I normally agree with ya Rich but there I don't.

On high end graphics card reviews when your testing the raw power these things have, efforts should be made to negate the possibility of getting results effected by CPU bottleneck. On this particular card it probably wouldn't matter so much but on the 4870X2 is mattered and from what I've read on the GTX 295, at stock CPU frequencies, it was getting quite a bit of bottleneck.

While I agree with the fact that the graphics card shouldn't be overclocked (unless its a factory overclocked card of course), having the CPU overclocked is something you should expect from readers seeing as this is an overclocking website and forum is it not?

Looking forward to the 260 review anyway :) will read later.

Any news on a 295 review?

Was the 4870x2 done on an i7 rig tho? i7 offers much more more than a Q6600 at stock.

The thing is if folk read my reviews and hopefully make an informed decision upon them. Sure it would be nice to show the possibilities but where do you stop? I have a i7 965 on the way, have a WC setup this will hopefully run 4ghz 24/7 but this represents a miniscule section of the market when you consider alot of folk are still running P4's and AMD chips (no offence intended). I can fully appreciate what you're saying m8 and believe me it's frustrating as hell to run everything at stock with all this lush kit but you have to draw a line and stick to it.

Imo we have to represent the larger market while still using the most up to date kit. Bit of a contradiction there I know but as the saying goes; You can please all the people some of the time, and some of the people all the time, but you cannot please all the people all the time.

I am open minded on this though and will bend to the greater majority - what do others think?

I vote both. Do the benchmarks on both a stock and OC'd system. It may a serious amount of time to do everything twice but it does point at both audiences.

Most reviews I've read always do a bunch of stock stuff, then a last page that's balls-to-the-wall. I'd be very disappointed to read a review that wasn't mainly run at stock speeds. But alternatively the "something for nothing" OC part of me always wants to see stuff at their limit.
VB

I guess it wouldn't hurt to run a 'balls to the wall' benchmark in future reviews, or maybe extend the overclocking section and run the full rig overclocked for a few benchies to appease the hard core?

That would certainly create a good image for the clockers :)

Was the 4870x2 done on an i7 rig tho? i7 offers much more more than a Q6600 at stock.

The thing is if folk read my reviews and hopefully make an informed decision upon them. Sure it would be nice to show the possibilities but where do you stop? I have a i7 965 on the way, have a WC setup this will hopefully run 4ghz 24/7 but this represents a miniscule section of the market when you consider alot of folk are still running P4's and AMD chips (no offence intended). I can fully appreciate what you're saying m8 and believe me it's frustrating as hell to run everything at stock with all this lush kit but you have to draw a line and stick to it.

Imo we have to represent the larger market while still using the most up to date kit. Bit of a contradiction there I know but as the saying goes; You can please all the people some of the time, and some of the people all the time, but you cannot please all the people all the time.

I am open minded on this though and will bend to the greater majority - what do others think?

I do agree to the large umm degree, and you're right that the I7 offers more than a C2Q. Also agree for anything up to this card then people aren't likely to have a 965 overclocked on water but if you're getting into the 4870X2's and GTX295's then the people these cards are aimed at are likely going to have components from the higher end overclocked and tuned to diminish as best they can any bottlenecks.

You are right that with this card and an I7 of any sort, you aint likely to run into bottlenecks apart from maybe Fallout 3, but every card seems to bottleneck on that even with an overclocked 965. I guess you've just gotta decide when it comes to the super heavyweights whether it's worth it to up the CPU a few hundred MHz. While this would take them out of whack with other results from lower cards, as long as the two competing flagships were done the same, they would be comparable.

That said mate, it's obviously up to you and Jimbo how to best do your reviews. Either way I'll still enjoy reading them and seeing what you have to say about them. Numbers can still only tell ya so much anyway... The general feel of how they played and if they bottomed out anywhere is always much more informative.

:)

Great review as always Webbo - I think it's good to see some stock speeds but also the potential the card can have with it's balls to the wall.

Nice one. :)

Maybe later drivers will increase the slim gap but I doubt it as what effects the 285 will also effect the 280 being they are essentially the same card with different clockspeeds (and of course the die shrink).

This is true.

I don't think the 280 drivers are done with yet tho tbh. 181 is a mid range-ish number, if u think 175 to 184. 184 will be the banker, b4 the next gpus come out.

Chances are the cd would have 180/179 drivers.

Must say I'm impressed with the frequency NVidia are releasing drivers (nightmare for us reviewers but meh).