tweaktown review GTX660Ti

[SirPauly]

Imho,

Gamers have different needs and what they deem important may be totally different from each other. Instead of trying to force my subjective view as some sort of rule for gamers, well, I simply share it and allow the market to decide what is truly important.

A little respect, kindness and tolerance goes a long way in forums.

The way I look at it is like this: nVidia has a strong brand name based on their pro-active nature and this encompasses a great deal to me; from developer relations, long term support, feature differentiation, trying to bring that bit more to improve the gaming experience. Some may disagree and it's all good but this pro-active nature may allow a bit of a premium for nVidia and some added advantage for the consumer.

nVidia, offers competitive price/performance but really never dominated it and really believe this pro-active nature is the differentiation from AMD. I also believe AMD is more reactive than proactive.

Since the 3XXX series, AMD has offered impressive price/performance and continues to do so now with stiff competition, but 'till AMD can somehow outwork and be more pro-active than nVidia -- see the same price/performance claims, nVidia owners are not too bright -- AMD clearly has the better price/performance and obviously the better buy.

Why does Kepler sell? Based on the nVidia brand name coupled with a balance of efficiency and performance; and also continuing to raise the bar of innovation and improved gaming experiences though GPU Boost and adaptive V-sync.

 

[RussianSensation]

ocre, I think you are misunderstanding where we are coming from. We don't care if GTX660Ti will sell well or not. On this forum we are not shareholders (maybe some people are). We discuss hardware and recommend it based on its technical and performance metrics and compare it based on price. If NV released a $3,000 GPU and it sold 120,000 units, most of us could care less about that. They did, it's called K10 and K20 Tesla series. I am sure NV will do really well on its earnings report August 9th because of those pre-orders as well as Tegra 3. But here, we would discuss how a $450 HD7970 has higher double-precision compute performance than K10 for a fraction of the price, and would deem K10 a massive failure for compute projects that are run on our boards because it costs 5-6x more.

If you want to discuss how well NV is doing there are lots of financial forums such as Trefis.com, SeekingAlpha and so on. They look at margins, growth rates, market share, etc. They care if GTX660Ti will sell well since they actually invest into the stock.

On this forum, when we are discussing that GTX660Ti may be shaping up to be a disappointment against the 7950, it's strictly from a technical and performance/$ perspective, and has nothing to do if it'll make $ for NV. That's why it's the Videocard and hardware forum, not financial / investor forum. Whether or not 10 million NV fanboys buy the card has no merit of what we think of it on AT. NV's stock price could increase 2x because of GTX660Ti, but that has nothing to do with whether or not GTX660Ti is worth buying at $300 for us gamers. Will it beat GTX570 OC? Will it tank with Anti-aliasing? These things matter for gamers looking to upgrade. Investors don't care about these things.

For example, GTX550Ti was garbage compared to HD6850 and yet last holiday season, it was the best selling GPU on Amazon. Who cares if it sold well. On this forum some of us want to educate people who don't follow hardware to not waste their hard earned dollars on garbage such as the GTX550Ti. Personally if I could convince those 10 million people to get a GTX460 over the GTX550Ti after visiting our forum, I would feel that I provided useful advice that got them the better product, which means in the end they would get a better gaming experience. It's not about recommending a person an NV or AMD card but allowing a person to spend their $ wisely and not overpay for poor products such as the 2900XT, GTX550Ti, Bulldozer, etc. It would be no different than discussing the prices and performance of high-end audiophile equipment at Head-fi.org for example. We praise and criticize certain products. Our perspective changes if drivers change performance, if new games favour one brand or the other, if someone wants specific features from each vendor, if prices change the standing of the cards, etc.

If this was a car forum and someone was looking to buy a fuel efficient car and we knew that Car A was better than Car B, but even if Car B sold 10x more units, why would we recommend Car B? Everyone here already knows that GTX660 series will sell like hot cakes. NV can put out a tin-can with NV logo and it will sell - FX5800/5900 series. Those were the worst videocards made in the history of NV, and became completely useless in all DX9 games and yet sold like hot cakes. 5200Ultra. *cough* *cough*. D:

For example, NV being late by 6 months isn't being criticized here from a company's strategy or market share perspective, but we discuss it from a point of view of what we expect a 6 months older GPU to cost and to perform relative to 6 months older 7850/7870/7950 cards.

 

[SirPauly]

We discuss hardware and recommend it based on its technical and performance metrics and compare it based on price.

Gaming is much more than just technical, performance metrics and price -- for me, it is about the gaming experience and what kind of enjoyment does the product offer? Immersion, fun, quality of image, quality of the AA and AF samples, what is being displayed, feature differentiation, flexibility of settings.

I've been posting since 1999 with the same Alias and came up in the 3dfx forums with Dave Baumann, when he was known as Wavey -- and not once have I ever recommended a GPU. What I like is as much data as humanly possible so gamers can make up their own minds.

 

[Jaydip]

Well one thing I find particularly stupid is branding anyone as a fanboy.I prefer nv products but that doesn't mean all amd card purchasers are amd fanboy.Thats just lame.Very few people read reviews and ask for help in tech forums before buying their card.They may be just ill informed.The fanboy crap sells like hotcake here.We are tech enthusiasts good but we should be civilized first to start a serious tech discussion.

 

[SirPauly]

Indeed!

 

[RussianSensation]

and not once have I ever recommended a GPU. What I like is as much data as humanly possible so gamers can make up their own minds.

Ok, for example, GTX460 with overclocking reached HD5870 performance in some cases. That's your information. Then someone comes in and asks what the best card is for $200. What do you say? Nothing? You could say GTX460 because it has overclocking headroom to 825-850mhz. You don't have to but some of us will and it's perfectly acceptable to state that GTX460 ~ HD6850 but if you are going to be overclocking it's probably the faster card overall. That's not biased against AMD in any way. Many of us recommended GTX460s for a long time based on its technical merits and price/performance.

I realize all those other things you mentioned are important as well but NV it's much harder to present those "soft differences" because they are so difficult to measure. I mean you could try proving to the forum that NV has superior image quality advantage with AA or AF at the moment, but it would be nearly impossible. You can measure SLI vs. CF micro-stutter or FPS though and that's quantifable. That's what it comes down to. No one ignores thigns such as PhysX or Adaptive Vsync but is Adaptive Vsync on a slower card a more important feature than performance? Not for 99% of people it's not.

For example, NV does have advantages:

1) 3d Vision surround (prob. not possible on GTX660Ti in modern games at high frames)
2) Triple monitor gaming from 1 kepler card (for sure not possible on the 660Ti)
3) PhysX - Name 1 good game in 2012 that used PhysX and 2 good games in 2011?
4) Better NV drivers...ya we've been over this myth for single GPUs.

You could bring all those up, but it would be just as useful as discussing Eyefinity for someone who only games on 1 monitor.

Well one thing I find particularly stupid is branding anyone as a fanboy.I prefer nv products but that doesn't mean all amd card purchasers are amd fanboy.Thats just lame.Very few people read reviews and ask for help in tech forums before buying their card.They may be just ill informed.The fanboy crap sells like hotcake here.We are tech enthusiasts good but we should be civilized first to start a serious tech discussion.

So how would you explain GTX550Ti that consumes more power than a 6850 and performs worse outselling it on Amazon last holiday season:

1) Uninformed consumer;
2) NV fanboys;
3) AMD drivers suck;
4) Person A, B, C, D told me NV is the premium GPU brand;
5) GTX580 is the fastest GPU, so GTX550Ti must also be better than whatever AMD has
6) But it has PhysX (GTX550Ti can't play modern games and PhysX at the same time).
7) I've been buying NV only and I don't want to switch.

I am not saying everyone who buys an NV/AMD card is an NV/AMD fanboy, not at all. But if someone still buys a GTX550Ti for the same price as the 6850 after seeing the information, they are either stupid, or brand biased (i.e., NV fanboy). What do you think a fanboy is? A fanboy isn't someone who buys a superior GTX580 over the 6870. A fanboy is someone who buys the slower GTX570 to play Metro 2033 or Crysis 2 over the 7950 because NV made it and they think it's "better." There is a big difference between preferring a brand because it actually brings something to the table over the competition and buying something only based on brand value.

If Fruit of the Loom t-shirts were taken off the assembly line and stamped as Chanel underwear, would you drop $100 on each? There are plenty of high-end brands that actually have better quality attached to them and then there are poseurs that charge a huge premium only because of the brand. NV and AMD both have good and bad cards in their line-ups. Sometimes the price is justifie and sometimes it isn't. A fanboy would buy an $500 FX5800U and not care that it's the worst high-end GPU ever made, until NV tells him so.

This is no differnet than someone buying FX8150 because they'll "never buy an Intel CPU just because [insert some reason]." Or they'll try to claim that all benchmarks where FX8150 loses were paid for by Intel or optimized for Intel and are being tested unfairly against the FX8150. That's not fanboism?

Stating that you prefer NV, for example, because you use PhysX extensively or run Badaboom or play with 3dVision in Surround on 1 GPU etc. etc. is providing objective reasoning for the statement that shows some merit because AMD cannot offer such features. Buying only 1 brand because "I won't buy AMD no matter what" actually is fanboysm. It's basically dismissing a competitor without any objective reasoning for the dismissal. That's exactly what fanboysm is.

This article explains it a lot better than I can:

The Science of Fanboyism

For example, in simplest terms, if GTX680 was preferable over the 7970 3 months ago because it was faster and cheaper, some people bought the 680, justifiably so. Then let's say HD7970 becomes faster and $100 cheaper over time. Some of those rational individuals will no longer look unfavourably towards the originally rejected alternative (i.e., 7970). Even though they rejected it initially, they can objectively reevalute it as conditions change. A fanboy would be someone who bought the 680 and will forever continue to recommend it over the 7970 because they have rejected the slower product in the first place, regardless of what happens to the performance of the 7970 or the price relative the 680 later on, etc.

So I guess by definition of "fanboyism", I would call myself a fanboy of Price/Performance. And I guess SirPauly would be Price vs. Performance + Features + Supporting a company that tries to make the gaming industry better through unique industry innovations.

Features are often very hard to quantify because they are too subjective. Price, performance, power consumption, game bundles, resale value, etc. can be quantified, which is why they are often easier to discuss and measure.

 

[Jaydip]

I believe 90% purchasers out there are just ill informed.They stick to the overall brand image of the manufacturer regardless of the underlying performance.They are not fanbois by any stretch.Also generally speaking NV cards cost more so they believe they are faster.Not so honest retailers have no reason to rectify that either for their own account.

 

[SirPauly]

Ok, for example, GTX460 with overclocking reached HD5870 performance in some cases. That's your information. Then someone comes in and asks what the best card is for $200. What do you say? Nothing? You could say GTX460 because it has overclocking headroom to 825-850mhz. You don't have to but some of us will and it's perfectly acceptable to state that GTX460 ~ HD6850 but if you are going to be overclocking it's probably the faster card overall. That's not biased against AMD in any way. Many of us recommended GTX460s for a long time based on its technical merits and price/performance.

I realize all those other things you mentioned are important as well but NV it's much harder to present those "soft differences" because they are so difficult to measure. I mean you could try proving to the forum that NV has superior image quality advantage with AA or AF at the moment, but it would be nearly impossible. You can measure SLI vs. CF micro-stutter or FPS though and that's quantifable. That's what it comes down to. No one ignores thigns such as PhysX or Adaptive Vsync but is Adaptive Vsync on a slower card a more important feature than performance? Not for 99% of people it's not.

For example, NV does have advantages:

1) 3d Vision surround (prob. not possible on GTX660Ti in modern games at high frames)
2) Triple monitor gaming from 1 kepler card (for sure not possible on the 660Ti)
3) PhysX - Name 1 good game in 2012 that used PhysX and 2 good games in 2011?
4) Better NV drivers...ya we've been over this myth for single GPUs.

You could bring all those up, but it would be just as useful as discussing Eyefinity for someone who only games on 1 monitor.



So how would you explain GTX550Ti that consumes more power than a 6850 and performs worse outselling it on Amazon last holiday season:

1) Uninformed consumer;
2) NV fanboys;
3) AMD drivers suck;
4) Person A, B, C, D told me NV is the premium GPU brand;
5) GTX580 is the fastest GPU, so GTX550Ti must also be better than whatever AMD has
6) But it has PhysX (GTX550Ti can't play modern games and PhysX at the same time).
7) I've been buying NV only and I don't want to switch.

I am not saying everyone who buys an NV/AMD card is an NV/AMD fanboy, not at all. But if someone still buys a GTX550Ti for the same price as the 6850 after seeing the information, they are either stupid, or brand biased (i.e., NV fanboy). What do you think a fanboy is? A fanboy isn't someone who buys a superior GTX580 over the 6870. A fanboy is someone who buys the slower GTX570 to play Metro 2033 or Crysis 2 over the 7950 because NV made it and they think it's "better." There is a big difference between preferring a brand because it actually brings something to the table and buying someone only based on the brand.

If Fruit of the Loom t-shirts were taken off the assembly line and stamped as Chanel underwear, would you drop $100 on each? There are plenty of high-end brands that actually have better quality attached to them and then there are poseurs that charge a huge premium only because of the brand. NV and AMD both have good and bad cards in their line-ups.

This is no differnet than someone buying FX8150 because they'll "never buy an Intel CPU just because [insert some reason]." That's not fanboism?

I don't recommend. I stay out of recommendation threads for the most part. On occasion I do and try to find out more about their subjective tastes and tolerances -- find out and ask questions -- what is important to the individual.

In my stereo3d thread that has around 220,000 views -- not once recommended stereo3d to anyone -- only shared my views -- pro's and con's --subjective tastes and tolerances.

 

[3DVagabond]

Agenda?

strong words there sir.

