[Fool.com] - NVIDIA Corporation's New Gaming Graphics Cards Seeing Fast Adoption

[Factum]

This might upset some people that doesn't understand statistics (and thus reject Steam-numbers), but for the rest of the world, this is actually pretty good information:
NVIDIA Corporation's New Gaming Graphics Cards Seeing Fast Adoption -- The Motley Fool

After looking at the Steam Hardware Survey results for September, it's clear that graphics specialist NVIDIA's (NASDAQ:NVDA) latest Pascal-architecture graphics processors are being adopted at an extremely rapid rate.

The numbers are quite telling, Pascal is on the way to be the most succesful GPU-arch from NVIDIA...ever.

 

[thesmokingman]

The Motley Fool owns shares of and recommends Nvidia

Don't forget their perspective...

 

[NeoNemesis]

It took AMD too long to get the RX 480 custom boards out.

I was definitely leaning towards AMD, but when I saw a GTX 1060 on sale for the price of a RX 480, I jumped on it.

The 1070 and 1080 numbers don't mean much because there isn't any competition at that level.

I personally feel that AMD is heading in the right direction, and I'm expecting big things out of them from Zen and Vega.

 

[Rvenger]

Everyone also thinks they need to beat Nvidia to stay alive. Also not true.

They just need to stay at 30% at a minimum

 

[razor1]

It took AMD too long to get the RX 480 custom boards out.

I was definitely leaning towards AMD, but when I saw a GTX 1060 on sale for the price of a RX 480, I jumped on it.

The 1070 and 1080 numbers don't mean much because there isn't any competition at that level.

I personally feel that AMD is heading in the right direction, and I'm expecting big things out of them from Zen and Vega.

the 1070 and 1080 keep in mind the 970 and 980 had no competition either....

I can see Zen being successful, but with Vega coming out 1st half of next year even if its variants are competitive with the 10 series performance and enthusiasts segments forget them gaining anything because the people that have already bought their graphics cards, unless Vega is significantly faster,10% or more with other metrics lining up too, they aren't going to be getting many sales at all.

 

[razor1]

Everyone also thinks they need to beat Nvidia to stay alive. Also not true.

They just need to stay at 30% at a minimum

Either nV's numbers (estimates) for projections for q3 are off or AMD's q3 estimates are off, one of them is going to be off because there is no way nV would have given out a 40% increase in gross sales without volume increase that total up to the loss of volume that they had in q2. So if that happens, AMD will not have 30%, they will be back down to 23% or so give or take a couple points. Yeah I don't trust AMD with their numbers they have always been too optimistic when it comes to forecasting and the opposite is true of nV they have always been conservative with their numbers.

 

[Rvenger]

There's no doubt that Pascal is a great architecture. I loved my GTX 1080 and I also have a GTX 1060 running my VR rig like a champ.

Vega can't come soon enough for AMD.

 

[NeoNemesis]

the 1070 and 1080 keep in mind the 970 and 980 had no competition either....

I can see Zen being successful, but with Vega coming out 1st half of next year even if its variants are competitive with the 10 series performance and enthusiasts segments forget them gaining anything because the people that have already bought their graphics cards, unless Vega is significantly faster,10% or more with other metrics lining up too, they aren't going to be getting many sales at all.

You're right, the ship has sailed for AMD going head to head against the 1070/1080. What they need to do is start turning people's heads and I think that they have a good shot at that with Vega.

You have to admit, the Vega 10 (Fury Pro) looks like an absolute beast (if the rumors are true).

 

[razor1]

I don't see it as an absolute beast, nV still has its advantages with less flops they can still do more and power advantage that we have never seen when ATI/AMD and nV has gone head to head. the 1060 vs the rx480.....

So if AMD doesn't have a flop advantage they are going to be lower performance. and with what we have now as rumors, they are just past what nV has, that isn't enough.

Just wrote a pretty big post over at B3D about Vega too.

Cut and pasted it for ya here.
"Perf/watt
the hd 4870 wasn't like that, and yeah the hd 5870 did, correct. but even that wasn't as great as this, it was the price of those cards that really hurt nV, Fermi had problems out of the gate hence the 6 months delay. And I have always stated if you EVER see more than one quarter difference between launch of cards from either IHV you better get ready for disappointment, something went wrong. But still the 5870 series started to loose traction soon after its Fermi's V2 release, which those problems were rectified some what.

