Personal computing discussed

Moderators: renee, Hoser

 
Techgoudy
Gerbil First Class
Posts: 142
Joined: Tue Oct 02, 2012 5:01 pm

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Fri May 10, 2013 11:01 pm

1280x720 Maximum
Score: 7659
Performance: Extremely High

    FINAL FANTASY XIV: A Realm Reborn Official Benchmark (Exploration)
    Tested on:5/10/2013 11:49:23 PM
    Score:7659
    Average Framerate:64.499
    Performance:Extremely High
    -Easily capable of running the game on the highest settings.

    Screen Size: 1280x720
    Screen Mode: Windowed
    Graphics Presets: Maximum
    General
    -Improve overall graphic quality. : Enabled
    -Disable rendering of objects when not visible. (Occlusion Culling) : Disabled
    -Use low-detail models on distant objects to increase performance. (LOD) : Disabled
    -Cache LOD data only when necessary. (LOD Streaming) : Disabled
    -Smooth edges. (Anti-aliasing) : Enabled
    -Increase transparent lighting quality. : Enabled
    -Grass Quality : High
    Shadows
    -Use low-detail models on shadows to increase performance. (LOD) : Disabled
    -Display : All
    -Shadow Resolution : High: 2048 pixels
    -Shadow Cascading : High
    -Shadow Softening : High
    Texture Detail
    -Texture Filtering : High
    -Anisotropic Filtering : High
    Effects
    -Naturally darken the edges of the screen. (Limb Darkening) : Enabled
    -Blur the graphics around an object in motion. (Radial Blur) : Enabled
    -Effects While in Motion : Display All
    -Screen Space Ambient Occlusion : High
    -Glare : Normal
    Cinematic Cutscenes
    -Enable depth of field. : Enabled

    System:
    Windows 7 Professional 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_gdr.130318-1533)
    Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
    8085.059MB
    AMD Radeon HD 7800 Series(VRAM 4095 MB) 8.17.0010.1191

    Benchmark results do not provide any guarantee FINAL FANTASY XIV: A Realm Reborn will run on your system.

    FINAL FANTASY XIV: A Realm Reborn Official Website http://na.finalfantasyxiv.com
    (C) 2010-2013 SQUARE ENIX CO., LTD. All Rights Reserved.

    Tweet
    http://sqex.to/ffxiv_bench_na #FFXIV Score:7659 1280x720 Maximum Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz AMD

    Radeon HD 7800 Series
 
XorCist
Gerbil First Class
Posts: 116
Joined: Wed Apr 18, 2007 6:54 pm
Location: Clovis, CA

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Fri May 10, 2013 11:54 pm

is it just me or is this low?


FINAL FANTASY XIV: A Realm Reborn Official Benchmark (Exploration)
Tested on:5/10/2013 9:53:11 PM
Score:4041
Average Framerate:32.647
Performance:High
-Easily capable of running the game. Should perform well, even at higher resolutions.

Screen Size: 1920x1080
Screen Mode: Full Screen
Graphics Presets: Maximum

General
-Improve overall graphic quality. : Enabled
-Disable rendering of objects when not visible. (Occlusion Culling) : Disabled
-Use low-detail models on distant objects to increase performance. (LOD) : Disabled
-Cache LOD data only when necessary. (LOD Streaming) : Disabled
-Smooth edges. (Anti-aliasing) : Enabled
-Increase transparent lighting quality. : Enabled
-Grass Quality : High
Shadows
-Use low-detail models on shadows to increase performance. (LOD) : Disabled
-Display : All
-Shadow Resolution : High: 2048 pixels
-Shadow Cascading : High
-Shadow Softening : High
Texture Detail
-Texture Filtering : High
-Anisotropic Filtering : High
Effects
-Naturally darken the edges of the screen. (Limb Darkening) : Enabled
-Blur the graphics around an object in motion. (Radial Blur) : Enabled
-Effects While in Motion : Display All
-Screen Space Ambient Occlusion : High
-Glare : Normal
Cinematic Cutscenes
-Enable depth of field. : Enabled

System:
Windows 7 Ultimate 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_gdr.130104-1431)
AMD Phenom(tm) II X4 955 Processor
8189.555MB
AMD Radeon HD 6800 Series(VRAM 4095 MB) 8.17.0010.1191


Benchmark results do not provide any guarantee FINAL FANTASY XIV: A Realm Reborn will run on your system.

FINAL FANTASY XIV: A Realm Reborn Official Website http://na.finalfantasyxiv.com
(C) 2010-2013 SQUARE ENIX CO., LTD. All Rights Reserved.

Tweet
http://sqex.to/ffxiv_bench_na #FFXIV Score:4041 1920x1080 Maximum AMD Phenom(tm) II X4 955 Processor AMD Radeon HD 6800 Series
 
Techgoudy
Gerbil First Class
Posts: 142
Joined: Tue Oct 02, 2012 5:01 pm

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sat May 11, 2013 12:16 am

XorCist wrote:
is it just me or is this low?


FINAL FANTASY XIV: A Realm Reborn Official Benchmark (Exploration)
Tested on:5/10/2013 9:53:11 PM
Score:4041
Average Framerate:32.647
Performance:High
-Easily capable of running the game. Should perform well, even at higher resolutions.

Screen Size: 1920x1080
Screen Mode: Full Screen
Graphics Presets: Maximum

General
-Improve overall graphic quality. : Enabled
-Disable rendering of objects when not visible. (Occlusion Culling) : Disabled
-Use low-detail models on distant objects to increase performance. (LOD) : Disabled
-Cache LOD data only when necessary. (LOD Streaming) : Disabled
-Smooth edges. (Anti-aliasing) : Enabled
-Increase transparent lighting quality. : Enabled
-Grass Quality : High
Shadows
-Use low-detail models on shadows to increase performance. (LOD) : Disabled
-Display : All
-Shadow Resolution : High: 2048 pixels
-Shadow Cascading : High
-Shadow Softening : High
Texture Detail
-Texture Filtering : High
-Anisotropic Filtering : High
Effects
-Naturally darken the edges of the screen. (Limb Darkening) : Enabled
-Blur the graphics around an object in motion. (Radial Blur) : Enabled
-Effects While in Motion : Display All
-Screen Space Ambient Occlusion : High
-Glare : Normal
Cinematic Cutscenes
-Enable depth of field. : Enabled

System:
Windows 7 Ultimate 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_gdr.130104-1431)
AMD Phenom(tm) II X4 955 Processor
8189.555MB
AMD Radeon HD 6800 Series(VRAM 4095 MB) 8.17.0010.1191


Benchmark results do not provide any guarantee FINAL FANTASY XIV: A Realm Reborn will run on your system.

FINAL FANTASY XIV: A Realm Reborn Official Website http://na.finalfantasyxiv.com
(C) 2010-2013 SQUARE ENIX CO., LTD. All Rights Reserved.

Tweet
http://sqex.to/ffxiv_bench_na #FFXIV Score:4041 1920x1080 Maximum AMD Phenom(tm) II X4 955 Processor AMD Radeon HD 6800 Series


No that sounds about right actually.
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sat May 11, 2013 12:19 am

8919 with everything maxed out at 1920x1080 (though I did enable all the extra options for the loader in terms of graphics).
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sat May 11, 2013 6:29 pm

I'm glad to have such a response to my thread, but I wish everyone would just post pictures or the results, rather than the entire long configuration and results dialog ... (⊙﹏⊙✿)

Everyone's results look more or less in-line except for the one guy that's obviously cheating.
 
VJB
Gerbil In Training
Posts: 4
Joined: Fri May 10, 2013 7:27 pm

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sat May 11, 2013 10:04 pm

That's also what I've said, I was joking around but I've also mention my real score wich is 11700 on Maximum :)
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sat May 11, 2013 10:13 pm

VJB wrote:
That's also what I've said, I was joking around but I've also mention my real score wich is 11700 on Maximum :)

680 SLI or a single 680?
Victory requires no explanation. Defeat allows none.
 
credible
Gerbil XP
Posts: 499
Joined: Fri Oct 07, 2011 7:47 pm

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 12, 2013 10:42 am

My boy needed a new card, so I put in a few bucks and got myself one of these:)

FINAL FANTASY XIV: A Realm Reborn Official Benchmark (Exploration)
Tested on:5/12/2013 11:35:23 AM
Score:7220
Average Framerate:60.433
Performance:Extremely High
-Easily capable of running the game on the highest settings.

Screen Size: 1920x1080
Screen Mode: Windowed
Graphics Presets: Maximum
General
-Improve overall graphic quality. : Enabled
-Disable rendering of objects when not visible. (Occlusion Culling) : Disabled
-Use low-detail models on distant objects to increase performance. (LOD) : Disabled
-Cache LOD data only when necessary. (LOD Streaming) : Disabled
-Smooth edges. (Anti-aliasing) : Enabled
-Increase transparent lighting quality. : Enabled
-Grass Quality : High
Shadows
-Use low-detail models on shadows to increase performance. (LOD) : Disabled
-Display : All
-Shadow Resolution : High: 2048 pixels
-Shadow Cascading : High
-Shadow Softening : High
Texture Detail
-Texture Filtering : High
-Anisotropic Filtering : High
Effects
-Naturally darken the edges of the screen. (Limb Darkening) : Enabled
-Blur the graphics around an object in motion. (Radial Blur) : Enabled
-Effects While in Motion : Display All
-Screen Space Ambient Occlusion : High
-Glare : Normal
Cinematic Cutscenes
-Enable depth of field. : Enabled

System:
Windows 7 Ultimate 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_gdr.130318-1533)
Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
8062.680MB
NVIDIA GeForce GTX 660 Ti(VRAM 4042 MB) 9.18.0013.1422

Benchmark results do not provide any guarantee FINAL FANTASY XIV: A Realm Reborn will run on your system.

FINAL FANTASY XIV: A Realm Reborn Official Website http://na.finalfantasyxiv.com
(C) 2010-2013 SQUARE ENIX CO., LTD. All Rights Reserved.

Tweet
http://sqex.to/ffxiv_bench_na #FFXIV Score:7220 1920x1080 Maximum Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz NVIDIA GeForce GTX 660 Ti
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Wed May 15, 2013 6:06 pm

Waco wrote:
VJB wrote:
That's also what I've said, I was joking around but I've also mention my real score wich is 11700 on Maximum :)

680 SLI or a single 680?
The benchmark doesn't support SLI. It's obviously still a falsified result. Sad that some people feel the need to cheat at a pointless benchmark when we're just trying to collect data.
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Thu May 16, 2013 10:24 am

auxy wrote:
Waco wrote:
VJB wrote:
That's also what I've said, I was joking around but I've also mention my real score wich is 11700 on Maximum :)

680 SLI or a single 680?
The benchmark doesn't support SLI. It's obviously still a falsified result. Sad that some people feel the need to cheat at a pointless benchmark when we're just trying to collect data.

Not supported != not working.

I could see 680 SLI getting a score like that if it was working well...but you're probably right.
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Fri May 17, 2013 4:28 pm

Waco wrote:
Not supported != not working.

I could see 680 SLI getting a score like that if it was working well...but you're probably right.
Well, in this case, "not supported" does equal "not working", because it doesn't work for anyone in SLI or Crossfire mode. I read a lot of forums. :3

I'd actually expect a higher score out of SLI 680s, though. I scored almost 10,000 on my OC'd Titan.
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Fri May 17, 2013 7:15 pm

Eh, with almost 9K on a single 7970 I'd say there's something else holding it back. :P
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Fri May 17, 2013 7:22 pm

Waco wrote:
Eh, with almost 9K on a single 7970 I'd say there's something else holding it back. :P

Yah, that's true; it's really CPU-limited.
 
Star Brood
Gerbil First Class
Posts: 189
Joined: Wed Mar 21, 2012 4:57 am

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Thu May 23, 2013 10:43 pm

I got around 4K with my 7850 2GB (stock clocks) and twin Xeon 5160's (~Core 2 Quad 6600). I did max settings and 1680x1050.
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Thu May 23, 2013 10:58 pm

auxy wrote:
Waco wrote:
Eh, with almost 9K on a single 7970 I'd say there's something else holding it back. :P

Yah, that's true; it's really CPU-limited.

One can hope that both new consoles being 8-way parallel will help developers finally see the light of SMP.


One can still...hope...right? :lol:
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Fri May 24, 2013 3:44 pm

Waco wrote:
One can hope that both new consoles being 8-way parallel will help developers finally see the light of SMP.


One can still...hope...right? :lol:
It's not as if developers these days don't want to parallelize things, it's just that some things can't really be solved or even helped by parallelization.

In the end, a simulation can only go as fast as its slowest part, and that means, when you dig right down into it, that single-thread performance is the most important thing for a simulation. Sure, you need to run all parts of the simulation, and massive parallelism allows you to run more complex simulations, but the speed of the simulation is still limited by your single-threaded performance.

Most games are relatively simple simulations, so you don't really need to parallelize the CPU work that much; the graphics part is the complex part, and thanks to the extremely discrete nature of graphics and CPU parts (with respect to the addressing of memory and such) it's almost impossible to have the CPU and GPU work effectively on the same data and task at the same time.

This is why AMD's pushing toward HSA and hUMA, but it remains to be seen how much of those concepts these machines (PS4 and XB1) actually implement. At a very high theoretical level, it would be optimal if the CPU cores can take over when there's some computationally-intense single-thread work to be done for the graphics, and if the GPU cores could take over when there is lots of relatively simple, easily-parallelized work to be done, and I think that's their aim. We can already do this on a machine with discrete graphics and CPU hardware, but it requires some very intelligent programming with foresight into what hardware you're using and exactly what kind of loads you're applying; I suspect AMD's goal is to automate all that.

In the end, I personally wonder if the actual compute throughput of the CPUs and GPUs of the next-gen consoles is really sufficient, even with the benefit of HSA/hUMA and lightweight console-game OSes (though rumor has it the XB1's OS takes up 3GB of memory?!) will be enough to make the next-gen console games significantly more impressive in any way than current-gen games. The machines are so weak...
 
Star Brood
Gerbil First Class
Posts: 189
Joined: Wed Mar 21, 2012 4:57 am

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 10:01 am

I've done a lot of physics engine programming as a hobby and I just have to say, there's no excuse to not take advantage of multiple cores. At the very least you can just thread your more complex physics codes and not have so many operations per thread.
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 3:02 pm

auxy wrote:
It's not as if developers these days don't want to parallelize things, it's just that some things can't really be solved or even helped by parallelization.

I'm a parallel software dev. :P Some things can't be solved or helped, no, but most game engine tasks are pretty easy to parallelize. Most developers (publishers and bosses really) don't care about getting the engine to run faster as long as it's "fast enough" though.
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 3:30 pm

Waco wrote:
I'm a parallel software dev. :P Some things can't be solved or helped, no, but most game engine tasks are pretty easy to parallelize. Most developers (publishers and bosses really) don't care about getting the engine to run faster as long as it's "fast enough" though.
"Most" game engine tasks outside of graphics and physics don't need to be parallelized, which was my point. (I guess you could also include pathfinding in that.) Sure, you CAN, but from what I have come to understand, it's not really helpful and generally more trouble than it ends up being worth.

My point was twofold; one, I don't think that expanding CPU parallelization is going to really make that much of a difference in the overall quality or capability of games moving forward. I know Cryptic, the developers of Champions Online, Star Trek Online, and the recent Neverwinter (Online, although it's actually not part of the title) games, said that they only run one game thread and one graphics thread because running more doesn't make their game run better. I don't know how true that is (seems like you could run at least a couple of graphics threads), but I do think it echoes the sentiments of a lot of developers. The Dolphin emulator team says that their emulator runs 3 threads at most (4 if you run low-level sound emulation in its own thread, but that's foolish) because there's just no use for more.

Leading on from that, then, the second point; I don't think the "next-gen" consoles are going to be all that impressive. I think they're going to peak (in terms of shiny graphics) really early due to the developer-friendly x86/GCN hardware (and the overall lack of utility of eight slow processors for gaming), and then people will lose interest in actually gaming on them. This is one place where I think Microsoft actually has the advantage, and I think in the end the XB1 might possibly end up being the more successful console financially simply because they're hedging their bets on the hardcore console gaming crowd losing interest in gaming and using the machine for other things.

Anyway, this is all offtopic to the thread, and it's possible I'm completely wrong about all of this -- I'm running on logic derived from hearsay and speculation after all, because I'm neither a game developer nor a parallel software developer.
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 3:33 pm

auxy wrote:
"Most" game engine tasks outside of graphics and physics don't need to be parallelized, which was my point.

If you have a fast CPU, sure, that's true.

If you have 8 low-power CPU cores you kinda do. :)
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 3:36 pm

Waco wrote:
If you have a fast CPU, sure, that's true.

If you have 8 low-power CPU cores you kinda do. :)
Yah, maybe so. I wonder how much benefit they can really get out of it, though. Amdahl's Law and all.
 
Star Brood
Gerbil First Class
Posts: 189
Joined: Wed Mar 21, 2012 4:57 am

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 3:48 pm

Waco wrote:
auxy wrote:
"Most" game engine tasks outside of graphics and physics don't need to be parallelized, which was my point.

If you have a fast CPU, sure, that's true.

If you have 8 low-power CPU cores you kinda do. :)


To echo this, I have programmed a lot for WarCraft 3. Its engine is limited to 1 running CPU core (no parallel threads) and each thread has an operation limit of 30K. In-order execution at its worst. Hitting this thread limit results in a thread crash with no catch functionality, a lot of coders have scratched their head at what happened to half their engine mechanics. If there was a way to break the war3 engine out and use more cores then the performance issue wouldn't be just so plain terrible. The fact that developers are getting paid to program and they choose not to break up their code is a slap to the face - I would have given good money to have an improved engine on that game.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 4:08 pm

Star Brood wrote:
To echo this, I have programmed a lot for WarCraft 3. Its engine is limited to 1 running CPU core (no parallel threads) and each thread has an operation limit of 30K.
That's scripting, not programming; you're just writing a script for an ancient game engine to process. The operation limit is just a limitation of the engine, not of single-core processing; this has nothing to do with parallelization, really.

Besides, keep in mind that when War3 came out, there was no such thing as a dual-core processor. Nobody had SMP, why would they have parallelized it?
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 4:11 pm

Star Brood wrote:
The fact that developers are getting paid to program and they choose not to break up their code is a slap to the face - I would have given good money to have an improved engine on that game.

They're getting paid to make things work.

If a single process/thread is fast enough for a given task on most machines they don't have a whole lot of reason to make it SMP-aware.


That's why I'm hoping the Jaguar cores just suck...:lol:
Victory requires no explanation. Defeat allows none.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 4:22 pm

Waco wrote:
They're getting paid to make things work. If a single process/thread is fast enough for a given task on most machines they don't have a whole lot of reason to make it SMP-aware.
Yep!
Waco wrote:
That's why I'm hoping the Jaguar cores just suck...:lol:
Did you see codedivine's comparison of Jaguar and Stars? Pretty enlightening; we were discussing it in another thread. Summary: VERY impressive given the power envelope, but overall not so great. Trying to play even current-generation serious games on a Kabini would be a joke; I don't see how the next generation is going to manage...

And Microsoft now says the Xbox One should be "fully capable" of gaming at 4K...
 
