The Ultimate Guide/TOOL to help determine ideal Price/Performance!

[nosfe]

but there's also the issue with 60fps or 120fps? as long as the minimum fps doesn't fall way below 30(tv's are at about 24-30, depending on the standard) you won't notice any difference

 

[kreacher]

Originally posted by: djayjp

I think I addressed the whole 5fps card issue already... such a comparison does NOT exist. It is fiction. Actually, now that I think about it, this comparison CAN NOT exist because card A's performance at that price would be impossible to achieve in comparison to the $100 card. That's how obscenely good of a price-performance ratio card A has compared to an (assumedly) decent $100 card.

Such comparisons do exist, maybe not 5fps but close enough. No offense, but your calculation looks targeted towards accountants rather than gamers. If you really want to create a useful guide for gamers you will have to give more weight to factors like resolutions, AA, future games, etc.

 

[djayjp]

Originally posted by: kreacher

Originally posted by: djayjp

I think I addressed the whole 5fps card issue already... such a comparison does NOT exist. It is fiction. Actually, now that I think about it, this comparison CAN NOT exist because card A's performance at that price would be impossible to achieve in comparison to the $100 card. That's how obscenely good of a price-performance ratio card A has compared to an (assumedly) decent $100 card.

Such comparisons do exist, maybe not 5fps but close enough. No offense, but your calculation looks targeted towards accountants rather than gamers. If you really want to create a useful guide for gamers you will have to give more weight to factors like resolutions, AA, future games, etc.

I don't think you appreciate why this is the 'Ultimate' guide-- it's because it is eternal (well, as long as moore's law holds), other wise, I would have to update this thing every friggin' month! You can use the methods here to ALWAYS guide your purchasing decisions. ALSO, a big issue is that each person has different needs... just like you said-- resolution, aa, games, etc. The point is that the guide gives YOU tools which will NEVER go out of date. thanks for the positive feedback, keep it comin' guys

 

[garritynet]

PLEASE... by all means, feel free to show me any section which could accurately be described as "nearly unreadable". thanks a bunch.

I could just quote your first post but I think we already know what I am talking about. Its needlessly dense, poorly formatted and reads like transcribed rant rather than a guide. I do not mean come across as insulting, just offering a helpful critique to improve your 'ultimate' guide.

 

[garritynet]

Actually, now that I think about it, this comparison CAN NOT exist because card A's performance at that price would be impossible to achieve in comparison to the $100 card. That's how obscenely good of a price-performance ratio card A has compared to an (assumedly) decent $100 card.

Your continued insistence that a card that delivers 5fps at any cost is somehow a good value to gamers demonstrates that your guide is flawed. A 5fps card is of no functional value to gamers. Its performance might be a technical 5fps but as far as practical applications in computer gaming it might as well be zero. Stick a zero in your equations and you'll end up with the same result as us.

Also: I don't think any disagrees that the 4670 is a decent budget card.

 

[Kelvrick]

This guide is horrible. I don't even want to spend the time saying how horrible it is.

 

[nyker96]

3dmark isn't good estimate. 9600gt scores like 9k, GTX 280 scored mere 11k but at high rez high AA/AF, 9600gt will crawl to a stop in real games while GTX280 will be playable. This is not a good test for what you getting in real life for games, especially at high rez/high setting/high AA AF.

but it's good effort nevertheless OP, good job!

 

[Munky]

Originally posted by: djayjp

Originally posted by: munky
The needs also depend on your monitor. I personally use a 27" screen at 1920x1200, so I need at least an 8800gt-level of performance

4670= 8800gs, but is MUCH cheaper!

Yes... but I was talking about a 8800gt

 

[bryanW1995]

Originally posted by: djayjp

Originally posted by: razor2025
I think the idea is great, except the method and formula is inherently faulty as far as consumer decisions go. Take for example 3850/4670 vs 9600GSO

http://www.anandtech.com/video/showdoc.aspx?i=3405&p=6

4670>3850

performance comparison between 4670 and 9600gso:

http://www.anandtech.com/video/showdoc.aspx?i=3405&p=7

Regarding the 9600gso, the gddr3 version retails for at least $105; the 4670 retails for $61. 4670 has significantly greater performance than the 3850. According to the formula (revised it would be 1.5x, though personally I now think 2.2x is more optimal), since 9600gso costs 72% more, its performance should at the very least be 108% higher to justify the greater cost. Case closed.

And what is this "chart in OP" that you refer to? If I will make a leap and assume you are talking about overall performance (?). I don't see such a chart on the link you posted.

And for sure a big part of one's decision should be based on the intended application, but the "intended application" is usually constantly changing regarding games and users upgrade frequently, thus some kind of overall performance stat is required to make an informed decision (also to help predict future overall performance). The purpose of this method is to determine where one stands on the price/performance curve and how much one is either gaining or losing each and everytime one upgrades. It also crucially argues that simply taking the dollar value and dividing by FPS is flawed on its own and that you can't say that getting 100% more performance is worth 100% more money.

you need to do more checking on your prices. I bought a 9600 gso for $37 AR a few months ago.

 

[djayjp]

Originally posted by: garritynet

PLEASE... by all means, feel free to show me any section which could accurately be described as "nearly unreadable". thanks a bunch.

I could just quote your first post but I think we already know what I am talking about. Its needlessly dense, poorly formatted and reads like transcribed rant rather than a guide. I do not mean come across as insulting, just offering a helpful critique to improve your 'ultimate' guide.

So you agree you were not being accurate then when you essentially said that it was barely intelligible. Myself, I tend to think that it's about 10x better written than the average forum post. Yeah, I suppose part of the problem is that I set people's expectations too high right off the bat with the title... but in a way it's true because it never becomes outdated, you just re-apply for each new generation.

 

[djayjp]

Originally posted by: Kelvrick
This guide is horrible. I don't even want to spend the time saying how horrible it is.

Thanks a whole lot for the hugely constructive feedback... very much appreciated! now get lost

 

[djayjp]

Originally posted by: munky

Originally posted by: djayjp

Originally posted by: munky
The needs also depend on your monitor. I personally use a 27" screen at 1920x1200, so I need at least an 8800gt-level of performance

4670= 8800gs, but is MUCH cheaper!

Yes... but I was talking about a 8800gt

Yeah, cuz we all know that 10% extra is sooo worth it! (96 vs 112 shaders... er unless it has half the mem bandwidth, can't remember) And actually, the 4670 has slightly higher performance than the 8800gs

 

[djayjp]

Originally posted by: nyker96
3dmark isn't good estimate. 9600gt scores like 9k, GTX 280 scored mere 11k but at high rez high AA/AF, 9600gt will crawl to a stop in real games while GTX280 will be playable. This is not a good test for what you getting in real life for games, especially at high rez/high setting/high AA AF.

but it's good effort nevertheless OP, good job!

You've got a decent point with 3dmark06, I just sort of bolted it on top there as a little extra frame of reference/different perspective (which actually ends up agreeing rather well with my prediction/assessment)

THANK YOU!!! That's the very first truly solid compliment I've received so far!

 

[Munky]

Originally posted by: djayjp

Originally posted by: munky

Originally posted by: djayjp

Originally posted by: munky
The needs also depend on your monitor. I personally use a 27" screen at 1920x1200, so I need at least an 8800gt-level of performance

4670= 8800gs, but is MUCH cheaper!

Yes... but I was talking about a 8800gt

Yeah, cuz we all know that 10% extra is sooo worth it! (96 vs 112 shaders... er unless it has half the mem bandwidth, can't remember) And actually, the 4670 has slightly higher performance than the 8800gs

LOL... 10% is what the 9600gt fanboys were claiming the deficit from a 8800gt was, and it turns out they were too optimistic when the going got tough on the video card. 8800gs is not even in the same league as a 8800gt.

 

[djayjp]

you need to do more checking on your prices. I bought a 9600 gso for $37 AR a few months ago.

OK, now that's what I call value!!! holy crap! BUT, is that the ddr2 version? Yeah, all I did was check newegg, the real-world comparison part isn't meant to be uber thorough, it's meant to show you trends in the market and product pricing vs relative performance, and just to get you used to applying the price/performance method which will never go out of date, unlike the seemingly day to day flux of prices.

 

[djayjp]

LOL... 10% is what the 9600gt fanboys were claiming the deficit from a 8800gt was, and it turns out they were too optimistic when the going got tough on the video card. 8800gs is not even in the same league as a 8800gt.

can you be more specific? What is it about the specs that would cause a significant decrease in performance? Is the mem bandwidth halved? In that case at your rez, I highly recommend the gt

 

[djayjp]

Originally posted by: garritynet

Actually, now that I think about it, this comparison CAN NOT exist because card A's performance at that price would be impossible to achieve in comparison to the $100 card. That's how obscenely good of a price-performance ratio card A has compared to an (assumedly) decent $100 card.