Guess what Vega is.....

And doesn't matter about tapeouts and all the other stuff, because in recent history, both of these companies ALWAYS have launched close to each other within a quarter, as they are prepared to do so. The only time they couldn't do it is when they knew they couldn't match up with something and it would hurt them.

PS keep this in mind, Tape out of Vega was q2 of this year, So why is it taking 3 or more quarters for it to come out? Why wasn't Vega also on the same time schedule as Polaris, was AMD not interested to go into the performance or enthusiast segment, the performance segment is by far the largest segment by volume and over all profits.

Do we soon forget the reasoning AMD gave for Polaris's launch (How about the r600, the FX series, Fermi V1, Fiji), and we now know why it took them longer to come out, something didn't go right. Every time we see these companies have to give reason for something that is delayed more than 2 q's over a competitor's product, that reason is most likely BS, there has been the underlying cause of "we are F'ed"

Then you look at multiple design teams, usually when you have multiple design teams you have one team working on what is coming out soon, and its iterations and the second team working on future architectures which won't see day of light for a while, yet we see the same design team working on Vega and Polaris with a staggered release, that is something we have never seen before if they are truly that much different. We all know the most that these companies can do when fast tracking a project like these its a quarter up, that's it. They can't move mountains and push up timetables of future products with this kind of complexity.

So lets say Vega was fast tracked, that means since Polaris's release, AMD was not expecting to come into the performance market for a year + another quarter to against nV? Does that sound like anything that is remotely possible, to give away so much money and the entire market segment for essentially an entire generation? That is a lot of money, around 5 billion dollars to say we are not interested in so we didn't plan for it. All the while they were so in tune with LLAPI's that they couldn't plan for future products that supported LLAPI's better then their competition? I see a disconnect there if that was the case.

More things to add, when ever either of the IHV's had a delay in current products, that never changed the time tables of future products, so we can't say Fiji's delay had something to do with Vega's delay because they are not bound to one another. "

Should have added in there, "entire generation and then some" because OEM/system builder deals will keep going after the launch of new cards, so the backlash of having no competitive cards is actually longer then the time they weren't available.

 

[Shintai]

I shouldn't surprise anyone. Vastly better products in all segments, alone in the performance and enthusiast segment.

Current sales of cards favour Nvidia in around ~85% to ~15%. With the GP107 release it will only tilt more in Nvidias favour.

Even for the late released GTX 1060. (mindfactory.de)
RX 460 995
RX 470 2440
RX 480 5370
GTX 1060 13765
GTX 1070 21520
GTX 1080 8595

AMD 8805 cards (16.7%). NVIDIA 43880 cards (83.3%).

 

[Factum]

Don't forget their perspective...

if you have any factual errors, please report...if not, you fall into the catagory of the few people I talked about that do not like facts.

 

[thesmokingman]

if you have any factual errors, please report...if not, you fall into the catagory of the few people I talked about that do not like facts.

Fool and seeking alpha have never posted anything bad about Nvidia because they're both stock shills. Not sure why you're picking my post out, to get hits on your thread up??

 

[Factum]

Fool and seeking alpha have never posted anything bad about Nvidia because they're both stock shills. Not sure why you're picking my post out, to get hits on your thread up??

Why are you posting?
You can point to no errors in the numbers, so you try and posion the well sprinkled with personal attack...says it all

 

[razor1]

Fool and seeking alpha have never posted anything bad about Nvidia because they're both stock shills. Not sure why you're picking my post out, to get hits on your thread up??

oh yes they do.

Please get that bias rubbish outta here, Seeking alpha is a bit different just many idiots think they know what semiconductor companies are all about they should stick with crap they know about and not the technical things they don't. I have not read a single good article from Seeking alpha, outside of business concepts from them. They try too hard to convince their audience they know the tech stuff, which yeah they don't, idiots lol.