Waco
Maximum Gerbil
Posts: 4850
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Sun May 26, 2013 4:36 pm

Yeah I saw that. :) That's why I'm hopeful for the SMP-ness of newer games.
Victory requires no explanation. Defeat allows none.
 
Star Brood
Gerbil First Class
Posts: 189
Joined: Wed Mar 21, 2012 4:57 am

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Tue May 28, 2013 11:14 am

auxy wrote:
this has nothing to do with parallelization, really.


I mentioned the OP limit on a thread because it made us need to branch our code out to multiple threads. A projectile system, for example, would crash the thread between 20-100 projectiles depending on how many physics are applied to the projectiles. I was using the OP limit of an example when we've needed to break things out into multiple threads, and as a testament to the opportunity for multiple cores to come into play. There's no reason why something like a projectile system's operations couldn't be split over multiple cores.

There are plenty of areas where this would improve performance in games - you mentioned pathing algorithms earlier. Pathing calculations apply to almost every game, but are almost always running on a single core despite the massive number of calculations needed to predict where a unit might go. Some developers may claim that it doesn't improve performance, but they must not be running the right simulations if they really think that. A single CPU in their "recommended system specs" should never be hitting 100% unless the remaining cores in that recommended CPU are running a substantial amount of subroutines already.
 
LostCat
Minister of Gerbil Affairs
Posts: 2107
Joined: Thu Aug 26, 2004 6:18 am
Location: Earth

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Wed May 29, 2013 12:33 pm

auxy wrote:
Yep!
Waco wrote:
That's why I'm hoping the Jaguar cores just suck...:lol:
Did you see codedivine's comparison of Jaguar and Stars? Pretty enlightening; we were discussing it in another thread. Summary: VERY impressive given the power envelope, but overall not so great. Trying to play even current-generation serious games on a Kabini would be a joke; I don't see how the next generation is going to manage...

And Microsoft now says the Xbox One should be "fully capable" of gaming at 4K...

On a side note, the GPU core on the X1/PS4 does not seem to be the same one on Kabini, and since AMD claimed a while back that they'd seperated their CPU and GPU core engineering they might indeed not be (at the very least the clock speeds are different, I don't know about the rest of it.) Kabini is also running with plain DDR3, where X1 has the SRAM and PS4 has GDDR5. I don't know if any other aspects of the consoles are different from Kabini but I wouldn't be surprised.
Meow.
 
auxy
Graphmaster Gerbil
Topic Author
Posts: 1300
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Mon Jun 03, 2013 9:02 pm

Savyg wrote:
On a side note, the GPU core on the X1/PS4 does not seem to be the same one on Kabini, and since AMD claimed a while back that they'd seperated their CPU and GPU core engineering they might indeed not be (at the very least the clock speeds are different, I don't know about the rest of it.) Kabini is also running with plain DDR3, where X1 has the SRAM and PS4 has GDDR5. I don't know if any other aspects of the consoles are different from Kabini but I wouldn't be surprised.
Well, if you look at the benchmarks codedivine did, they were comparing CPU performance (thus "Jaguar vs. Stars", not "Kabini vs. Llano") specifically and that was my concern (since the topic of the thread has been the CPU-heavy nature of this benchmark.)
 
LostCat
Minister of Gerbil Affairs
Posts: 2107
Joined: Thu Aug 26, 2004 6:18 am
Location: Earth

Re: Final Fantasy XIV: A Realm Reborn Official Benchmark

Mon Jun 03, 2013 9:18 pm

auxy wrote:
Well, if you look at the benchmarks codedivine did, they were comparing CPU performance (thus "Jaguar vs. Stars", not "Kabini vs. Llano") specifically and that was my concern (since the topic of the thread has been the CPU-heavy nature of this benchmark.)

I don't think the consoles are as light as they look on CPU considering they're custom procs and you're not likely to have AAA stuff optimized for dual core/lower amounts of RAM. At least one of his benchmarks are optimized for dual core from what I can tell.
Meow.

Who is online

Users browsing this forum: No registered users and 1 guest
GZIP: On