Big Navi is coming

Gamer X said:
Says the angry person^ who is "getting a RTX2060".... but constantly hanging out in a Radeon thread.

Why haven't you got your new video card yet..? Is it because the RTX2060 SUPER sucks ballz.. and you are not done trying to convince the stupid people to buy one..?




The only Card that Nvidia sells that is worth anything, is the RTX2080 SUPER. And that is about to be jabaited too...
Click to expand...

I never said I'm gettings a RTX 2060 Super, what I said is that IF i was to upgrade, I would consider it. And my geforce GTX1070Ti is serving me pretty well so as I've also said I may skip this generation and see what 2020 brings us.

And please don't trash this thread also. I mean it was kind of fun the first time but its getting old.
 
ManofGod said:
Well clearly, that is they way you think and maybe you are right. However, for me, I was referring to specifically and only the steam survey, which should be universal and across the board, since steam has access to every computer it is installed on. (At least the hardware and software information, anyways.) I have been on steam since the beginning and as far as I can recall, I have had no requests for a survey, which tells me that is not particularly legitimate, unless it is done in the background, of course.
Click to expand...

Nah, too easy...
 
Gamer X said:
nly 251mm^2 and beats the much bigger Vega20 (also 7nm) chip in many games. "Big-Navi" is essentially 64 CU Navi... and would be roughly the same size as Vega20. That means Big-Navi would be 30%+ increase over Radeon XT 5700XT in games. And can manufacture these 335mm^2'ish chips much cheaper than the Mi50 chip in the Radeon 7.
Click to expand...
Gamer X said:
Travm...

I wouldn't get daunted by those talking heads. Since being here, I've been character attacked by the SAME usual suspects, over and over. None of them (attacking you, or Me) have facts to back up their claims, but demand you support your claims.... while not reading the reviews.

As such, when the "Understood You" in the room is that little $299 Navi is superior to $500 Turing. Yet these same clown still hide behind the fact that haven't read any reviews or educated themselves on Navi, or RDNA. So they ignorantly toss around memes and such, not caring they have no facts to work with.

Just point out Navi's price/performance... and these cheerleaders get all angry & start talking about the past.. !



Big Navi is coming.... and as I predicted (weeks ago), it will be sooner than anyone thinks.
Click to expand...

:ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO:
 

Attachments

  • gamerxnavi.png
    gamerxnavi.png
    3 MB · Views: 0
ManofGod said:
Oh, it is not be dismissed as insubstantial right now, I think it is not particularly relevant, at all. (Fun but, not relevant.)
Click to expand...

We'll see..

if it shows no adoption of 5700, the Green folks will jump all over it. The Red folks will say "gaming cafes in China"

If it shows crazy 5700 adoption, the Red folks will paste it all over the net and the Green folks will say "I never see anyone get that survey, it's Fake news"
 
Brian_B said:
We'll see..

if it shows no adoption of 5700, the Green folks will jump all over it. The Red folks will say "gaming cafes in China"

If it shows crazy 5700 adoption, the Red folks will paste it all over the net and the Green folks will say "I never see anyone get that survey, it's Fake news"
Click to expand...

Not me, I truly believe that a steam "survey" should be an across the board thing, with all users being surveyed, unless they opt out.
 
I don't trust steam survey cause I have seen surveys pop up for me.... and it goes like this.

I spend the week at one place and weekends at another. All week when I'm on my main LInux AMD box I never see a survey. On the weekends I am sometimes on a machine with Linux and NV... I have seen a survey pop up on that one a couple times the last year. I do very very occasionally play on a very old windows box of my wives that has both NV in it and windows. That machine gets a survey pop up damn near monthly. lol

Don't trust the survey at all.... it might be interesting, but highly suspect imo.
 
Factum said:
Sample size is big enough...even if no one you know participate FYI.
Click to expand...
I'm glad you *know* that the sample size is big enough, but that doesn't make me convinced just because you say so... I have 5 desktops and 3 laptops running steam at the house... haven't gotten on survey. Maybe just unlucky? I honestly don't know how encompassing their survey is and unless we get details it's hard to say whether it's relevant or not.
 
ManofGod said:
Not me, I truly believe that a steam "survey" should be an across the board thing, with all users being surveyed, unless they opt out.
Click to expand...

I agree with this, seems silly to poll only select people when you could easily just get all of this information.
 
ChadD said:
I don't trust steam survey cause I have seen surveys pop up for me.... and it goes like this.

I spend the week at one place and weekends at another. All week when I'm on my main LInux AMD box I never see a survey. On the weekends I am sometimes on a machine with Linux and NV... I have seen a survey pop up on that one a couple times the last year. I do very very occasionally play on a very old windows box of my wives that has both NV in it and windows. That machine gets a survey pop up damn near monthly. lol

Don't trust the survey at all.... it might be interesting, but highly suspect imo.
Click to expand...

This is why I'm very hesitant, you're not the first person to report similar behavior and until it's either explained or they change to just reporting all hardware (unless opted out), I will continue to be skeptical of all surveys.
 
While I do have my doubts about accuracy, its the only source (pun intended :D:D ) that lets you see the survey by card model.

I'm sure STEAM does analysis and data mining on the survey results so its gotta have some degree of fidelity.

I've used steam for years and used multiple hardware, and have never been asked for a survey, nor anyone I know. That doesn't mean its wrong, but it doesn't mean its right either. So just get a huge rock of salt when you look at it.
 
Auer said:
Obviously the guy who does the Steam hardware survey comes over to [H], writes down the names of all AMD GPU users and then excludes them from the survey.
Click to expand...
He doesnt just go to [H], he goes to other forums too.. all the forums actually.



But seriously, steam hardware survey does gather information, but on here it gets applied beyond the scope of the data gathered. How many people have steam installed on old platforms they still use, but dont game on? Its a skewed metric because all the previously sold cards are still there. What people here are doing is pulling this out as if it prooves something, when all its really doing here is showing us that Nvidia has sold more gpus over the last 10 years than everyone else combined. Technically not wrong.... but when we're talking about sales of RX 5700 vs whatever, its totally irrelevant. It is not the right type of survey for this information. Pulling it out and saying HEERE LOOK! is stupid.
 
Auer said:
Obviously the guy who does the Steam hardware survey comes over to [H], writes down the names of all AMD GPU users and then excludes them from the survey.
Click to expand...

Why do they have to do that when their software already knows what your running before it pops up the hey take a survey button.

It wouldn't be hard for steam to selectively make their survey say whatever the fuck they want. I have no proof they are doing that. They probably aren't. I just find it odd that the 20 or so other Linux gamers I talk to report almost the same experience as I do.... can count on one hand the number of times we have seen the survey pop up on our Linux machines, yet we log our accounts onto a windows box and boom survey pops up. It's hard to call that random when you hear the same from people you have gamed with for years... I game probably 15-25 hours a week on average 99% of that on one of my Linux boxes. Yet the few hours a month I spend on windows is when steam chooses to pop up surveys. Ok once... but I swear I have seen the stupid survey pop up on my wives windows 3 or 4 times this year alone, in comparison I saw it on one of my Linux boxes once this year. So 550 or so hours gaming on Steam Linux this year... 1 survey, mabey 30-40 hours on a windows box and at least 3 pops.

Obviously I'm one person... but I have heard the same type of stories from other Linux gamers.

Seeing as how the Steam hardware survey has become a major sales tool, despite Valve claiming its not intended to be. I am sure someone over there has thought... hmmm we could make a lot of money here.
 
Brian_B said:
We'll see..

if it shows no adoption of 5700, the Green folks will jump all over it. The Red folks will say "gaming cafes in China"

If it shows crazy 5700 adoption, the Red folks will paste it all over the net and the Green folks will say "I never see anyone get that survey, it's Fake news"
Click to expand...

I keep using it despite what numbers it shows.
Just like these numbers:
Add-in-Board-GPU-Market-Share-2002-to-Q1-2019.png
 
Factum said:
A little education about sample sizes:
https://www.google.com/search?q=survey+sample+size

Those pesky facts always get in the way of a subjective opinion...
Click to expand...