4 Signs NVIDIA's High-Growth Days Are Over -- The Motley Fool

How Risky Is NVIDIA Corporation? -- The Motley Fool

Motley Fool Guru Analysis of NVIDIA Corporation (NVDA)

 

[thesmokingman]

Why are you posting?
You can point to no errors in the numbers, so you try and posion the well sprinkled with personal attack...says it all

Shrugs, the first post was a PSA. And now, here comes the redirect...

 

[razor1]

Shrugs, the first post was a PSA. And now, here comes the redirect...

Tell ya one thing, if my predictions of AMD dropping a good deal of marketshare this coming Q come true, you think Montly Fool is mistaken? And at this point I am very sure I am going to be right I would say in the neighborhood of 90%, to this point I shorted AMD stock for a day after the release of the Q numbers......

For me that 10% risk, I'm willing to take because I already got a 25% increase on AMD in the past few months

 

[StormClaw]

I shouldn't surprise anyone. Vastly better products in all segments, alone in the performance and enthusiast segment.

This.

I sincerely wanted to give AMD a chance, because i wanted a particular Freesync monitor, but just couldn't find any worthy video card from them. Everything they got out there is severely behind Nvidia.

I ended up with Nvidia and Gsyc display and felt like i was forced into it because of lack of other options. There is just nothing there on the 1080 / Titan XP level.

 

[CSI_PC]

Also if you look at the Steam survey figures they show that from a gaming perspective Nvidia had a more successful launch (look at June/July) for all models and that they had less problems than reported by such as SemiAccurate regarding supply, 1060 was a bit later but its growth is very strong.
Caveat yeah this is only steam but it has a massive user base.
Steam Hardware & Software Survey

Maybe saving grace for 480 was/will be crypto mining (IMO it did not impact sales too much), but aside from Etherium (niche on Nvidia as require Linux to overcome the Windows issue it has) the 1060/1070 are also nice mining cards.
The big challenge for AMD is that the potential target audience seemed to had purchased the R9 390 series when it was discounted just before the release of Polaris and this figure has not dropped (still at 0.43% for the last 2 months), meaning gamers are seeing little incentive to upgrade from Hawaii to Polaris, and this is a problem because we can see Nvidia is easily selling the 1060 to gamers comparing their monthly footprint on Steam.
Cheers

 

[primetime]

pure fucking troll thread.....thanks for the updates of the ignore list

 

[Dayaks]

I shouldn't surprise anyone. Vastly better products in all segments, alone in the performance and enthusiast segment.

Current sales of cards favour Nvidia in around ~85% to ~15%. With the GP107 release it will only tilt more in Nvidias favour.

Even for the late released GTX 1060. (mindfactory.de)
RX 460 995
RX 470 2440
RX 480 5370
GTX 1060 13765
GTX 1070 21520
GTX 1080 8595

AMD 8805 cards (16.7%). NVIDIA 43880 cards (83.3%).

Based on the 1060 vs 480 AMD might not be too bad off if they actually had a 1070 competitor.

Interesting how the 1070 sold more cards than the 1060+480. I realize the releases are staggered but in the recent past some people act like the ~$400 range is a tiny market.

 

[Shintai]

Based on the 1060 vs 480 AMD might not be too bad off if they actually had a 1070 competitor.

Interesting how the 1070 sold more cards than the 1060+480. I realize the releases are staggered but in the recent past some people act like the ~$400 range is a tiny market.

As time goes the different release date for Polaris 10 and GP106 should even out. Remember these numbers comes from Germany who by history have been pro AMD due to the fabs.

Anyone knew since Maxwell and even before that 300-400$ was a growing fast expanding market. Some people tho decided to get fooled by some PR slides. Despite the clear show that people move up in SKUs, not down. Intels constant quarter after quarter record K sales was also a pointer, in case anyone needed it bended in neon.

Intel also upped its guidance with 700M$, Nvidia with a record Q3 outlook and AMD...well not so much in any.

 

[Factum]

pure fucking troll thread.....thanks for the updates of the ignore list

Like I said, some people do not like facts...thanks for the constructive contribution to the thread

 

[CSI_PC]

As time goes the different release date for Polaris 10 and GP106 should even out. Remember these numbers comes from Germany who by history have been pro AMD due to the fabs.

