The Official Graphics Card and PC gaming Thread

The HDD area is very cool because the drives have one of the giant 230mm fans to themselves. As for passively cooled chipset, you make it sound like there were actively cooled chipsets! LOL!
The VF1000 has a 2500rpm 11-blade 80mm fan fitted. At full speed, this is going to be a noisy fan. The VF1000 does come with a FanMate, however, and the rated minimum speed when using the fan mate is 1400rpm, which should hopefully be sufficient to cool the 4870 (I can't guarantee this, but I expect it will be) and will be a little more acceptable for noise.
The fact that the fan is 11-blade means it will produce a higher pitched noise than the Sapphire 4870 cooler.
You never need more than one Crossfire bridge connector for each join. There's a reason why only one bridge connector is fitted to an X2
Cheesepie: Game hasn't been leaked yet, so I couldn't give any estimates. Depends on the sort of graphics settings and performance you're after. I'm sure the game will run, but how well is yet to be seen.

 

I love the drive cooling and I especially like that the drives mount with rubber bushings. That was an extremely thoughtful touch to include it stock.

The one on my 680i board was and it still ran hotter than all get-out

Does the VF1000 include memory sinks? The MEMIO(Memory) temp in GPU-Z on the top card idles at 70-ish and loads near 95. I know th cards are supposed to run hot but isn't this a bit toasty? I think aftermarket may be in order :S

 

The rubber bushings were a nice touch, but they're completely useless, the drive vibration in the HAF is no better than the NZXT Lexa, which just mounts using plastic. Contrastingly, the NZXT Whisper which uses thick rubber bushings has almost no vibration at all. I can just about hear the 90Hz from the nine drives that produce it.
As for the 680i board, special case, they used more power (and thus produced more heat) than the CPU in such systems. Not joking, the TDP of the E6600 was 65W, the 680i I believe was 67W. The 780i I think was the highest at 70 something. Typical passive NB temps for the 780i were in the eighties celsius from what I heard.
As for the VF1000, pretty sure it includes memory sinks, but I thought MEMIO was part of the GPU, not the temperature of the memory chips themselves. My MEMIO temps reach three figures just about, and it's not caused me any issues.

 

They're not perfect, but needless to say they are a large step up from my previous setup which did not suspend the drives on the rubber, only dampened the vibration a bit.

Glad I got rid of it fast for a proper Intel chipset Gigabyte board

Woo comforting to know someone else has that. Playing Crysis I've seen mine hit 100. Anything else hits about 95.


Seeing better performance from BC2 than I expected. Running full Dx10 maxed with all settings cranked with no AA seems smooth. I'll be back with performance numbers later.

 

This I wanted to find out, as I could not differentiate between it being DX10 or the Anti-Aliasing that was ruining performance. I would have expected DX10 personally, but if it's the AA, maybe there's hope after all.

The MSRP has been stated for the 4GB HD5970 cards and all partners will be selling them at $999 or about £825. Looks like I'm not buying one after all

Plan J I think it is, I'm going to buy a solo 5870 to put with my Q9550 so I get better single-GPU performance for when crossfire isn't working. Then as far as my gaming PC goes, I will have to stick with 4870s for the time being, perhaps until autumn time. DX11 stuff is outrageously expensive due to nvidia's incompetence...

 

so what card would one put on top?

(considering my mobo lays flat due to my case)

 

It probably makes little difference, but I'd stick the Zalman card at the "top" because it relies on dumping the heat into the case, and you don't want a card that does that right next to your PSU in my opinion, and secondly the air may be more cramped at the bottom and rear-exhaust coolers are better at dealing with such situations, so the Sapphire would be better off at the bottom.
This of course assumes your PC is laid out as in the picture. Is that your PC?

 

lol no thats the PC of the guy thats selling me the card.

 

Just Cause 2 Demo
Minimal: Radeon HD2900 Pro/HD3800 series/HD4670/HD5570 or higher, Geforce 8800 series/9600 series/GT220 or above
Reduced: Radeon HD3870/HD4700 series/HD5700 series or higher, Geforce 8800GT/9800GT/GT240 or above
Moderate: Radeon HD4870/HD5800 series or higher, Geforce GTX260 or above
Good: Radeon HD5870 or higher, Geforce GTX295 or above, GTX470 (estimated)
Optimal: Radeon HD5870 CF or higher, Geforce GTX280 Tri-SLI or above, GTX470 SLI (estimated)
Extreme: N/A

Just Cause 2 Demo - PhysX enabled
Minimal: Geforce 8800 series/9600GSO G92/GT240 or above
Reduced: Geforce 8800GTS G92/9800GTX/GTS250 or above
Moderate: Geforce GTX275 (not 280) or above
Good: Geforce GTX295 OC or above, GTX470 SLI (estimated)
Optimal: Geforce GTX285 Quad-SLI, GTX470 Tri-SLI (estimated)
Extreme: N/A

Depressingly, another game slightly optimised for nvidia cards isn't able to make the 'Extreme' criteria even with estimated figures for GTX480 Quad SLI taken into consideration (6.15x GTX260-216).