By garritynet: "Your continued insistence that a card that delivers 5fps at any cost is somehow a good value to gamers demonstrates that your guide is flawed. A 5fps card is of no functional value to gamers. Its performance might be a technical 5fps but as far as practical applications in computer gaming it might as well be zero. Stick a zero in your equations and you'll end up with the same result as us."

my response: First of all, you are wrong... it's NOT "at any cost", it's at $1... this point is crucial. And it doesn't have an ABSOLUTE performance of 5fps... just in that ONE game! That leaves one able to play tons of games that are 1-2 years older!

Ok, here's the --full-- low-down on this whole uber-hypothetical mess. First, however, perhaps the comparison itself here is supremely flawed to begin with... no way! impossible! Like how about the fact that it's with ONE mysterious game X that nobody might care about running at some mysterious settings...

Alright, here we go-- Games are designed to be able to be played on several different generations of hardware. Take a look at crysis: low, medium, high, very high... that's AT LEAST 4 generations worth of performance range there (probably more if you turn on and off various settings combinations; also, that's with the game in stock mode, whereas if you customize a config, you can get WAY better performance-- I can get the game to run on what looks virtually indistinguishable from High on my 8400m GT laptop card! Considering this point, you can probably squeeze 8 generations worth of performance range out of crysis!). SO, if card B gets an ethereal 45fps, which is 9x the performance of card A we might assume that it's running at the highest spec, in which case we can see that Low is gen 1 running at 1x speed, we multiply that by 2x2x2 and you get a factor of 8, therefore, you can take the 5 and multiply it by 8 and... voila! 40fps at lowest settings! ALL FOR A MIRACULOUS $1. Imagine if ALL computers shipped with decent graphics like that for essentially free (compare this 9:1 performance delta vs intel's MORE EXPENSIVE integrated gfx turds). You really seem to be missing the point about moore's law-- you constantly are getting large increases in performance for free, if you go ahead of that curve, you are always losing (UNLESS, like I said, you have certain specific NEEDS which must be met, these would outweigh the price/performance benefits and is at the discretion of the individual buyer to decide the best value for THEM... AS I CLEARLY STATE TIME AND TIME AND TIME AGAIN IN THE FRIGGIN GUIDE!!!!) Sorry, i'm just getting tired of repeating the same point.... That being said, I've been considering adding a weighted part to the formula which rewards fps' that are close to 30fps-- Like: if one card gets 15fps and another gets 30, the second card will be considered >2x

 

[garritynet]

By "at any cost" I simply meant that it could cost .01 and it would still be without value. I probably should have put some commas in there.

When I stated "I could quote your first post" I was not admitting that I had made an inaccurate claim. I was stating that your entire first post is what I meant by "unreadable".

And it doesn't have an ABSOLUTE performance of 5fps... just in that ONE game! That leaves one able to play tons of games that are 1-2 years older!

No body upgrades their video card to play games from two years ago at 45fps.

That being said, I've been considering adding a weighted part to the formula which rewards fps' that are close to 30fps-- Like: if one card gets 15fps and another gets 30, the second card will be considered >2x

I appreciate your efforts to come up with some math formula to help people pick video cards. The fact is that there is already a pretty easy "rule of thumb" as it is.

I just tend to go with the cheapest card I can get that will run the games I'm currently playing well enough

It's not that complicated. Buy the cheapest card that will give you >30 fps at the native resolution of your monitor in the games you want to play.

or the ubiquitous

Buy middle of the road. Stay middle of the road.

In order to figure out the FPS for your formula you have to find up benchmarks. So if your gonna look at bench marks just look at the ones for games you want to play. See what cards play it well at the resolution that best matches your monitor and your done. Then see if any of those are particularly cheap. No need to divide anything.

Honestly though, the people of the Video Card forum do a pretty good job keeping up with the best values as it is. Anyone can post a quick thread and within hours have several solid recommendations. No math needed.