That was educational. Thank you.

Given that the SHS gives out %'s to the hundredths place, and shows distinctions of 0.01%, I think it's plausible to infer that the sample size is at least 10,000.

I'll let others run the calculators for themselves, as I'm sure people here will not agree with my assumption. It was enlightening to me though. Thanks.
 
Those claiming that the SHS is irrelevant are either feigning ignorance or have never been within earshot of a statistics class.

Yes, AMDs lackluster releases in recent years have lead to a heavy decline in their marketshare. If only they weren't so determined to continue the trend.
 
Brian_B said:
That was educational. Thank you.

Given that the SHS gives out %'s to the hundredths place, and shows distinctions of 0.01%, I think it's plausible to infer that the sample size is at least 10,000.

I'll let others run the calculators for themselves, as I'm sure people here will not agree with my assumption. It was enlightening to me though. Thanks.
Click to expand...

It also kind of invalidates the "RTX card are TOO expensive" argument...those peske facts again.
 
IdiotInCharge said:
Those claiming that the SHS is irrelevant are either feigning ignorance or have never been within earshot of a statistics class.

Yes, AMDs lackluster releases in recent years have lead to a heavy decline in their marketshare. If only they weren't so determined to continue the trend.
Click to expand...
Not irrelevant, just saying suspect and to use it as a tool, not the end all be all. Then again, I'm suspect of anything that I don't know details about be it Intel, Nvidia or AMD or any other third party. Do you believe every poll you read? I don't and don't intend to start now. It gives a generalized overview, but without the specifics, like how often do results drop off, do they have to be active gamers or just logged in once? Why survey at random when you already know the GPUs... Why not just ask everyone to allow anonymous statistics (opt out at install or something) and allow people to see actual usage. Just seems silly to use a poll of a few random users when you have access to the actual data your trying to extrapolate.
 
Auer said:
Obviously the guy who does the Steam hardware survey comes over to [H], writes down the names of all AMD GPU users and then excludes them from the survey.
Click to expand...

Once again, if that is what you believe and accept, that is cool. Me? I find the Steam Hardware survery inaccurate at best, entertaining at worst but otherwise, entirely useless.
 
Ready4Dis said:
Not irrelevant, just saying suspect and to use it as a tool, not the end all be all. Then again, I'm suspect of anything that I don't know details about be it Intel, Nvidia or AMD or any other third party. Do you believe every poll you read? I don't and don't intend to start now. It gives a generalized overview, but without the specifics, like how often do results drop off, do they have to be active gamers or just logged in once? Why survey at random when you already know the GPUs... Why not just ask everyone to allow anonymous statistics (opt out at install or something) and allow people to see actual usage. Just seems silly to use a poll of a few random users when you have access to the actual data your trying to extrapolate.
Click to expand...

You should read the link...as the sample size crosses into tens of thousands...the added accuracy does not justify the added cost of the effort.
It is basic statistics 101 and the only one having problems with objective scientific facts are people with emotinal subjective views.

And that is quite nicely demontrated in this thread.

This is quite evident in groups like flat-earthers, moon-hoaxers, anti-vaxxers ect.
 
Factum said:
You should read the link...as the sample size crosses into tens of thousands...the added accuracy does not justify the added cost of the effort.
It is basic statistics 101 and the only one having problems with objective scientific facts are people with emotinal subjective views.

And that is quite nicely demontrated in this threads.
Click to expand...

What added cost of effort? You mean they have to call or go door to door to get their stats? You mean they have to send and email or mailer to get information? At best, you are referring to maybe 1% or less of the actual steam members, which is nothing more than a statistical anomoly, a blip, if you will.
 
ManofGod said:
Sample size on a service like this needs to be near 100% or it means nothing. Fact is, I have never been surveyed, that I can recall, and I have had Steam since it's release.
Click to expand...
Had about 3 or 4 hardware surveys since release personally.
 
bricks-100003009-64_1000.jpg
ChadD said:
Why do they have to do that when their software already knows what your running before it pops up the hey take a survey button.

It wouldn't be hard for steam to selectively make their survey say whatever the fuck they want. I have no proof they are doing that. They probably aren't. I just find it odd that the 20 or so other Linux gamers I talk to report almost the same experience as I do.... can count on one hand the number of times we have seen the survey pop up on our Linux machines, yet we log our accounts onto a windows box and boom survey pops up. It's hard to call that random when you hear the same from people you have gamed with for years... I game probably 15-25 hours a week on average 99% of that on one of my Linux boxes. Yet the few hours a month I spend on windows is when steam chooses to pop up surveys. Ok once... but I swear I have seen the stupid survey pop up on my wives windows 3 or 4 times this year alone, in comparison I saw it on one of my Linux boxes once this year. So 550 or so hours gaming on Steam Linux this year... 1 survey, mabey 30-40 hours on a windows box and at least 3 pops.

Obviously I'm one person... but I have heard the same type of stories from other Linux gamers.

Seeing as how the Steam hardware survey has become a major sales tool, despite Valve claiming its not intended to be. I am sure someone over there has thought... hmmm we could make a lot of money here.
Click to expand...

ManofGod said:
Once again, if that is what you believe and accept, that is cool. Me? I find the Steam Hardware survery inaccurate at best, entertaining at worst but otherwise, entirely useless.
Click to expand...
 
ManofGod said:
Sample size on a service like this needs to be near 100% or it means nothing. Fact is, I have never been surveyed, that I can recall, and I have had Steam since it's release.
Click to expand...
So you failed at the statistics class, we get it.

I don't know the methodology or sample size, but you can get very accurate results with "low" samples, and as the methodology matures you may need even smaller samples without loosing meaningful accuracy.
And Steam probably sells that info to customers, so they it better be accurate.

But don't worry, wether you took the survey or not, you are part of the statistic. :D:D
 
Factum said:
It also kind of invalidates the "RTX card are TOO expensive" argument...those peske facts again.
Click to expand...

Well, nVidia is doing well... but RTX isn't. An awful lot of Pascal at the top of the list. nVidia holds all Top 10 spots, with 6 of those as Pascal.

The top 4 cards - all Pascal models (1060, 1050Ti, 1050, 1070) - represent almost 70% of the entire card base. I think that shows where the market is - those most expensive of those cards barely gets you into the bottom end of RTX territory. There are ~a lot~ of 1060's out there (provided you believe the reports).

AMD 580 comes in at #11 though, and growing - although that's probably due to the recent firesales on Polaris given the impending release of Navi, whereas Pascal already had it's firesale moment and we haven't seen much movement on RTX pricing until just recently with the Super introduction.

A quick and dirty search in the July 2019 numbers doesn't even show Turing in there at all. Could be the July survey is published using numbers previously gathered though, or maybe RTX just isn't popular in Chinese Gaming Cafes, or maybe the agents in black helicopters are specifically excluding those people who wn RTX from responding to the survey, or maybe they are getting misreported as a different model. Or maybe RTX ownership is just pretty low.
 
Brian_B said:
Well, nVidia is doing well... but RTX isn't. An awful lot of Pascal at the top of the list. nVidia holds all Top 10 spots, with 6 of those as Pascal.

The top 4 cards - all Pascal models (1060, 1050Ti, 1050, 1070) - represent almost 70% of the entire card base. I think that shows where the market is - those most expensive of those cards barely gets you into the bottom end of RTX territory. There are ~a lot~ of 1060's out there (provided you believe the reports).

AMD 580 comes in at #11 though, and growing - although that's probably due to the recent firesales on Polaris given the impending release of Navi, whereas Pascal already had it's firesale moment and we haven't seen much movement on RTX pricing until just recently with the Super introduction.

A quick and dirty search in the July 2019 numbers doesn't even show Turing in there at all. Could be the July survey is published using numbers previously gathered though, or maybe RTX just isn't popular in Chinese Gaming Cafes, or maybe the agents in black helicopters are specifically excluding those people who wn RTX from responding to the survey, or maybe they are getting misreported as a different model. Or maybe RTX ownership is just pretty low.
Click to expand...

