5870 vs 470 for an EVGA Fanboy

[Keysplayr]

Next time I won't link to the specific page I get a quote from if that's better for you?
The story doesn't change just because it's the 9th game, it's just the quote which expressed it was on the 9th game's page.

Sorry, I thought I had linked to the source page and not the graph, but apparently not, so here is the specific page it's from:
http://www.xbitlabs.com/articles/video/display/geforce-gtx-480_12.html#sect0

The HD5870 is "0" so anything above 0 is a win for the GTX480 (with margin of the victory) and anything below is a loss (HD5870 gets higher frame rates).
As for 4xAA (8xAA), not sure, I assume that means that indeed it was different AA for some games (whatever the setting they chose was).

http://www.hexus.net/content/item.php?item=24000&page=16
Hexus added their frame rates across the 5 games they tested, and at 1920 the GTX480 had a 26% higher score. At 2560 it was reduced to 18%

Techpowerup also has a relative performance page:
http://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_480_Fermi/32.html
The HD5870 is rated for ~89% of a GTX480 at 1680x1050, 90% at 1920x1200 and 93% at 2560x1600.

The general trend, as expressed in the "cherry picked" quote from an AT page, is that as resolution increases the HD5870 closes the gap to the GTX480.

That's great! Thanks! And now I'm sure you'll tell us all which the most widely used and popular gaming resolutions are today. And even if it was 25x16, the GTX is still ahead. It would be nice to know the exactly levels of AA used in each game without having to sift through every page in the Xbit article though. Those reviews of theirs are HUGE.

 

[Dark4ng3l]

Just beware of the opinions posted here the focus group has come out in force it seems. You can't take what they say at face value.

 

[BababooeyHTJ]

I'm not sure how you can trust a poll of a card that has barley hit the hands of the end users yet.

 

[Lonyo]

That's great! Thanks! And now I'm sure you'll tell us all which the most widely used and popular gaming resolutions are today. And even if it was 25x16, the GTX is still ahead. It would be nice to know the exactly levels of AA used in each game without having to sift through every page in the Xbit article though. Those reviews of theirs are HUGE.

You said "the GF100 series has more staying power" and "[Cranking things up] really does show the strength of GF100."

I refuted that claim by showing that 1920x1200 results vs 2560x1600 (including with various levels of AA), the gap between the two cards in fact decreases, indicating that your "staying power" concept is incorrect.
Now, since we're starting off with a GTX470 card, one which is slower than the GTX480 (I am sure I don't need to show evidence for THAT), it seems that since the gap between the GTX480 and HD5870 declines with increasing resolution (and, it seems, with at least some various AA levels), that the gap between the GTX470 and the HD5870 would conversely increase as the resolution increases, giving the HD5870 more staying power over the GTX470, at least in terms of average frame rates, if not minimum.

Now, given that the OP is using a resolution between 1920x1200 and 2560x1600, the HD5870 is the one which is going to be the better bet since the average performance gap will be high. If it was 1680x1050 the gap would be lower, and the GTX470 might be the better buy.

What the most used resolution is doesn't matter when this particular user is choosing a card for his particular resolution. If the most widely used resolution was 1280x1024, would you recommend a card based on relative performance at that level? No, you find out the specifics of the user to give the best recommendation, and since the OP is using a reasonably high resolution, the HD5870 seems a reasonable choice over the GTX470 given the average performance at higher resolutions.

Of course, if the OP is more concerned about minimum frame rates, then the GTX470 might be a more preferred solution, if there are specific games the OP plays and GTX470 scores well.

Maybe you could try addressing the OPs question with statements which are somewhat helpful when it comes to picking a card rather than arguing about which page of a review a quote comes from, or talking about "GF100" as an overall architecture (and making incorrect statements about it) when the OP asked a question about two specific cards and a specific resolution.
Unfortunately many sites don't seem to have done GTX470 tests.

 

[Keysplayr]

You said "the GF100 series has more staying power" and "[Cranking things up] really does show the strength of GF100."

I refuted that claim by showing that 1920x1200 results vs 2560x1600 (including with various levels of AA), the gap between the two cards in fact decreases, indicating that your "staying power" concept is incorrect.
Now, since we're starting off with a GTX470 card, one which is slower than the GTX480 (I am sure I don't need to show evidence for THAT), it seems that since the gap between the GTX480 and HD5870 declines with increasing resolution (and, it seems, with at least some various AA levels), that the gap between the GTX470 and the HD5870 would conversely increase as the resolution increases, giving the HD5870 more staying power over the GTX470, at least in terms of average frame rates, if not minimum.

Now, given that the OP is using a resolution between 1920x1200 and 2560x1600, the HD5870 is the one which is going to be the better bet since the average performance gap will be high. If it was 1680x1050 the gap would be lower, and the GTX470 might be the better buy.

What the most used resolution is doesn't matter when this particular user is choosing a card for his particular resolution. If the most widely used resolution was 1280x1024, would you recommend a card based on relative performance at that level? No, you find out the specifics of the user to give the best recommendation, and since the OP is using a reasonably high resolution, the HD5870 seems a reasonable choice over the GTX470 given the average performance at higher resolutions.