 

[razor2025]

Originally posted by: djayjp

http://www.anandtech.com/video/showdoc.aspx?i=3405&p=6

4670>3850

performance comparison between 4670 and 9600gso:

http://www.anandtech.com/video/showdoc.aspx?i=3405&p=7

In that review, there's no data on CoD4 Performance. Sure, 4670 is a better card if you looked at ONLY the Crysis scores. Again, it goes to my point that you buy hardware tailored to your software requirements. Not the other way around. 9600GSO is quite competetive against 4670 in other applications.

Originally posted by: djayjp
Regarding the 9600gso, the gddr3 version retails for at least $105; the 4670 retails for $61. 4670 has significantly greater performance than the 3850. According to the formula (revised it would be 1.5x, though personally I now think 2.2x is more optimal), since 9600gso costs 72% more, its performance should at the very least be 108% higher to justify the greater cost. Case closed.

Where do you get $61 figure for 4670? Average price for 4670 I've seen are ~$70 shipped. For 9600GSO GDDR3 version, I found this:
http://www.newegg.com/Product/...x?Item=N82E16814500083
$72 shipped. And if you know anything about 9600GSO, the older DDR2 version is actually FASTER. Which can be hard for $78:
http://www.newegg.com/Product/...x?Item=N82E16814150320

If you're going to compare pricing, please make it fair. Apples to Apples. Real world cost of 9600GSO vs 4670 is roughly the same. Not this magical "72%" you speak of.

Originally posted by: djayjp

And what is this "chart in OP" that you refer to? If I will make a leap and assume you are talking about overall performance (?). I don't see such a chart on the link you posted.

And for sure a big part of one's decision should be based on the intended application, but the "intended application" is usually constantly changing regarding games and users upgrade frequently, thus some kind of overall performance stat is required to make an informed decision (also to help predict future overall performance). The purpose of this method is to determine where one stands on the price/performance curve and how much one is either gaining or losing each and everytime one upgrades. It also crucially argues that simply taking the dollar value and dividing by FPS is flawed on its own and that you can't say that getting 100% more performance is worth 100% more money.

I'm referring to the Ultimate guide chart. The purpose of the guide is commendable, but the methods and formula used to create the data points are faulty. The analysis and the formula over-generalizes the performance of all the cards; it uses 3Dmark06, which anyone experienced with video card performance analysis knows that it's FAR from reliable indicator of performance. If they're going to produce a chart, make it useful. It seems like a simple slap job to me. If you're going to call it something with "Ultimate" in it, it better be ultimate in comprehensiveness. I want high-resolutions and actual GAME performance/$.

I do think that 4670 is an excellent mid-range card that provides useful performance in low-power fashion. However, to judge it as being astronomically better than closest Nvidia offering is simply being disingenuous.

 

[LOUISSSSS]

i too think the guide is horrible, people are subject to their opinion and the waste of time doing all these calculations isn't even worth the $10 i MIGHT lose buying the card i WANT.

people will STILL buy the card that gives them the performance they want for the game they play.

your guide only allows people to compare one game from one source. big waste of time imo

 

[djayjp]

Originally posted by: garritynet
By "at any cost" I simply meant that it could cost .01 and it would still be without value. I probably should have put some commas in there.

ANYTHING that gives you decent performance (crysis on low at 40fps) for $1 or .01 is NOT w/o value.. stop your weak attempts at warping the truth, you'd make a terrible lawyer.

When I stated "I could quote your first post" I was not admitting that I had made an inaccurate claim. I was stating that your entire first post is what I meant by "unreadable".

Everyone else seemed to be able to read it... you're saying you can't??? once again, stop your weak attempts at warping the truth.

And it doesn't have an ABSOLUTE performance of 5fps... just in that ONE game! That leaves one able to play tons of games that are 1-2 years older!

No body upgrades their video card to play games from two years ago at 45fps.

Once again, it plays a new game (crysis warhead) at more than 40fps on low. Stop your weak attempts at warping the truth.

That being said, I've been considering adding a weighted part to the formula which rewards fps' that are close to 30fps-- Like: if one card gets 15fps and another gets 30, the second card will be considered >2x

I appreciate your efforts to come up with some math formula to help people pick video cards. The fact is that there is already a pretty easy "rule of thumb" as it is.

I just tend to go with the cheapest card I can get that will run the games I'm currently playing well enough

It's not that complicated. Buy the cheapest card that will give you >30 fps at the native resolution of your monitor in the games you want to play.

or the ubiquitous

Buy middle of the road. Stay middle of the road.

