programs, I didn't thumb you guys down for it... But we have to remember that the editors were writing this article based on this specific game, and thus talk about it in the context of such. :)

+1'ed. I agree with both your points in the 1st paragraph.

I'd have to say though that it would've been interesting to see how the GTX 680 would do since it has more memory bandwidth compared to the GTX 670. I'm not sure about this as it has been a while since I read on it, but that might be their only difference. :)

It must be the HTT then. I have heard that the i3's HTT proves helpful in games (generally-speaking), unlike with the i7's (possibly because they already have 4 cores to handle 4 threads at a time without it). :)


The Core i3 performing better than the Pentium could be explained by the possible fact that this game is well-threaded enough to take advantage of at least 4 CPU threads, but the fact that the dual-module Bulldozer and quad-module Piledriver outperformed the quad-core Stars... Could this mean that this game does a lot more integer operations or something? Also, if the quad-module/octa-core Piledriver performed that well, could this mean this game is even more well-threaded that it can take advantage of more than 4 threads?! Hm... Quite exciting!
I wish we could see the i7-3960x spread its legs more... 😍
 
Try having a gander at this. 😀 The performance and technology of Far Cry® 3 (AMD Blogs)
 
[citation][nom]army_ant7[/nom]For those mentioning how the FX-8350 being better at heavily-threaded productivity programs, I didn't thumb you guys down for it... But we have to remember that the editors were writing this article based on this specific game, and thus talk about it in the context of such. +1'ed. I agree with both your points in the 1st paragraph.I'd have to say though that it would've been interesting to see how the GTX 680 would do since it has more memory bandwidth compared to the GTX 670. I'm not sure about this as it has been a while since I read on it, but that might be their only difference. It must be the HTT then. I have heard that the i3's HTT proves helpful in games (generally-speaking), unlike with the i7's (possibly because they already have 4 cores to handle 4 threads at a time without it). The Core i3 performing better than the Pentium could be explained by the possible fact that this game is well-threaded enough to take advantage of at least 4 CPU threads, but the fact that the dual-module Bulldozer and quad-module Piledriver outperformed the quad-core Stars... Could this mean that this game does a lot more integer operations or something? Also, if the quad-module/octa-core Piledriver performed that well, could this mean this game is even more well-threaded that it can take advantage of more than 4 threads?! Hm... Quite exciting!I wish we could see the i7-3960x spread its legs more...[/citation]
Well when you search about a product on the net and they tell you is not a good ideea to get a car because you can`t climb rocks with it ... the conclusion should have been that the game is not really CPU bound .... since most of CPUs out there will be able to run it, but the author said crap about I3 being better than an FX, that`s the deal about it.
 
omg i can't believe these comments... it must be a pita sometimes to get an article out.

This article is about a game called Far Cry 3 and its performance numbers with specific hardware. It's only responsibility is to that game and the hardware tested. The article does not reach beyond that. Everything is presented properly. It is not Tom's responsibility to account for every possible little complaint that may come up from the community. There's a million, remember. It's the reader's responsibility to take the information provided and do something with it, or nothing at all.

When I saw the CPU chart, the first thing I noticed was the i3. Who didn't think in their head what was stated? Everybody, or nearly everyone. The comment was perfectly fine as it was relating directly to this game and this test and didn't wander outside of that box. So what's the problem? Smart people are worried about dumb people? So you get it, you're just worried about dumb people. Uh, thanks for looking out. I sense a bit of arrogance there.

It's a good article. It does what it says it'll do. What more could you ask for? If they had more hardware around to test with, they probably would have, and would it have been worth it to stall the article for more time to gather more hardware for more testing that'd just produce obvious results anyway? Hmm...
 
This game is a candidate for GOTY for me. I still think graphics are better looking Crysis overall in terms of vegetation physics, levels of detail in places. Far Cry 3 is a 2012 game and it can't outright beat out Crysis 1 in graphics? That's being generous as in videos online Crysis 1/Warhead look better to me:

http://www.youtube.com/watch?v=E3ohqbEn1v8

Not only does Crysis look more realistic in terms of lighting/sun shafts/vegetation physics but it runs way better too. I agree with other posters that while Far Cry 3 does look good, it isn't as well optimized as it should have been. Going from High to Ultra drops performance like a rock and the game barely looks better. Small changes from HBAO to HDAO hardly change the image quality but drop performance by dozens of frames.
 
It would be nice to see 1 generation old card in the mix like the 580,6950,560...so we can gt a better handle on what upgrades we may want to look at. For now I have to make a fuzzy guess that my 560s in sli that are about a 580 that is about a 7870 would run this game at high/ultra with AA set to low or off.

As for the FX vs i3...in the context of a game....a cpu that is 90 bucks less but runs just as good is a fair comment but an incomplete one because the value of the 90 bucks comes in the OC and the multi core non gaming plus of the FX chip. I am also left wondering if multiplier tests would impact where the CPUs would stand.
 


what are you talking about ? the pentium and i3 BOTH have the same amount of L3 cache .
and the i3 does NOT have 2 extra cores. its 2 physical core and 4 threads because each core is hyperthreaded.
however toms had done some tests last year showing that hyperthreading barely benefits frame rates in most games so am really surprised that difference betwen i3 and pentium is so much.
 
