Street Fighter 4 PC Benchmark Released

[Seika]

http://www.shacknews.com/onearticle.x/59143

Download Links: http://www.eventhubs.com/news/2009/jun/14/pc-street-fighter-4-benchmark-tool-released/

Torrent: http://www.demonoid.com/files/details/1959793/2325666/

Whoa it looks nice, this is a first day buy for me.
My friend was running it at 300FPS on a Quad-Core with a GTX280, everything at Max.:eek:


Some one with a good rig post screenshots! (Don't have photoshop on Win7 :()

I stole these from another thread....
http://img44.imageshack.us/img44/709/streetfighterivbenchmarj.png

Ok, my results with everything at max.
http://img41.imageshack.us/img41/4267/streetfighterivbenchmarves.jpg

[hunterx280]

Downloading to my laptop now. This should run just fine, I believe. I want to give it a test run to make sure.

[H.Bogard]

Same as the Devil May Cry 4 benchie they released.

I'm getting ~150 frames per second. :eek:

[JRR1285]

http://img34.imageshack.us/img34/2097/notenoughpower.jpg

New video card is needed I think, now I have a reason. lol

I just realized I built this computer almost 2 years ago... it has served me well.

[hunterx280]

My bench keeps crashing on WIN 7 every time I change a setting. Anyone else run into this problem? I'm using a 9600M GT on my laptop but I need to update to the latest drivers that came out a week ago. I wonder if that's the problem. Sadly I'm at work and I don't want to download a 136MB file over the work network that isn't related to my job.

[JRR1285]

Tell them it would help your computers stability and increase productivity.;)

[hunterx280]

Tell them it would help your computers stability and increase productivity.;)

Sadly they only use the Intel GMA onboard video cards. I have someone with a wireless air card who might be willing to download it for me. It'll just take a little while. Less time then it will if I had to wait until I got home. That's another 7 hours :D

[muddi900]

The menu is borked for under win 7. It does not recognize any mouse input and when I use the keybard, it keeps looping(the menu, not the test)

EDIT: Restart made it work. Benchmark is weird as I ran the test with the same settings multiple times and it gave me different results. Saved only one of them:

SCORE: 7422
AVERAGE: 71.54FPS
OS: Windows 7 Ultimate
CPU: Intel(R) Core(TM)2 Duo CPU E7200 @ 2.53GHz
Memory: 2046MB
Graphics Card: ATI Radeon HD 4800 Series
Display Setting: 1440x900 60Hz 4xAA

(c)CAPCOM U.S.A., INC. 2008, 2009 ALL RIGHTS RESERVED.


Screens:
http://i385.photobucket.com/albums/oo299/muddi900/StreetFighterIV_Benchmark2009-06-1.jpg

http://i385.photobucket.com/albums/oo299/muddi900/STIV.jpg

http://i385.photobucket.com/albums/oo299/muddi900/StreetFighterIV_Benchmark2009-06-16.jpg

[hunterx280]

I just updated to the latest Nvidia drivers for Win 7 and this app is borked. Any changes to the settings cause it to crash. I hope the full version works better on release.

[kyrieee]

New video card is needed I think, now I have a reason. lol

I just realized I built this computer almost 2 years ago... it has served me well.

Nah, you don't need really AA at that resolution and you barely notice it in this game anyway. Don't use QAA though, if you want to max out go for 8xFSAA. If you have the old GTS you might want to upgrade though just because there are so damn many good GPUs at good prices at the moment. 'At the moment' being the last 18 months lol =D

I still have a 8800GT which works fine for most games since I'm not a resolution whore and use a CRT, but damn if I'm not tempted to buy a new GPU just to get 80-ish FPS in games. But I don't need it. I just need to convince myself of that =D

This benchmark was nice. A huge improvement over Capcom's previous PC efforts which I've all ragequit after 5 minutes

[total]

Mmmmmm....must test the SF4 box when I get home.

[OrangePulp]

Taito Type X²

* OS: Microsoft Windows XP Embedded SP2 [3]
* CPU: Intel Socket T CPU. Supported CPUs include Celeron D 352, Pentium 4 651, Intel Core 2 Duo E6400
* Chipset: Intel Q965 + ICH8
* Video output: 640x480 (VGA), or 1280x720 (HDTV 720p)
* RAM: 166/200MHz DDR2 SDRAM. Supported capacities 512MiB, 1GiB, 4GiB.
* GPU: PCI Express x16-based graphics. Supported GPU include ATI RADEON (x1600Pro, x1300LE) or NVIDIA GeForce (7900GS, 7600GS, 7300GS)
* Sound: Onboard Realtek HD 7.1 channel Sound (supports add-in sound cards)
* LAN: 1000BASE-T 10/100BASE-TX
* I/O ports: 1x JVS, 4x USB 2.0, 1x serial (max 2), 1x parallel port, 2x PS/2, 2x SATA
* Audio inputs: AKG C535EB Stage Microphone, line-in (Surround 7.1)
* Audio outputs: 7.1, SPDI/FX
* Expansion Slots: 1x PCI Express x16 (used by video card), 1x PCI Express x4, 2x PCI
* Storage: SATA 3Gbit/sec Hard Drives