I am not trying to convince anyone here to buy anything. I am not trying to push or persuade, nothing of the like. I find it really interesting you would say such things when in this very thread there seems to be this a sort pitch being thrown over and over page after page. What if i was on a mission and would just repeat buy the 7950 and overclock it to 680 levels. Whatever you do dont the buy the 660ti when you can buy a more expensive card and overclock it to the 680 levels. What if i repeated this over and over, page to page. with charts and graphs, trying to prove the 7950 was a better buy for you. You probably would have no issue if i was doing these things now would you?

Look dude, I love tech, love discussing technology and seeing things progress and unfold. I like to see the companies play their hands and i try to imagine how their hand will play. The 660ti to me is Nvidia's next move. Every bit of anything i have said in this thread has to do with how well of a hand it is. If it will play well or not. I am not trying to convince "enthusiast" that the 660ti (or the 7950) is the card they should get. I am not repeating a pitch over and over in hopes that someone might actually do as i say or do. I am not telling anyone to buy anything.

I am trying to speculate on the new hardware that is coming into the market. Whether it will do well, or not. Whether its will be successful or not. Why? because i have followed the PC since as long as i could remember. Since before i was in high school i have had a great interest in the technology.

When i try to discuss the 660ti here, its about how the card may do. How well it could perform in the market its being molded to. Its a card priced like the 7870 but can outperform the 7950. It will do extremely well from where i see this. But i keep hearing how the more expensive card in the next price bracket can overclock to 680 levels and no *should* buy this 660ti. What? then i here from you that the 670 performs better, which is another card that will cost more, in a higher bracket as well. What? of course it will, why shouldnt it?

Why would you keep throwing up the 670 anyway? why is all this overclock the 7950 to 680 levels coming up now? In discussing the real market where the 660ti will be sold, i cannot see how any of this fits in at all. How is this an indication of the 660ti,and whether it will be a successful launch or not. In the real market 95% of people buy GPUs to play games. Thats it. They buy the performance they need. They dont care about benchmarks or even know how to see their fps. If they want 680 performance, they buy the gtx680. They sure in heck dont buy a 7950 and try not max overclock it to 680 levels. Very few people do this. Very few. The majority of GPU owners just want to have it work, play their games and not have to worry about it.

But i guess your right, the 670 is more powerful. So the 660ti just sucks and you can overclock the 7950 so there is no point in nvidia even releasing the 660ti or whatever.

i serously disagree and i will bet you money right now, real money, that the 660ti will be a success launch and will make a splash. That is what i have been speculating. I will bet that no matter how much you bring up the more expensive 670, or others try to sell the more expensive 7950 overclocked to the max, the 660ti will be very successful. If a person made it there night and day obsession to convince ppl on this forum not to buy the 660ti, they might could manage to get 50 people to buy AMDs 7950 instead. They might get people to try that overclocking thing. But in the grand scheme, do you really think these more expensive cards are gonna somehow stop the success of the 660ti?

agenda????
This is what i am discussing: Whether this new GPU from nvidia will be successful or a flop. I am not asking anyone here if i should buy it or not. I am not telling anyone if they should buy it or not. I am not wondering if the 670 is more powerful, or even if its 20% more powerful. I mean my god, the 670 is supposed to be more powerful. WOW! i guess maybe the higher number means its more powerful? maybe?

How much faster is something if it's 100% faster? It's 2x as fast. Now, how fast is something that's 160% faster? More than 2x as fast. Why would someone try and convince people that a card is more than 2x as fast as another one when it isn't and he knows it isn't?

I've explained why I'm comparing it to the 670. It's because there isn't enough of a sampling of games to compare it properly with a card from a different arch. The 670 and the 660ti are the exact same chip with reduced resources. It is much easier to compare it to the 670 which is a known commodity in relationship to AMD cards. The 670 is 20% faster than the 660ti. The 670 is also 20% faster than the 7870. There's a real good chance that the 660ti and the 7870 will be comparable, performance wise. People are not saying the 660ti will be priced comparable to the 7870 though, they are saying it's going to be more comparably priced to the 7950. I don't know how else to explain my point.

I've already said that I won't be surprised if the 660ti outsells the 7870 and/or the 7950. I fully expect it to. Most of the people who buy it will be real happy about it to. Until AMD puts a real effort into marketing their cards, which they don't, this will continue to happen. I find it educating to watch nVidia do it's thing in the market. People just eat it up. It doesn't mean that all of us fall for it though. I'm personally glad that AMD struggled with the pricing on So. Islands. I might actually be able to justify buying a 7950 for Open CL rendering soon. The 7800 falls off a bit much and it's not like a Kepler is worth the powder to blow it up with in Open CL.

