Best Graphics Cards For The Money: January 2012 (Archive)

Page 34 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.


In the passmark chart GTX 660 has 4111, higher than the GTX 590 at 4111.
Notice too though that this is also below the GTX 580 at 4968.
Further up the chart again, the GTX 680 beats the GTX 690.
It looks like dual GPU cards to not fair as well in this specific benchmark.
 


So you agree with me in saying that GTX660 should be above the GTX590 ? Then why is the hierarchy chart http://www.tomshardware.com/reviews/gaming-graphics-card-review,3107-7.html show 590 is topping 660?

 

It doesn't... you're the one getting it backwards.

That list gives the GTX 660 a rating of 4111, the GTX 590 only gets 4034.
This article puts the GTX 590 two tiers above the GTX 660.

And this just shows how inaccurate passmark is. The GTX 590 ranks below the 580 AND the 570 in their list, which is completely ridiculous. It's nearly twice as powerful as the 580, which also means it's a lot more powerful than the 660.
 

That data is useless. They have the GTX 590 well behind the GTX 580 AND the GTX 570, which is ridiculous. The GTX 590 has two of the same GPUs as the 580, so obviously it should be much faster than the 580, let alone the 570. Just ignore anything that says Passmark, that's the lesson here.
 


No, the GP Mark chart does not accurately reflect the performance of dual GPU cards in games.
The GTX 590 is faster than the GTX 580 or GTX 660 in games.
 


I bought my CPU based on their rating -.-. Can you show me a more accurate chart?
 


I believe you because your name says gfx card master :)
 
I am asking to include the HD 4400, HD 4600 and HD 5200 in the hierarchy chart. Though they have insignificant impact in game performance but they are not negligible. Without them the list is incomplete.
 
I am asking to include the HD 4400, HD 4600 and HD 5200 in the hierarchy chart. Though they have insignificant impact in game performance but they are not negligible. Without them the list is incomplete.
 

The problem is that those integrated GPUs are used in CPUs with different TDPs. Which means that performance can vary a lot.

For example, both the Core i3-4130 and the Core i3-4130T have HD Graphics 4400, but the 4130 has a TDP of 54W while the 4130T has a TDP of only 35W. That means the HD Graphics 4400 is much more likely to reduce its clock frequency to stay within the allowed TDP. Lower clocks = lower performance, thus HD Graphics 4400 isn't just HD Graphics 4400.
 

This is a valid point, but I think they should be tested on the higher TDP chips; i.e. their chart position represents the best that could be expected from them. After all, ALL GPUs will throttle under the right (wrong?) circumstances.
 
On the hierarchy chart, perhaps you could expand with an estimate of integrated graphics. Particularly of the stronger offerings such as HD4600 and amd apu.
A prospective buyer might want to know at which point a discrete graphics card would be significantly better.
 
Status
Not open for further replies.