LOL Nvidia

[baseline bum]

The GTX 960 looks like it's going to be a piece of .



I like what AMD had to say about it:

We’re all aware of rumors swirling around an upcoming product launch this week. What does this product offer?
•A value oriented 128-bit memory interface
•A maximum 2GB frame buffer
•Questionable performance uplift over its predecessor, that launched over a year and a half ago
If this all sounds unappealing…it’s because it is. Better products already exist and are widely available in the market from AMD. The AMD Radeon™ R9 280 Series graphics are still gamers’ first choice, in this class, with features including, but not limited to:

•More memory bandwidth, with 256-bit (and higher) parts
•Beyond HD gaming experience with MSAA and AF, for superior performance
•Smooth game play at 4K resolutions in top-10 games like DOTA2, CS:GO, TF2, LoL and more
•Support for AMD technologies like Mantle, Eyefinity, TrueAudio (R9 285) and Freesync (R9 285)
•A starting price of 179 €.

Read more: http://wccftech.com/nvidia-geforce-g...-radeon-r9-285

[baseline bum]

GTX 970 was ing amazing, but god the 960 looks like . LOL barely being better than a 760, when a 760 is already a card in comparison to an R9 280.

[baseline bum]

What's funny is I bet Nvidia tries to price this bull at $250 so there isn't some enormous gap between the 960 and 970 in price.

[lefty]

Their tablet is pretty good tbh

[leemajors]

Their tablet is pretty good tbh

You got one?

[lefty]

You got one?

A friend does

It also got some good reviews

(Nvidia Shield)

[DJR210]

Meh, I prefer the "80" and up models if I'm gonna buy a card anyway.

[baseline bum]

Meh, I prefer the "80" and up models if I'm gonna buy a card anyway.

I think the 980 isn't good value at $580 or so. The 970 is pretty nice at $350 though, at least until the R9 380x comes out and rapes everything.

[DJR210]

I think the 980 isn't good value at $580 or so. The 970 is pretty nice at $350 though, at least until the R9 380x comes out and rapes everything.

I'm still gaming at 1080p 60 FPS, so my 680 is still doing ok for the games I play.

[baseline bum]

I'm still gaming at 1080p 60 FPS, so my 680 is still doing ok for the games I play.

Was there a big difference between 670 and 680? I know the difference between 770 and 780 was pretty sizable, but 970 vs 980 seems like a pretty small gap (~15%).

[DJR210]

Was there a big difference between 670 and 680? I know the difference between 770 and 780 was pretty sizable, but 970 vs 980 seems like a pretty small gap (~15%).

Well the peasants will lead you to believe that the 680 wasn't a good buy because "you can just OC the 670 to 680 levels" which doesn't make sense considering you can boost the 680 too. But yeah, it's like 5-10 FPS better on average, but considering it doesn't get all games maxed out locked above 60, I'm glad I have those extra frames at this point.

[baseline bum]

Well the peasants will lead you to believe that the 680 wasn't a good buy because "you can just OC the 670 to 680 levels" which doesn't make sense considering you can boost the 680 too. But yeah, it's like 5-10 FPS better on average, but considering it doesn't get all games maxed out locked above 60, I'm glad I have those extra frames at this point.

Nice.

[anakha]

I'm never buying any device with an NVIDIA SoC again.

Tegra 2? Piece of .
Tegra 3? Piece of .
Tegra 4? They couldn't sucker OEMs into using that

K1? Two different versions on two different devices. The Shield uses a SoC version that NVIDIA only assembled (processor was outsourced). The Nexus 9 SoC uses a processor that NVIDIA manufactured.

Guess which one ended up being a piece of .

[baseline bum]

Damn the GTX 960 is in epic fail territory. It's even behind AMD's ty R9 285 offering, which is their worst price to performance card.

[baseline bum]

nevermind. The 960 is still though. I hope nobody here buys it.

[baseline bum]

Well me, the memory bug on the GTX 970 was acknowledged by Nvidia as "Oops, we gave the wrong specs on the GTX 970 for 5 months". Instead of being a 4GB card with 64ROPs, it's a 3.5GB card with a really slow 512MB memory par ion (28GB/s vs 150GB/s in the good par ion) and with 56ROPs.



you Nvidia, I wish I would have bought an R9 290x now. What a piece of company.

