Thursday, January 10th 2019

NVIDIA CEO Jensen Huang on Radeon VII: "Underwhelming (...) the Performance is Lousy"; "Freesync Doesn't Work"

PC World managed to get a hold of NVIDIA CEO Jensen Huang, picking his thoughts on AMD's recently announced Radeon VII. Skirting through the usual amicable, politically correct answers, Jensen made his thoughts clear on what the competition is offering to compete with NVIDIA's RTX 2000 series. The answer? Vega VII is an "underwhelming product", because "The performance is lousy and there's nothing new. [There's] no ray tracing, no AI. It's 7nm with HBM memory that barely keeps up with a 2080. And if we turn on DLSS we'll crush it. And if we turn on ray tracing we'll crush it." Not content on dissing the competition's product, Jensen Huang also quipped regarding AMD's presentation and product strategy, saying that "It's a weird launch, maybe they thought of it this morning."
Of course, the real market penetration of the technologies Jensen Huang mentions is currently extremely low - only a handful of games support NVIDIA's forward-looking ray tracing technologies. That AMD chose to not significantly invest resources and die-space for what is essentially a stop-gap high-performance card to go against NVIDIA's RTX 2080 means its 7 nm 331 mm² GPU will compete against NVIDIA's 12 nm, 545 mm² die - if performance estimates are correct, of course.
The next remarks came regarding AMD's FreeSync (essentially a name for VESA's Adaptive Sync), which NVIDIA finally decided to support on its GeForce graphics cards - something the company could have done outright, instead of deciding to go the proprietary, module-added, cost-increased route of G-Sync. While most see this as a sign that NVIDIA has seen a market slowdown for its G-Sync, added price-premium monitors and that they're just ceding to market demands, Huang sees it another way, saying that "We never competed. [FreeSync] was never proven to work. As you know, we invented the area of adaptive sync. The truth is most of the FreeSync monitors do not work. They do not even work with AMD's graphics cards." In the wake of these word from Jensen, it's hard to understand the overall silence from users that might have their FreeSync monitors not working.

Reportedly, NVIDIA only found 12 out of 400 FreeSync-supporting monitors to support their G-Sync technology automatically in the initial battery of tests, with most panels requiring a manual override to enable the technology. Huang promised that "We will test every single card against every single monitor against every single game and if it doesn't work, we will say it doesn't work. And if it does, we will let it work," adding a snarky punchline to this matter with an "We believe that you have to test it to promise that it works, and unsurprisingly most of them don't work." Fun times.
Source: PC World
Add your own comment

270 Comments on NVIDIA CEO Jensen Huang on Radeon VII: "Underwhelming (...) the Performance is Lousy"; "Freesync Doesn't Work"

#101
Anymal
Mr.Mopar392Even if vega 7 was faster you still wouldn't buy it, your distain for amd is showing!
Well its not faster, thats why. You can also imagine how lame is to release top 7nm gpu with performance as top 16nm from 2 years ago, with higher tdp.
Posted on Reply
#102
Vayra86
SasquiI do believe that RTX was thought of long before NVidia and implemented in DX12 by MS. If it weren't for that, then it likely wouldn't have happened.

They do get kudos for pushing the envelope and make it part of their marketing strategy. Will it pan out? Maybe someday.
I think RTX was in the pipeline since Volta and since Pascal pushed the performance level to a comfy Ultra 1080p/60 for midrange, which means there is little ground left to cover except going along with the resolution increases of a niche of gamers. Nvidia knows just like everyone else that we're entering the realm of diminishing returns, they simply can't keep scaling things up the old way. That on its own was never going to be enough to keep the Nvidia profit train going, so they needed 'added value', and hey look, we already have Tensor cores, let's let 'em do RT. And that resulted in Quadro... and a new reason to push it on content developers. Boom new desire, new market created and datacenter tech repurposed.

