Friday, October 9th 2020

AMD Big Navi Performance Claims Compared to TPU's Own Benchmark Numbers of Comparable GPUs

AMD in its October 8 online launch event for the Ryzen 5000 "Zen 3" processors, provided a teaser of the company's next flagship graphics card slotted in the Radeon RX 6000 series. This particular SKU has been referred to by company CEO Lisa Su as "Big Navi," meaning it could be the top part from AMD's upcoming client GPU lineup. As part of the teaser, Su held up the reference design card, and provided three performance numbers of the card as tested on a machine powered by a Ryzen 9 5900X "Zen 3" processor. We compared these performance numbers, obtained at 4K UHD, with our own testing data for the games, to see how the card compares to other current-gen cards in its class. Our testing data for one of the games is from the latest RTX 30-series reviews, find details of our test bed here. We obviously have a different CPU since the 5900X is unreleased, but use the highest presets in our testing.

With "Borderlands 3" at 4K, with "badass" performance preset and DirectX 12 renderer, AMD claims a frame-rate of 61 FPS. We tested the game with its DirectX 12 renderer in our dedicated performance review (test bed details here). AMD's claimed performance ends up 45.9 percent higher than that of the GeForce RTX 2080 Ti as tested by us, which yields 41.8 FPS on our test bed. The RTX 3080 ends up 15.24 percent faster than Big Navi, with 70.3 FPS. It's important to note here that AMD may be using a different/lighter test scene than us, since we don't use internal benchmark tools of games, and design our own test scenes. It's also important to note that we tested Borderlands 3 with DirectX 12 only in the game's launch-day review, and use the DirectX 11 renderer in our regular VGA reviews.
With Gears 5, AMD claims performance of 73 FPS at 4K, with its DirectX 12 renderer, with the "Ultra" preset. This number ends up 16.24 percent faster than that of the RTX 2080 Ti, which scores 62.8 FPS on our test bed. The RTX 3080 is 15.61 percent faster than the AMD card, at 84.4 FPS.

Call of Duty: Modern Warfare was never added to our VGA review test selection, but we tested the game separately at launch (find its test bed information here). In this testing, we found the RTX 2080 Ti to score 77.9 FPS at Ultra Settings, with RTX-off. In comparison, AMD claims Big Navi scores 88 FPS, making it 12.96 percent faster.

We know this is a very coarse and unscientific way to compare AMD's numbers to ours, and AMD has probably cherry-picked games that are most optimized to its GPUs, but it lends plausibility to a theory that Big Navi may end up comparable to the RTX 2080 Ti, and trade blows with the upcoming RTX 3070, which NVIDIA claims outperforms the RTX 2080 Ti. The RX 6000 "Big Navi" ends up with a geometric mean of 21% higher frame-rates than the RTX 2080 Ti in these three tests, which would imply almost double the performance of the RX 5700 XT. Big Navi is rumored to feature double the CUs as the the RX 5700 XT, so the claims somewhat line up.
Add your own comment

262 Comments on AMD Big Navi Performance Claims Compared to TPU's Own Benchmark Numbers of Comparable GPUs

#201
b1k3rdude
Untill TPU gets their hands on an actual RX6000 card and runs it in the same system as a 2080ti/3080, this article is noithing more than pointless fluff.
Posted on Reply
#202
INSTG8R
Vanguard Beta Tester
The only things that can be compared is it was the same game and at 4K the rest is irrelevant as a comparable.
Posted on Reply
#203
Jism
sepheronx
I do not know why it's so hard for you guys to wait till reviews are out to share judgement.
So many people wrote that the Big Navi woud'nt be nothing more then a 2x 5700 in relation of performance. Well they appearantly got more out of it at this point. And the "large" cache players a significant role along with it. I'm curious in the tech, only just a few more weeks, to see what rabbit AMD pulls out of the hat this time.
Posted on Reply
#204
Nkd
nguyen
Yeah Navi21 selling at 600usd +-50usd can already be called high end gaming.
The elephant in the room here is AMD would like to make profit that could later turn into R&D budget, AMD can't do that if they go into a price war against Nvidia (whom already maintain godly profit margin). Therefore the only way AMD can do is slot their GPU into the gap between Nvidia price brackets, selling as many as they could while maintain as high profit margin as they can (5700/5700XT ring any bells ? ) .

