• 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.

660ti or a 7950 for 1920*1200 gaming?

I would get a GTX670 over a GTX660TI in the long run as the price difference is not enough IMHO. The GTX660TI does have limited bandwidth which is shown up by more intensive forms of AA(the chart shows that),and one has to think with more graphically intensive games in the next two years,whether it will start to cause an issue. Now if the GTX660TI was closer to £200,that would not be an issue,but is around £240 to £270 and GTX670 cards can be had for £280 to £300.

The GTX660TI reminds me of the 8800GT 256MB. The latter did relatively well against the 512MB version at the time but began to fall behind as time progressed,and I can see the reduced bandwidth of the GTX660TI doing the same when compared to the GTX670,but that is my personal view.

If you want to get a GTX660TI wait until the price drops nearer to £200.

If it were my choice and I had no more than £250 I would probably edge towards the HD7950,more for the fact they overclock a decent amount(although power consumption will increase) and HWBot has an average of around 1166MHZ for them from the stock 800MHZ,hence, even a poorly overclocking card should do quite well. OcUK have a HIS card for around £240 which looks decent value for money IMHO.

There is noise of a price drop on the HD7870 and HD7950 - so I would wait a few weeks IMHO as it should affect GTX660TI prices too.
 
Last edited:
Agreed. The 7950 is a no brainer in the ~£240 bracket.

660Ti isn't competing. If you factor in the overclock you're getting a real decent bit of kit for the cash.
 
Rusty0611 said:
Humbug: if you'd actually read my post you'd see that is the average of all reviews. It's entertaining to keep seeing you revert back to the Tom's review. I'm not saying its invalid I'm saying that the average of all reviews is a far better barometer.

If you have issues with it I suggest you take it up with CAT THE FIFTH as he took the time to compile them.
Click to expand...

Its not cat's review, its here http://www.3dcenter.org/artikel/lau...nch-analyse-nvidia-geforce-gtx-660-ti-seite-2

Most of them are using Factory overclocked GTX 660ti against Reference Radeon's.

Some are using factory overclocked cards from both sides.

It's the worst possible pool to get an average of the cards performance from as most of them are running overclocked card against reference cards

Tom's review and Canucks are the only two using reference settings on all cards, and they confirm each other.

This is a lot of money we are talking about here, other people's money.

You have to be sure its right before you start encouraging them to part with with that money, especially if what they have is already better.
 
Last edited:
Rusty0611 said:
It doesn't matter who did the averaging it's still an average.
Click to expand...

Oh dear.... it has nothing to do with who the average comes from, its about what the average comes from.

You take 10 overclocked cards, 2 reference cards and take an average from them all, what you will find is the average is biased towards the overclocked cards.
 
You're the only person in the world arguing that the 660Ti DOESN'T sit between the 7870 and 7950 performance wise.

If you're talking about recommending then I wouldn't recommend this card in a million years! It stinks. It's poor value for money at ~£240 and there are better overall options above AND below it.
 
Rusty0611 said:
You're the only person in the world arguing that the 660Ti DOESN'T sit between the 7870 and 7950 performance wise.
Click to expand...


Look at all the reviews in context, And, Well obviously i'm not the only one, again. Look at the bloody reviews ;)

According to the settings we chose on a per-game basis, picked to maximize visual quality at playable frame rates, the GeForce GTX 660 Ti is close to, if not slightly slower than a Radeon HD 7870. We already know this runs counter to Nvidia's expectations, which put the new card between AMD's Radeon HD 7870 and 7950. However, after comparing lab results, the outcome of our testing appears tied to the way we picked settings for each game, likely taxing the 660 Ti's memory bandwidth more than less-demanding options would.

Perhaps our settings favour the Radeon cards. Perhaps theirs favour the GeForce-based boards. And maybe the most real-world outcome lies somewhere in between. But, I believe the truth is that GeForce GTX 660 Ti performs within 5%, plus or minus, of the Radeon HD 7870.

Frankly, a few percentage points doesn't sway our opinion one way or the other. Here's the real bottom-line: using our benchmark suite and settings, the GeForce GTX 660 Ti and Radeon HD 7870 trade blows. That fact alone justifies Nvidia's £250 price tag.

Your choice comes down to strengths, weaknesses, and features, then. If you don't tend to use anything more than 4x MSAA, the GeForce GTX 660 Ti will perform better than many of our results indicate, since you won't be taxing its memory subsystem as hard. Conversely, if you're a fan of lots of anti-aliasing, the Radeon cards tend to pull ahead in this price range.
Click to expand...
 
You're posting the same thing over and over again from Tom's. Repeating the same thing doesn't make it have more resonance each time.

680/7970>670>7950>660Ti>7870>7850.

Value for money in my eyes goes:

7850>7950>7970/670>7870>680>660Ti

nVidia cards have never really been what you'd coin value for money but that's not what you're paying for.
 
Last edited:
Rusty0611 said:
You're posting the same thing over and over again from Tom's. Repeating the same thing doesn't make it have more resonance each time.
Click to expand...

Because its right, they are so close to eachother you can use Nvidia optimised games to make the GTX 660ti look better, in the same way you can use AMD optimised games to make the 7870 look better, the same thing goes for using different settings in those games, if you take that out, and don't use factory overclocked cards against reference cards what you actually find is that they are equal, the 7870 is not better than the GTX 660ti which in turn is not better than the 7870.

Beyond that it just depends on what games you do play.

What Tom's are saying is pretty pragmatic, and is in my opinion absolutely bang on.

Do you seriously think using factory overclocked cards against reference cards is a fair showing?

@ Final8y. i already linked Anandtech review earlier in this thread and used it to show them as one benching overclocked GTX 660ti's against reference radeon's

Its here http://forums.overclockers.co.uk/showpost.php?p=22595610&postcount=12
 
Last edited:
No it isn't a fair comparison but reviewers can only review what's on the market.

Reviews of mid range cards are generally rubbish anyway. The settings they're reviewed at are settings which are barely playable.

I'm sure I can find a review which backs my point up as well and keep quoting constantly from it but it's not really going to move on from that is it?

The 7870 isn't a slow card. It's just slightly slower than the 660Ti at the settings at which it's realistically going to used at. It's not surprising that the 7870 is better with AA on but bad FPS is still bad FPS. It's a midrange card for a reason.

It is funny to see mid range cards sold at north of £200 which can't handle AA properly.
 
humbug said:
Look at all the reviews in context, And, Well obviously i'm not the only one, again. Look at the bloody reviews ;)
Click to expand...

aside from performance though, toms review site is american and where your quote says £250, this should be $300. the 7870 and 660 might cost the same in us but they don't in uk.
 
Rusty0611 said:
No it isn't a fair comparison but reviewers can only review what's on the market.

Reviews of mid range cards are generally rubbish anyway. The settings they're reviewed at are settings which are barely playable.

I'm sure I can find a review which backs my point up as well and keep quoting constantly from it but it's not really going to move on from that is it?

The 7870 isn't a slow card. It's just slightly slower than the 660Ti at the settings at which it's realistically going to used at.

It is funny to see mid range cards sold at north of £200 which can't handle AA properly.
Click to expand...

Look i'm sick of arguing. i really don't want to argue with people but this guy is asking if he should dump his 7950 for a GTX 660ti, i think you and i agree on this: HELL NO!!!!!!

Yes i also agree the GTX 660ti is a little faster than the 7870 (with AA or any heavy duty work, off) so its down to preference.

Now, i do like all of that stuff and given the 7870 is a little faster with it, its the one for me.
 
Not that im particularly interested in stock performance but here:

Anandtechs results are with the reference 660ti:

http://www.anandtech.com/bench/Product/548?vs=647

I find some of anandtechs results dubious and im not sure which driver they used for the 7870 scores but overall the 660ti looks slightly faster to me.

Im sure max overclocked the 7870 would smash its back doors in though:p
 
You should now take back your nVidia fan boy remark from the other thread after you've seen some of my 660ti thrashing ;)

@psycho - depends how high the Ti clocks :p. Not that it's a reason to get one. The thing is disgustingly bad.
 
Last edited:
Rusty0611 said:
Humbug you're absolutely clueless mate. You keep posting the same thing all over the forum while completely ignoring the overall average analysis that CAT THE FIFTH did. I don't know if you're upset that your card is slower than a 660Ti or something...

Here is how the 660Ti actually performs:
Click to expand...

Sorry mate but the info Humbug posted from Toms is pretty much bang on. Yes yours was an average of many reviews but an average of many biased reviews, Toms used reference cards versus reference cards then took an average of all tests, its the best average anyone has produced so far.
 
Last edited:
Cant we just say they are pretty much equal then, stock vs stock within margin of error.

A different game suite would produce slightly different results.

Also you cant just dismiss reviews just by calling them biased.
 
Also for what its worth, I dont think the tomshardware review is that fair ir typical of expected use. A lot of games tested and benched there are run at 8X MSAA, which I think is unreasonable and unnecessary, especially on a mid-high end graphics card.

If tomshardware did all the tests at 4XMSAA Im sure the 660ti would have fared slightly better.

At its typical resolutions and settings, which I believe would be 1080p, high-ultra, 2-4X MSAA I believe the 660ti is as fast if not very slightly faster than the 7870 whwn both at stock.

They are close enough that we can call them equal though at these settings.
 
You must log in or register to reply here.
Back
Top Bottom