[ H ]: AMD Gains 4.8% Market Share in Q1 2016

[BFG10K]

Some good news for the red team, and good news for all consumers due to increased competition.

I think they also need to pursue profits with more aggressive licensing deals.

http://www.hardocp.com/news/2016/08/19/amd_gains_48_market_share_in_q1_2016#.V7f6srGxWHs

 

[biostud]

Also 330% stock increase this year. (Although coming from a very low value )

http://www.tomshardware.com/news/amd-zen-cpu-stock-market,32520.html

 

[sandorski]

sweet

 

[ViRGE]

The recent issuance of both Mercury Research and Jon Peddie Research market share data for Q2 2016

Huh? JPR's Q2'16 Add In Board Report is not out yet.

 

[Sonikku]

Well statistically speaking it was kind of hard to go lower at this point. Though I wish that market share was more with gamers than with just miners.

 

[zinfamous]

Also 330% stock increase this year. (Although coming from a very low value )

http://www.tomshardware.com/news/amd-zen-cpu-stock-market,32520.html

330% is still 330%. the starting value doesn't matter to the investors, except that it allows them much greater gains on if smart and lucky enough to hold/buy those shares back when they were under $2. Things are going great right now, as I think most of the short sellers are cleared out.

AMD traded at the highest volume on the market on Friday.

 

[Arachnotronic]

These numbers are misleading.

First of all, they only talk about units, not revenue share. Given that NVIDIA has scaled back its efforts to win non-gaming OEM designs (i.e. MacBook Pro, those Carrizo laptops with iGPU + Radeon, etc.), it's not hard for AMD to increase its share of the low-value dGPU market.

Last quarter, NVIDIA reported that revenue from its gaming GPU biz was up 14% quarter-over-quarter and 18% year-over-year. Last quarter, AMD reported that its total computing and graphics segment was down 5% quarter over quarter and up 15% year-over-year "driven by higher client notebook processor and graphics sales."

If AMD's total GPU +CPU biz was up 15% year-over-year and higher notebook APU sales were part of that growth, and if NVIDIA's gaming GPU sales were up 18% year-over-year, then this means that NVIDIA actually gained revenue share in the gaming dGPU market.

Next, AMD is guiding to a ~$200 million increase in revenue for Q3 over Q2 thanks to graphics/semicustom, while NVIDIA is guiding to a $250 million quarter-over-quarter increase in Q3 over Q2. Part of this could be Tegra, but it's probably mostly coming from dGPUs.

So, again in terms of revenue share, NVIDIA should actually gain share on AMD in the coming quarter.

 

[kondziowy]

Well, I take your opinion with an enormous grain of salt, as I noticed 98% of what you say always leads to conclusion Nvidia > AMD.

 

[zinfamous]

These numbers are misleading.

First of all, they only talk about units, not revenue share. Given that NVIDIA has scaled back its efforts to win non-gaming OEM designs (i.e. MacBook Pro, those Carrizo laptops with iGPU + Radeon, etc.), it's not hard for AMD to increase its share of the low-value dGPU market.

Last quarter, NVIDIA reported that revenue from its gaming GPU biz was up 14% quarter-over-quarter and 18% year-over-year. Last quarter, AMD reported that its total computing and graphics segment was down 5% quarter over quarter and up 15% year-over-year "driven by higher client notebook processor and graphics sales."

If AMD's total GPU +CPU biz was up 15% year-over-year and higher notebook APU sales were part of that growth, and if NVIDIA's gaming GPU sales were up 18% year-over-year, then this means that NVIDIA actually gained revenue share in the gaming dGPU market.

Next, AMD is guiding to a ~$200 million increase in revenue for Q3 over Q2 thanks to graphics/semicustom, while NVIDIA is guiding to a $250 million quarter-over-quarter increase in Q3 over Q2. Part of this could be Tegra, but it's probably mostly coming from dGPUs.

So, again in terms of revenue share, NVIDIA should actually gain share on AMD in the coming quarter.

Notice that the report is about market share, and not revenue share. Not sure why you are floating a different topic here. Not saying it isn't worth considering, but your point isn't the same topic.