Nvidia already knew 3080/3090 would have no direct competition; the 3070 is the one that Nvida is prepping to counter Big Navi. If AMD decide to price Navi21 at 500usd then Nvidia would have to lower the 3070 price accordingly, though I doubt AMD would do that.

Overall Nvidia is touting 3080 as their flagship gaming GPU, there is no shame for Navi21 being slightly slower. Remember that Nvidia only need TU104 to beat AMD flagship before.
If its 5% slower on average then 3080 that doesn't actually make it 3070 competitor though. So not sure what everyone is getting when they call it 3070 competitor. Its more like 3080 competitor.
Posted on Reply
#205
Zach_01
Jism
So many people wrote that the Big Navi woud'nt be nothing more then a 2x 5700 in relation of performance. Well they appearantly got more out of it at this point. And the "large" cache players a significant role along with it. I'm curious in the tech, only just a few more weeks, to see what rabbit AMD pulls out of the hat this time.
And the biggest mistake a lot of people made is that they thought that it was 2x5700XTs glued together on some kind of crossfire mode and and we know that cant even make a +70~80% performance. I'm expecting to be a real close 3080 competitor with more gap in price compared to performance gap, and with less power draw also. Wont beat it in all aspects but still.

Edit:
typo
Posted on Reply
#206
INSTG8R
Vanguard Beta Tester
Nkd
If its 5% slower on average then 3080 that doesn't actually make it 3070 competitor though. So not sure what everyone is getting when they call it 3070 competitor. Its more like 3080 competitor.
And these numbers are being extrapolated from a completely inaccurate chart so there are no reliable comparisons yet...
Posted on Reply
#207
hathoward
Can't wait to see AMD GPU event. If they can dethrone the 3090 it'll be the first full team red rig I've built in over 20 years. (Yes I know that was before the Radeon acquisition...)
Posted on Reply
#208
quadibloc
I know there are a lot of people saying that the card they teased wasn't their absolute top-end card.
On the one hand, why not tease the absolute top end card in a situation like that, since doing so would give bigger numbers, which would be more impressive - and nothing else, really, but those three numbers, was revealed.
On the other hand, one would expect AMD to make an effort, if only for prestige purposes, to have something out there that would beat the 3090 or at least the 3080.
Nvidia's 3090 is "overpriced" at least in the sense that it is significantly behind the 3080 in price versus performance. So if you want the very best, you will have to pay a premium - and the 3090's performance will only yield dividends if one is gaming at a really high resolution.
So if nearly all the sales, and hence, nearly all the profits, are going to be made from sales of cards like the 3070, then AMD does only really need to compete in that arena.
And just because the new Nvidia cards are better bargains than Turing doesn't mean that AMD couldn't undercut them the same way they undercut Turing: skimp on the ray tracing, since frame rate, and not good looks, is what gamers tend to be concerned about.
Big, but budget priced, and better focused. Nvidia can have the ultra top-end to itself.
But that's only one possibility, and I think AMD will try to offer some ray-tracing capability that sort of looks comparable, and it will try as hard as it can to have something comparable even to Nvidia's top-tier offerings if it can. Maybe in the form of dual-GPU cards if it has to.
Posted on Reply
#209
arbiter
Even if card is say within 15% which i think AMD numbers claim. They couldn't sell it for around same price or even hell 100$ still doesn't make it the better buy without question. All gotta do is look at software suite nvidia is offering with their cards right now. The Green screen, DLSS, nvenc, RTX voice and that is only naming a few. Unless you don't need those things for people that stream for example all those features make nvidia card better buy. AMD needs more then just coming close to nvidia atm. Truthfully the within 15% of rtx3080 is suspect at best just cause AMD's history of things.
Posted on Reply
#210
Vayra86
nguyen
"a billion dollars is a small amount of money because that is less than 1% of what Bezos made" :)
this is what your statement implied.
Does Bezos himself think 1bil is a small amount of money ? definitely not, since he has sound mind after all.

oh well since I dont bully disabled kid I just forgive all your insults, gave me good laugh also.
So the financial situation of an individual, relative to all the money in the world, is the same as a market segment now?

