5870 vs 470 for an EVGA Fanboy

[MrK6]

Nope. 1920x1200 w/ Medium Setting, DX=9. 50-60 fps

Yeah, "High" settings, HBAO, and 8xAA exist for a reason.

And has it been confirmed that the slow loading is a DX11 issue or an ATI issue yet?

 

[scooterlibby]

It's weird OCGuy's unabashed defense of Fermi and implication that anyone that disagrees is in the can for ATI. He thinks GTX480 is a good deal for him - that's fine! Not everyone ranks preferences in the same manner.

I had GTX260's in SLI (and many other prior Nvidia setups) and and am a huge fan of EVGA's fantastic customer support, but this generation does not seem like a good deal to me. I really enjoy the 5970, but will happily go back to Nvidia in a generation or two from now if it is warranted.

Again, this is just my opinion. A reasonable argument could be made otherwise without being intrinsically biased.

 

[Madcatatlas]

It's weird OCGuy's unabashed defense of Fermi and implication that anyone that disagrees is in the can for ATI. He thinks GTX480 is a good deal for him - that's fine! Not everyone ranks preferences in the same manner.

I had GTX260's in SLI (and many other prior Nvidia setups) and and am a huge fan of EVGA's fantastic customer support, but this generation does not seem like a good deal to me. I really enjoy the 5970, but will happily go back to Nvidia in a generation or two from now if it is warranted.

Again, this is just my opinion. A reasonable argument could be made otherwise without being intrinsically biased.

+1


edit: voted 5870 out of the two options.

 

[jrphoenix]

I have been waiting it out for Fermi, but I'm not really impressed with Nvidia's offering here. I'm considering switching to ATI, but I've only had one bad experience with a laptop card.

So, as I see it, I have two options:

5870:

• Quieter
• Cooler
• Generally better performance

470:

• EVGA/Nvidia product
• Cheaper

I have dual DVI monitors. I've heard the 480 has idle problems with dual monitors - does this affect the 470? I also heard there is a dual monitor flicker issue with the 5870 - any truth to this?

Also, any suggestions on a particular brand of ATI card? Who's the EVGA of ATI?

I just built a new system. I was waiting for Fermi to come out for prices to drop. Now I just don't think it was worth the wait. Prices are staying and it may be near impossible to get a Fermi for months. I just went x-fire with 2 5870's

 

[Axon]

All the hatin' in the GPU forum lately makes me a sad Dino

 

[OCGuy]

All the hatin' in the GPU forum lately makes me a sad Dino

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.

 

[Magusigne]

The whole point of being a fanboi is to go against common sense and to blindly support your company of choice. Regardless of what anybody tells you.

This being said.

eVGA 470 all the way man.

Sure 5870 has been proven to be better in most things. But if your a fanboi thats not supposed to matter. Your supposed to champion your card!

 

[huddy]

I voted 5870, but I would recommend a 5850. Solid card. OC's quite well. You could always have some fun and try to flash it to 5870 speeds (edit, forgot that you can't unlock the shaders on the 58xx series)

 

[MrSpandex]

There are some important considerations here:
What's your CPU?
What resolution are you playing at?
How long do you hope to keep the card?
Do you have a Crossfire/SLI motherboard (and would you want to use Crossfire/SLI)?

I run 2048 x 1152 (I usually compare performance at 1920x1200) as far as resolution.
CPU is a Core i5 750 - so CPU bottleneck shouldn't be too much of an issue.
Not interested in SLI - had a dual-gpu 7950 and was not impressed.
Also, not interested in overclocking.

The 470 runs about 5 degrees hotter then a 8800GT(2 degrees cooler then a 8800GT running Furmark).

Under load, the 470 is just over 2db louder then a 5870(7db louder then a 8800GT).

The more I think about it the more I like just going with nvidia. In the end its only a couple degrees hotter than many cards I wouldn't be a fanboy if I didn't (though those poll results are pretty epic).

Feel free to convince me otherwise. I've probably changed my mind 4 times in the past 24 hours.

 

[Keysplayr]

I run 2048 x 1152 (I usually compare performance at 1920x1200) as far as resolution.
CPU is a Core i5 750 - so CPU bottleneck shouldn't be too much of an issue.
Not interested in SLI - had a dual-gpu 7950 and was not impressed.
Also, not interested in overclocking.


The more I think about it the more I like just going with nvidia. In the end its only a couple degrees hotter than many cards I wouldn't be a fanboy if I didn't (though those poll results are pretty epic).

Feel free to convince me otherwise. I've probably changed my mind 4 times in the past 24 hours.

GF100 series seems to have a bit more "staying power" when things get really cranked up. That is generally what I have found in my own testing. Maintain better minimum framerates especially when all the candy is turned on.

 

[Mr. Pedantic]

Not interested in SLI - had a dual-gpu 7950 and was not impressed.

To be honest, the 7950 wasn't exactly all that impressive anyway. SLI and Crossfire have come a long way since then.

Also, not interested in overclocking.

Why not? Because aside from all the excellent benefits to minimum frame rates you get from overclocking, if you aren't interested and you care more about gaming performance than other things, I would have suggested you get a faster i5 that runs at least 3GHz.

GF100 series seems to have a bit more "staying power" when things get really cranked up. That is generally what I have found in my own testing. Maintain better minimum framerates especially when all the candy is turned on.

Including the resolution? Where the 20% lead GTX480 has over the 5870 at 1680x1050 completely dries up?

 

[Axon]

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 guess? My personal highlight today was "fanboi fail," was rofling for a bit

 

[thilanliyan]

Between a 5870 and a 470 I'd choose a 5870. But between a 5850 and a 470 that would be a harder decision.

 

[Keysplayr]

To be honest, the 7950 wasn't exactly all that impressive anyway. SLI and Crossfire have come a long way since then.


Why not? Because aside from all the excellent benefits to minimum frame rates you get from overclocking, if you aren't interested and you care more about gaming performance than other things, I would have suggested you get a faster i5 that runs at least 3GHz.


Including the resolution? Where the 20% lead GTX480 has over the 5870 at 1680x1050 completely dries up?

And 19x12. I can't speak for 25x16, I don't have a monitor that big.
And also, what is needed is to look for reviews that reallly crank things up. As much as possible. This really does show the strength of GF100.

 

[1h4x4s3x]

5870 obviously

 

[Madcatatlas]

Now that you mention what your resolution is, and the fact that you dont overclock, it looks like a factory overclocked 5870 would be win for you. The MSI Lightning is factory overclocked 5870, no idea about cost.
However, if you are going to SLI or Crossfire, the options even out pretty much and RIGHT now, the greater tesselation power of 470, which we cant be naive about, will probably be of more use in games coming out this year.

Ofcourse we dont know what ATI is planning in terms of making up for the lower tesselation power, but thats nothing you can do something about RIGHT now.
GL in your choice!

 

[SHAQ]

If you are really a fanboy you will get a 470. Otherwise only get a XFX version 5870, unless you can wait for the 5890. thx

 

[blanketyblank]

GF100 series seems to have a bit more "staying power" when things get really cranked up. That is generally what I have found in my own testing. Maintain better minimum framerates especially when all the candy is turned on.

If you're making statements like that you should probably tell him the details or else it's misleading. For instance your own testing was with a GTX 480 and NOT a 470.
This makes a huge difference since the 480 is undisputedly the fastest single gpu so it really isn't a surprise the minimums and even averages would be higher than a 5870.
Are you saying a 470 which you did not test has more "staying power" than a 5870?
If so please explain why and what criteria you use for that statement.

 

[ArchAngel777]

If you're making statements like that you should probably tell him the details or else it's misleading. For instance your own testing was with a GTX 480 and NOT a 470.
This makes a huge difference since the 480 is undisputedly the fastest single gpu so it really isn't a surprise the minimums and even averages would be higher than a 5870.
Are you saying a 470 which you did not test has more "staying power" than a 5870?
If so please explain why and what criteria you use for that statement.

Not that I think Keys was trying to mislead, but I do agree with this post. We can't use results from a 480 and apply them to the 470. While they share the same arch, there is still some significant differences.

 

[Rezist]

If you're making statements like that you should probably tell him the details or else it's misleading. For instance your own testing was with a GTX 480 and NOT a 470.
This makes a huge difference since the 480 is undisputedly the fastest single gpu so it really isn't a surprise the minimums and even averages would be higher than a 5870.
Are you saying a 470 which you did not test has more "staying power" than a 5870?
If so please explain why and what criteria you use for that statement.

Well he mentioned GF100 not the 470 or 480 so technically what he said is not wrong. As long as GF100=GTX480.

 

[Lonyo]

And 19x12. I can't speak for 25x16, I don't have a monitor that big.
And also, what is needed is to look for reviews that reallly crank things up. As much as possible. This really does show the strength of GF100.

Um, no it doesn't.
Most reviews seem to show the gap between the GTX480 and HD5870 closing as you increase resolution.

The most distinguishing result in this benchmark is that it once again shows the Radeon and GeForce performance gap closing with resolution

http://www.anandtech.com/show/2977/...x-470-6-months-late-was-it-worth-the-wait-/18

http://www.xbitlabs.com/misc/pictur...eforce-gtx-480/charts/30_480vs587_big.png&1=1
In 17 of 20 of the non-AA games the HD5870 closes the gap going from 1920 to 2560.
In 11 of 17 of the AA tests, the HD5870 closes the gap from 1920 to 2560.