[citation][nom]loops[/nom]It would be nice to see 1 generation old card in the mix like the 580,6950,560...so we can gt a better handle on what upgrades we may want to look at.[/citation]
Maybe you are like myself, and dont tend to stray to far from Tom's. I'm guessing there is at least one other site that benched this game with a 580. When you find it, come back and tell me about it, lol.
 


I was pretty sure they had cache differences. Oh well.



Yeah, yeah... Bad wording. Jeez...

AVX could be the cause here. If they show a big difference in the Dirt 3 engine, then every game compiled to take advantage of AVX will show this disparity it seems.

This one is for the owners of FC3 to answer, since there are usually 2 exe's for this.

Cheers!

EDIT: It could be something in the IMC as well. For instance, the G6xx doesn't have DDR3 1333 support, whereas the G8xx has. The i3 could have a better IMC as well, or something like that.
EDIT2: AVX could be indeed the cause, since if you take a look, Ph II has a big difference with the FX line. Uhm...
 
[citation][nom]dscudella[/nom]The only difference between the G860 & i3-2100 is the clock multiplier (30 vs 31) and the i3-2100 has HT. Other than Hyper Threading and 100mhz their is no difference between the two CPU's. I don't understand how 100mhz effected the playability that much. Their must be something else.Pentium G860 & i3-2100 tech specs[/citation]
Yeah, the more to it is Hyper Threading, The i3 can execute 4 threads simultanously, and FC3 can address a max of 4 threads i think because it's not scaling with thread count beyond 4, or at least so it seems.

Look at almost any modern game bench and that's what you'll find most of the time.
 
Does anyone notice the the last gen fx quad core slaughters the sandbridge pentium and phenom ii x4. Maybe people need to adjust their opinions on the phenom ii and the pentiums. Even the i3 shows major performance gains over the pentium, who only major difference I believe is the lack of hyper threading. And on the whole mocking of AMD's 8 core processors, this game is obviously not meant to take advantage of anything over 4 cores. They might as well mock Intel 6 core processor for not giving much of a performance boost for costing around $500.
 
WoW You guys must have the secret formula for getting this game to run. I purchased the retail DVD and cannot play more than 15 min of the game. It automatically upgraded to V1.03 and still locked up.
Cannot get out of truck on first outpost mission. Submitted a report to UBI support and they have not responded.
AMD Radeon 12.11beta Catalyst Software; AMD Phenom X4 965BE 3.4Ghz; 8Gig DDR3 1333 memory;
Radeon 5870 2 Gig DDR5 Eyefinity Graphics.

Game runs smooth up to that point.
 
[citation][nom]BigMack70[/nom]Wow soooooooo many AMD fanboys with hurt feelings in these comments...Come on guys, get over yourselves... Intel is the way to go for gaming CPUs anymore. That's not Toms trying to make AMD look bad, that's just how things are today.[/citation]
That's why AMD is within spitting distance of Intel in this game ? OC on the 8350 and the AMD is on par although it's not like it was struggling in this instance to begin with.
 
[citation][nom]sugetsu[/nom]"The good news for folks with Piledriver-based processors is that the FX-8350 is nearly as quick as Intel's Core i3-2100 (never mind the fact that the Core i3 costs $90 less)."My God... Are the reviewers of this website paid to make AMD look bad? .[/citation]

I dont think you need to pay anyone to make amd look bad. They do that themselves just fine.
 
The CPU section seems a little half arsed, where's the same CPU at the same clock but with different core counts? Or the same CPU with the same cores at different clocks?
 
This review missed an opportunity to say something very important about the technology behind this game.
I played most of this game for the last two days on a configuration that was well below spec but still enjoy;ed the game.

The review went from medium settings to ultra, completely dropping the numbers for a few cards, notably the HD 7770 and 7850 I was watching out for.


My system specs were the following:
i5-2500k (only 3 cores, clocked at 1600 Mhz, iGPU seemed to make no use of higher clocks), HD3000 O/Clocked to 1800Mhz


In Game settings were the following:

Dx9, 1280x720, 0AA, 8xAF, vsyvc=off
Textures: high
Ambient Lighting: high
Shadow: medium
Post FX: medium
Geometry: very high
Vegetation: high
Terrain: low
Water: medium
Environment: high

Average frames = 17

The point is, I was shocked at the visual quality I was able to squeeze out of this system. This is probably, by far, the most scaleable and optimized graphics engine EVER!! Crytek should be proud, and so too the people at Ubisoft that optimized the game. The game itself was too many levels of awesome to state here. I can't wait for a play-through with a better GPU requiring all 3.3Ghz of i5-2500k.

Cheers an bench.
 
[citation][nom]BigMack70[/nom]Note also I'm not saying that AMDs chips suck... just that there is absolutely zero reason to build a gaming rig around one in 2012, unless you make that decision purely on the fact that you like AMD/hate Intel. If you are brand neutral, no compelling reason exists to build a 2012 gaming rig with an AMD CPU, at any price point.[/citation]

Between the i3 and i5 price range is were AMD has the advantage. Even though games most games today do not take advantage of 6 or 8 core processors the list is growing. In games like Battlefield 3 multiplayer matches, the FX 6 and 8 core processor show considerable gains over intels i3's and pentiums. That said there are other factors to consider like power draw, but this isn't that bad since their idle draw is competitive.

 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom