• 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.

Hmm... GTX480 and unheavenly performance?

CPU limited with a 3.5GHz Q6600 in a graphics benchmark? Really?

To the extent that I'm getting similar scores to a similarly-clocked GTX470 with the same CPU?

What could cause a card to post poorer performance like this? Hard to believe it's a hardware error. Wouldn't any such error be likely to give more problems like crashing, locking up, etc?

Any other possibilities? Power? Faulty connectors?
 
strumpusplunket said:
CPU limited with a 3.5GHz Q6600 in a graphics benchmark? Really?

To the extent that I'm getting similar scores to a similarly-clocked GTX470 with the same CPU?
Click to expand...

ye why not? at the extreme scale of it, i got, no joke, a 200% performance bump at some points in black ops from just upgrading the cpu, from 20 to 60 (3D). whilst that is obviously a one off, i would say 90% of games received some form of performance gain with perhaps 30% of those being a large performance gain.
 
i ran the benchmark to show its not just you at the same settings and got 50.8 average. now im running my card at 915 core so you would expect me to get nearly 580 performance. im running a phenom 2 at 3.6ghz. theres my problem.
 
Buckster said:
reviews say .. nope - hardly any difference at all - here is Crysis for example

http://www.bjorn3d.com/read.php?cID=1379&pageID=5565
Click to expand...

thats why you should never put all your faith in benchmark sites, real world experience from normal users is far more valuable. they often just use a very small section of gameplay from every game, which does not give a good overall picture of the performance
 
buttkinz said:
i ran the benchmark to show its not just you at the same settings and got 50.8 average. now im running my card at 915 core so you would expect me to get nearly 580 performance. im running a phenom 2 at 3.6ghz. theres my problem.
Click to expand...
I just did a quick test with my 580's
(MSI afterburner reported my cards useage something like = GPU1 90%, GPU2 85% )

Untitled-39.jpg
 
Last edited:
bee said:
thats why you should never put all your faith in benchmark sites, real world experience from normal users is far more valuable. they often just use a very small section of gameplay from every game, which does not give a good overall picture of the performance
Click to expand...

The commonly made comparison that an overclocked 470 is similar performance wise to a 480 has been shown in this thread (this being real world experience from normal users)

However, all things being linear, wouldn't the difference be the same if the cpu speed on both machines be increased by the same amount (say for example both of us upgraded to 2600K chips and clocked them at the same speed)

Strumpusplunket is wondering why his particular 480 is performing so badly, someone needs to explain to me how a faster cpu will increase performance relative to a GTX470 with the same cpu.
 
Last edited:
Yes, that's pretty much my thinking.

I can accept similar scores between michael's 470@850MHz and my stock 480, since that would be expected. But similar scores to a 480@825MHz with the same CPU? And buttkins' 470@915MHz beats mine with a Phenom II X4, which won't be much better than my 3.5GHz Q6600...
 
Read it and weep you Fermi boys :D

FPS: 31.2
Scores: 786
Min FPS: 18.1
Max FPS: 61.5

Edit, same as op : 1680x1050, normal tess, 16xAA, 4xAA
 
Last edited:
:D

So what settings are those at? I would have expected a clocked 5870 to post far better results than that.

With my 480, I've now tried benching at 1680x1050 without AA and with default 4xAF. Now I get 59.2average fps, and a score of 1492.

This puts it just a touch behind a 570 from the Heaven benchmark thread with a similar Qxxxx processor at the same settings. Again, even at stock, a 480 should beat a 570, so my overclocked 480 falling short seems another indication of something up.
 
I'm just guessing here, so don't have a go if it turns out to be silly, but could the resolution here be playing a part?
I often notice when reading graphics card reviews that at lower resolutions there is less difference between cards. The difference at 2560*1600 is usually more pronounced than at 1680*1050.
 
Yeah, that is true, because as you lower the res the cards have less work to do, and the weaker ones tend to catch up to the stronger ones.

It's not the same issue, though, as a stronger card failing to perform as well as it should at lower res.

Any chance this could be related to power connectors? I didn't really like the look of my 6+2 PCI-E connectors. Just felt a bit flimsy. Though I'd imagine a problem here would result in crashing or even failing to POST rather than poor performance?
 
nope, its my 470 im running at 915, its under water :) i think that because we run cpu's that arent in the same league as the i7's we cant compare our benchmarks, for example for a chap running a 470 at my speeds in 3dmark11 using an i7 he gets 6200 odd points, whereas i only get 5200 odd. from comparing the phenom to an i7 of my friends he is always getting much higher fps than me, yet the cpu is only the difference really :( hopefully bulldozer will be competative in the future so prices for i7's etc will drop a bit :)

edit: what timings and speeds are you running your ram at by the way strum?

also my scores are quite a bit higher than if i hadnt tweaked the phenom, iv managed to get a good 10-20% extra performance by having a really high NB speed and very fast and tight ram timings
 
Last edited:
It's at 392MHz, 5-5-5-18, 2T. Not very fast. I've never really bothered about clocking/tightening RAM, and I was always led to believe it made such an insignificant difference. Is that perhaps an outdated view now?

I do get what you mean about the difference between i7 and C2Q performance. I know using PCI-E 1.1 will also lose me maybe 2-3%. However, I probably wouldn't worry too much, except for the fact that my 480 seems to be posting worse (or at best equal) results than people with similar CPUs and weaker graphics cards.
 
GoogalyMoogaly said:
I'm just guessing here, so don't have a go if it turns out to be silly, but could the resolution here be playing a part?
I often notice when reading graphics card reviews that at lower resolutions there is less difference between cards. The difference at 2560*1600 is usually more pronounced than at 1680*1050.
Click to expand...

I agree with this. The majority of benchmarks I see around are at 1920 x 1200. Perhaps at that resolution the difference between my 470 and the 480 would be somewhat larger.
 
tbh im not sure faster ram makes much difference, it does for certain benches but in real world stuff iv never noticed. only the NB has made visible changes to standard. can you clock the q6600 any further and try the benchmark again, if there is a difference in fps that scales reasonably with clock speed then that may show a reason for lower speeds. although apart from different ram speeds etc/ram divider setting I cant really explain why yours is slower than a chap with the same cpu and gpu.
 
You must log in or register to reply here.
Back
Top Bottom