I didn't count the games where either there is no graph or the results can't be read (Batman AA specifically in that last category). Feel free to go through the review and add that game in though from the numbers.


Hence my point earlier of asking what resolution he would be playing at.
And then there's the CPU limiting performance. Someone with a GTS250 currently probably isn't going to have a 4GHz i7, so the differences in performance might not even be discernable between either the GTX470 and HD5870, or the HD5850 and other cards. And if he's at low resolutions, the GTX470 might be a better idea since the performance gap will be lower, although again an HD5850 might be plenty of power especially if he has a medium level CPU.

 

[Keysplayr]

Um, no it doesn't.
Most reviews seem to show the gap between the GTX480 and HD5870 closing as you increase resolution.


http://www.anandtech.com/show/2977/...x-470-6-months-late-was-it-worth-the-wait-/18

http://www.xbitlabs.com/misc/pictur...eforce-gtx-480/charts/30_480vs587_big.png&1=1
In 17 of 20 of the non-AA games the HD5870 closes the gap going from 1920 to 2560.
In 11 of 17 of the AA tests, the HD5870 closes the gap from 1920 to 2560.

I didn't count the games where either there is no graph or the results can't be read (Batman AA specifically in that last category). Feel free to go through the review and add that game in though from the numbers.


Hence my point earlier of asking what resolution he would be playing at.
And then there's the CPU limiting performance. Someone with a GTS250 currently probably isn't going to have a 4GHz i7, so the differences in performance might not even be discernable between either the GTX470 and HD5870, or the HD5850 and other cards. And if he's at low resolutions, the GTX470 might be a better idea since the performance gap will be lower, although again an HD5850 might be plenty of power especially if he has a medium level CPU.

Not for anything Lonyo, but there were 9 other games benched "before" Wolfenstein in the AT review. Any reason you went to the 9th game out of all of them? Cherry picking?

And about the Xbit link. First off, I want to make sure I'm reading the graphs properly. The center vertical line is the baseline performance in any given game for the 5870 with the settings for all up top. And anything of the left side of that line is less performance than a 5870 and anything on the right side of that line is greater performance than the 5870.

And the "AA 4(8)x" is 8xAA only when available, otherwise 4xAA. I get anything wrong yet?
Let me know.

 

[blanketyblank]

Well he mentioned GF100 not the 470 or 480 so technically what he said is not wrong. As long as GF100=GTX480.

That is precisely why it is misleading. GTX480 is a part of GF100, but GF100 != (not equal) GTX480 since it encompasses the GTX 470 and eventually GTX 460 and maybe others as well.
It is especially since misleading since this thread is asking specifically about the GTX 470 and NOT the GTX 480. Therefore unless the statement applies to BOTH the 470 and 480, it is more wrong than right. Thus Keys should have used 480 in place of GF100 or he should provide evidence that it applies to both which he did not but made it sound like he did.

As an example it is like saying ATI cards are faster than NVidia.
It is true if you compare 5970 to one of NVidia's cards, but it is false if you pick other card comparisons i.e. 5770 vs 480. Unless it holds true for every element of the group the statement is false.

 

[Lonyo]

Not for anything Lonyo, but there were 9 other games benched "before" Wolfenstein in the AT review. Any reason you went to the 9th game out of all of them? Cherry picking?

And about the Xbit link. First off, I want to make sure I'm reading the graphs properly. The center vertical line is the baseline performance in any given game for the 5870 with the settings for all up top. And anything of the left side of that line is less performance than a 5870 and anything on the right side of that line is greater performance than the 5870.

And the "AA 4(8)x" is 8xAA only when available, otherwise 4xAA. I get anything wrong yet?
Let me know.

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.

 

[Keysplayr]

If you're making statements like that you should probably tell him the details or else it's misleading. For instance your own testing was with a GTX 480 and NOT a 470.
This makes a huge difference since the 480 is undisputedly the fastest single gpu so it really isn't a surprise the minimums and even averages would be higher than a 5870.
Are you saying a 470 which you did not test has more "staying power" than a 5870?
If so please explain why and what criteria you use for that statement.

blanketyblank, you reeeeeeeealy need to simmer down about all this. Very uptight and you're suggesting that I am suggesting that a GTX470 does better than a 5870. Hey, in some cases, it certainly does going by AT's review. But for the most part I am MOST certain you knew what I was talking about.

I was speaking relatively. The GF100 SERIES A.K.A. GTX470 and GTX480 have more staying power and better overall minimum framerates than 5850 and 5870 respectively.

I hope this clears that up, and ends your insistence to battle everything I say just for me saying it. Chill dude. It's graphics cards, not organ transplants.