RTX 3xxx performance speculation

Earings report coming up, may get a hint of some roadmap for the next releases. I still don't believe that their RTX line is doing as well as the 'gaming division' numbers are glossing over. Investors aren't stupid, they will want a crumb to work with regarding the new architecture buildup.
 
So they pretty much confirmed what I was speculating, but I was only partially correct. A LOT of the gaming revenue right now is coming in from their Tegra sales inside switch hardware. It just didn't make any sense that RTX cards were selling on the desktop in any way shape or form that could be compared to the 9 or 10 series gpus. The pricing is just way too high to merit mainstream sales numbers, and the price performance ratio has essentially gone unchanged since the 1080gtx release years ago. Yes the super cards have dropped in a hair more bang for your buck, but not 4 years of gpu growth worth. The other front they are making money is laptop gpus. Nvidia has really had the market cornererd in laptop builds due to their partnership with Intel that has spawned so many design wins. With more gamers buying outside of igpu solutions than ever, this is another massive success for Nvidia, however further underlining the decline in Desktop RTX sales. I ran out of fingers to count how many times Jensen said how super excited he was about the excellent RTX desktop sales during the earnings report call, another red flag on the bullshit meter.

TL;DR RTX desktop sales are bad, but Nvidia is still making money. They may not be as motivated to price the next gen cards aggressively if not forced by AMD. If they are happy with their huge margins, and don't mind the reduced sell through, we might reach year 5 of no price/performance benefit to upgrading.
 
So they pretty much confirmed what I was speculating, but I was only partially correct. A LOT of the gaming revenue right now is coming in from their Tegra sales inside switch hardware. It just didn't make any sense that RTX cards were selling on the desktop in any way shape or form that could be compared to the 9 or 10 series gpus. The pricing is just way too high to merit mainstream sales numbers, and the price performance ratio has essentially gone unchanged since the 1080gtx release years ago. Yes the super cards have dropped in a hair more bang for your buck, but not 4 years of gpu growth worth. The other front they are making money is laptop gpus. Nvidia has really had the market cornererd in laptop builds due to their partnership with Intel that has spawned so many design wins. With more gamers buying outside of igpu solutions than ever, this is another massive success for Nvidia, however further underlining the decline in Desktop RTX sales. I ran out of fingers to count how many times Jensen said how super excited he was about the excellent RTX desktop sales during the earnings report call, another red flag on the bullshit meter.

TL;DR RTX desktop sales are bad, but Nvidia is still making money. They may not be as motivated to price the next gen cards aggressively if not forced by AMD. If they are happy with their huge margins, and don't mind the reduced sell through, we might reach year 5 of no price/performance benefit to upgrading.


It looks more like you can't accept RTX success, and are attempting trying to rationalize away contrary evidence.
Massive jump in gaming revenue: Must be SoC in the switch. :rolleyes:
CEO proclaiming great RTX sales that correspond with high revenues: Must be lying. :rolleyes:

The old SoC in the Switch will be a VERY cheap part, and isn't the driver of gaming revenue that RTX parts are.

The simple and obvious answer here, is that RTX cards are selling well as the CEO said, and that corresponds to the revenue jump.

56% YOY gaming revenue increase, is not an indicator of RTX sales decline in any reality.
 
It looks more like you can't accept RTX success, and are attempting trying to rationalize away contrary evidence.
Massive jump in gaming revenue: Must be SoC in the switch. :rolleyes:
CEO proclaiming great RTX sales that correspond with high revenues: Must be lying. :rolleyes:

The old SoC in the Switch will be a VERY cheap part, and isn't the driver of gaming revenue that RTX parts are.

The simple and obvious answer here, is that RTX cards are selling well as the CEO said, and that corresponds to the revenue jump.

56% YOY gaming revenue increase, is not an indicator of RTX sales decline in any reality.

Yeah, people trying to FUD the numbers are boring:
https://www.jonpeddie.com/press-rel...t-from-massive-global-survey-finds-41-percent
 

I'm not sure I'd call 4500 wccftech users the best source for ascertaining Nvidias sales. Nvidia having a 55% market share seems low, but saying 27% of that market share is RTX parts sounds crazy high.

I'd call the steam hardware survey a substacially more realistic source. For January 2020, the top 10 GPUs were all Nvidia. Only one of them was an RTX part, 2060, #7 spot @ 2.12% (1060 is #1 at 13%). All the RTX parts add up to roughly 7%... That's means all the RTX parts combined add up to just over half of the 1060 alone. They also give GPUs by manufacturer, of which 75% is Nvidia.

Point being, Steams numbers, which are a pretty reliable metric, don't mirror your source. They show Nvidia substancially stronger in market share but also show a pretty low adoption of RTX parts. A lot more in line with what I would have guessed.

Sure, Nvidia has sold some cards, but nowhere near what your link would suggest. Nvidia is riding high because their lack of competition allows for a healthy markup, not because they are moving a lot of units.
 
  • Like
Reactions: spine
like this
So they pretty much confirmed what I was speculating, but I was only partially correct. A LOT of the gaming revenue right now is coming in from their Tegra sales inside switch hardware. It just didn't make any sense that RTX cards were selling on the desktop in any way shape or form that could be compared to the 9 or 10 series gpus.

I just read the earnings call transcript and Nvidia said nothing about Tegra revenue dominating.

What they actually said was that RTX is doing very well driven by healthy notebook sales and Tegra also added to the overall success in the quarter.

Our gaming business -- the end-market demand is really terrific, it's really healthy. It's been healthy throughout the whole year. And it's pretty clear that RTX is doing fantastic, and it's very -- it's super clear now that ray tracing is the most important new feature of next-generation graphics. We have 30 -- over 30 games that are -- that have been announced, 11 games or so that have been shipped. The pipeline of ray tracing games that are going to be coming out is just really, really exciting. The second factor -- and one more thing about RTX, we finally have taken RTX down to $299. So it's now at the sweet spot of gaming. And so, RTX is doing fantastic, its sell-through is fantastic all over the world.

The second part -- the second part of our business that is changing in gaming is this, the amount of notebooks sales and the success of Nintendo Switch has really changed the profile of our overall gaming business. Our notebook business, as Colette mentioned earlier, has seen double-digit growth for eight consecutive quarters, and this is unquestionably a new gaming category, like it's a new game console. This is going to be the largest game console in the world, I believe.

And the reason for that is, because there are more people with laptops and they are of any other device. And so the fact that we've been able to get RTX into a thin and light notebook -- a thin and light notebook is really a breakthrough. And it's one of the reasons why we're seeing such great success in notebook. Between the notebook business and our Nintendo Switch business, the profile of gaming overall has changed and it has become more seasonal. It's more seasonal because devices, systems like notebooks and Switch are built largely in two quarters, Q2 and Q3.

And they are build largely in Q2 and Q3 because it takes a while to build them and ship them, and put them into the hubs around the world, and they tend to build it ahead of the holiday season. And so, that's one of the reasons why Q3 will tend to be larger, and Q4 will tend to be more seasonal. Q1 will tend to be more seasonal than the past. But the end demand is fantastic. RTX is doing great, and part of it is just a result of the success of our notebooks.
 
I'm not sure I'd call 4500 wccftech users the best source for ascertaining Nvidias sales. Nvidia having a 55% market share seems low, but saying 27% of that market share is RTX parts sounds crazy high.

I'd call the steam hardware survey a substacially more realistic source. For January 2020, the top 10 GPUs were all Nvidia. Only one of them was an RTX part, 2060, #7 spot @ 2.12% (1060 is #1 at 13%). All the RTX parts add up to roughly 7%... That's means all the RTX parts combined add up to just over half of the 1060 alone. They also give GPUs by manufacturer, of which 75% is Nvidia.

Point being, Steams numbers, which are a pretty reliable metric, don't mirror your source. They show Nvidia substancially stronger in market share but also show a pretty low adoption of RTX parts. A lot more in line with what I would have guessed.

Sure, Nvidia has sold some cards, but nowhere near what your link would suggest. Nvidia is riding high because their lack of competition allows for a healthy markup, not because they are moving a lot of units.

I think you should look at my link again...it's NOT wccftech....fail harder?

And Steam has the following numbers:
RTX Turing: 7,35%
GTX Turing: 4,96‬%

That gives Turing a 12,31% marketshare.
 
Last edited:
I think you should look at my link again...it's NOT wccftech....fail harder?

And Steam has the following numbers:
RTX Turing: 7,35%
GTX Turing: 4,96‬%

That gives Turing a 12,31% marketshare.

Surely you are fucking with me?

Maybe you should read what you are linking. The second half of the FIRST SENTENCE...

"...held a worldwide survey of PC gamers from the 3.5 million Wccftech site visitors that resulted in 4,477 responses to 26 questions in October 2019"

Tell me again who failed here.
 
lets be real here at what point did you see the 2080ti at MSRP prices when it first launched?
I wasn't looking?

I don't have a habit of buying when products first launch. The only exceptions have been the 9700Pro way back, and abandoning my HD6950 CFX for the GTX670.
 
I just read the earnings call transcript and Nvidia said nothing about Tegra revenue dominating.

What they actually said was that RTX is doing very well driven by healthy notebook sales and Tegra also added to the overall success in the quarter.

It looks more like you can't accept RTX success, and are attempting trying to rationalize away contrary evidence.
Massive jump in gaming revenue: Must be SoC in the switch. :rolleyes:
CEO proclaiming great RTX sales that correspond with high revenues: Must be lying. :rolleyes:

The old SoC in the Switch will be a VERY cheap part, and isn't the driver of gaming revenue that RTX parts are.

The simple and obvious answer here, is that RTX cards are selling well as the CEO said, and that corresponds to the revenue jump.

56% YOY gaming revenue increase, is not an indicator of RTX sales decline in any reality.

We have 2080ti's, rVII, 5700xt, 1080ti, and 2070 in my household in use at the moment. I'm nuetral regarding both amd and Nvidia, and was sharing my personal opinion. If you don't agree, I appreciate hearing what your take is on the situation. I haven't seen any real evidence of this RTX success, only the opposite. You only have to look as far as the steam hardware survey, and compare historical sell through to see the reduced adoption rate of the 2060 and up cards. I typically look for hard data to base my opinions on, and not the sales pitch of a CEO trying to work his investors. He has had a great track record as Nvidia CEO but In this particular situation it is my opinion that he was really trying to push a narrative that the cards are selling better than they are. Thank you spending the time to share your personal opinion with me, not everyone is going to agree all of the time, especially with politics and business

Have you seen any real numbers comparing 9xx, 10xx, or 20xx series sales data? Other than some of the public surveys out there, it's hard to get really defined data, which leads to conversations like the one we're pondering right now.

I wish you the best of luck, cheers.
 
We have 2080ti's, rVII, 5700xt, 1080ti, and 2070 in my household in use at the moment. I'm nuetral regarding both amd and Nvidia, and was sharing my personal opinion. If you don't agree, I appreciate hearing what your take is on the situation. I haven't seen any real evidence of this RTX success, only the opposite. You only have to look as far as the steam hardware survey, and compare historical sell through to see the reduced adoption rate of the 2060 and up cards. I typically look for hard data to base my opinions on, and not the sales pitch of a CEO trying to work his investors. He has had a great track record as Nvidia CEO but In this particular situation it is my opinion that he was really trying to push a narrative that the cards are selling better than they are. Thank you spending the time to share your personal opinion with me, not everyone is going to agree all of the time, especially with politics and business

Have you seen any real numbers comparing 9xx, 10xx, or 20xx series sales data? Other than some of the public surveys out there, it's hard to get really defined data, which leads to conversations like the one we're pondering right now.

I wish you the best of luck, cheers.

I am going by Occam's razor. The simplest/most reasonable explanation is the most likely.

They are making a pile of gaming money, and the CEO repeatedly says RTX cards are selling well. I have no reasonable reason to question that.

Who are you to call him a liar with no evidence? If you want to claim that, then the burden is on you to prove your accusations.

Calling CEO's liars without evidence, weakens your argument, and makes it look more like a rant, than a reasoned argument.

Selling well doesn't have to mean selling equal to the 10 series. Prices have increased and that will have an effect on sales, but it hasn't been the devastation that some expected, or hoped for, which seems to be your case.

Turing also has a LOT more more models which will split the counts between more models this generation, making it more difficult for individual models to rise in ranks.

Even so the RTX 2060 is already in the top 10 of the Steam Survey. No AMD cards at all in the top ten, There are SIX RTX cards ahead of the first Navi card.

If the Steam Survey shows you that RTX in trouble with results like that, then relatively it must be showing you that AMD is going out of business.
 
I think you should look at my link again...it's NOT wccftech....fail harder?

And Steam has the following numbers:
RTX Turing: 7,35%
GTX Turing: 4,96‬%

That gives Turing a 12,31% marketshare.
Isn't those percent numbers also considering IGPU's? It has Intel IGPU's all over the place, the survey looks at what is being used to game on GPU wise and counts it. Basically if just discrete cards were calculated, the percentages would be higher for Turing.
https://store.steampowered.com/hwsurvey/videocard/
 
Isn't those percent numbers also considering IGPU's? It has Intel IGPU's all over the place, the survey looks at what is being used to game on GPU wise and counts it. Basically if just discrete cards were calculated, the percentages would be higher for Turing.
https://store.steampowered.com/hwsurvey/videocard/

I don't think the presence of Intel GPU's on there really changes what is being disputed. Per Steam's metric, the breakdown is (roughly) 75% Nvidia, 15% AMD, 10% Intel as far as market share is considered. I don't think removing Intel from that metric would drastically change the ratio of Nvidia to AMD distribution (which is one of the elements of Factums link that I was disputing by bringing up Steam). In the same sense, while removing Intel's 10% would obviously raise the distribution of all other parts, I don't think you'd see any one SKU (or GPU family) get an appreciable rise. We're talking fractions of a percent here, it doesn't change the data enough to change the argument. My point stands that per Steam (which I find to be a much more credible source), both the notion that Nvidia only has a 55% market share and the notion that 27% of that is made up of RTX parts (take note here: article said RTX, not Turing) is not credible or an accurate depiction of the market.

Nvidia can obviously say and do whatever they please, and I believe that RTX sales are strong like they claim. I also believe the word strong is being misinterpreted. Strong doesn't have to mean selling a lot of units, it could just mean making a substantial profit for every unit sold. I believe with RTX cards, Nvidia very much means the later. And to be clear, because their seems to be a lot of confusion when the price of the RTX cards gets called into question... I don't blame Nvidia for doing it the way they have. It was obviously effective. That doesn't mean I have to like it, or that I have to be blind to the fact that if AMD were more competitive, Nvidia wouldn't be able to get away with the pricing they do. Nvidia very much rules the GPU market, and they pretty much get to do what they want because of it.
 
  • Like
Reactions: Auer
like this
I don't think the presence of Intel GPU's on there really changes what is being disputed. Per Steam's metric, the breakdown is (roughly) 75% Nvidia, 15% AMD, 10% Intel as far as market share is considered. I don't think removing Intel from that metric would drastically change the ratio of Nvidia to AMD distribution (which is one of the elements of Factums link that I was disputing by bringing up Steam). In the same sense, while removing Intel's 10% would obviously raise the distribution of all other parts, I don't think you'd see any one SKU (or GPU family) get an appreciable rise. We're talking fractions of a percent here, it doesn't change the data enough to change the argument. My point stands that per Steam (which I find to be a much more credible source), both the notion that Nvidia only has a 55% market share and the notion that 27% of that is made up of RTX parts (take note here: article said RTX, not Turing) is not credible or an accurate depiction of the market.

Nvidia can obviously say and do whatever they please, and I believe that RTX sales are strong like they claim. I also believe the word strong is being misinterpreted. Strong doesn't have to mean selling a lot of units, it could just mean making a substantial profit for every unit sold. I believe with RTX cards, Nvidia very much means the later. And to be clear, because their seems to be a lot of confusion when the price of the RTX cards gets called into question... I don't blame Nvidia for doing it the way they have. It was obviously effective. That doesn't mean I have to like it, or that I have to be blind to the fact that if AMD were more competitive, Nvidia wouldn't be able to get away with the pricing they do. Nvidia very much rules the GPU market, and they pretty much get to do what they want because of it.

Yeah totally this. They won't say they're weak because why would they want to spook investors? Almost all of their current cards for sale are RTX cards so even if they're not selling near as many as they did in Pascal's time it doesn't matter, they are making a nice bump in profit. I think the 3xxx series will have much better sales than the 2xxx cards.
 