RTX is an afterthought. This was never about realtime RT, it was about producing content more efficiently. And that is proven by the complete lack of RTRT content. RTX is a complete gamble and it was cheap for Nvidia to do, they had to offload those faulty Quadros somewhere anyway. In that sense the risk they take is low, and you can rest assured they have a non-RT alternative escape route. We already heard rumors about GTX 11xx...
Posted on Reply
#103
oxidized
AnymalWell its not faster, thats why. You can also imagine how lame is to release top 7nm gpu with performance as top 16nm from 2 years ago, with higher tdp.
You call that "lame" i call that ridiculous lol.
Posted on Reply
#104
Anymal
MarkoszIf I were to choose between AMD and NVIDIA for exactly the same performance for the same money, I'd choose AMD since NVIDIA as a company is disgusting.
And thats without being an AMD fanboy, I'm angry at them for releasing the RX 590 as they did, because that card for that money is such a bullshit.
Thats a lot of bullshit.
Posted on Reply
#105
Casecutter
NxodusBesides, who cares about CEO's nowadays? they are replacable like batteries. He's just a 3D printed bureoucrat
You don't know who Jensen Huang is he's just not a "figure-head"! It be like saying China could just dump Xi Jinping. But nice to hear an "up-swell" from people feeling he might find himself to close to the door.
AnymalIf you can choose betwen VII and 2080 both priced at 750usd, and I believe we all have that power to make our own decision, why the hell would you choose VII?
First In America the more realistic scenario in a couple of weeks is Vega 7 could be a $660 USD card working a Rebate. For perspective today, the RTX 2080 has 2-3 SKU's starting $695, then ramps up to $730-750 (3-5 SKU's), while the bulk (14) are $780 on up to $900.

Let revisit this say mid-March-April and see what plays out.
Hitman_Actualsparks a fire under AMD's butt's
AND let's not forget Nvidia, they've been complacent also... RTX 2060 is a dud for $350 being 2019.
Posted on Reply
#106
Mr.Mopar392
PrunyBoycot ngreedia, who cares about tdp, just buy Radeon
I'll be getting a vega 7, I will never own another nvidia card as long amd a is around, and seeing some of these nvidiots comments make my choice to choose amd much easier
Posted on Reply
#107
Anymal
Vayra86I think RTX was in the pipeline since Volta and since Pascal pushed the performance level to a comfy Ultra 1080p/60 for midrange, which means there is little ground left to cover except going along with the resolution increases of a niche of gamers. Nvidia knows just like everyone else that we're entering the realm of diminishing returns, they simply can't keep scaling things up the old way. That on its own was never going to be enough to keep the Nvidia profit train going, so they needed 'added value', and hey look, we already have Tensor cores, let's let 'em do RT. And that resulted in Quadro... and a new reason to push it on content developers. Boom new desire, new market created and datacenter tech repurposed.

RTX is an afterthought. This was never about realtime RT, it was about producing content more efficiently. And that is proven by the complete lack of RTRT content. RTX is a complete gamble and it was cheap for Nvidia to do, they had to offload those faulty Quadros somewhere anyway. In that sense the risk they take is low, and you can rest assured they have a non-RT alternative escape route. We already heard rumors about GTX 11xx...
You really need to read more about rtx. Compare it to screen space reflections, whats the diff.
Mr.Mopar392I'll be getting a vega 7, I will never own another nvidia card as long amd a is around, and seeing some of these nvidiots comments make my choice to choose amd much easier
Well, now, who is a fanboy and a hater.
Posted on Reply
#108
eidairaman1
The Exiled Airman
RH92Sure sure but those Radeon Rays are nowhere to be seen in actual products so the least we can say is that AMD is a full generation behind on this topic + 1 full generation behind pure rasterisation performance since they are catching in 2019 what is 2017 1080Ti perf ( at similar MSRP on top of that so they don't even have price advantage anymore ) , and don't even get me started with energy efficiency ! I mean things are far from looking great on their GPU department ..........
They came to realize the perf hit for the detail level provided wasnt worth launching fully or advertising it, so still under development. Nv is getting skewered for it.
Posted on Reply
#109
Vayra86
AnymalYou really need to read more about rtx. Compare it to screen space reflections, whats the diff.


Well, now, who is a fanboy and a hater.
You're new / not frequent here, its okay, but you can search my post history and see how well versed I am. I know precisely what it is compared to everything else.
Posted on Reply
#110
PanicLake
champsilvaThis is mining reason, same happened to amd.