I found this in 3 seconds?
upload_2019-8-3_21-48-30.png


And next month should hav the "Super"'s if NVIDIA's driver history keeps up with their normal modus.
 
Factum said:
I found this in 3 seconds?
View attachment 178367

And next month should hav the "Super"'s if NVIDIA's drive r history keeps up with their normal modus.
Click to expand...

I swear those aren't in my list lol

July 2019 summary, Sorted by % Share

Capture.JPG

Makes me think maybe the Steam Survey does only show what you want it to show...
 
Last edited:
Factum said:
You should read the link...as the sample size crosses into tens of thousands...the added accuracy does not justify the added cost of the effort.
It is basic statistics 101 and the only one having problems with objective scientific facts are people with emotinal subjective views.

And that is quite nicely demontrated in this thread.

This is quite evident in groups like flat-earthers, moon-hoaxers, anti-vaxxers ect.
Click to expand...

I'm not sure how using actual data that's available costs more than specifically generating a sample set. I understand statistics but I also understand how easily manipulated they are. You can use statistics to tell many stories and it has been done many times. Most polls and statistics give you how many people, what the criteria is, and what the nominal % of accuracy is. Without knowing the sample size, who is chosen, etc, I'm not going to 100% trust it. This is nothing like scientific facts, this is a company trying to make money giving you a result of their poll. As I mentioned, the data isn't useless but should be taken with some skepticism unless they detail exactly how the polls are done. Ever notice how to oposing views point at the same statistics and say it proves both sides? Statistics with controls I can believe (at least trust), statistics without controls are just numbers.
 
Oh, btw, I do suspect those results they have aren't seriously misconstrued, however I still take the results with some caution due to the source and lack of transparency. It just seems if they really wanted to make the data useful it would actually take less effort on their part, which begs the question why make it more difficult to have less precise data. The skeptic in me says one is easily manipulated and the other isn't. Realistically it's probably just because they didn't want to scare people with telemetry stuff after the outcry to microsoft for doing so.
 
Ready4Dis said:
I'm not sure how using actual data that's available costs more than specifically generating a sample set. I understand statistics but I also understand how easily manipulated they are. You can use statistics to tell many stories and it has been done many times. Most polls and statistics give you how many people, what the criteria is, and what the nominal % of accuracy is. Without knowing the sample size, who is chosen, etc, I'm not going to 100% trust it. This is nothing like scientific facts, this is a company trying to make money giving you a result of their poll. As I mentioned, the data isn't useless but should be taken with some skepticism unless they detail exactly how the polls are done. Ever notice how to oposing views point at the same statistics and say it proves both sides? Statistics with controls I can believe (at least trust), statistics without controls are just numbers.
Click to expand...

You are saying Steam has an interest in scewing the numbers on a product they do not sell?
 
Factum said:
You are saying Steam has an interest in scewing the numbers on a product they do not sell?
Click to expand...

A lot of people base purchases based on popularity. You have never heard anyone argue buying the most popular hardware is logical as its what developers will target for years ? And arguments about future driver quality if large numbers of cards are out there.

Steams survey is a sales tool. Valve public line is they doesn't intend it to be... however. As designed developers target hardware based on what that survey says. End customers also base their purchases based on what it says. The incentive for any company to game that survey is huge. Its no accident that NV courts big game cafe clients with dirt cheap cards. Do they directly pay Valve... probably not, but if Valve hasn't considered it their stupid. (and I have no doubt they have been offered cash by at least one of the players)

I am not saying Valve is selling survey tills.... BUT they could. Zero doubt. Nvidia AMD Intel and MS would without question pony up if Valve offered. (assuming they are not already) lol ;) MS would gladly pay millions for the steam survey to say windows 10 is growing like a bad virus and windows 7 is dying hard. NV would likewise pay millions to have the survey say RTX is blowing up. For that matter AMD would pay well to show big navi adoption as well.
 
You must log in or register to reply here.
Back
Top