[baseline bum]

http://hexus.net/tech/news/graphics/...mory-problems/

Nvidia stung into action?

We've learned that the GeForce GTX 970 is more than an SMM-reduced version of the GTX 980, which is a fact that will come into play once that 0.5GB is routinely accessed. We've learned that Nvidia has chosen not to impart the GTX 970's differences until Internet rumours surfaced on a potential memory-bandwidth issue, and we have learned that Nvidia has known all along that the information passed along to reviewers - ROP counts, L2 cache, etc. - has been wrong... and has done nothing about it until forced to do so when speculation grew too rife.

There are hundreds of behind-the-scenes tweaks made to each iteration of a GPU architecture. Some of the more important ones, particularly with regards to high-level architecture, should be disclosed from the get-go. Nvidia chose not to with the highly-accomplished GeForce GTX 970, wrongly in our opinion, and while the under-the-hood tweaks don't materially change performance by any degree, Nvidia should have been very clear that the GeForce GTX 970 and GTX 980 use different back-end technologies, more so for buyers who can tax the main 3.5GB framebuffer when gaming at 4K.

What do you, the gamers, make of Nvidia's initial reluctance to share how the GeForce GTX 970 really works? Or do you even care, because performance hasn't changed one iota from yesterday to today?

[Cry Havoc]

AMD has a real chance to do some damage here if their next line of cards are impressive. Nvidia really the bed here.

[baseline bum]

Meh, I prefer the "80" and up models if I'm gonna buy a card anyway.

Man you're right, with Nvidia you get ed on lower than 80-class GPUs.

[baseline bum]

AMD has a real chance to do some damage here if their next line of cards are impressive. Nvidia really the bed here.

Nvidia is teflon and AMD marketing is .

[DJR210]

Man you're right, with Nvidia you get ed on lower than 80-class GPUs.

you're on like a two week delay for responses? I'm honored you chose mine to quote .

I don't think the 70 models are bad, I just figure if I'm gonna invest in something like that might as well get the best one (at least best right below 690/ an)

[baseline bum]

you're on like a two week delay for responses? I'm honored you chose mine to quote .

I don't think the 70 models are bad, I just figure if I'm gonna invest in something like that might as well get the best one (at least best right below 690/ an)

These suckers gave fake specs to all the review sites who then went on to hype the 970 as a barely cut down 980. And when they got caught on their the response is "oops, PR ed up".

[DJR210]

These suckers gave fake specs to all the review sites who then went on to hype the 970 as a barely cut down 980. And when they got caught on their the response is "oops, PR ed up".

So it's not a close a they led everyone to believe?

[baseline bum]

So it's not a close a they led everyone to believe?

They segmented the memory into a fast 3.5GB par ion and a slow 0.5GB one on the 970 after disabling one of the eight L2 cache units, and disabled 8 of the 64 ROPs from the full GM204 chip in the 980. None of this was public knowledge before today when the card has been out since September. It can never hit the 224GB/s bandwidth it was advertised as having since it can't use the 3.5GB and 0.5GB par ions in parallel. Nvidia made it sound like the 970 was just a 980 with some CUDA cores disabled, less texture units, and a lower clock. At least according to the specs they gave every review site. The 980 doesn't have this memory segmentation into a fast 150GB/s pool and a slow 28GB/s pool like the 970, and thus doesn't stutter at 3.5GB VRAM usage like a lot of people have been reporting with the 970.

[DJR210]

They segmented the memory into a fast 3.5GB par ion and a slow 0.5GB one on the 970 after disabling one of the eight L2 cache units, and disabled 8 of the 64 ROPs from the full GM204 chip in the 980. None of this was public knowledge before today when the card has been out since September. It can never hit the 224GB/s bandwidth it was advertised as having since it can't use the 3.5GB and 0.5GB par ions in parallel. Nvidia made it sound like the 970 was just a 980 with some CUDA cores disabled, less texture units, and a lower clock. At least according to the specs they gave every review site. The 980 doesn't have this memory segmentation into a fast 150GB/s pool and a slow 28GB/s pool like the 970, and thus doesn't stutter at 3.5GB VRAM usage like a lot of people have been reporting with the 970.

I'm too drunk for all these stats, I'll read the post after I my gf and take a nap