My bar is simple, they haven't had to reduce any RTX card prices.

You might argue the Super range effectively was a price cut, but they were certainly shifting cards whilst doing it! :cool:

I view everything going on in the RTX market as a direct result of competition from AMD. RTX cards came out high because AMD didn't have anything to compete with them. Few months later, AMD comes up with something that does compete in the mid-range segment, and suddenly we have the Super cards. Not necessarily a price drop, but certainly a move to make Nvidia's cards more attractive in a space that suddenly became competitive. Meanwhile, nothing happened with the 2080 Ti. No performance adjustments, no price adjustments. Why? Because they didn't have to. Because there was still nothing else like it on the market.. It remained the only card that I truly view as a 4K 60FPS capable (without sacrifices). If you want that level of performance, you have no choice. You pay what Nvidia is asking, or you don't get what you want. Like I said above.. I get it. Doesn't mean I like it.

Yeah totally this. They won't say they're weak because why would they want to spook investors? Almost all of their current cards for sale are RTX cards so even if they're not selling near as many as they did in Pascal's time it doesn't matter, they are making a nice bump in profit. I think the 3xxx series will have much better sales than the 2xxx cards.

I kinda look at the 2000-series as Nvidia testing just what they can get away with. Were a few generations into their substantial performance lead over the competition, and now they are throwing in new tech on top of that. With no real threat from AMD, it's a good opportunity to see just how much people are willing to pay, and see if a low volume, high profit business model can be more effective than a low profit, high volume one. Obviously it was for this generation. That doesn't mean it'll always be so. If AMD can put a little pressure on Nvidia, hopefully that will lead to less expensive 3XXX cards. We'll see.
 
Last edited:
  • Like
Reactions: Auer
like this
Well better hope AMD has some very potent and high reaching cards to drive those prices down. I see AMD hitting a Nvidia wall with technology advancement, mostly dealing with AI and what that can do for graphics, gaming and probably a paradigm shift for many ways in doing games rendering and otherwise. While AMD will probably have some fun stuff coming out, maybe first, AMD has mastered 7nm node, hopefully Nvidia has as well or that could add its own bumps in the road. Not saying Nvidia will stumble with 7nm node but until the GPU hits the buyers that to me is a possibility. AMD probably should invest a lot in two areas, AI and quantum computing which I don't think they are.
 
Well better hope AMD has some very potent and high reaching cards to drive those prices down. I see AMD hitting a Nvidia wall with technology advancement, mostly dealing with AI and what that can do for graphics, gaming and probably a paradigm shift for many ways in doing games rendering and otherwise. While AMD will probably have some fun stuff coming out, maybe first, AMD has mastered 7nm node, hopefully Nvidia has as well or that could add its own bumps in the road. Not saying Nvidia will stumble with 7nm node but until the GPU hits the buyers that to me is a possibility. AMD probably should invest a lot in two areas, AI and quantum computing which I don't think they are.

I only half tuned into the earnings call, and I don't really pay and mind to what Nvidia is doing outside consumer GPUs... do we know how lucrative things like AI, quantum computing, or even professional rendering gear even are? The vibe I got from the earnings call is gaming is still the bread and butter.

I think the biggest thing AMD has going for them right now is they are kicking ass in the CPU spectrum, something we haven't really been able to say since the Core 2 Duo released 14 years ago. The strong Ryzen sales should mean they have a little bit of extra cash to fund other areas. I think it's also earned them back some brand recognition (and hopefully loyalty) that they haven't had in a while. I still remember a time when buying an Intel CPU and an Nvidia GPU felt dirty. I also remember basically a whole decade when you'd be a fool to do anything else. Point being, enthusiasts are looking at AMD right now in a way that they haven't in a seriously long time. I think they are winning a significant number of CPU customers back. With luck, it'll cause people to give their GPUs a closer look also.
 
I think you should look at my link again...it's NOT wccftech....fail harder?