We still need to look at Q3 and actual physical sales of units for Polaris/Pascal to get a better picture of what is going on this year, but if what all of the vendors are saying is true and reflected in the data, then AMD vastly outsold nVidia in # of units shipped. nVidia will still crush them in margins and revenue, obviously, because nVidia charges $100 premiums for shiny, mediocre shells on top of their increasingly expensive hardware. But again, they have the power to do that, so they do it.

 

[Arachnotronic]

Well, I take your opinion with an enormous grain of salt, as I noticed 98% of what you say always leads to conclusion Nvidia > AMD.

You can check all of the numbers and facts I'm using, they are public information. Nice personal attack though

 

[zinfamous]

You can check all of the numbers and facts I'm using, they are public information. Nice personal attack though

To be fair, I would call that more of a simple observation than a personal attack, unless standards in VC&G are vastly different--that's probably true.

And FWIW, you seem to me the more honest, reasonable, and ready with solid data nVidia partisans on these boards. I personally find your posts of greater value than any of the other partisans.

 

[jpiniero]

Well statistically speaking it was kind of hard to go lower at this point. Though I wish that market share was more with gamers than with just miners.

Yeah, that's the thing. How long can AMD reasonably expect this second GPU mining fad to continue?

 

[Erithan13]

These numbers are misleading.

That's what some of us have said about the Steam hardware survey. Speaking of which, you are on record as a believer in the steam numbers and even stated in that thread that you did not expect AMD to gain marketshare. Now that they have, it's interesting to note that you are now talking about revenue share, am I to interpret that as a subtle shifting of the goalposts?

Indeed, there looks to be very troubled times ahead for anyone standing by the steam hardware survey. I predict multiple sources are going to be telling us over the next few months that AMD have clawed back market share while the steam numbers don't change to reflect that. As well as all the other problems identified with the steam survey, that should surely be the final nail in the coffin for the validity of those numbers.

Yeah, that's the thing. How long can AMD reasonably expect this second GPU mining fad to continue?

I can think of three things that will bring an end to Ethereum mining:

1) Ethereum is planned to move to proof of stake, sometime, maybe....if/when it happens that should be the end of mining entirely.
2) Ethereum collapses completely and becomes literally worthless.
3) The growth of the difficulty curve isn't matched by the increase in value of Eth, meaning GPU mining rapidly hits the pointless diminishing returns phase and then becomes unprofitable (can't break even with power costs). Arguably, with the current price and difficulty of Eth, those of us with expensive energy are already hitting that point.

Which of these if any will kill mining and when will that happen? What effect will that have on AMD? Might there be another cryptocurrency that rises in place of Eth? Get your crystal ball out and make a guess....

 

[AtenRa]

These numbers are misleading.

http://www.dictionary.com/browse/misleading

verb (used with object), misled, misleading.
1.
to lead or guide wrongly; lead astray.
2.
to lead into error of conduct, thought, or judgment.
verb (used without object), misled, misleading.
3.
to be misleading; tend to deceive:
vague directions that often mislead.

Thread Topic = [ H ]: AMD Gains 4.8% Market Share in Q1 2016

Actually, it is your numbers that are misleading and not the Market Share numbers.

 

[jpiniero]

Indeed, there looks to be very troubled times ahead for anyone standing by the steam hardware survey. I predict multiple sources are going to be telling us over the next few months that AMD have clawed back market share while the steam numbers don't change to reflect that. As well as all the other problems identified with the steam survey, that should surely be the final nail in the coffin for the validity of those numbers.

If the AMD cards are being sold to mine they aren't going to be gaming on them. So any share gain wouldn't be reflected in a Steam survey.

 

[Abwx]

These numbers are misleading.


Last quarter, NVIDIA reported that revenue from its gaming GPU biz was up 14% quarter-over-quarter and 18% year-over-year.

This Nividia s number is misleading, yes...

By selling whole cards rather than the GPU only they artificialy inflated their revenues numbers, for instance rather than getting say 200$ for a GPU they got 400$ for a whole cards, of course the added 200$ do not increase signficantly the margin in $$ and add of course 0 to the maketshare, but this double the apparent revenue, they did so by selling themeselves their new cards...