You're totally not grasping at straws now, are you? You're also not bullying at all, following up with a clear insult as you just did ;)
All I gave you was a perspective on your view of statistic. You took that as an insult and ran with it, you still didn't let that go. And you still didn't get the slightest idea about what Steam stats do and don't tell you. Now I'm a 'disabled kid' Cool, man. You're the real adult here.
Posted on Reply
#211
r9
Me personally I liked that comparison as I thought doing it myself, so this article saved me some work.
I did not expect to be hard proof but a ballpark and it should be taken as that.

The fact is if the test was showing that Big Navi was 20% faster than RTX 3080 it would have been taken as a rock solid proof that it is faster than RTX 3080 by the same people that bitch and moan in this thread right now.
Posted on Reply
#212
INSTG8R
Vanguard Beta Tester
r9
Me personally I liked that comparison as I thought doing it myself, so this article saved me some work.
I did not expect to be hard proof but a ballpark and it should be taken as that.

The fact is if the test was showing that Big Navi was 20% faster than RTX 3080 it would have been taken as a rock solid proof that it is faster than RTX 3080 by the same people that bitch and moan in this thread right now.
That’s the problemi tho the s not even ball park close comparison its being used to declare victory/defeat when there is zero comparison data available from the short test we saw
Posted on Reply
#213
Zach_01
It was just a “tease”. Perhaps to show to people that 6000 is not only a 2080Ti/3070 competitor but something more.
If they wanted to be clear about the 3080 they would have done it already. I’m thinking they don’t rush things about it because they know that RTX30 series is on short supply right now.

I fully understand that many do not believe that AMD may have something worthy at hand because of the so called history. But who also believed 3-4 years ago that AMD would catch and get ahead the CPU game in such short time. Intel failures also helped but still, AMD alone has done some worthy job, increasing their CPU performance per watt more than 200% from prior to 2017.

Yes we should be prudent about next AMD announcement, but there is a lot of evidence that things are working well inside AMD. And they are focused on their goals.

By the claims that rumors state about the enhanced node (density) and enhanced/refined architecture = +50% performance/watt they could actually be very close to top challenging nVidia.

No I don’t believe that they can repeat the CPU ZEN3 story, but RDNA2 could be their ZEN2 time in the GPU world, as I’ve said a few times.
Posted on Reply
#214
wolf
Performance Enthusiast
Arpeegee
People must either be in denial or lack good analytical skills if you think AMD would show their BEST card at a CPU event when they have a GPU event at the end of the month.

They've been mum on details all year with Navi and if I were to place a bet I would think the performance is closer to the RTX 3080 then people want to admit. People also need to keep in mind that even if it doesn't absolutely win over the 3080, if it's priced $100 cheaper than the "MSRP $699" then Nvidia has lost this battle. No regular person would pay $100-$200 more for only 5% to 10% better performance with a power hungry card (though the RTX 3090 sales kinda undermine my point :p).
Whether it's the fastest Big NAVI card or not, who knows, opinion is pretty divided on that and for now you can believe what you want to believe based on your interpretation and perception.

As for if it goes punch for punch with a 3080 but sells for cheaper "then Nvidia has lost this battle. No regular person would pay $100-$200 more for only 5% to 10% better performance", that might be true if both offerings were from the same company, but even then people are willing to pay more than double for ~10-15% more with the 3090. Then don't forget that Nvidia has the market and mindshare, even at a price/performance disadvantage, they will sell a buttload of cards. Beyond that we have no indication of relative RT performance and whether than can answer DLSS with anything better than a 70% render scale with contrast adaptive sharpening. If AMD steps it up and can match the performance (or even take the crown) and match the feature set, even then I'd wager Nvidia cards still flying off the shelves, these things take a long time to slowly change.

