• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

660ti worth an extra $50 over the 7870?

Armageddon415

Golden Member
Sep 24, 2000
1,608
0
0

aaksheytalwar

Diamond Member
Feb 17, 2012
3,389
0
76
1. They perform very similar, especially once overclocked if at all the 7870 might be a bit faster. The 660 ti can cost more only because of physx etc, not really because of more performance.

2. Neither is good enough for 1440p.
 
Jul 29, 2012
100
0
0
The 7870 overclocked can only approach 660 ti stock performance. 660 ti and for that matter 7950s tend to be great overclockers, and more so than 7870. Proportionally, that $50 is worth it if you get a 660 ti that overclocks well.

Battlefield 3 performance is also superior
 
Feb 19, 2009
10,457
8
76
Stock the 660ti is 7% faster. OC vs OC, about the same.

Is 7% worth $50, or 20% of the products value? If you play BL2 and like Physx in games, then yes, its worth it.
 

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
Actually the graph shows both stock and o/c's 660Ti is closer to stock and overclocked 7950 than it is to the 7870. That's with AA.
Without AA it actually pulls ahead of the 7950 stock and o/c'd.
 

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
Extremely game dependent, in that list, remove Hard Reset Demo and its a tie. ;)

Just shows how close they are OC vs OC.
How come you don't suggest removing Dirt Showdown then? Please be fair Silverforce. If you don't, I will begin to do the math and show the percentages of each game the 7870 or 660Ti is faster over each other. Then it really looks like a landslide difference.
 
Feb 19, 2009
10,457
8
76
How come you don't suggest removing Dirt Showdown then? Please be fair Silverforce. If you don't, I will begin to do the math and show the percentages of each game the 7870 or 660Ti is faster over each other. Then it really looks like a landslide difference.
Because its a new DX11 game and Hard Reset is dx9 and old?

I guess you could if you wanted to emphasize my point, that benchmarks are very game dependent. Lets just bench a single synthetic and nothing else ok??

But please do the maths for the individual games/benches. I'm seeing average of 60.4 fps vs 56.8.. OC vs OC.
 

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
Because its a new DX11 game and Hard Reset is dx9 and old?

I guess you could if you wanted to emphasize my point, that benchmarks are very game dependent. Lets just bench a single synthetic and nothing else ok??

But please do the maths for the individual games/benches. I'm seeing average of 60.4 fps vs 56.8.. OC vs OC.

Heh, are there any other pre-DX11 games in that list?
Ok, stand by.
 
Last edited:

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
GPU 660Ti (o/c) 7870 (o/c) fps faster % faster

STalker CoP.... 63.............. 61............. 2.............. 3.5%

Metro 2033..... 28............. 31.............. 3.............. 10%

Just Cause 2... 87.............. 76............. 11............. 15%

Alien vs. Pred. 53............. 54.............. 1.............. 2%

Lost Planet 2.. 73.............. 50............. 23............. 46%

Civ 5............ 49............. 65.............. 16............. 31%

Shogun 2....... 57.............. 40............. 17............. 43%

Crysis 2......... 62.............. 53............. 9.............. 17%

Hard Reset..... 91.............. 66............. 25............. 38%

Batman AC...... 73............. 76.............. 3.............. 4%

Battlefield 3..... 73.............. 62............. 11............. 18%

DiRT Showdown.. 55............. 82.............. 27............. 50%

Sniper Elite 2.... 21............. 23............. 2.............. 10%


Anyone notice anything?

Also, anyone notice anything interesting about the additional graphs Silverforce and Rvenger presented? I got the cigars right here.
 
Last edited:

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Super Moderator
Apr 6, 2004
6,282
5
76
What that Nvidia is cheating on LP2 and AMD is cheating on Dirt Showdown? These benches look like a wash to me and I don't know what you are trying to prove.
 
Last edited:

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
What that Nvidia is cheating on LP2 and AMD is cheating on Dirt Showdown? These benches look like a wash to me and I don't know what you are trying to prove.
Yes you do. Nobody said anything about cheating. You'd see the pattern if you actually wanted to.

Look at the differences in performance on a PER GAME basis, and it's over. Where the 7870 beats the 660Ti, the games are fewer and the percentage are far smaller in comparison. So, it's impossible, if you actually want to see and want understand these numbers, to call it a wash. Not even close.
Anyway. The numbers are there for anyone who wants to really see them. That is all that matters to me.

Also, the graph that you and Silverforce presented to all of us do not even include 660Ti o/c numbers.

Somebody check my math if you care to. Appreciated.
 
Last edited:

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
That benchmark is for the 660, not the 660Ti. No cigar needed - I don't smoke. ;)

The truth is that the cards are all pretty similar. I think it's pretty clear that the 7870 is superior to the 660, and given how close they are in price, the 660 just has to drop. The 660Ti, in my opinion, is slightly faster than the 7870. Whether it's worth $50-70 more, I can't really say. Don't people pay that much all the time for an 8% improvement - like 680 vs. 670 or 7970OC to 7950OC?
 
Last edited:

Grooveriding

Diamond Member
Dec 25, 2008
9,082
1,224
126
7870.

Those averages are skewed by ancient DX9 games that both cards are getting well more FPS than they need in. In the demanding games, the picture becomes clear that the 7870 is better.

At the point you are paying the $300 for a 660ti, you should just get the 7950 which is about the same cost and a much better card.
 

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Super Moderator
Apr 6, 2004
6,282
5
76
GPU 660Ti (o/c) 7870 (o/c) fps faster % faster

STalker CoP.... 63.............. 61............. 2.............. 3.5% Nvidia barely wins

Metro 2033..... 28............. 31.............. 3.............. 10% AMD has a small lead

Just Cause 2... 87.............. 76............. 11............. 15% Nvidia has a small lead

Alien vs. Pred. 53............. 54.............. 1.............. 2% AMD barely wins

Lost Planet 2.. 73.............. 50............. 23............. 46% Nvidia wins big

Civ 5.......... 49............. 65.............. 16............. 31% AMD wins big

Shogun 2....... 57.............. 40............. 17............. 43% Nvidia wins big

Crysis 2....... 62.............. 53............. 9.............. 17% Nvidia has a small lead

Hard Reset..... 91.............. 66............. 25............. 38% Nvidia wins big

Batman AC...... 73............. 76.............. 3.............. 4% AMD barely wins

Battlefield 3.. 73.............. 62............. 11............. 18% Nvidia has a small lead

DiRT Showdown.. 55............. 82.............. 27............. 50% AMD wins big

Sniper Elite 2. 21............. 23............. 2.............. 10% AMD has a small lead


Anyone notice anything?

You brought up the cheating speculation in another thread. I still don't know what you are trying to prove. They look pretty equal in performance especially if you pick a wider variety of games.


It all boils down to if you want a copy of BL2 and Physx features vs save $50 and get the 7870 with Sleeping Dogs. They are both great cards, period. Just think about it, once one company releases another driver update the tables will turn yet again. This is a cat and dog chase for crying out loud.
 
Last edited:

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
7870.

Those averages are skewed by ancient DX9 games that both cards are getting well more FPS than they need in. In the demanding games, the picture becomes clear that the 7870 is better.

At the point you are paying the $300 for a 660ti, you should just get the 7950 which is about the same cost and a much better card.
I agree that the 7950 is better than the 660Ti, for people willing to OC and who have space for a 11"+ card. That's probably most people on this forum, but not everyone. So even at the same price, the 660Ti, which is actually faster than a 7950 stock vs. stock, is a reasonable choice in many cases.
 
Last edited:

96Firebird

Diamond Member
Nov 8, 2010
5,654
272
126
Armageddon, since nobody else has asked and this thread turned into a battle over things you may not care about, let me ask you a few questions...

Any other games you play?

What does the rest of your system look like?

Do you plan on OCing, or keeping it stock?
 
Thread starter Similar threads Forum Replies Date
moonbogg Graphics Cards 70

ASK THE COMMUNITY