Speaking of AMD, the company is on a roll as it also announced a new high-end consumer graphics card, which it says can take on Nvidia’s RTX 2080 graphics card.
As it turns out, the video card wars are going to charge into 2019 quite a bit hotter than any of us were expecting. Moments ago, as part of AMD’s CES 2019 keynote, CEO Dr. Lisa Su announced that AMD will be releasing a new high-end, high-performance Radeon graphics card. Dubbed the Radeon VII (Seven), AMD has their eyes set on countering NVIDIA’s previously untouchable GeForce RTX 2080. And, if the card lives up to AMD’s expectations, then come February 7th it may just as well do that.
[…]At a high level then, the Radeon VII employs a slightly cut down version of AMD’s Vega 20 GPU. With 60 of 64 CUs enabled, it actually has a few less CUs than AMD’s previous flagship, the Radeon RX Vega 64, but it makes up for the loss with much higher clockspeeds and a much more powerful memory and pixel throughput backend. As a result, AMD says that the Radeon VII should beat their former flagship by anywhere between 20% and 42% depending on the game (with an overall average of 29%), which on paper would be just enough to put the card in spitting distance of NVIDIA’s RTX 2080, and making it a viable and competitive 4K gaming card.
AMD has managed to shake up the processor market with their Zen architectures, and it’s high time the same happens to the video card market. Nvidia has basically had this market all to itself for several years now, so hopefully this new Radeon card can shake things up a bit and hold us over until 2020, when Intel will be entering the dedicated graphics card market as well.
Looks like Nvidia boss is unimpressed.
https://www.pcworld.com/article/3332025/computers/nvidia-disses-the-radeon-vii-vowing-the-rtx-2080-will-crush-amds-underwhelming-gpu.html
Also interesting is the swipe at freesync, especially considering the recent addition of support..
Steve Balmer was also unimpressed then he saw iPhone…
Right, except this is almost the same ineffective, hot and expensive Vega as we know it, but with 4 HBM2 stacks and double memory bandwidth.
Twice as many ROPs, much better process…
I needed a new GPU to start developing compute software, and I had no preconceptions about which brand I would buy, but on paper (and benchmarks) nvidia trounced AMD, so I bought an RTX 2080 TI in december.
This 2019 CES news is good for AMD, but the review seems to completely ignore the existence of 2080 TI and focuses exclusively on the RTX 2080. It would be fine if they considered it in a different class, but I find the omission somewhat misleading in their characterization of the high end market.
Here are some real-world benchmarks from last year that include RX Vega 64, GTX 1080 TI, RTX 2080, RTX 2080TI (obviously they don’t include the unreleased Raedon VII, but based on AMD’s charts it will be similar to the RTX 2080).
https://www.forbes.com/sites/antonyleather/2018/09/19/nvidia-rtx-2080-and-2080-ti-versus-gtx-1080-and-1080-ti-ultimate-upgrades-or-fantastic-flops/
Nvidia’s response will very likely be to lower the price of RTX 2080 and RTX 2080 TI (which is widely regarded as way over-priced with nvidia exploiting the lack of competition). So while this competition is great for consumers and I’m glad to see someone’s making headway with 7nm tech, I still think AMD has more catching up to do with performance. Then again, maybe they don’t care about the very top of the market?
Either way, it’s a great development for gamers. I’m looking forward to independent parties reviewing it!
AMD most definitively cares about the top of the market, since there’s where the halo and high margins reside. The problem for them is that they simply can’t execute at the same level as NVIDIA. Which is a pity.
AMD also has very strong compute architecture, but their software ecosystem is so behind NVIDIA that their HW can not make up for it. Which is another pity. Since NVIDIA has gone crazy with their pricing this generation, that’s why near monopolies are dangerous for consumers.
Most people can’t afford the 2080 TI. AMD has an opportunity with those people. Even making a card that competes with a 1080 is a huge improvement for them. My wife’s 1080 is faster than my crossfire r9 fury nitro setup and my cards beat vega 64s in some benchmarks. AMD really missed the boat on Vega 64 in several ways, including the fact that their older cards won memory benchmarks against them.
laffer1,
Yes I agree, there’s nothing wrong with competing for more average consumers who care about affordability more than having the best specs. It seems like it might be a great card, but I would have worded the article differently to explain that.
For my needs, I wasn’t looking just at the raw performance, I was sold on the performance per watt.
https://www.phoronix.com/scan.php?page=article&item=nvidia-2080ti-linux&num=6
The RTX 2080 ti really stood out on this metric. It’s what convinced me to go with it over 1080 TI or vega 64 SLI, which I would have been willing to try. I kind of feel bad for not supporting the underdog, but using more energy only to loose performance doesn’t make sense to me.
The Raedon VII could be interesting assuming they drastically improve the performance per watt over the vega 64. I guess it’s possible going from 14nm to 7nm, but we’ll need to wait until these are independently reviewed.
AMD should focus the mid range because nVidia’s current new mid range offerings are a joke.
that 250 to 350 USD range is the sweet spot that most gamers fall in when buying cards.
The deal breaker for me is that AMD supports free drivers, while NVidia doesn’t. I choose to vote with my wallet and don’t have to deal with proprietary stuff.
I won’t even consider NVIDIA anything until they have open source drivers and developer documentation for their GPUs. Pretty much everything they do is underwhelming and lousy.
Sodki,
0brad0,
I agree with you both. I feel very strongly about open computers, yet it’s not always easy to act in accordance with one’s ideologies when market realities are tilted against you. For example, I suspect that we all are using x86 computers where the most privileged execution levels are running proprietary management engine code that we can’t audit or change regardless of if we choose AMD or Intel.
https://boingboing.net/2016/06/15/intel-x86-processors-ship-with.html
Ideally there’d be a lot more competition fighting to attract open computing advocates like us, but without adequate competition it’s pretty easy for corporations to just ignore us because they know our pragmatic options are limited.
IMO that’s a different issue all together.
ME / PSP issues do not affect my everyday use of my computer.
Security issues and how poorly designed so many embedded devices are goes way beyond just Intel ME / AMD PSP and other vendors embedded management chips; only a few years earlier BMCs (IPMI) were the big issue. Practically everything in the embedded space has / had issues. This is more about the mindset of how products are designed and how security is pretty much always an afterthought.
0brad0,
Well, in both cases the hardware is proprietary and we’re forced to run 3rd party proprietary software to use the hardware. It’s not the exact same scenario but I feel similarly restricted in either case.
It’s problematic for both security and openness. I worry that hardware manufacturers are making more regressions than progress. The problem is that voting with your feet isn’t always an effective strategy when there isn’t enough competition to being with. Market consolidation is extremely detrimental, IMHO.