And Steam has the following numbers:
RTX Turing: 7,35%
GTX Turing: 4,96‬%

That gives Turing a 12,31% marketshare.
So you honestly believe these are Real Numbers on Steam representing all PC Gamers??? Haven't logged onto Steam since over a year when I was on my 1080Ti and switched to a 2080Ti. Quit being one of the millions of Steam Lemmings and following their numbers. They're bullshit. Nvidia's gaming sector made hundreds of millions last quarter. This is 100,000,000 million. This is what my 2080Ti costs $1309. Add the lower tier cards from $699 down to $249. Now go back to your widdle numbers from Steam. Does that equal 100's of Millions??? Nope.

Don't come on here talking market share with your fanboy bullshit again till you have a full grasp of what that actually is or someone like me who is well invested at 44, will out right embarrass your ass. Wanna talk Stock options? 401K, IRA, returns, and dividends? Investment futures? Are you WOKE yet?

Steam surveys=fake news that doesn't represent the whole pc gaming community.
 
So you honestly believe these are Real Numbers on Steam representing all PC Gamers??? Haven't logged onto Steam since over a year when I was on my 1080Ti and switched to a 2080Ti. Quit being one of the millions of Steam Lemmings and following their numbers. They're bullshit. Nvidia's gaming sector made hundreds of millions last quarter. This is 100,000,000 million. This is what my 2080Ti costs $1309. Add the lower tier cards from $699 down to $249. Now go back to your widdle numbers from Steam. Does that equal 100's of Millions??? Nope.

Don't come on here talking market share with your fanboy bullshit again till you have a full grasp of what that actually is or someone like me who is well invested at 44, will out right embarrass your ass. Wanna talk Stock options? 401K, IRA, returns, and dividends? Investment futures? Are you WOKE yet?

Steam surveys=fake news that doesn't represent the whole pc gaming community.

Why so grumpy? Half the shit you just spewed doesn't even apply here... did you just show up to start a fight?

Nobody is claiming that Steam's numbers are a depiction of the sales numbers of any given SKU, so I'm not sure what nonsense your talking about the numbers not adding up to hundreds of millions. We're talking about a loose approximation of market share, and I'd say Steam putting Nvidia at 75% of the market is probably pretty damn accurate. Don't be so obtuse (based on your attitude, I realize this is asking a lot)... just because you don't use Steam anymore does not change the fact that it's the dominant platform in PC gaming, and thus the closest were going to get to an accurate snapshot of the distribution of hardware.
 
So you honestly believe these are Real Numbers on Steam representing all PC Gamers??? Haven't logged onto Steam since over a year when I was on my 1080Ti and switched to a 2080Ti. Quit being one of the millions of Steam Lemmings and following their numbers. They're bullshit. Nvidia's gaming sector made hundreds of millions last quarter. This is 100,000,000 million. This is what my 2080Ti costs $1309. Add the lower tier cards from $699 down to $249. Now go back to your widdle numbers from Steam. Does that equal 100's of Millions??? Nope.

Don't come on here talking market share with your fanboy bullshit again till you have a full grasp of what that actually is or someone like me who is well invested at 44, will out right embarrass your ass. Wanna talk Stock options? 401K, IRA, returns, and dividends? Investment futures? Are you WOKE yet?

Steam surveys=fake news that doesn't represent the whole pc gaming community.
Don't know about him but yes, what are the advantages of rolling over a retirement account (ESOP) to a Broker Account NUA (Net Unrealized Appreciation) Strategy where company paid low but stock is now very high, very low cost basis for tax to open account? Compared to a traditional IRA?
 
Wanna talk Stock options? 401K, IRA, returns, and dividends? Investment futures? Are you WOKE yet?

You know any actual finance professional will fall over laughing at reading this right? I know I did.

Steam surveys=fake news that doesn't represent the whole pc gaming community.

In what way is it not representative? Ever since the steam survey came out people have been saying this. And nobody has explained yet why it’s not a reasonable proxy for the market.
 
You know any actual finance professional will fall over laughing at reading this right? I know I did.



In what way is it not representative? Ever since the steam survey came out people have been saying this. And nobody has explained yet why it’s not a reasonable proxy for the market.
Sample size, how determine; as in not counting same machine twice, what regions or regions, how often accomplished. Just too much information we do not know about this survey -> For me it is not credible or can stand on it's own. Age breakdowns would be useful information as well. Does it sorta reflect Valve/Steam gaming platform hardware used? Maybe.
 