As the name implies, Type X² is a natural evolution of Type X and X+ that allows the use of newer PC technologies introduced after the unveiling of its predecessors. Like its predecessors, the specs of Type X² are comparable to that of a contemporary PC.

Those are the specs of the arcade cabinet that runs SF4, so I'd imagine the game would run easy on all modern systems. Considering the original game is basically a PC game, it annoys me that it's taken them so long to get it out on PC.

[Kenzington4short]

Sigh. Ran this last night and it ran like crap on my laptop, even after turning everything way down. I'm a little surprised because I thought I was above the recommended specs. Looks like I may have to cancel my order. So much for having a portable SFIV machine. That would've been so awesome.

[Lance Uppercut]

Sigh. Ran this last night and it ran like crap on my laptop, even after turning everything way down. I'm a little surprised because I thought I was above the recommended specs. Looks like I may have to cancel my order. So much for having a portable SFIV machine. That would've been so awesome.

I'm having some trouble running this on my brother's new laptop as well, which is strange considering that it's more powerful than my desktop.

[DeathtollWRX]

downloading now..
Time to whoop!

[DeathtollWRX]

Here's mine
Radeon 4890
virus scanner on
steam on
http://img134.imageshack.us/img134/5655/sf4f.jpg
Shot at 2009-06-16

the fight scenes were all 230FPS plus
the scene with all the cars was 120 plus...

with MAX settings I got
SCORE: 9681
AVERAGE: 110.44FPS
OS: Windows XP Professional
CPU: Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz
Memory: 2046MB
Graphics Card: ATI Radeon HD 4800 Series
Display Setting: 1680x1050 60Hz 8xAA

(c)CAPCOM U.S.A., INC. 2008, 2009 ALL RIGHTS RESERVED.

[mesh]

Whoa it looks nice, this is a first day buy for me.


I'd change that to 'Whoa it looks nice to those that like cartoon shaders'. I don't and it looks like crap to me.

Shame too, the pc ain't exactly overloaded with decent fighters.

[DeathtollWRX]

I'd change that to 'Whoa it looks nice to those that like cartoon shaders'. I don't and it looks like crap to me.

Shame too, the pc ain't exactly overloaded with decent fighters.

This would be the ULTIMATE arguement for Gameplay over graphics


btw the graphics rock

[Lance Uppercut]

I'd change that to 'Whoa it looks nice to those that like cartoon shaders'. I don't and it looks like crap to me.

Shame too, the pc ain't exactly overloaded with decent fighters.

You haven't played SFIV, have you?

[J Arcane]

I ran three tests with the benchmark, on my Athlon X2 3800+/Radeon 4670 machine.

Test 1 was on complete default settings, with V-Sync on. Kept a rock solid 59.xxx the whole time, no noticeable slowdown.

It was during this test that I realized V-sync was on, so for Test 2, I kept all settings the same, but turned that off. 77+ FPS average, A Rank, 12,000+ points.

For test three, I cranked everything up. All settings on the highest values, highest AA and AF, the works. Ran about 30FPS on average, looked great, frame rate consistent and the game still looked smooth.

All these results were fairly predictable to me, given my hardware and such.

What wasn't predictable however, was the game speed for each of these tests. SFIV seems to directly tie the speed of the game to the current frame rate. Higher frame rate, faster game. So the fights in my Test 2 were over faster than Test 1, which were over faster than Test 3 (which were so slow as to cause occasional issues with the video lagging behind the audio, Hadoken ending it's sound before the animation and such).

I was surprised, for one thing because I don't think I've ever seen a game do this, at least not since the early days of Turbo buttons and things. I don't see how multiplayer will work unless it defaults the game speed to the frame rate of the slowest player, and over all the changing in game speeds would get jarring as fuck in that case. Outside of online, V-sync would seem the best way to keep the game speed consistent, so long as your rig can consistently push the 60fps needed.

IS there some setting I missed or something that fixes this, or was this intentional?

[Seika]

The benchmark seems to be very dependent on CPU clock speed.
On my PC at least, when I pushed the clock speed of CPU higher, I got a substantial increase in FPS.
That's probably why people with laptops are not getting the desired results.

@J Arcane
That happened to me too.
Weird, it's like they used old style game programming, but I can't believe that. :confused:
Now I'm more interested on how the MP will work.

[Panthera]

It was probably done like that for... wait for it... benchmarking.

[Ancalagon]

@J Arcane

If its a benchmark its probably intentional. If they are stupid enough to have that in SP and MP however....

Running the game faster like that might be a good way of making sure that the CPU and GPU are both equally taxed.

[Jackel]

How big is the benchmark download?

[Seika]

136MB (too short blah)

[Jackel]

K. thats not bad, I'll grab this and see how my machine handles it.

[kyrieee]

What wasn't predictable however, was the game speed for each of these tests. SFIV seems to directly tie the speed of the game to the current frame rate. Higher frame rate, faster game. So the fights in my Test 2 were over faster than Test 1, which were over faster than Test 3 (which were so slow as to cause occasional issues with the video lagging behind the audio, Hadoken ending it's sound before the animation and such).

IS there some setting I missed or something that fixes this, or was this intentional?

It's intentional and it's standard procedure in benchmarks
You want to make sure that the benchmark renders exactly the same thing and the same number of frames every time. It's not indicative of the game

[H.Bogard]

It's intentional and it's standard procedure in benchmarks
You want to make sure that the benchmark renders exactly the same thing and the same number of frames every time. It's not indicative of the game

I can see the reasoning but I don't agree with it being 'standard procedure' though.
I've been whoring to 3DMark and NV/ATI techdemos for 6 years now.

[J Arcane]

I can see the reasoning but I don't agree with it being 'standard procedure' though.
I've been whoring to 3DMark and NV/ATI techdemos for 6 years now.
Yeah, I was gonna say, I don't spend much time with benchmarks but I can't say as I recall one that did what SFIV does with the game speed.

[kyrieee]

I can see the reasoning but I don't agree with it being 'standard procedure' though.
I've been whoring to 3DMark and NV/ATI techdemos for 6 years now.

Okay, I should say game benchmarks then :p
Most of the in-game benchmark I've done has behaved this way.
The Unreal engines do it (the ones I've tried), id's engines do it, the Cryengine does it and I'm sure I could list more if I thought about it for a while. Even if it's not standard it's a very common practice

[H.Bogard]

Okay, I should say game benchmarks then :p
Most of the in-game benchmark I've done has behaved this way.
The Unreal engines do it (the ones I've tried), id's engines do it, the Cryengine does it and I'm sure I could list more if I thought about it for a while. Even if it's not standard it's a very common practice

Cryengine wasn't running above normal speed, if I recall correctly. Though it does have a frame dump option, but I think that's for when its running really slow.

Anyway, semantics. Bah to it! Anyone know when a playable demo will be out for this?

[Wackman3000]

Here's me:

http://i3.photobucket.com/albums/y71/Accordman/Clipboard01-6.jpg

[kyrieee]

Cryengine wasn't running above normal speed, if I recall correctly. Though it does have a frame dump option, but I think that's for when its running really slow.

Anyway, semantics. Bah to it! Anyone know when a playable demo will be out for this?

Not to keep this going :p but the reason you didn't notice is probably because Crysis doesn't go fast enough to notice on most machines hehe

But here's a fast benchmark http://www.youtube.com/watch?v=QGqEp9irDuE

[violent]

SCORE: 7789
AVERAGE: 87.82FPS
OS: Windows XP Professional
CPU: Intel(R) Core(TM)2 Duo CPU E6850 @ 3.00GHz
Memory: 3582MB
Graphics Card: ATI Radeon HD 4800 Series
Display Setting: 1680x1050 60Hz 8xAA

Thems my current score though I could probably clean a few things. This is everything at max for reference.

[Esquilax1138]

SCORE: 6056
AVERAGE: 57.42FPS
OS: Windows XP Home Edition
CPU: Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
Memory: 2046MB
Graphics Card: NVIDIA GeForce 8800 GTX
Display Setting: 1920x1200 60Hz 2xAA

Not bad for an older rig, and 2xAA is good enough at 1900x1200 with everything at max. If I go with no AA it avgs 95 fps. And it looks amazing, just waiting for the pre-orders to open up on Steam to give them my money.

[Jackel]

After looking at the size of the file (400 mb) I'm going to not do this. Stupid low bandwidth.

[total]

And it looks amazing, just waiting for the pre-orders to open up on Steam to give them my money.

Wait it is getting a Steam release? Well fucking right on.