GTX460 or 470

[glenimp617]

I too thought Dirt2 was an Nvidia game. Suppose its because it always seems to do better on nVidia hardware.

 

[Marine-RX179]

Dirt 2 is an ATI endorsed and supported game, it came out even before the fermi hardware, how can it be nvidia biased...because it's DX11??

I know, but the game does seem to favour the GTX cards more. May be bias is not the right word to use...Nvidia does seems to optimised their cards to run very well particularly on that game, or simply because the game engine itself work better with Nvidia's hardware spec, despite the game is supposed to be ATI optimised.

 

[RavenXXX2]

How can you claim it's an nvidia game just because it runs great on fermi, do all games need to run at the same performance level on either brand to compare without bias. Dirt 2 was given away with ATI cards, it is featured in their marketing campaigns, it's an ATI game, so it's completely fair to compare performance with Dirt 2. It shows the Dirt 2 engine with DX11, tessellation ect runs good on ATI but better on Nvidia, no doubt F1 2010 will show the same performance gap.

 

[Marine-RX179]

How can you claim it's an nvidia game just because it runs great on fermi, do all games need to run at the same performance level on either brand to compare without bias.

If you actually bother read what clarified in my 2nd post post, I didn't 'claim' Dirt 2 is a Nvidia game. I said the game run extremely well with the Nvidia cards (despite it is an ATI game). The result is quite unusual so it is not a good representation. The point I was trying to make was games such as BCBF2 would be better representation for pitting cards against cards.

Also, mikeo is calling me BSing for saying 5850 is roughly 10% faster than the GTX460 1GB, and the GTX470 is roughly 10% faster than the 5850, which was based on the AnandTech results from the AndandTech thread. So are you agreeing with him on that?

 

[RavenXXX2]

Also, Dirt 2 is a Nvidia biased game,

That's as good as saying it's an TWIMTBP title " an nvidia game " just because it performs better on fermi you would prefer to rule it out for comparison purposes and that makes no sense, but because BC2 performance is close between ATI and Nvidia, that's perfectly fine to compare. So if we go by your methodology, any games that perform significantly better on one brand should not be trusted as it's a biased game lol.

Regarding performance between the mention cards, yes they seem about right but there will always be exceptions when some game will perform better than others.

 

[Marine-RX179]

That's as good as saying it's an TWIMTBP title " an nvidia game " just because it performs better on fermi you would prefer to rule it out for comparison purposes and that makes no sense, but because BC2 performance is close between ATI and Nvidia, that's perfectly fine to compare. So if we go by your methodology, any games that perform significantly better on one brand should not be trusted as it's a biased game lol.

I already admitted bias was the wrong word to use in #62. Quit trying to pick bone from an egg.

 

[glenimp617]

eggs

 

[Marine-RX179]

eggs

haha, very funny

 

[arc@css]

I don't get the joke though by now I will have moved on to the cheese and biscuits, coffee and cigars.

 

[mikeo]

Now who's the one trying to raise the bait here? You obviously haven't see this thread:
http://forums.overclockers.co.uk/showthread.php?t=18176548&page=2&highlight=AnandTech

Also, Dirt 2 is a Nvidia biased game, so it's not a good representation. You want to compare the cards on fair ground, try results of more neutral games such as BFBC2.

And see here look at the result of 1920x1200 4xAA 16xAF of the Palit GTX 460 Sonic Platinum (800MHz) compare to 5850:
http://www.techspot.com/review/299-palit-inno3d-geforce-gtx-460/page6.html
how am I wrong to say GTX460 1GB roughly need to be around 800MHz to be around 5850 speed?

Fair cop Guv. But we are talking single digit frame rate differences here!

And I would never accuse anyone of "BSing". I would never be so rude friend.

But interesting link there though. Not read this particular review. Though it does some what back up my comment regarding an overclocked 460 being about equal to a stock 470. Which was pretty much the thrust of my original post. And I do have personal experience to back that up (well at least in the games we share).

The bottom line is. Whichever of these three cards anyone buys (460/470/5850) I don't think anyone would be too short changed. And all perform decently and swap places in reviews almost at random depending on game/resolution/settings.

Though this page of that review is interesting. Shows very little between any of the three cards (with a 460 @ 860MHz). Though obviously the other two cards would move away if OC'd themselves.

http://www.techspot.com/review/299-palit-inno3d-geforce-gtx-460/page12.html

If I've offended you, sorry.