Nvidia's President and Chief Executive Officer says Turing adoption is faster than Pascal's.

Jensen Huang
-- President and Chief Executive Officer

If we launched -- we launched well, first of all, the answer is that RTX adoption is faster than Pascals adoption. If you normalize to time zero of launch. The reason for that is because Pascal launched top to bottom, on the same day. And as you guys know, we weren't able to do that for Turing. But if we did that for Turing, if we did that for Turing, the adoption rate is actually faster and it's to me it's rather sensible. And the reason for that is because Pascal was basically DX12 and Maxwell was DX12 and Turing is the world's first DXR, their first ray tracing GPU. Brand new functionality, brand new API and a lot more performance. And so I think it's sensible that that Turing's adoption is going to be rapid. The second element of Turing is something that we've never talked about before, and we're mentioning it more and more because it's such an exciting growth market for us is notebooks.

The installed base of Pascal has very, very little notebook in it. And the reason for that is because, in the past, we were never able to put a high-performance gaming GPU into a thin and light notebook, until we invented Max-Q and in combination with our energy efficiency. We were able to -- we're now able to put a 2080 into a laptop and it's still beautiful.
 
Sample size, how determine; as in not counting same machine twice, what regions or regions, how often accomplished. Just too much information we do not know about this survey -> For me it is not credible or can stand on it's own. Age breakdowns would be useful information as well. Does it sorta reflect Valve/Steam gaming platform hardware used? Maybe.

Well first of all the survey's credibility has nothing to do with whether you personally have reviewed the sampling process. We know that the sample size is massive and statistically relevant. Let's assume the survey is flawed due to the reasons you mentioned. Ask yourself this very basic question.

Why would flaws in the steam survey benefit one graphics card manufacturer over another? Why would it double count Nvidia hardware more than it would double count AMD hardware?

The flawed survey argument doesn't hold water if you think about it for 10 seconds.
 
Well first of all the survey's credibility has nothing to do with whether you personally have reviewed the sampling process. We know that the sample size is massive and statistically relevant. Let's assume the survey is flawed due to the reasons you mentioned. Ask yourself this very basic question.

Why would flaws in the steam survey benefit one graphics card manufacturer over another? Why would it double count Nvidia hardware more than it would double count AMD hardware?

The flawed survey argument doesn't hold water if you think about it for 10 seconds.
How do you know sample size is massive? That tells me zero. Second is if done in a region(s) where one company has an advantage over the other, Vat tax differences, manufacturing locally etc. would skew the results. What countries contribute to the numbers? Which ones are left out? Without information telling us how it is sampled, verified, measures to prevent corruption, I do not see the validity or the accuracy can be trusted. Saying "We Know . . ." in your context = we do not have a clue how done, validity, what data is analyzed or if corrupted (integrity of data), time periods of sampling etc.
 
How do you know sample size is massive? That tells me zero. Second is if done in a region(s) where one company has an advantage over the other, Vat tax differences, manufacturing locally etc. would skew the results. What countries contribute to the numbers? Which ones are left out? Without information telling us how it is sampled, verified, measures to prevent corruption, I do not see the validity or the accuracy can be trusted. Saying "We Know . . ." in your context = we do not have a clue how done, validity, what data is analyzed or if corrupted (integrity of data), time periods of sampling etc.

That is conspiracy and conjecture.

You didn’t answer the simple question. Why would any of those factors skew the results to the benefit of one manufacturer?
 
Second is if done in a region(s) where one company has an advantage over the other, Vat tax differences, manufacturing locally etc. would skew the results.

I know that this happens in smaller countries, but it's been my impression (that I'd happily have corrected) that in larger markets, the price structure seen in the US, UK, and Eurozone are largely representative.

As much as there may be differences in specific markets, on the whole, it is difficult to fault the Steam survey for including market differences.
 
That is conspiracy and conjecture.

You didn’t answer the simple question. Why would any of those factors skew the results to the benefit of one manufacturer?
Corrupted manipulated data is one that would skew it obviously. The data is inconsistent, GPU percent goes down then up the following month with a large variance. Some months the data is absent - what all those GPU's where turned off? I just don't find it trustworthy in the end. Does it hint at hardware being used - yes. How accurate I have not a clue.
 
Back
Top