[QUOTE="Blue-Sky"]

Can you link your PSU? Or exact model name.

Hard to imagine a 750W can't accomadate 2 PCI-E connectors.

shred

Actually he needs four PCIe 6 pin connectors since each card uses two 6 pin connectors.

Can you link your PSU? Or exact model name.

Hard to imagine a 750W can't accomadate 2 PCI-E connectors.

Blue-Sky

Mobo:

Ideal would be getting a Z68 motherboard, but a P67 is basicly the same just without graphic a connectors for the IGP on the CPU.

MSI's P67/Z68 motherboards have had good ratings from reviewers, and when it comes to service/support/RMA they are friendly and helpfull with is a good selling point in my opinion.

(In that I can tell that the folks at ECS dont even know their own products, it's tough to get help from the support when they dont even know what a motherboard is... ...)

Case:

As I always do... I would like to point out the thing most people don't care squat about: How lovely it's to have a fully capable gaming rig being dead silent when your not playing games!

To make that happen you'll need to buy a case without mesh all over, the HAF and most other coolermaster case's are therfor out of question, the fractial design cases are mostly all very quiet and are very slick looking too, the Fractial Design XL if you want a fulltower or the Fractial desing R3 if you just need a big midtower.

A CM212+ have no trouble fitting in these two cases, Ideal I would get the FD XL fulltower because of it's great looks and it's large roomy workspace.

Fratial's fans are very silent but dont move a ton of air, but the cooling capabilety of these cases are in the top segment.

PSU:

I am not quite but nearly a corsair fanboy atleast when it comes to PSU, but stick to the OLD series if you can find one (vx/CX/HX/TX) and rather get those over the V2 of the very same PSU. (Some verions are "refined" but with less quality components! O.o )

Skip the "gamer" versions if you want the very best quality.

This is the old verion and a really great PSU. (@ newegg) It will run your rig without the slightest of problem when you go berserk with the voltage and trying to bend the rules of what is possible with a HD6950 and a 2500k.

How are the new drivers ? anyone updated them yet.

xeffectx

I have a problem with my Power Mac G5. Everytime I press the power button I hear the drives boot-up, and the fans boot-up, but nothing shows up on the monitor. If I let it continue the fans get really loud, and still nothing shows up on the screen. Can anyone helpRetro_gamer0123

[QUOTE="swehunt"]

[QUOTE="ferret-gamer"]Witcher 2 doesn't use tessellation, having that option enabled or disabled would be irrelevant. And as he mentioned right in the original post he plays on 1400x900, A low resolution.

Here is a chart showing the performance differences in Witcher 2:

Now 1400x900 is a lower resolution than 1280x1024, so does his claim really seem that far fetched?

ferret-gamer

Sorry ofc. It does not use tesselation don't know what I were on about? :? Anyway: I've been trying to replicate a bit above HD6870's performance while recording with fraps, the settings were all max'ed out with the gtx460 OC'ed to 900 on the core and much less agressive OC to the memory the phenom2 X4 is no bottleneck @ 4Ghz, to not have a bottle in the HDD's while recording i have the witcher 2 on the SSD while recording on another larger mecanical 1TB F3 drive.

The results was about 18-20FPS running some different savegames, some dips were even lower. Sure, it's fully possible to obtain a few FPS more (this game flow alot better over 20fps) But holding a steady framerate over 30 feel like a far fetc at this point.

I know people often exsagerate their performance in various games, people claiming running this and that on their cards, usally I let that pass since it's none of my concern but I have been playing witcher 2 alot and many other folks in this thread and if the HD6870 is that much better than all other cards I would start reccomend it over other higer performaing cards.

That would be the end of it I thought, but now theres another problem, I have made these recordings made on the missformantion that the resolution was 1600*900, using a 1440*900 obviously gives a bit better result and the whole thing seems much more real, now testing with 1440*900 the results went up a few FPS more.

I agree that it's possible to run witcher 2 @ almost 30FPS with a gtx460 on a 1440*900 resolution all max'ed out if it's xtreme OC'ed, to be honest im not sure how a HD6870 performs in withcer 2 opposed to a heavily OC'ed gtx460 but my guess is that they would be quite near eachother.

