-
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.
7950 vs 660 ti
- Thread starter davidparry1000
- Start date
Thats all true if you don't overclock. But as the 7850 can over clock by 40%+ on the core, once all over clocked it will be a different picture. We know a fully over clocked 7850 is close to 7950 performance, which is out of reach for a 660 I'm sure.
Lol at quoting the tweak town benchmarks.
Not sure why everyone thinks the tweaktown benchmarks are wrong?
I actually get higher than what they qouted with mine
http://www.tweaktown.com/articles/4...er_edition_video_card_overclocked/index3.html
My score on the same card
http://www.3dmark.com/3dm11/4330169
Nah, the review seems fine to me. Quite in depth actually for a preview and as said no proper drivers for the card and no voltage control so even higher scores come release
Perhaps the fact that according to them the 660ti is faster than a 7970 in dirt 3? Alarm bells ringing yet?
How about how the GHz edition gets a 30% performance boost from the 7970 in dirt 3? Bit fishy to say the least, but believe it if you like.
Only the MSI power edition is likely to have voltage control, and AMD drivers have improved a lot in the 4 releases since the drivers used in that preview.
How about how the GHz edition gets a 30% performance boost from the 7970 in dirt 3? Bit fishy to say the least, but believe it if you like.
Only the MSI power edition is likely to have voltage control, and AMD drivers have improved a lot in the 4 releases since the drivers used in that preview.
The Stock 660ti scores higher than the stock 7970 in 3dmark so why is it so hard believe that it beats it in some of the games.
I have the card and can confirm the tweaktown reviews scores are correct also I have tested it in the games shown and both heaven and 3d mark.
The scores given for the 660ti on tweaktown are correct no matter how unbelievable you find them.
I can post screen shots of my benchmark scores if you want?
Considering the card doesn't have proper drivers yet and has no voltage tweaking atm until the overclocking apps support it I think it's a little too early to judge.
Well for one 3DMark is a synthetic app and has 0 relevance to gaming. But more importantly.
Tweaktown always have nvidia cards performing above average review or the AMD cards performing under average review, they are also always first with a new nvidia cards and always before release
in that particular review they say they are using 12.4 / 12.7. if they are using 12.4 (which are very old drivers) then it explains why that difference as 12.4 drivers are the worst and slowest drivers you could possibly use, and 12.7 are BETA drivers which are not optimised for much at all.
Basically, Tweaktown reviews stink. and trust me i'm not the only one who thinks that, some of my friends are hardened nvidia fans and even they have looked there a couple of times and called Tweaktown dodgy.
Last edited:
I don't think you can discard a review from the off. If it's blatantly out of sync with what the others are, then fine but until then I believe we should all just wait and see
.3DMark is not that out of sync, altho it is 7% higher for the GTX 660ti than it is here for example http://www.guru3d.com/article/inno3d-geforce-gtx-660-ti-ichill-review/19
3DMark is very synthetic. if you take another synthetic bench, like Unique Heaven for example my 7870 is around GTX 660ti and not for off a GTX 670, just to show you the variation between them.
It's the drivers they used for the whole review that sets alarm-bells ringing loud, 12.4 especially are not good drivers, why use such bad 6 moth old drivers if your wanting to show a fair comparison?
3DMark is very synthetic. if you take another synthetic bench, like Unique Heaven for example my 7870 is around GTX 660ti and not for off a GTX 670, just to show you the variation between them.
It's the drivers they used for the whole review that sets alarm-bells ringing loud, 12.4 especially are not good drivers, why use such bad 6 moth old drivers if your wanting to show a fair comparison?
but can't seem get the core clocked highly at the same time.How much?
Humbug - AMD can only blame themselves for the poor performance of their drivers relative to now and reviewers can only review what they had there and then.
The problem is that when they're comparing to other cards they're often just using the original review for those comparison points rather than re-running the tests on today's drivers.
It is a large time (and money) sink to re-review a card again when a different card comes out in the future. As I said, the blame lies in one place only that reviews aren't necessarily relative to today. While people will engage in debates on forums, I would image that a lot of people don't and simply aggregate the reviews available when purchasing. By having drivers which haven't unleashed the full potential of the cards until now... AMD would have undoubtedly lost sales.
The problem is that when they're comparing to other cards they're often just using the original review for those comparison points rather than re-running the tests on today's drivers.
It is a large time (and money) sink to re-review a card again when a different card comes out in the future. As I said, the blame lies in one place only that reviews aren't necessarily relative to today. While people will engage in debates on forums, I would image that a lot of people don't and simply aggregate the reviews available when purchasing. By having drivers which haven't unleashed the full potential of the cards until now... AMD would have undoubtedly lost sales.
Around £180, not bad if the performance is between 7850 and 7870.
It does make me chuckle when people say "but they have not used the latest drivers, so the bench is pointless". I am not having a dig at anybody but I do remember a while back people saying how great the 12.4's were. It takes a long time to do benches as myself and Rusty have done so and I wouldn't expect reviewers to keep loading up new drivers each time they are released and re-running the tests.