Of course, if the OP is more concerned about minimum frame rates, then the GTX470 might be a more preferred solution, if there are specific games the OP plays and GTX470 scores well.

Maybe you could try addressing the OPs question with statements which are somewhat helpful when it comes to picking a card rather than arguing about which page of a review a quote comes from, or talking about "GF100" as an overall architecture (and making incorrect statements about it) when the OP asked a question about two specific cards and a specific resolution.
Unfortunately many sites don't seem to have done GTX470 tests.

Staying power, as in, minimum framerates. All you had to do was ask.

 

[Dark4ng3l]

Don't forget that it has staying power by using twice the power to do the same job.

 

[OCGuy]

Suddenly everyone is Al Gore.

 

[Unkle_Tar]

The poll says it all, overwhelming majority for the 5870. You can ignore peoples bickering back and forth.

 

[BababooeyHTJ]

snip

 

[Dark4ng3l]

Yeah the focus group yells louder because they get stuff in exchange for those services but the huge majority of the forum bought ATI cards and are too busy using them to bother marketing them to you.

Enough with the trolling already. -Admin DrPizza

 

[Lonyo]

Don't forget that it has staying power by using twice the power to do the same job.

We're not comparing the GTX480, it's the GTX470, which uses quite a lot less power than the GTX480.
It also has a little less staying power. In the two games which AT has minimum, or simulated minimum, frame rates, the GTX470 beats out the HD5870 by a small margin, and only uses a bit more power to do so, while costing less.

However, using GTX480 minimums to say the GTX470 has better minimums isn't really appropriate. And obviously nor is comparing GTX480 power when the card in question is the GTX470.

http://www.hardocp.com/article/2010/03/25/nvidia_fermi_gtx_470_480_sli_review/
[H]Ocp seems to indicate that HD5870 has better minimums than the GTX470 sometimes, although because they don't always use the same settings it's not always easy to compare.
AT has the opposite results on its two tests (Crysis and BFBC2), although the cards are close.

What some people don't seem to realise is that the GTX480 isn't the GTX470, so talking about how the minimums of the GTX480 are better than the HD5870 doesn't mean much when the card in question is the GTX470. Nor is the power consumption of the GTX480 in any way relevant.
Unfortunately many sites don't or haven't benchmarked the GTX470, or have done so and compared it more to the HD5850 (which with the [H] review style makes things awkward).

 

[BababooeyHTJ]

Yeah the focus group yells louder because they get stuff in exchange for those services but the huge majority of the forum bought ATI cards and are too busy using them to bother marketing them to you.

Sounds like you're doing a good enough job.

 

[Dark4ng3l]

I was talking about the power draw at idle with dual screens. Yeah I knew id get called out by the marketeers for that comment and I was prepared to back it up.

Enough pro nvidia shouting around here you don't need to waste your time on me when you can use it on them.

 

[Unkle_Tar]

We're not comparing the GTX480, it's the GTX470, which uses quite a lot less power than the GTX480.
It also has a little less staying power. In the two games which AT has minimum, or simulated minimum, frame rates, the GTX470 beats out the HD5870 by a small margin, and only uses a bit more power to do so, while costing less.

However, using GTX480 minimums to say the GTX470 has better minimums isn't really appropriate. And obviously nor is comparing GTX480 power when the card in question is the GTX470.

http://www.hardocp.com/article/2010/03/25/nvidia_fermi_gtx_470_480_sli_review/
[H]Ocp seems to indicate that HD5870 has better minimums than the GTX470 sometimes, although because they don't always use the same settings it's not always easy to compare.
AT has the opposite results on its two tests (Crysis and BFBC2), although the cards are close.

What some people don't seem to realise is that the GTX480 isn't the GTX470, so talking about how the minimums of the GTX480 are better than the HD5870 doesn't mean much when the card in question is the GTX470. Nor is the power consumption of the GTX480 in any way relevant.
Unfortunately many sites don't or haven't benchmarked the GTX470, or have done so and compared it more to the HD5850 (which with the [H] review style makes things awkward).

I don't find their reviews awkward at all, in fact I think they are the best on the net (from the ones I've seen of course). They run the games at the highest settings the cards can handle and play it to see if it runs well enough.

 

[BababooeyHTJ]

I was talking about the power draw at idle with dual screens. Yeah I knew id get called out by the marketeers for that comment and I was prepared to back it up.

Enough pro nvidia shouting around here you don't need to waste your time on me when you can use it on them.

I would rather hear someone who actually owns the card bash it. Not someone who had a bias before reading a review. Judging by a quick glance at your recent post history I'm starting to wonder who are these makreteers that you are referring to. Speaking from experience?

What was that about power draw with dual screens? :biggrin: I've heard enough threads where people need to up their memory clocks on certain 5xxx cards for that to work properly. I'm pretty sure that doesn't help power consumption. 5870 is not without it's flaws either.

 

[thilanliyan]

Suddenly everyone is Al Gore.

Some people actually care about power consumption and hence cooling (of the card and possibly the room you're in).