 

Do any of you guys play MKV 720 files with MPC Home cinema 1.3.1249.0, and have strange kind of frame lag? What I mean is, it'll be smooth for minutes at least, then it stutters a great deal, then its fine again. Cpu usage is negligable, ram usage seems to be high at 75%+. Not sure why it would need so much. Perhaps they figure video rendered from ram, is quicker, then that of a HDD. Which is logical. Surely the Ram usage is not the problem...
Any and all opinions welcome

 

I play 720p MKVs with HomeCinema, but haven't run into this issue, and the program rarely uses any more than 70MB, not 3 gig!
However, I'm using 1.2.1081.0.

 

Noted. I may try an older version tomorrow now. I shut down the program, which freed up over 2Gb of ram. I know, I'm in awe about this too. Now I'm hovering around 27% again. Really weird...

Even with the ram usage down, it didn't help. And yet, all the files play fine on another player. Arcsoft Total Media Theater.

 

I installed the K-Lite codec pack with Home Cinema built in. Then I replaced the Home Cinema exe with the regular version exe renamed to match. Home cinema seems to have infrequent issues with certain file formats where normal has no issues at all.

 

The best program I've seen yet IS MPC. It allows control of multiple audio channels, and Subtitles in a FILE. Where as Total Media Theater seems to have zero control over the audio channels when there are more than one. Unless of course its a BD you're playing. I certainly like Total Media Theater when the Files only have one audio channel Smooth playback, zero stutters across the board.
Thanks for your input Jeff

 

Bad Company 2. Very interesting results. As long as I don't use AA it seems to be very playable.

Performance: 7.5/10, 6/10 w/ AA

Single card:

All maxed Dx10 1920 x 1200, no AA, HSAO/HBAO on. Getting 40-ish FPS with dips into the 20s. Drop to 1680 x 1050 I average 55-60-ish dipping into the high 30s.

Dual card:

Same settings. Getting 65-70-ish average(highly variable) with dips into the high 40s, low 50s. GPU-Z is showing about 80% usage on the second card. Only using 2xAA my FPS drops to 50-55 average with dips into the low 40s, high 30s.

AA seems to be our killer here. Performance is actually decent with maxed settings and Dx10. AA is doable, but only with very high end setups. At my res I think 5850s could do it just fine.

Interesting to note. AA doesn't work without Dx10 on. Dx9 gives you 4 or 5 extra FPS with nearly the same visuals. So if you want a few FPS and don't intend to run AA, Dx9 seems to be acceptable. Dx10 seems to only have some better smoke/dust effects and some very minor lighting effects.

Overall it runs better than I had hoped, but it still doesn't run as well as the engine is apparently capable of. The option to turn Dx9/Dx10 on and off in the actual menu would have been nice. Right now you have to edit an INI file.

That it takes advantage of Crossfire is appreciated, but the AA performance is dismal.

Graphics: 9/10

The graphics are good. Some minor parts of the environment have a few blurry textures but overall the game is very high-res friendly. The models are some of the best I've ever seen and the character textures challenge Crysis. The overall look of the game is very reminiscent of Crysis. Namely the foliage looks identical and it uses similarly high res characters and vehicles, similar vehicle/interaction mechanics. The guns were a special treat for me. Not seen gun models quite this clean very often. Only MW2 matches BC2 in the detail on the guns and sheer variety. I won't include AA in my rating because that's more performance related. The actual AA effect works well.

That said the game looks fantastic. There are a few slightly rough spots here and there, but overall I'm quite impressed.

Sound: 10/10

The sound design is my favorite part of the game. You can tell the devs worked very hard on it. For once a game has managed to do sound occlusion and reverb properly. Firing a gun inside a building is markedly different to firing it outside. The audio is noticeably higher bitrate than your average game and the detail and directional effects are perfect. I actually scored a kill through some bushes last night. I couldn't see him, but I could hear him perfectly

Oh yeah, the guns sounds are perfect as well

Gameplay: 10/10

The first Bad Company had a terribly mediocre campaign. I felt burdened playing it. You can tell they've worked MUCH MUCH MUCH harder on it this time around. The result is a game that takes many cues from others like Crysis, CoD4/MW2, Drake's Fortune and the like. This isn't bad though because it actually manages to do everything pretty well. The campaign is action-packed and the story(while nearing the edge of my ability to suspend my disbelief) gives you good motivation to slug on. My only real complaint is that the story is only 5-6 hours long. Maybe 7 is you take your time. I beat it in one sitting.

But that's a minor complaint as the multiplayer is the real focus here. It pays heavy homage to its roots and revisits a few older maps and concepts. IMO the gameplay and map design challenge BF2 and the advancement system s a bit more streamlined. I dare say I think I've found a new favorite multi game

Mechanics. First-off the shooting mechanic slaughters the one in BF2. The guns are easier to aim, the hits feel more solid, and everything is a bit less cumbersome. They've allowed you a bit more health this time around so the famous insta-death from BF2 is less likely to happen. All the vehicle controls have been refined and all of the vehicles handle better than any Battlefield before it. Seriously, play some classic Conquest mode and you'll see it lives up to the Battlefield name for sure

Oh yeah and fully destructible buildings in large-scale multiplayer are sweet. That is truly a ground-breaking feature.

Overall: 9/10

This is Battlefield reborn. Everything has been improved in every way possible. Other than some small performance and graphics issues here and there, I have zero complaints about the game.


If you count performance a bit more meticulously than me you might rate it a bit lower.

 

im gettting about 50 avg everything on the lowest details, dx9 no aa/af at 1920 x1200.

seeing your results, i cant wat to jump to CF and a Quad core! me= happy

i just hope this case doesnt overheat. if it does looks like the case upgrade im planning along with the corsair H50-1 watercooling kit will come very soon

it truely is a fantastic game though, and one worth the upgrade. loving it more than i liked MW2, not sure if it tops COD4/2 for me btu then this is ALOT more realistic and squad based, rather than the other 2.

BTW i cba withthe SP how are the frames on that jefF?

 

Shaff: If necessary, just stick a really noisy fan in there for the interim, sort the case out later.
Jeff: I did wonder about DX9/DX10, but yeah I already knew you can't use AA without being in DX10. I will try the game without AA to see how well it runs.
As for graphics, I'll look again, but I found the SP graphics very disappointing, easily beaten by the textures and effects in COD4, the game looked very antiquated to me. But I didn't play for very long due to the lag.

 

seriously cod4?

ouch, the MP looks fantastic! seriously breathtaking graphics. esp on the jungle maps. though i have foudn when sunlight goes between tree leaves, if there is a rock itll go through that too lol.


and i have everything on low yet it textures on the floor keep switching between a decent level and fail textures. i want it to be constant fail ones to i get a FPS boost. dont know if thats just me though,

 

Played it through again briefly, and after the intro scene the graphics are much, much better. Reminiscent of Crysis in style. They're better in some parts, worse in others.
Microstutter does almost entirely eliminate the game's immersive feel, as you can't help but notice it as your character jerks and lunges along the path, but the frame rate itself, with AA off never seems an issue.
Attached a frame rate graph. For some reason, the intro text with 'this is what really happened that day' is limited to 8fps.
CPU-wise, the game is using 35-45% of my Core i5, so if this scales linearly I imagine it would use 80-100% of a stock Q6600.
GPU-wise, the game uses between 260 and 290% of my GPUs, with fan speeds adjusting to 80% and 60% for the two cards.
The frame rate graph dips at the end sharply. This isn't an artifact of the benchmark, the game did lag quite noticeably when a huge explosion went off in my vicinity. This drop can therefore be seen as 'typical' of when action heats up.
The huge spike at the start is when the menu was opened.






In other news, played Aliens vs Predator briefly (or tried to) - the benchmarks are accurate. Worse, the game requires an absolute 60fps minimum frame rate, as even at 45fps it runs horrifically. The tragic thing is, I get 41fps at 1920x1200 in the starting room, and GPU scaling is an astonishing 390%.

 

Haha the intro scene is horrific. The rest of the game I thought was quite good looking though. The character models especially are downright impressive I think Crysis did foliage a bit better but the overall graphics standard is right up there.

On some systems there is a slight microstutter and maybe even some minor input lag. Try turning HBAO/HSAO off. This has worked for some.

Sam, try adding "+fullproc" at the end of the target line. This worked for some in the beta. I'm using it too though not entirely sure if it's working or not. Not a magical fix but it might help in your case seeing as it isn't using the CPU fully nor the cards. If the cards are scaling so badly maybe it's because they're CPU limited, seeing the low usage.

Hmm I suspected as much. Scaling isn't exactly great for me either but better than that I think. I get about 160-170% scaling.

I think it was already agreed that AVP is a steaming pile. But hey, at least it uses Crossfire, LOL.