So I have changed my mind, the result's seem quite possible according to my testings, If I had a youtube account I would upload some of the videos but meh... maybe another time. :)

That chart was using a GTX 460, and is directly from Nvidia.

Witcher 2 doesn't use tessellation, having that option enabled or disabled would be irrelevant. And as he mentioned right in the original post he plays on 1400x900, A low resolution.

Here is a chart showing the performance differences in Witcher 2:

Now 1400x900 is a lower resolution than 1280x1024, so does his claim really seem that far fetched?

ferret-gamer

Sorry ofc. It does not use tesselation don't know what I were on about? :? Anyway: I've been trying to replicate a bit above HD6870's performance while recording with fraps, the settings were all max'ed out with the gtx460 OC'ed to 900 on the core and much less agressive OC to the memory the phenom2 X4 is no bottleneck @ 4Ghz, to not have a bottle in the HDD's while recording i have the witcher 2 on the SSD while recording on another larger mecanical 1TB F3 drive.

The results was about 18-20FPS running some different savegames, some dips were even lower. Sure, it's fully possible to obtain a few FPS more (this game flow alot better over 20fps) But holding a steady framerate over 30 feel like a far fetc at this point.

I know people often exsagerate their performance in various games, people claiming running this and that on their cards, usally I let that pass since it's none of my concern but I have been playing witcher 2 alot and many other folks in this thread and if the HD6870 is that much better than all other cards I would start reccomend it over other higer performaing cards.

That would be the end of it I thought, but now theres another problem, I have made these recordings made on the missformantion that the resolution was 1600*900, using a 1440*900 obviously gives a bit better result and the whole thing seems much more real, now testing with 1440*900 the results went up a few FPS more.

I agree that it's possible to run witcher 2 @ almost 30FPS with a gtx460 on a 1440*900 resolution all max'ed out if it's xtreme OC'ed, to be honest im not sure how a HD6870 performs in withcer 2 opposed to a heavily OC'ed gtx460 but my guess is that they would be quite near eachother.

So I have changed my mind, the result's seem quite possible according to my testings, If I had a youtube account I would upload some of the videos but meh... maybe another time. :)

[QUOTE="swehunt"]

[QUOTE="silversix_"]Just tried and i get 56-68 FPS without uber/vsync rest all maxed 1680x1050. With uber i'm down to 17-24FPS. I see no point of Uber, it looks awesome without anyway.Mehdi-Y

We dont know the setting nor the framerate that Mehdi-Y use.

OMG can you not see the chart? It clearly says 4870 512M. And I did NOT say that it is more than twice as fast, it's 65% you really dont know maths do you? :?

And now let me show you what sort of a staggering difference VRAM can make.

The GTX 260 is more than twice faster let alone the 1GB 4870. Infact even stepping down the resolution it seems 512MB is a serious bottleneck with this game.

So having less VRAM isnt just a problem if you are on high resolution.

These are just two games, they dont bench the newer games on these cards anymore, I am pretty sure there are more games that bottleneck with 512MB VRAM.

Gambler_3

:lol: Guru3D stalker HD4870 512mb own gtx260 @ 2560*1600

Farcry 2 @ tomshardware tells another story than your chart @ 1900*1200 with 4AA and 8AF and very high settings HD4870 512mb gets 40.4FPS where the 1GB vins with 45.5FPS, can't you see that your charts are sqreewed up?

Have a look those two farcry benchmarks then you head over to tomsH/GURU3D or anandtech and check for crossrefference because their card (or the testers) is OFC malfunktion! Have a look at the reviews there then you start talking about the "staggering difference" the total sum of benchmarks from Tomshardware in all the quality settings (1900*1200+ AA) the 512mb HD4870 gets 1338 and the 1GB version gets 1381, so 1338 vs. 1381THATS the staggering difference? :?

In my opinion and ALL other reviewers the two is performing nearly identical in just about any game if the card has trouble with drivers in a particular game that may be it but how much of a fool makes a chart on it and post it in his review when the result is obviously not even near correct?

I for one own a HD4850 512mb and it kicks their HD4870 512 by a enourmus margin in farcry 2 using the same settings, that is just silly!