In order to figure out the FPS for your formula you have to find up benchmarks. So if your gonna look at bench marks just look at the ones for games you want to play. See what cards play it well at the resolution that best matches your monitor and your done. Then see if any of those are particularly cheap. No need to divide anything.

The point is that with my method, you can ACTUALLY KNOW, not guess or go by anecdotal recommendations; video card manufacturers constantly try to manipulate you into upgrading... one of the central points with my method is that it makes it entirely OBJECTIVE (if you want it to be) so that you don't accidentally get fooled into paying more for something that merely looks better on a graph chart-- the whole popular misconception regarding the value of a 1:1 performance/price ratio is a great example of this... once again, it is a guide, or tool, for you to use IN CONJUNCTION WITH YOUR NEEDS!!!

 

[djayjp]

Originally posted by: LOUISSSSS
i too think the guide is horrible, people are subject to their opinion and the waste of time doing all these calculations isn't even worth the $10 i MIGHT lose buying the card i WANT.

people will STILL buy the card that gives them the performance they want for the game they play.

your guide only allows people to compare one game from one source. big waste of time imo

Thank you SOOO much for the constructive feedback! Much appreciated..., now, unless you want to provide constructive criticism with fair, balanced, and just feedback, please get lost.

First of all, it takes as long as it does to read a couple reviews.

Where do you see that anywhere it says you're only allowed to compare one game?

The point is is that it saves you from being one of those people who goes... oooh wow a "9500gt" sounds great! When, in fact, for $5 more you get nearly 3x the performance. ONCE AGAIN, it is a guide/tool to let you OBJECTIVELY determine if you are getting the best bang for your buck, rather than guessing or looking at pretty charts. The ONLY ultimate value is determined by moore's law, if you always go beyond that point EVERYTIME you upgrade... is it really a waste of time? At the very least, it seems valuable determining where one's actual purchased card falls on this curve, or do you prefer the ignorance is bliss approach? This is a tool to help make you an informed, empowered consumer, not a victim of marketing (like thinking 1:1 performance/price ratio is good value), or vague recommendations.

 

[djayjp]

Originally posted by: razor2025

Originally posted by: djayjp

http://www.anandtech.com/video/showdoc.aspx?i=3405&p=6

4670>3850

performance comparison between 4670 and 9600gso:

http://www.anandtech.com/video/showdoc.aspx?i=3405&p=7

In that review, there's no data on CoD4 Performance. Sure, 4670 is a better card if you looked at ONLY the Crysis scores. Again, it goes to my point that you buy hardware tailored to your software requirements. Not the other way around. 9600GSO is quite competetive against 4670 in other applications.

I think the point is that in most applications it is ever so slightly better-- whatever, let's call it a tie, the point in the original post is that the price sucks vs the 4670.

Originally posted by: djayjp
Regarding the 9600gso, the gddr3 version retails for at least $105; the 4670 retails for $61. 4670 has significantly greater performance than the 3850. According to the formula (revised it would be 1.5x, though personally I now think 2.2x is more optimal), since 9600gso costs 72% more, its performance should at the very least be 108% higher to justify the greater cost. Case closed.

Where do you get $61 figure for 4670? Average price for 4670 I've seen are ~$70 shipped. For 9600GSO GDDR3 version, I found this:
http://www.newegg.com/Product/...x?Item=N82E16814500083
$72 shipped. And if you know anything about 9600GSO, the older DDR2 version is actually FASTER. Which can be hard for $78:
http://www.newegg.com/Product/...x?Item=N82E16814150320

If you're going to compare pricing, please make it fair. Apples to Apples. Real world cost of 9600GSO vs 4670 is roughly the same. Not this magical "72%" you speak of.

Sorry, I just looked at newegg and saw that the norm for the 9600gso is its ddr3 version with DOUBLE the mem bandwidth, so I figured the ddr2 version was some cheap, cut down version (like I saw one on there with gddr3 w/ 48 shaders... ugh!). I think I only saw ONE ddr2 version, so I didn't count that one. I didn't look at shipped figures, I looked at the retail price including rebates and before tax. IF that model you are referring to could be considered close to the reference model used in reviews, THEN, you're right, it IS in fact a
good buy. HOWEVER, look at the WHOLE list!:

http://www.newegg.com/Product/...iption=9600gso&x=0&y=0

What may be reference is far more expensive at $105, I dunno, I'm too lazy to look into it at the moment, but if you can find me the reference card stats that match this lower cost 9600gso, then yes, it MIGHT just be as good as the 4670.

Originally posted by: djayjp

And what is this "chart in OP" that you refer to? If I will make a leap and assume you are talking about overall performance (?). I don't see such a chart on the link you posted.

And for sure a big part of one's decision should be based on the intended application, but the "intended application" is usually constantly changing regarding games and users upgrade frequently, thus some kind of overall performance stat is required to make an informed decision (also to help predict future overall performance). The purpose of this method is to determine where one stands on the price/performance curve and how much one is either gaining or losing each and everytime one upgrades. It also crucially argues that simply taking the dollar value and dividing by FPS is flawed on its own and that you can't say that getting 100% more performance is worth 100% more money.

I'm referring to the Ultimate guide chart. The purpose of the guide is commendable, but the methods and formula used to create the data points are faulty. The analysis and the formula over-generalizes the performance of all the cards; it uses 3Dmark06, which anyone experienced with video card performance analysis knows that it's FAR from reliable indicator of performance. If they're going to produce a chart, make it useful. It seems like a simple slap job to me. If you're going to call it something with "Ultimate" in it, it better be ultimate in comprehensiveness. I want high-resolutions and actual GAME performance.

Hey there Razor, it seems the formatting on the post got messed up, look within it for my whole answer.

Actually, that 3dmark06 post was just added on top and was NOT the basis for the whole guide... I wish, that would make it a whole lot faster coming up with those numbers! No, since there was such a gamut of cards that I went through, I had to go through a number of different benchmarks (OF COURSE only comparing two cards in the SAME benchmark!). Several benchies were taken from anandtech. I agree about 3dmark06-- it's not very reliable, and I prefer real-world game tests to get my numbers from, as is the case with ALL numbers in the guide. ONCE AGAIN, the purpose of this guide is not to maintain a current price/performance guide, those are just EXAMPLES there to help you get familiar with the method for determine optimal price/performance... if I were to keep it current, it would have to be nearly continuously updated (due to the flux in prices and new cards upsetting ratios). What makes it so 'Ultimate' compared to any other guide is that it never becomes obsolete (the examples practically do though overnight!), that is until moore's law does!

 

[garritynet]

ANYTHING that gives you decent performance (crysis on low at 40fps) for $1 or .01 is NOT w/o value.. stop your weak attempts at warping the truth, you'd make a terrible lawyer.

Who is talking about running Crysis at 40fps for $1. As I recall we were talking about running a game a 5 fps.

Everyone else seemed to be able to read it... you're saying you can't??? once again, stop your weak attempts at warping the truth.

It is very poorly composed. I think that you would be able to communicate your ideas more effectively with some editing and abridging of your original post.

Once again, it plays a new game (crysis warhead) at more than 40fps on low. Stop your weak attempts at warping the truth.

What are you talking about here? I was referring to this comment by you:

SO, if card B gets an ethereal 45fps, which is 9x the performance of card A we might assume that it's running at the highest spec, in which case we can see that Low is gen 1 running at 1x speed, we multiply that by 2x2x2 and you get a factor of 8, therefore, you can take the 5 and multiply it by 8 and... voila! 40fps at lowest settings!

In this comment you did some bizarre math where you figured that this 'card A', the one dollar card with 5fps will now run Crysis at 40fps. I mean you are literally saying that because 2^3 multiplied by 5 equals 40 then the imaginary card that I4AT created will also run the most graphic intensive game available at 40fps on low. On what reality is this grounded?

You really seem to be missing the point about moore's law-- you constantly are getting large increases in performance for free, if you go ahead of that curve, you are always losing (UNLESS, like I said, you have certain specific NEEDS which must be met, these would outweigh the price/performance benefits and is at the discretion of the individual buyer to decide the best value for THEM

No one buys video cards just so they can stay in the optimal position of a curve some guy on the internet made up. We all have 'specific needs' which 'outweigh' any imagined 'benefits' that you might be referring to.

 

[LOUISSSSS]

1. so what card according to your math would give my system the best performance/$$ to play cod3 @ 16x10 @ 8xaa 8xaf
see the system in my signature (minus the video card of course) for reference.

2. same question as above, but what if my system had a Pentium D 3.2ghz?

3. What card according to your math would give best performance/$$ to play cod3 @ 16x10 @ 0xaa 0xaf