I don't think so! The difference is that nVidia was stable and just dropped, while AMD had a "bubble" and then came back to usual values, is even higher than "before the bubble".
nVidia : red
AMD : green

Posted on Reply
#111
eidairaman1
The Exiled Airman
yakkEveryone is replaceable, but the public face of a company is integral to the position CEO.

This was more akin to a scripted pre-event smack talk, not a multi-national business corporation.
He is just another muppet, trying to minimize the truth.
Posted on Reply
#112
Valantar
Nxodusbut AMD on the other hand is so nice, secretly they are a non-profit NGO spending all the profits to help the elderly and clean the oceans.
Well, no, they've got less power to abuse, so it's quite logical that there are fewer abuses of power from that side. That doesn't take away Nvidia's responsibility for their actions though. Far from it. They're still entirely to blame for being the anti-consumer, proprietary tech-promoting, wannabe monopolists that they are. AMD on the other hand supports open standards across the board, and while this is (obviously) done for both marketing reasons and out of necessity (can't establish a proprietary standard if you're 1/3rd of your competitor's size), it's still a better and more consumer-friendly stance.


As for the VII, it looks decent. Interesting to see where final (in-game) clocks end up.
Last gen:
Vega 64, mid-to-large die, 8gb expensive HBM2, pushed to the limit, 275W TDP.
trailed or matched
GTX 1080 mid-to-small die, 8GB relatively cheap GDDR5X, OC potential, 180W TDP.
A clear win for Nvidia in both cost and performance.

This gen:
Radeon VII, mid-to-small die, 16GB (slightly less) expensive HBM2, unknown OC potential (likely not much), 300W TDP
reportedly roughly matches the
RTX 2080, with a huge die, 8GB (roughly equally) expensive GDDR6, less OC potential than last gen, 225W TDP.

Even counting the process jump, AMD is gaining on Nvidia. Nvidia does have RT and Tensor cores, but they're not really useful, and might never really be (for this level of hardware). Traditional rendering is still king, and AMD is hopefully progressing with a new arch, but as a stop-gap, this looks surprisingly good. Didn't think we'd see 7nm Vega for consumers, but given that the die likely costs half of an RTX 2080 die and HBM2 and GDDR6 cost roughly the same per GB, this sure is a role reversal from last year when Nvidia made more money selling faster parts for similar prices.

Have to wonder what Navi will bring to the table. My Fury X lives on, but who knows what next year will bring?
Posted on Reply
#113
Argyr
Vayra86We already heard rumors about GTX 11xx...
Those rumors are 100% false. Seriously, who would believe that any sane company would cannibalize its own products?
Posted on Reply
#114
Tsukiyomi91
the amount of salt is just... wow.

@Nxodus let alone "nerfing" the already expensive & hard to produce Turing core that has Tensor & RT Cores baked into the silicon. No one is sane enough to spend another billion or trillion dollar just to make another silicon that has the same codename but without those 2 key features that one has already put into motion for nearly a decade.
Posted on Reply
#115
iO
He's pissed that AMD only had to port their ancient arch to 7nm to become somewhat competitive again.


And the only thing RTX crushes in it's current form is the framerate...
Posted on Reply
#116
Sasqui
Vayra86RTX is an afterthought. This was never about realtime RT, it was about producing content more efficiently. And that is proven by the complete lack of RTRT content.
I am just going to have to disagree with you there, at least part of it :p the dream of Real Time RT has been around for a while. It's been that, just a dream thanks to the fact that no consumer hardware was capable of handling it. That said, it'll be interesting to see who is willing to put the money into developing content with a very limited audience.
Posted on Reply
#117
Argyr
ValantarAMD on the other hand supports open standards across the board
No one said AMD were dumb. Very smart move. They are forced to do open standards, those bring you the most virtue-signaling points.
Posted on Reply
#118
mouacyk


GTX 11XX series will artifact with the wallpaper patterns behind him.
Posted on Reply
#119
Sasqui
NxodusThey are forced to do open standards, those bring you the best virtue-signaling points
Even better, they foster new markets. Everyone wins, including the consumer.
Posted on Reply
#120
RH92
eidairaman1They came to realize the perf hit for the detail level provided wasnt worth launching fully or advertising it, so still under development.
Truth is litle do you ( or anyone else outside of AMD ) know about that so unless you work there and you have secrets to share yeah ....... pretend at least to be objective you are not even trying ....
Posted on Reply
#121
Valantar
NxodusNo one said AMD were dumb. Very smart move. They are forced to do open standards, those bring you the most virtue-signaling points.
Not among insecure gamer bros; what Huang is doing here is nothing more than virtue-signaling how "tough" he is. The funny thing is, only fundamentally insecure people act like this. Oh well.