 

[ZimZum]

Some people actually care about power consumption and hence cooling (of the card and possibly the room you're in).

Thats hippie talk.

 

[Mr. Pedantic]

I don't find their reviews awkward at all, in fact I think they are the best on the net (from the ones I've seen of course). They run the games at the highest settings the cards can handle and play it to see if it runs well enough.

As I said before, while I applaud them for doing this (someone has to, it's all for our benefit, after all), and that there are definite advantages to their testing methodology, the lack of comparable numbers between IQ levels means that you can't compare between GPU configurations that haven't been tested - sure, they tested 480 SLI, but what about tr-SLI? What about quadfire 5970s? Tri-fire 5870s, etc? Whereas if you go with the 'normal' method of GPU benchmarking, you can apply, say, a given scaling factor (say, 60&#37 and get numbers to compare multi-GPU setups to each other.

Suddenly everyone is Al Gore.

My computer is in my bedroom. It is bloody small. I have a 4870 and a Core i7 running 3GHz @ 1.136V. Running a game for an hour increases the temperature in my room by 5C. And you're telling me that I shouldn't care that a certain video card uses 100W more?

 

[Patrick Wolf]

Some people actually care about power consumption and hence cooling (of the card and possibly the room you're in).

True. But, I think the HD 5xxx crowd is more impressed by the performance per watt (efficiency) compared to the better-late-than-never Fermi.

 

[thilanliyan]

Thats hippie talk.

Yes I'm a hippie tree hugger.

 

[Red Storm]

True. But, I think the HD 5xxx crowd is more impressed by the performance per watt per dollar (efficiency) compared to the better-late-than-never Fermi.

Added one important detail for ya.

 

[bryanW1995]

Are you kidding? Most of the posts have been civil even if they disagree. There are a few people who are just trolling, but this is nothing compared to the GT200/4XXX launches.

I'm semi-impressed actually.

I agree. However, I view this as a negative. I'm looking for flamethrowers and hand grenades, but all I'm seeing is matches and firecrackers.

I voted for 470 because I think it's a better choice of the two at a $50 price difference. However, I would pick 5850 over either card in a heartbeat at current prices.

Suddenly everyone is Al Gore

lol

 

[3DVagabond]

The general trend, as expressed in the "cherry picked" quote from an AT page, is that as resolution increases the HD5870 closes the gap to the GTX480.

Just so nobody calls you out on it I want to clarify, The gap closes until you get someone running AA to walk the tightrope between the 5870 and 480 memory amounts. In other words, you can run out of mem on the 5870 and force swapping, which slows it dramatically, while the 480 is still chugging along. These are typically at unrealistic settings where neither card is giving playable frame rates, but % will swing dramatically to the 480. At playable settings the performance gap does indeed close between them.

That said this is about 5870 vs. 470 and that's a no brainer IMO, unless you need/want the proprietary nVidia features. The 5870 is just a better gaming card.

 

[Tempered81]

That said this is about 5870 vs. 470 and that's a no brainer IMO, unless you need/want the proprietary nVidia features. The 5870 is just a better gaming card.

I used to rag on Nvidia about their "proprietary features". But now I'm starting to respect them somewhat for this. It is their strong point. Gpu accelerated physics, in-game MSAA in Batman, their ray tracing, ocean, hair, & rocket sled demos, twimtbp titles, the advanced game profiles menu in their driver control panel.

Ati has great engineers that are really surpassing Nvidia on the technical front. They produce more performance/watt, performance/area, performance/dollar etc... and they produce the fastest card. But the aesthetics of their driver control panel "CCC" are weak compared to Nvidia's "proprietary features". Ati is weak on this front, while Nvidia is very aggressive. Nvidia silicon engineers look weak in the face of ati's design team, while Nvidia's software is far superior to Ati's regurgitated hot garbage in a zip container.

Just like Uriah or Ujesh or whatever his name is could take a hint from Killbrew, Makedon could take a hint from the Nvidia programmers....I mean the CCC looks worse than Ray Adam's program made 5 years ago. In the end they both have problems when the wheels touch ground... the overheating 196.25 driver that broke fans, the overheating VRMS in furmark on 4000 series cards, the lack of detail in the AF on Ati cards in crysis, the piss poor blurred MSAA on Nvidia cards in Metro2033, list goes on.

It's a crap shoot, but in the end i like performance/dollar since I consider myself an enthusiast more than anything. Otherwise I would be a console player. Theres a lot of factors to consider too... No AA in Farcry2 puts a 4870X2 and 5970 way beyond Nvidia cards, and Nvidia cards kick ati's ass in Borderlands performance.

You'll get more efficiency & performance with the 5870 over the 470. But the 470 is a smoother more feature rich card that is slightly cheaper than the 5870. Tough call.

 

[Seero]

Just a few questions.

How many people will play 3 displays? 6 displays?

Who here say 5870 is better but believe that PhysX will have no future if it doesn't work on a ATI card?

How many people are going to get a new 120 LCD in the near future?

Who will likely to check out 3d Gaming in the near future?

Who hate the sound from PC, going for SLI/xfire with the newest, greatest, hotest video cards, yet not using liquid cooling?