 

[RussianSensation]

I don't recommend. I stay out of recommendation threads for the most part. On occasion I do and try to find out more about their subjective tastes and tolerances -- find out and ask questions -- what is important to the individual.

In my stereo3d thread that has around 220,000 views -- not once recommended stereo3d to anyone -- only shared my views -- pro's and con's --subjective tastes and tolerances.

Wow nice, you got a link to that thread? :thumbsup:

I totally agree with what you are saying but when threads like these come up, we don't have time to write a 100 page essay on all the features AMD and NV offer and how they differ and their subjective and measurable differences. This isn't trying to win a Gov't contract with a 100 page request for proposal document with 20 pages of appendices. It's just sharing general knowledge of hardware.

What video card for i5-2500 3.3GHz? $200 range?

Imagine if each time people spent time discussing PhysX, 3dVision, EyeFinity, 5 differnet resolutions, 0/2/4/8 AA, SSAA, DX10+SSAA, OpenGL vs. DirectX performance, 1 vs. 2 vs. 3GB VRAM limit, performance with mods, support in older games going back to 3, 5, 7, 10 years, etc.

It would be a nightmare. Also, most people have Google to read up on any NV/AMD feature in great detail. If they want a quick recommendation, they aren't asking for how Batman AC plays at 2560x1600 with SSAA and PhysX on a GTX550Ti. It's usually very general like the thread I linked.

 

[SirPauly]

http://www.rage3d.com/board/showthread.php?t=33932576

 

[BlockheadBrown]

What video card for i5-2500 3.3GHz? $200 range?

Thanks for showing the video card pricing and providing the links (possibly another thread).

 

[KingFatty]

Even though they rejected it initially, they can objectively reevalute it as conditions change.

Then they would be ridiculed for being a so-called "flip-flopper" who lacks consistency.

 

[RussianSensation]

Then they would be ridiculed for being a so-called "flip-flopper" who lacks consistency.

I haven't seen anyone being ridiculed for "flip-flopping" lately. This isn't the Olympic games where consistency matters.

Consistency imo is objectively re-evaluating the changing GPU landscape. If suddenly one GPU or another becomes a better buy, then why shouldn't the recommendation change? We can't keep using outdated MSRP prices and outdated launch drivers from old reviews to compare how cards measure up today.

If for example GTX670 drops $100 to $300, suddenly it would make 7970s overpriced. At $400-430, it is arguably overpriced against an aftermarket $320 HD7950. I guess it just depends on the perspective using current market prices and performance. I remember people didn't want to buy the GTX480 for $500 but when Newegg had dropped its prices to $175-200 vs. $260-270 GTX570s, some people considered it a better buy.

I don't think of someone as a flip-flopper if they change their mind because prices change the standing of GPUs or any other component. If we didn't care about price, we'd all be running GTX680 4GB Tri-SLI and HD7970 GE Tri-CF.

 

[ocre]

ocre, I think you are misunderstanding where we are coming from. We don't care if GTX660Ti will sell well or not. On this forum we are not shareholders (maybe some people are). We discuss hardware and recommend it based on its technical and performance metrics and compare it based on price. If NV released a $3,000 GPU and it sold 120,000 units, most of us could care less about that. They did, it's called K10 and K20 Tesla series. I am sure NV will do really well on its earnings report August 9th because of those pre-orders as well as Tegra 3. But here, we would discuss how a $450 HD7970 has higher double-precision compute performance than K10 for a fraction of the price, and would deem K10 a massive failure for compute projects that are run on our boards because it costs 5-6x more.

If you want to discuss how well NV is doing there are lots of financial forums such as Trefis.com, SeekingAlpha and so on. They look at margins, growth rates, market share, etc. They care if GTX660Ti will sell well since they actually invest into the stock.

On this forum, when we are discussing that GTX660Ti may be shaping up to be a disappointment against the 7950, it's strictly from a technical and performance/$ perspective, and has nothing to do if it'll make $ for NV. That's why it's the Videocard and hardware forum, not financial / investor forum. Whether or not 10 million NV fanboys buy the card has no merit of what we think of it on AT. NV's stock price could increase 2x because of GTX660Ti, but that has nothing to do with whether or not GTX660Ti is worth buying at $300 for us gamers. Will it beat GTX570 OC? Will it tank with Anti-aliasing? These things matter for gamers looking to upgrade. Investors don't care about these things.

For example, GTX550Ti was garbage compared to HD6850 and yet last holiday season, it was the best selling GPU on Amazon. Who cares if it sold well. On this forum some of us want to educate people who don't follow hardware to not waste their hard earned dollars on garbage such as the GTX550Ti. Personally if I could convince those 10 million people to get a GTX460 over the GTX550Ti after visiting our forum, I would feel that I provided useful advice that got them the better product, which means in the end they would get a better gaming experience. It's not about recommending a person an NV or AMD card but allowing a person to spend their $ wisely and not overpay for poor products such as the 2900XT, GTX550Ti, Bulldozer, etc. It would be no different than discussing the prices and performance of high-end audiophile equipment at Head-fi.org for example. We praise and criticize certain products. Our perspective changes if drivers change performance, if new games favour one brand or the other, if someone wants specific features from each vendor, if prices change the standing of the cards, etc.

If this was a car forum and someone was looking to buy a fuel efficient car and we knew that Car A was better than Car B, but even if Car B sold 10x more units, why would we recommend Car B? Everyone here already knows that GTX660 series will sell like hot cakes. NV can put out a tin-can with NV logo and it will sell - FX5800/5900 series. Those were the worst videocards made in the history of NV, and became completely useless in all DX9 games and yet sold like hot cakes. 5200Ultra. *cough* *cough*. D:

For example, NV being late by 6 months isn't being criticized here from a company's strategy or market share perspective, but we discuss it from a point of view of what we expect a 6 months older GPU to cost and to perform relative to 6 months older 7850/7870/7950 cards.

But i dont think anyone here is looking to buy any 660ti. This isnt a "which should i buy" thread even though you seem endless determined to force feed what could be used as a perfectly constructed sales pitch.

What i am trying to discuss is the hardware itself, and not nvidias stock numbers. Instead of discussing this generations mid grade (nvidia's), you are bound and determined to keep repeating the overclocking of a more expensive card all together. Who cares how the 660ti compares to what we seen from previous generations, or how it actually could match or even beat the last gens flagship (gtx580). I guess thats an "agenda" as one would suggest. Or perhaps it doesnt fit into one? Maybe i should get with the program and follow the herd. If its not an overclocked 7950 then its not worth buying.......even though its cheaper.

We should be able to have discussions a little more meaningful, with more substance than what you would find from an everyday salesman. But since you keep insisting this is the only way it can be, i will question much.

What i cannot understand is how your missing the most fundamentally basic factor that is often the highest in importance. It seems to escape your ability to reason that this GTX660ti looks like it could be priced in the range of current 7870s. And if it is in this sub 300$ price range, this: "we are discussing that GTX660Ti may be shaping up to be a disappointment against the 7950" is completely out of place.

So now we can only wait for more concrete information, and until then i think its pointless to carry on back and forth.

 

[Final8ty]

Well one thing I find particularly stupid is branding anyone as a fanboy.I prefer nv products but that doesn't mean all amd card purchasers are amd fanboy.Thats just lame.Very few people read reviews and ask for help in tech forums before buying their card.They may be just ill informed.The fanboy crap sells like hotcake here.We are tech enthusiasts good but we should be civilized first to start a serious tech discussion.

Indeed.
The fanboy word is used because its easier to pull that card out from ones crack than to form a reasoned comment of explanation but there are exceptions where it is used in well reasoned comments like RussianSensation.
Personally i don't use the word at all, i would say pro NV or Pro AMD.

 

[SolMiester]

LOL, this thread got a bit out of hand....got to wonder about agendas when we have people bagging an unreleased GPU with unconfirmed pricing and then comparing it against the wrong competitor GPU?....

 

[Gordon Freemen]

lol, this thread got a bit out of hand....got to wonder about agendas when we have people bagging an unreleased gpu with unconfirmed pricing and then comparing it against the wrong competitor gpu?....

right ^ +1

 

[ocre]

How much faster is something if it's 100% faster? It's 2x as fast. Now, how fast is something that's 160% faster? More than 2x as fast. Why would someone try and convince people that a card is more than 2x as fast as another one when it isn't and he knows it isn't?

I've explained why I'm comparing it to the 670. It's because there isn't enough of a sampling of games to compare it properly with a card from a different arch. The 670 and the 660ti are the exact same chip with reduced resources. It is much easier to compare it to the 670 which is a known commodity in relationship to AMD cards. The 670 is 20% faster than the 660ti. The 670 is also 20% faster than the 7870. There's a real good chance that the 660ti and the 7870 will be comparable, performance wise. People are not saying the 660ti will be priced comparable to the 7870 though, they are saying it's going to be more comparably priced to the 7950. I don't know how else to explain my point.

I've already said that I won't be surprised if the 660ti outsells the 7870 and/or the 7950. I fully expect it to. Most of the people who buy it will be real happy about it to. Until AMD puts a real effort into marketing their cards, which they don't, this will continue to happen. I find it educating to watch nVidia do it's thing in the market. People just eat it up. It doesn't mean that all of us fall for it though. I'm personally glad that AMD struggled with the pricing on So. Islands. I might actually be able to justify buying a 7950 for Open CL rendering soon. The 7800 falls off a bit much and it's not like a Kepler is worth the powder to blow it up with in Open CL.

what i meant was 670 could cost 160% of what the 660ti cost. I established that clearly when i said 1.35 - 1.6x.

As for the 670, what you are saying still not working out. The 670 will be more powerful than the 660ti and it will cost more. whopping crazy isnt it?

And for your fancy way to reduce the 660ti to 7870 performance..... its strange that you would come to such a conclusion. I think its completely not correct. Your manipulating the numbers. because the 670 can be *up to* 20% faster than the 660ti doesnt make the 7870 as fast as the 660ti. There are cases in where the 670 is more than 20% faster than the 7870 so.......really your case is lost just by looking at the results from the TT review itself. In the 1600p test with max AA, the 660ti is still trading blows with the 7950. The 660ti still wins some just as it losses some. The 7950 is steady faster than the 7870. They dont trade blows (if they do its a rare event). The 660ti still is capable of trading blows with the 7950 with maxed out settings. This seems to conflict the 660ti being only as fast as the 7870.

We really need more data. I predict that the 660ti will be generally faster than the 7870, and even faster than the 7950 in some cases. I also think it could dip down to the 7870 in some cases but not the norm. I believe it could vary little more than other gk104 models. This is my guess from the data already. i am speculating now, and i guess we will all find out soon enough.

 

[3DVagabond]

what i meant was 670 could cost 160% of what the 660ti cost. I established that clearly when i said 1.35 - 1.6x.

As for the 670, what you are saying still not working out. The 670 will be more powerful than the 660ti and it will cost more. whopping crazy isnt it?

And for your fancy way to reduce the 660ti to 7870 performance..... its strange that you would come to such a conclusion. I think its completely not correct. Your manipulating the numbers. because the 670 can be *up to* 20% faster than the 660ti doesnt make the 7870 as fast as the 660ti. There are cases in where the 670 is more than 20% faster than the 7870 so.......really your case is lost just by looking at the results from the TT review itself. In the 1600p test with max AA, the 660ti is still trading blows with the 7950. The 660ti still wins some just as it losses some. The 7950 is steady faster than the 7870. They dont trade blows (if they do its a rare event). The 660ti still is capable of trading blows with the 7950 with maxed out settings. This seems to conflict the 660ti being only as fast as the 7870.

We really need more data. I predict that the 660ti will be generally faster than the 7870, and even faster than the 7950 in some cases. I also think it could dip down to the 7870 in some cases but not the norm. I believe it could vary little more than other gk104 models. This is my guess from the data already. i am speculating now, and i guess we will all find out soon enough.

We are all speculating, or more accurately deducing, the performance from the limited data we have. I completely accept that you disagree with my interpretation of the data and I also freely admit that I could have it wrong. I've explained how I've come up with my evaluation just so people can evaluate it themselves.

For example though, how have you decided that "the 670 can be *up to* 20% faster than the 660ti"? The 670 is actually more than 20% faster in some instances, tested Dx11 titles w/AA, for example. The 20% is the avg. that the 670 is faster when AA is used in the games tested in the review.

Depending on what benches you look at the 7870 does trade blows with the 7950 at reference clocks, 800MHz for the 7950 and 1000MHz for the 7870. Overall though it is a bit slower and quite a lot slower once you equalize clocks.

If the 660ti comes in at 7870 performance and pricing, that's fine. We're reading that it's ~$300, though. If it is then it had better be able to trade blows with the 7950 overall. That's with AA, because the 7950 is very capable of running @ 1080 w/AA. From looking at this review, I don't see it doing that. We'll see when more and better reviews come out.

 

[BenSkywalker]

What do you think a fanboy is?

Normally the type of person that makes absurdly stupid comments to prop up their argument, such as-

But here, we would discuss how a $450 HD7970 has higher double-precision compute performance than K10 for a fraction of the price, and would deem K10 a massive failure for compute projects that are run on our boards because it costs 5-6x more.

DP HPC calculation performance without ECC? Really?

My apologies, I'll just start talking out my rectum like the rest of the posters in this thread to blend in.

Since we all know that the 660Ti is going to retail for $5 why are people talking about the 7950 being better in terms of price/performance?

 

[RussianSensation]

Normally the type of person that makes absurdly stupid comments to prop up their argument, such as

That is totally uncalled for. If you are going to disagree, I would appreciate it that you focused on the topic at hand and not resort to personal insults. Bringing up ECC performance as a red-herring when my comment was related to the fact that some of us look at about price/performance and IF we were discussing DP performance, then a $500 7970 vs. $3k K10 would come up, as opposed to how much $ it makes for NV. Anyone who needs ECC won't be looking at an HD7970 so obviously that comment I made would be irrelevant to such an individual. I was not in any way insinuating that HD7970 is a better card than K10 is for all professional applications, but for general consumer compute its price/performance is certainly better.

Price/ performance of GTX660Ti vs. 7950 came up because based on rumored pricing, the 2 chips should be priced fairly close. Part of the discussion also revolved around consideration for AA/overclocking since many people use those "bonus features" on our forum. This is especially relevant this generation since AMD underclocked their videocards. Again, if you don't use MSAA and don't overclock videocards, then just skip pg. 13 of the Tweaktown review and overclocking comparisons. That doesn't mean we can't discuss those aspects in an unbiased manner.

Also, the price of the GTX660Ti is 99% confirmed to be $299, so no it won't be $5. Next time before being so condescending, maybe you could actually read the review and go to Page 13 and see that GTX660Ti tanked with AA/AF on, something that was discussed in this thread. Furthermore, from a theoretical point of view, it's also not a secret that it will be a cut-down GK104 chip with 24 ROP / 192-bit bus setup. That means theoretically speaking, the card won't be able to keep up with a GTX670. At the same time we know with 100% certainty that an overclocked HD7950 can surpass a 680 and there are plenty of factory preoverclocked 880-950mhz HD7950s for $320-350. With recent price drops on the 7950, it's not out of line to start comparing a $299 GTX660Ti against a $320-350 HD7950 with or without overclocking.

Since HD7870 cards can now be found for as low as $260, that positions the GTX660Ti between the 7870 and 7950. Since from early benchmarks, a GTX660Ti beat a 7870 without AA and lost to a 7950 with AA in demanding games, it's not biased to discuss how a GTX660Ti will perform against a stock or an overclocked 7950, or how it will perform agains the 7870. However, it actually is biased to ignore the 7950 and focus only on the 7870 since 660Ti looks to come between those 2 cards in terms of price.

If you have a problem when enthusiasts on this forum discuss overclocking or compare 3 cards priced closely to each other, I can't help you. We have discussed a similar situation not long ago in a peaceful manner when HD5850 cost ~ $270-285, GTX470 came out at $350 and HD5870 was $380-390. No one accused anyone of bias or claimed that overclocking shouldn't be considered because it's unfair or that AA should be ignored. Overclocking among those cards was widely discussed on our forum and all 3 were compared despite none of them priced exactly the same simply because of the pricing proximity among those 3 SKUs.

Interestingly enough, GTX460's most stand-out aspect on our forum was its overclocking capability. I don't recall not even once you bringing this technical advantage against the 460, which in stock form was not any better than the HD6850. Many of us praised the GTX460's overclocking headroom and it was a killer feature of that card against the similarly priced 6850. Further, aftermarket GTX460's had immense overclocking performance for the price, with the likes of EVGA GTX460 FTW! Edition and MSI GTX460 HAWK Edition. Not even once I recall you ever mentioning that those cards shouldn't be compared against a stock 6850 or against a 6870. I am trying to understand why suddenly you have a problem with this aspects when AMD card is now in a similar spot. Instead, here you are pointing fingers at me for being biased or having an agenda and trying to prop-up my argument. I am being completely fair in the same wasy as I brought this up when GTX460 OCed beat the 6850.

 

[3DVagabond]

Normally the type of person that makes absurdly stupid comments to prop up their argument, such as-



DP HPC calculation performance without ECC? Really?

My apologies, I'll just start talking out my rectum like the rest of the posters in this thread to blend in.

Since we all know that the 660Ti is going to retail for $5 why are people talking about the 7950 being better in terms of price/performance?

You think Russian is a fanboy? I've seen him take both sides of the red vs green debate depending on his position on the subject. At one point I even commented asking him if two people used his account.

 

[3DVagabond]

This is interesting. Videocardz.com has posted a leak about the 660ti.

The most interesting part of GPU-Z screenshot is the memory capacity of 3GB. While we can assume that some of the GPU-Z reading may not be correct, this one is almost certainly correct, but anyway, let’s wait for more vaild source.

They also say that the performance of this card compares favorably with the 670! Can this be right though? TT reported 2gig vRAM? This report says that the 3gig is 2x the reference amt.

 

[hyrule4927]

This is interesting.
[snip]
They also say that the performance of this card compares favorably with the 670! Can this be right though? TT reported 2gig vRAM? This report says that the 3gig is 2x the reference amt.

Isn't 1.5 or 3GB what was to be expected with a 192 bit memory bus? All the reports of 2GB had struck me as a little odd.