Anyone knew since Maxwell and even before that 300-400$ was a growing fast expanding market. Some people tho decided to get fooled by some PR slides. Despite the clear show that people move up in SKUs, not down. Intels constant quarter after quarter record K sales was also a pointer, in case anyone needed it bended in neon.

Intel also upped its guidance with 700M$, Nvidia with a record Q3 outlook and AMD...well not so much in any.

Problem is the Steam figures for AMD, not sure how you see 480/470 evening out with the 1060 when growrth is weak for those Polaris cards and strong for Nvidia for all Pascal.

Nvidia 1060: JUL 0.02%, AUG 0.17%, SEPT 0.46%
AMD 480: JUL 0.05%, AUG 0.11%. SEPT 0.18%

Nvidia 1070 is nearly as strong in Steam footprint as the 1060 in terms of recent month-month growth.

The problem for AMD is that there was a strong increase for the R9 390 series just before/at the launch of Polaris.
AMD 390: JUN 0.18%, JUL 0.41%, AUG 0.43%, SEPT 0.43%

So AMD consumers seem to be reluctant to upgrade from Hawaii to 480/470, where for Nvidia it seems the 1060 does have appeal to its core consumers.
Cheers

 

[Shintai]

Problem is the Steam figures for AMD, not sure how you see 480/470 evening out with the 1060 when growrth is weak for those Polaris cards and strong for Nvidia for all Pascal.

I never said I expected them to even out. But time would put them closer to their real selling as the sales lead time for Polaris 10 gets even out.

The problem for AMD is that there was a strong increase for the R9 390 series just before/at the launch of Polaris.
AMD 390: JUN 0.18%, JUL 0.41%, AUG 0.43%, SEPT 0.43%

Look at R9 200 numbers. There wasn't a strong increase for R9 390

1.20%
0.94%
0.48%
0.49%
0.47%

Vs R9 390 series

0.00%
0.18%
0.41%
0.43%
0.43%

And R9 380 series

0.00%
0.16%
0.38%
0.43%
0.42%

1.20 became 1.32 combined over the 5 months, Some related to Windows 10 adoption.

So AMD consumers seem to be reluctant to upgrade from Hawaii to 480/470, where for Nvidia it seems the 1060 does have appeal to its core consumers.
Cheers

You have to be a fool to go from Hawaii to Polaris. The main group upgrading is those with lower cards. Same applies for Nvidia. The problem for AMD is they dont have anything better for the performance and enthusiast segments.

 

[CSI_PC]

I never said I expected them to even out. But time would put them closer to their real selling as the sales lead time for Polaris 10 gets even out.



Look at R9 200 numbers. There wasn't a strong increase for R9 390

1.20%
0.94%
0.48%
0.49%
0.47%

Vs R9 390 series

0.00%
0.18%
0.41%
0.43%
0.43%

And R9 380 series

0.00%
0.16%
0.38%
0.43%
0.42%

1.20 became 1.32 combined over the 5 months, Some related to Windows 10 adoption.



You have to be a fool to go from Hawaii to Polaris. The main group upgrading is those with lower cards. Same applies for Nvidia. The problem for AMD is they dont have anything better for the performance and enthusiast segments.

You are comparing 200 series (that covers a broad range of GPUs) numbers to specifically the 390 series, also if you think 0.43% is weak with a jump from 0.18% in June, then what do you make of 480 at averall 0.18% to date...
And yes you emphasise my point people are not upgrading from Hawaii to Polaris, who is going to buy the 470/480?
However as I mentioned people did upgrade to the 390 in July rather than buy the 480, the 390 footprint increase that month was 0.23%, nearly as good as seen with the current Pascal GPUs.
In fact you also see the same jump with the 380 albeit not quite as strong.

And if you look closely, you can see the drop in 200 series actually aligns with the growth of the 390 and the 380, meaning AMD consumers upgraded from Hawaii to Hawaii rather to Polaris, primarily because it is the 200 series and not just 290/290x.
This does not look good for Polaris in terms of the 470/480 IMO because they have lost the core AMD support for those 2 models that selected the 390/380 instead and some jumped to the 1070/1080 it seems for the higher Hawaii models.
However the 1060 still managed to have strong growth (notice this is without affecting the 970 footprint) and is a card comparable in peer and performance to the 480.
The trend is same even when looking at older systems using the various cards.
Cheers

 

[Shintai]

You are comparing 200 series (that covers a broad range of GPUs) numbers to specifically the 390 series, also if you think 0.43% is weak with a jump from 0.18% in June, then what do you make of 480 at averall 0.18% to date...
And yes you emphasise my point people are not upgrading from Hawaii to Polaris, who is going to buy the 470/480?
However as I mentioned people did upgrade to the 390 in July rather than buy the 480, the 390 footprint increase that month was 0.23%, nearly as good as seen with the current Pascal GPUs.
In fact you also see the same jump with the 380 albeit not quite as strong.

And if you look closely, you can see the drop in 200 series actually aligns with the growth of the 390 and the 380, meaning AMD consumers upgraded from Hawaii to Hawaii rather to Polaris, primarily because it is the 200 series and not just 290/290x.
This does not look good for Polaris in terms of the 470/480 IMO because they have lost the core AMD support for those 2 models that selected the 390/380 instead and some jumped to the 1070/1080 it seems for the higher Hawaii models.
However the 1060 still managed to have strong growth (notice this is without affecting the 970 footprint) and is a card comparable in peer and performance to the 480.
The trend is same even when looking at older systems using the various cards.
Cheers

R9 380 and R9 390 series are "new" entries. In other words, they was counted as R9 200 series before. There was virtually no growth.

In terms of the overall progress, you can simply look on the installed base changes. They continue the trend.

 

[CSI_PC]

R9 380 and R9 390 series are "new" entries. In other words, they was counted as R9 200 series before. There was virtually no growth.

In terms of the overall progress, you can simply look on the installed base changes. They continue the trend.

There is no growth now but that is not the point, again they grew at the time the whole of the 200 series declined June-to-July, the trend is the move away from 200 series to the 390/380 rather than the 480 or waiting for 470.
Yes it is flat now because it is old product and one should not expect it to expand further with the launch of Polaris, but the context is the biggest growth for AMD was with the 390 and 380 rather than launched Polaris, the big jump for 390/380 was WHEN Polaris launched and not the 1st two months the 380/390 appeared in the list (increase was July and they appeared in the list May).
We do not see the same with Nvidia because 970 maintained its footprint over the same months and the 1060 has very strong growth.

Cheers

 

[Shintai]

There is no growth now but that is not the point, again they grew at the time the whole of the 200 series declined June-to-July, the trend is the move away from 200 series to the 390/380 rather than the 480 or waiting for 470.
Yes it is flat now because it is old product and one should not expect it to expand further with the launch of Polaris, but the context is the biggest growth for AMD was with the 390 and 380 rather than launched Polaris, the big jump for 390/380 was WHEN Polaris launched and not the 1st two months the 380/390 appeared in the list (increase was July and they appeared in the list May).
We do not see the same with Nvidia because 970 maintained its footprint over the same months and the 1060 has very strong growth.

Cheers

You do know that 380 series and 390 series cards was incorrectly counted as R9 200 series? Also remember you look at DX12 only.

 

[CSI_PC]

You do know that 380 series and 390 series cards was incorrectly counted as R9 200 series? Also remember you look at DX12 only. That got a Windows 10 requirement.

Ah I never knew that, any link where they said the 380/390 was counted as R9 200 and whether it took 2 months rather than 1 to correct; my posts and context is around the figure of 0.52% and 0.54% rather than the month before.

Even if you look lower than DX12 report the trend is the same in Table 2 (DX11 capable and the major Operating Systems).
,,,,,,,,,,,,,,,,,,,,,,.May.....June...July.....Aug.....Sept
R9 200 series: 1.74% 1.33% 0.68% 0.67% 0.64%
R9 390 series: 0.00% 0.22% 0.52% 0.53% 0.52%
R9 380 series: 0.00% 0.23% 0.54% 0.58% 0.58%
Thanks

For reference, compare that to 480 (again table 2 DX11 and associated Operating Systems).
RC480 series:.. - .......0.00% 0.06% 0.12% 0.21%
GTX1070:........ - .......0.05% 0.33% 0.71% 1.03%
GTX1080:........0.00% 0.10% 0.30% 0.49% 0.64%
GTX970...........5.13% 5.24% 5.21% 5.25% 5.17%
Thanks

Edit:
If going by what TaintedSpirit said then you would see a high proportion of 7970 as well for the 2xx being misreported but that does not happen.
I think he also forgot that AMD was down to around the low 20%s for discrete market footprint (with Hawaii) compared to Nvidia as reported by two leading analyst firms on sales shipments.

 

[Shintai]

Well a big hint to begin with is that 300 series launched in mid 2015 and you first see it in mid 2016?

But yes, AMDs installed base keeps declining because each generation sells less and less compared to Nvidia.

 

[CSI_PC]

Well a big hint to begin with is that 300 series launched in mid 2015 and you first see it in mid 2016?

But yes, AMDs installed base keeps declining because each generation sells less and less compared to Nvidia.

So my point is still valid because we can only say the correction was June and July increase is an accurate figure
Which would make sense because there was a sales drive to get rid of old stock but at the detriment of Polaris 480, that is my context and point along with that it shows there was no supply issues for Nvidia relative to AMD with Pascal/Polaris.
It did not help that initial benchmark reviews just before launch of 480 had it behind a 390X and for quite a few games the 390, which also would align with the sales spike for 390 and 390x.

Cheers

 

[Shintai]

So my point is still valid because we can only say the correction was June and July increase is an accurate figure
Which would make sense because there was a sales drive to get rid of old stock but at the detriment of Polaris 480, that is my context and point along with that it shows there was no supply issues for Nvidia relative to AMD with Pascal/Polaris.
It did not help that initial benchmark reviews just before launch of 480 had it behind a 390X and for quite a few games the 390, which also would align with the sales spike for 390 and 390x.

Cheers

So AMD had sold 0 R9 300 series cards until a year after release, then they suddenly sold a lot? Ok.....

 

[CSI_PC]

So AMD had sold 0 R9 300 series cards until a year after release, then they suddenly sold a lot? Ok.....

SO now your saying the 0.22% June figure is not an accurate correction and somehow the process managed to correct that the next month in July and was just a co-incidence that 380 and 390 had large discounted sales while also the 390 series was shown in reviews generally to be faster than the 480

I guess then it would mean hardly anyone buying Hawaii, which is strange considering the 2 leading market analysts reported AMD managing to claw back discrete market for last quarter (Polaris cannot be part of that as it is losing out in every way to each model of Pascal from a Steam userbase perspective, in theory the quarter would at most include July).
Cheers

 

[Shintai]

SO now your saying the 0.22% June figure is not an accurate correction and somehow the process managed to correct that the next month in July and was just a co-incidence that 380 and 390 had large discounted sales while also the 390 series was shown in reviews generally to be faster than the 480

I guess then it would mean hardly anyone buying Hawaii, which is strange considering the 2 leading market analysts reported AMD managing to claw back discrete market for last quarter (Polaris cannot be part of that as it is losing out in every way to each model of Pascal from a Steam userbase perspective, in theory the quarter would at most include July).
Cheers

AMD didn't gain market shareas such. If you look at the numbers you know why. AMD didn't ship more GPUs. It was Nvidia shipping a lot less due to the inventory change.

For steam, you can see the change in R9 200 series and R9 300 series is pretty much identical because its the exact same cards.

 

[Palladium@SG]

SO now your saying the 0.22% June figure is not an accurate correction and somehow the process managed to correct that the next month in July and was just a co-incidence that 380 and 390 had large discounted sales while also the 390 series was shown in reviews generally to be faster than the 480

I guess then it would mean hardly anyone buying Hawaii, which is strange considering the 2 leading market analysts reported AMD managing to claw back discrete market for last quarter (Polaris cannot be part of that as it is losing out in every way to each model of Pascal from a Steam userbase perspective, in theory the quarter would at most include July).
Cheers

The only reason AMD "gained" marketshare in Q2 is because NV cannibalized itself in their Maxwell->Pascal transition. Even the most diehard of NV fanboys won't buy Maxwell if Pascal is just around the corner.

 

[Shintai]

The only reason AMD "gained" marketshare in Q2 is because NV cannibalized itself in their Maxwell->Pascal transition. Even the most diehard of NV fanboys won't buy Maxwell if Pascal is just around the corner.

No, it was inventory clearance.

Q2 Q1
AMD 2,820,000 2,717,000
Nvidia 6,600,000 9,200,000

"Moreover, the company decided to clear out some of the inventory of older model cards amid the launch of Pascal-based GeForce 1070 and 1080 graphics cards, according to JPR. This greatly affected actual shipments and market share of NVIDIA: unit sales decreased by 14% YoY and by 28% QoQ, whereas market share declined to 70%."

 

[Cr4ckm0nk3y]

Fool and seeking alpha have never posted anything bad about Nvidia because they're both stock shills. Not sure why you're picking my post out, to get hits on your thread up??

With all due respect, you are wrong. Fool has not only posted articles that could be bad for NV, as razor gave you examples above, but they have also posted articles highlighting the positives of AMD and recommended buying AMD stock numerous times. If they were indeed stock shills then every article would say buy NV and sell AMD. Not recommending holds or even selling NV.

Like a majority of news sources Fool owns stock in a company they are reporting on. Frankly NV stock has been a good investment and you can't fault them for buying. At least they are honest and reported they own stock.

 

[CSI_PC]

The only reason AMD "gained" marketshare in Q2 is because NV cannibalized itself in their Maxwell->Pascal transition. Even the most diehard of NV fanboys won't buy Maxwell if Pascal is just around the corner.

That is part of my point, AMD fans did buy 390 and 380 series that cannibalized Polaris, it also does not help they are missing a 1070 equivalent.
Cheers

 

[pxc]

There may be some relationship between the Steam hardware survey statistics and card popularity, but it's not as simple as Ashraf makes it out to be.

First, Steam users are a subset of the PC gaming market. You can argue that it has a good cross-section of users, but we're talking about such small percentages that it matters how each generation is represented. Second, the survey is self-selecting since it's opt-in. There are various reasons some people may choose to participate or not. That skews the distribution of card owners. Basically, the data is fairly worthless for working out card adoption rates. There's a reason why analysis by JPR and others takes a while to figure out GPU growth and share, because it's a lot more complex than looking at a poll. It takes investigation beyond financial disclosures, like looking into supply chains and design wins.

Lastly, Ashraf is a click-bait troll, and doesn't care the least bit about inaccuracies and pretty much laughs in peoples' faces on other boards when he is called on out garbage he writes. He is often corrected and doesn't seem to care at all. He knows what gets him clicks, and that's what he lives on.

 

[CSI_PC]

There may be some relationship between the Steam hardware survey statistics and card popularity, but it's not as simple as Ashraf makes it out to be.

First, Steam users are a subset of the PC gaming market. You can argue that it has a good cross-section of users, but we're talking about such small percentages that it matters how each generation is represented. Second, the survey is self-selecting since it's opt-in. There are various reasons some people may choose to participate or not. That skews the distribution of card owners. Basically, the data is fairly worthless for working out card adoption rates. There's a reason why analysis by JPR and others takes a while to figure out GPU growth and share, because it's a lot more complex than looking at a poll. It takes investigation beyond financial disclosures, like looking into supply chains and design wins.

Lastly, Ashraf is a click-bait troll, and doesn't care the least bit about inaccuracies and pretty much laughs in peoples' faces on other boards when he is called on out garbage he writes. He is often corrected and doesn't seem to care at all. He knows what gets him clicks, and that's what he lives on.

That is true.
But the relationship sort of aligns, look at the primary GPU survey page (cannot currently use this for 480) and how many more 970 and 980 there are to both AMD R9 200 and R9 300 models
This is an easier to follow survey breakdown:
........May.....Jun.....Jul......Aug....Sept
970: 6.93% 7.21% 7.36% 7.32% 7.13%
980: 1.40% 1.45% 1.44% 1.40% 1.37%

R9 models:
200: 2.37% 1.86% 0.99% 0.96% 0.92%
380: 0.00% 0.32% 0.76% 0.80% 0.79%
390: 0.00% 0.31% 0.73% 0.73% 0.72%

Adding the 980ti and also Fury models would take this to around the 4x ratio.
Steam Hardware & Software Survey

Cheers