So many people here seem so sure of AMD and this product, and I've seen this cycle release after release. "Nvidia is worried about AMD this time!", "Nvidia will lose this battle, [insert architecture here] is so powerful/efficient this time!" etc etc... and I just find it so strange that release after release the cycle continues with the same flavour of statements. I guess if they do it enough times, they're bound to be right eventually. Don't get me wrong, I desperately want AMD to to contend with the top cards and even be top dog again, but I've learned to be cautiously optimistic rather than talking so surely of myself when really, I actually know next to nothing.
Posted on Reply
#215
INSTG8R
Vanguard Beta Tester
I hate the current attude that the battle is already over from a 5 second clip
i will only share my ”worst case”
it will perform between 70 and 8 6and90n the b3070 nd 80 will be more efficient actros the board A competitive privce they filled a gap they exclusively
Posted on Reply
#216
Totally
INSTG8R
I hate the current attude that the battle is already over from a 5 second clip
i will only share my ”worst case”
it will perform between 70 and 8 6and90n the b3070 nd 80 will be more efficient actros the board A competitive privce they filled a gap they exclusively
This is what I'm thinking too. Thinking back on past releases they probably have two SKUs. First card with two variants: First variant, similar to slightly better performance than the 3070 performance, better efficiency at load (never seen AMD have better Nvidia's efficiency at at idle or basic tasks), priced accordingly and a second slightly beefier version with higher clocks and/or more VRAM; Second card, Full die heavily oc'd out the box, worse efficiency due to oc and increased power limits, perf that is a few % lower than 3080, priced accordingly.
Posted on Reply
#217
Vayra86
INSTG8R
I hate the current attude that the battle is already over from a 5 second clip
i will only share my ”worst case”
it will perform between 70 and 8 6and90n the b3070 nd 80 will be more efficient actros the board A competitive privce they filled a gap they exclusively
I think something went wrong there, but I agree :D
Posted on Reply
#218
INSTG8R
Vanguard Beta Tester
Vayra86
I think something went wrong there, but I agree :D
i

Yeah don’t know hat/ happened there...eloquent in bonsense... I already have issue typing on thite. predicttive spell spell checkm, formatting all esch other. it wouldn’t churn out something alooetue
Posted on Reply
#219
Vayra86
INSTG8R
i

Yeah don’t know hat/ happened there...eloquent in bonsense... I already have issue typing on thite. predicttive spell spell checkm, formatting all esch other. it wouldn’t churn out something alooetue
I stundtanerd comelptyle!
Posted on Reply
#220
kapone32
nguyen
Can't read ? too poor for education ?
What part of "almost 1% of Steam Hardware Survey" is too hard to understand ?
Steam has a player base of 90 millions active users as of april 2019, 1% of that means 900 000 players own 2080 Ti. Yeah it's niche alright.
I have 1 laptop and 2 PCs with my Steam account. Which one does Steam report as mine? Steam is not a reliable source for GPU numbers. One of the best examples of a place to look would be a German retailer to see historical data on sales. You can find it for CPUs so the same should be true for GPUs.
Posted on Reply
#221
EarthDog
kapone32
I have 1 laptop and 2 PCs with my Steam account. Which one does Steam report as mine? Steam is not a reliable source for GPU numbers. One of the best examples of a place to look would be a German retailer to see historical data on sales. You can find it for CPUs so the same should be true for GPUs.
it should ask from the same pc it was installed on. Cookies.

Link?
Posted on Reply
#223
nguyen
kapone32
I have 1 laptop and 2 PCs with my Steam account. Which one does Steam report as mine? Steam is not a reliable source for GPU numbers. One of the best examples of a place to look would be a German retailer to see historical data on sales. You can find it for CPUs so the same should be true for GPUs.
Steam choose the user base for the survey at random at the begining of each month. Steam will ask if you would like to participate in the survey.
Steam only asked for my participation like once a year.

Does it matter if which machine was registered ? each account only get 1 participation after all, less chance for duplicates.
Amazon or Mindfactory vs Steam ? How about you just go and ask your local store :). Steam has over 100 million active accounts globally, its statistic should be closest to the truth out there.

Germany has about equal GPU allocations between Nvidia/AMD, so Mindfactory data is a good representation but only for first world country. Here in third world country, Nvidia outsold AMD by 9:1 because AMD distribution is very poor.

So what would you prefer ? Regional data or global ?
Posted on Reply
#224
EarthDog
If you went by this, people's disbelief in demand looks even more asinine, lol.
Posted on Reply
#225
nguyen
I would say demand for 3080 is about the same for 1080 Ti during the mining era in 2018, everyone just want a piece of that sweet cake :D.
Posted on Reply
Add your own comment