960: 2GB or 4GB

[vembutech]

yep

[typographie]

Shadow of Mordor runs alright on ultra on a 4 GB GTX 770, which is not much faster. But I don't think that one very strange example is a good reason to get a 4 GB version of a $200 video card. At the resolution and settings a GTX 960 is appropriate for, it is overwhelmingly rare for 4 GB to make a difference.

[SteveGrabowski]

Who the hell is expecting to play Shadow of Mordor on ultra on a 960 though? I think you're being unreasonable, tbh, to demand a £150 GPU to run games like that at those graphics settings without compromises.

 

 

I never said ultra, if you read my post again you'll see I was talking about running it on high. I don't run Shadow of Mordor on ultra on my 970 because they recommend 6GB for ultra textures. I use high textures because they recommend 3GB for it. I'd like to see if the 960 4GB is capable of running the game better now with the recommended amount of VRAM for high textures. In playing with high textures on my 970 the VRAM usage seems to be around 2.8GB or so. Shadow of Mordor is one of the few games an R9 280 outperforms a GTX 960 2GB, and I'd like to see if the extra 2GB gives the 960 the win over the 280 on that game. You guys can all talk about the 128 bit bus, but you're not Nvidia engineers and thus you don't know. There were so many people so sure the 980 wasn't that great at launch because of its 256 bit bus, and yet it has stood the test of time. Everyone saying the 4GB card isn't worth it is just making an educated guess, but you don't know until someone trustworthy actually tests that 4GB of memory, which Tech of Tommorow completely failed to do. Why would anyone test a 4GB 960 against a 2GB 960 using only settings that don't push memory usage over 2GB?

[othertomperson]

I never said ultra, if you read my post again you'll see I was talking about running it on high. I don't run Shadow of Mordor on ultra on my 970 because they recommend 6GB for ultra textures. I use high textures because they recommend 3GB for it. I'd like to see if the 960 4GB is capable of running the game better now with the recommended amount of VRAM for high textures. In playing with high textures on my 970 the VRAM usage seems to be around 2.8GB or so. Shadow of Mordor is one of the few games an R9 280 outperforms a GTX 960 2GB, and I'd like to see if the extra 2GB gives the 960 the win over the 280 on that game. You guys can all talk about the 128 bit bus, but you're not Nvidia engineers and thus you don't know. There were so many people so sure the 980 wasn't that great at launch because of its 256 bit bus, and yet it has stood the test of time. Everyone saying the 4GB card isn't worth it is just making an educated guess, but you don't know until someone trustworthy actually tests that 4GB of memory, which Tech of Tommorow completely failed to do. Why would anyone test a 4GB 960 against a 2GB 960 using only settings that don't push memory usage over 2GB?

 

Because having really high quality textures and every other graphics setting turned low doesn't make much sense. Personally I think the 2GB vram and the power of the GPU itself seem quite sensibly balanced. If you need more vram, you probably could do with a better GPU anyway, and those options are available.

 

People seem to have not noticed that Nvidia have shifted their x60 product into the 750 Ti's price point.

[SteveGrabowski]

Because having really high quality textures and every other graphics setting turned low doesn't make much sense. Personally I think the 2GB vram and the power of the GPU itself seem quite sensibly balanced. If you need more vram, you probably could do with a better GPU anyway, and those options are available.

 

People seem to have not noticed that Nvidia have shifted their x60 product into the 750 Ti's price point.

The 750 Ti launched at $150. And the 960 is no bargain at $200 just because the 760 was $250 at launch, as the 960 isn't a cut down 970 while the 760 was a cut down 770.

[othertomperson]

The 750 Ti launched at $150. And the 960 is no bargain at $200 just because the 760 was $250 at launch, as the 960 isn't a cut down 970 while the 760 was a cut down 770.

 

They both cost £150 for me. Even for you the price difference is like £30, so they're pretty much the same price point.

[SteveGrabowski]

They both cost £150 for me. Even for you the price difference is like £30, so they're pretty much the same price point.

 

That's like saying the GTX 960 and R9 290 were at the same price point when the 960 launched since the difference in price was the same. 

[othertomperson]

That's like saying the GTX 960 and R9 290 were at the same price point when the 960 launched since the difference in price was the same. 

 

No it isn't?

 

The 290 launched at double the price of the 960 more or less. Where the hell were you getting £200 290s!?

[SteveGrabowski]

No it isn't?

 

The 290 launched at double the price of the 960 more or less. Where the hell were you getting £200 290s!?

 

No, I mean an R9 290 was $40 more expensive than a GTX 960 in January. Same price point by your argument.

[othertomperson]

No, I mean an R9 290 was $40 more expensive than a GTX 960 in January. Same price point by your argument.

 

Do you mean that the launch price of the 960 was only $40 less than the end of life price of the 290 immediately preceding the 300 series release? Because if you're seriously equating the two...

[SteveGrabowski]

Do you mean that the launch price of the 960 was only $40 less than the end of life price of the 290 immediately preceding the 300 series release? Because if you're seriously equating the two...

 

I'm saying if you consider $150 and $200 the same price, then $200 and $240 should be the same too.

[othertomperson]

I'm saying if you consider $150 and $200 the same price, then $200 and $240 should be the same too.

 

The difference is marginal, but you're comparing an end of life price with a launch price whereas I'm talking about launch prices. Do you seriously not see the difference?

 

If your comparison were valid and they both launched at those prices then they would be the same price point. There is variation between different 290s greater than that price difference, you wouldn't say that the MSI 290 and the Sapphire 290 were completely different price points just because they are priced £13 apart.

[SteveGrabowski]

The difference is marginal, but you're comparing an end of life price with a launch price whereas I'm talking about launch prices. Do you seriously not see the difference?

 

You're saying 150 = 200. I disagree.

[othertomperson]

You're saying 150 = 200. I disagree.

 

Oh I get it, you just don't know what a "price point" is. Read up about it before straw-manning this hard.

[SteveGrabowski]

Oh I get it, you just don't know what a "price point" is. Read up about it before straw-manning this hard.

 

Explain it to me Adam Smith.

[othertomperson]

Explain it to me Adam Smith.

 

It means the price area at which a retailer sells a given product, for a given supply-demand curve. It is not the exact price that it ends up being sold at, but rather a consideration of the price's marketing.

 

The cards being sold off due to them being end of life dramatically alters their supply and demand from considerations at launch so to compare these two completely different contexts is skewing your view of the prices of these products massively.

 

So in the context of graphics cards £13 either way is not really enough to consider it a jump in price point imo, wheras the difference between the 960 and 970, and between the 970 and 980 are very definitely different price points.