Also, the entire term "virtue signaling" is BS - doing good is doing good, regardless of your intent is to be seen doing good. That might be seen as a character flaw, but your actions are still beneficial in the end. Acting like an ass just to look tough, on the other hand, just makes you look insecure, all the while benefiting nobody at all. Which of these is better, would you say?
Posted on Reply
#122
eidairaman1
The Exiled Airman
SasquiEven better, they foster new markets. Everyone wins, including the consumer.
Yep because NV was forced to by sales slumps, they didn't want to commit corporate suicide.
Posted on Reply
#123
M2B
ValantarWell, no, they've got less power to abuse, so it's quite logical that there are fewer abuses of power from that side. That doesn't take away Nvidia's responsibility for their actions though. Far from it. They're still entirely to blame for being the anti-consumer, proprietary tech-promoting, wannabe monopolists that they are. AMD on the other hand supports open standards across the board, and while this is (obviously) done for both marketing reasons and out of necessity (can't establish a proprietary standard if you're 1/3rd of your competitor's size), it's still a better and more consumer-friendly stance.


As for the VII, it looks decent. Interesting to see where final (in-game) clocks end up.
Last gen:
Vega 64, mid-to-large die, 8gb expensive HBM2, pushed to the limit, 275W TDP.
trailed or matched
GTX 1080 mid-to-small die, 8GB relatively cheap GDDR5X, OC potential, 180W TDP.
A clear win for Nvidia in both cost and performance.

This gen:
Radeon VII, mid-to-small die, 16GB (slightly less) expensive HBM2, unknown OC potential (likely not much), 300W TDP
reportedly roughly matches the
RTX 2080, with a huge die, 8GB (roughly equally) expensive GDDR6, less OC potential than last gen, 225W TDP.

Even counting the process jump, AMD is gaining on Nvidia. Nvidia does have RT and Tensor cores, but they're not really useful, and might never really be (for this level of hardware). Traditional rendering is still king, and AMD is hopefully progressing with a new arch, but as a stop-gap, this looks surprisingly good. Didn't think we'd see 7nm Vega for consumers, but given that the die likely costs half of an RTX 2080 die and HBM2 and GDDR6 cost roughly the same per GB, this sure is a role reversal from last year when Nvidia made more money selling faster parts for similar prices.

Have to wonder what Navi will bring to the table. My Fury X lives on, but who knows what next year will bring?
You're forgetting something very important, Vega 64 VS. GTX 1080 was 14 VS 16nm. Radeon VII VS. RTX 2080 is a completely different story, they're not gaining on nvidia, they're falling further behind.
"WhyCry" from VideoCardz says he has many sources stating that Navi was not good enough to compete and is delayed.
If the Radeon VII is the most AMD can offer within 2 years, their flagship will get beaten by nvidia's next gen XX70 class card, which is just sad.
Posted on Reply
#124
Mr.Mopar392
AnymalYou really need to read more about rtx. Compare it to screen space reflections, whats the diff.


Well, now, who is a fanboy and a hater.
Not a hater trust me, a fanboy yes. I Really don't care what anyone wants to buy its their money, i just don't care for Nvidia cards or their tech thats it. last nvidia cards i own was a pair of gtx 570 in sli they ran like garbage and color look wash out. Never own since then, not for the foreseeable future.
Posted on Reply
#125
Valantar
eidairaman1Yep because NV was forced to by sales slumps, they didn't want to commit corporate suicide.
Still have to love (grudgingly) adopting a standard one day, and the next pissing on it as "not working" when talking about a competitor. Either you then adopted a non-working tech, or you're lying to make yourself look good. Neither are particularly smart.

Nvidia would require proprietary PSUs if they could pull it off. Oh, damn, I probably gave them an idea now. Sorry!
Posted on Reply
Add your own comment
Dec 23rd, 2024 06:53 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts