Many many years of experience and some bias towards nVidia.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
470 sli or 6870 CF
- Thread starter JAPears
- Start date
Many many years of experience and some bias towards nVidia.
I'm glad you mentioned MLAA... 470 SLI gives you 64x4SS TRAA at playable fps. MLAA can't even touch that image quality.
- I run a dual monitor setup with 2 1080p monitors.
- Use my pc for design work primarily, using the adobe creative suit
- i play mainly WoW and SC2
- though i would like to dabble into some other games which are more demanding, including black ops, crysis etc
Hey Jimmy,
A pair of cards looks a bit OTT for your needs? . . . I understand its always tempting to splash the cash on shiny GPU's but it seems that graphic card investments nearly always lose value fast as a new "soup of the day" get released and your left with your expensive over-hyped hardware . . . the fickle OcUK GPU forum hype-machine moves on to the next latest and greatest GPU mania and the serious GPU e-Slongers scramble to sell their older expensive GPU's (often at half price) so that they can also buy the new thing and tell everyone they have the new thing . . . it all seems very silly but perhaps I am missing something!
Most normal mortals will cap their GPU budget at approx £150 . . . whether that be a new single card purchase or a pair of used yesteyears cards in SLI/CrossFire . . . you seem to be lining up a potential £380 GPU investment which puts you in a tiny, tiny "minority" of super-high-end power users with special needs . . . I question if you really "need" to be spending such a large amount of money or its just a case of "want" (have cash, will spend) . . . if you don't actually need to spend so much then you should also consider a single card around £150 (HD6850/GTX460 etc) or maybe a pair of the more affordable cards . . . anything you save today can be used to buy yourself something else nice or put towards an upgrade in the future!
The GTX 470 is a power-hungry, hot & noisy card unless you spend even more money to get a non reference models that has tweaked circuitry and special coolers to address these obvious flaws . . . I do believe nVidia will be updating the GTX 470 and releasing a tweaked version in the near future . . . with these current non desirable "Features" personally would not be investing in a pair . . . perhaps if my needs meant I could "justify" spending over £150 on a GPU I may get a single GTX 470 but the noise, heat and power would bug me!
Radeon HD 6850 (single or XFire), Geforce GTX 460 (single or SLI), Radeon HD 6870 (single) or Geforce GTX 470 (single) are the cards/solutions I find interesting at the moment, the HD6870/GTX470 as a single card set-up and the GTX 460/HD6850 as the multiGPU set-ups!
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
From the reviews i have seen 6870 crossfire is faster than gtx470 sli overclocking may be a mute point as you would need to overclock the 470's in sli to make up for the defecit. Overclocking gtx470 in sli will add more heat and noise to the package also. Here is the guru review has gtx470 sli v 6870 crossfire.
http://www.guru3d.com/article/radeon-hd-6850-6870-crossfirex-review/13
My vote goes to 6870 crossfire.
http://www.guru3d.com/article/radeon-hd-6850-6870-crossfirex-review/13
My vote goes to 6870 crossfire.
Last edited:
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
There is nothing to back this statement up when comparing the 6 series to the 4 series. I have not seen any threads on here yet with people having any 6 series multi gpu problems. mybe i have missed them but i have not seen any.
it is a know fact that when it comes to multi gpu support for games and bugs etc
nvidia and their drivers are ahead of AMD
Caporegime
- Joined
- 18 Sep 2009
- Posts
- 30,511
- Location
- Dormanstown.
6850 crossfire 
I would concurr with sem - crossfire software wise has dragged behind SLI for years and theres nothing so far to indicate anything different. They do seem to have addressed some of the hardware issues with the 6 series, but ignoring the min/max/avg performance numbers SLI puts up more consistant performance in most cases with a far more stable baseline to the fps.
I have used both solutions and do use GTX470 SLI myself, I don't hide the fact I have a preference for nVidia but neither do I have any reason to go out of my way to try and discredit AMD/ATI, I'm just calling it like I see it from years of experience with PC hardware.
Last edited:
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
Look at the results above rroff and tell me in crysis warhead which has the lower min fps out of 6870 crossfire and gtx470 sli. The last review i seen that went out its way with fraps during gameplay to see if this myth was true concluded that it was not the case at all a 5850 was giving better mins than a gtx470 at the time.
Last edited:
Soldato
- Joined
- 24 Aug 2009
- Posts
- 2,931
They maybe better on drivers, but both company's do have there ups and downs.
Have you got a link to this review mate ?
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
A quote from the introduction.
"Another reason why we wish to revisit this high-end graphics card battle is to settle a belief of green team fans that the GeForce GTX 480 is far superior to the Radeon HD 5870 when measuring minimum frames per second. This is an interesting argument as the average fps results did not suggest a huge difference in minimum frame rate performance when we last tested, though it could be possible, and therefore we wanted to find out if this was true or not.
However, simply showing the flat minimum frame rates is not enough considering that this could be the result of something else going on within the system that causes a split second drop in performance. Therefore we have recorded a frames per second timeline to reveal just how often the performance drops and for how long. Shiny new graphs coming your way, read on... "
http://www.techspot.com/review/283-geforce-gtx-400-vs-radeon-hd-5800/
These guys tested through gameplay not canned benchmarks to see if indeed the minimum frame advantage was a myth and thats exactly what was found that it is indeed a myth. We all know that canned benchmarks can be made to look good through drivers but gameplay does not lie. It looks to me during gameplay you would be hard pushed to tell the difference between the 2 company's if average fps was around the same.
Could it also be said that nvidia has had a performance advantage over the years so off course they will be better in the minimum frame rates. When the performance is equal though i don't think this is the case. This could be why nvidia fans percieve that nv has better minimum frame rates but in fact they don't its just that the amd card say 5870 is around 18% slower in the average so the gtx480 should automatically give a higher minimum.
Last edited:
I'm talking about graphs like in your link in later posts - but for multi GPU not single GPU, in a variety of games. A single min fps number on its own means nothing.
Fraps runs can be inconsistent in themselves, playing through the game can bring up different results each time, at least with a canned benchmark you're guaranteed accurate results for each card.
Here is the same site with more recent results, notice how the 470 now has the much better min FPS than the 5850.
Cheers mate, i will have a read tonight!
That's highly irrelevant, who sits there watching canned benchmarks? They're often not a true representation of actually playing the game, which you just inadvertently pointed out. To make results consistent though, they will do a run 3 times and then average the FPS results to give the most consistent results.
You really can't argue for using a canned benchmark when you often see higher performance than you would in game, there's seriously just no point to it. It's like trying to compare lap times around a circuit while disregarding the fact that there is a big difference between a solo lap, and a lap with other cars on the track. The solo lap is always going to be better because it's full predictable if you know the track well enough.
As for that Techspot review, are we reading the same thing? Because outside of Metro 2033, the minimum FPS of a GTX470 really isn't "much better" than the 5850, they're pretty much within a few FPS of each other for the most part, but it seems like you've forgotten that SLi versus Crossfire is being discussed here.
Last edited:
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
Gameplay is where its at raven canned benchmarks are just that, most don't even give you a decent view of what the game is going to play like. The sites that use fraps imo are what i go on as there benchmarks are the closest results as to what you will see in game. That review you posted shows us nothing really. The review i posted up gave a real time frame rate graph.