 

[IEC]

If the AMD cards are being sold to mine they aren't going to be gaming on them. So any share gain wouldn't be reflected in a Steam survey.

I have Steam installed on mining rigs, but it only reports 1 of 6 cards for the purposes of the hardware survey, and never reported more than 1 card on my gaming rig either (which at one point had R9 290 Tri-Fire). So Steam Hardware survey is hardly an accurate representation anyways.

 

[zinfamous]

I have Steam installed on mining rigs, but it only reports 1 of 6 cards for the purposes of the hardware survey, and never reported more than 1 card on my gaming rig either (which at one point had R9 290 Tri-Fire). So Steam Hardware survey is hardly an accurate representation anyways.

Steam hardware survey also tends to report IGP on intel chips over discreet AMD GPUs in systems, right? I'm getting to the point where I ignore any posts using steam hardware survey as an argument.

 

[Shivansps]

I have Steam installed on mining rigs, but it only reports 1 of 6 cards for the purposes of the hardware survey, and never reported more than 1 card on my gaming rig either (which at one point had R9 290 Tri-Fire). So Steam Hardware survey is hardly an accurate representation anyways.

Actually it is because it directly reflects the hardware that steam platform users have, and steam platform is the biggest plaform to distribute games on PC. Mining rings are not representation of gamers hardware.
SO unless MS start doing something similar with its market app, its gona stay as the most accurate way to meassure hardware. And AMD is doing worse than ever on the July one, lets see what August may bring.

People like to dimiss it just because just because is not on AMD favor. If someone have a better way to get more accurate results online just point me at it.

 

[IEC]

Did you read the latter part of my post? It never accurately reflected multi-card gaming setups or even cases where you have IGP + discrete, like with some laptops using nVidia's Optimus implementation. So if you are trying to calculate market share (which has a specific definition) from the Steam hardware survey (which has a different, flawed methodology) you are comparing apples and kumquats.

 

[AtenRa]

What people dont understand is that Steam Hardware Survey numbers are relative to Steam userbase ONLY. They ONLY represent the percentage share for each company ONLY on Steam and thats it.

Global AIB market share cannot be represented by the Steam Hardware Survey, only independent research firms numbers from JPR, Mercury Research and more, are legitimate for comparison.

 

[zinfamous]

Actually it is because it directly reflects the hardware that steam platform users have, and steam platform is the biggest plaform to distribute games on PC. Mining rings are not representation of gamers hardware.

Actually, it doesn't, and that is the real problem. It specifically does not reflect hardware accurately across all users, as in many cases it defines an IGP in systems over discreet GPU that are installed. It doesn't matter if that favors one brand over the other--it matters because it shows that the survey simply doesn't work. If it is broken in one direction, regardless of direction, it is not a useful metric.
http://forums.steampowered.com/foru...0da5e2ff1c856e49ef277&p=26196972&postcount=10

It doesn't properly distinguish between generations of GPUs
https://www.reddit.com/r/Amd/commen...ies_show_up_in_steam/?st=is3duap6&sh=b61ecd13

https://techreport.com/forums/viewtopic.php?t=117431

Look, it doesn't even properly reflect Linux/Steam's own OS
http://www.pcworld.com/article/3045...hier-than-steams-hardware-survey-implies.html


And, simply enough: Steam doesn't publish the methodology of how their survey works. I work in science, I read research articles all the time. I can tell you that any metrics used for any data in a report, where the methodology is not published and therefore verifiable by other investigators, is not acceptable. This wouldn't even pass approval to go into peer review. It would be rejected flat. So, it's simply an unacceptable method in its current use. About as reliable as the polygraph.

Look, this data and information has been out there long enough, widely available to the tech community-basically anyone that claims enough interest in these topics to post in these threads--that the unreliability of the steam hardware survey is a simple fact of the industry. It strikes me that anyone that posts in these pages using steam hardware survey data are either intentionally misleading, or guilty of the cardinal sin of any real enthusiasts discipline: failure to inform themselves of openly available, salient information.

To put a finer point on it: Using the steam hardware survey as a metric for hardware adaptation rates in the PC space is, quite literally, bringing the results of a phrenology exam into a diagnosis of a person's health and personality.

 

[IllogicalGlory]

Does anyone know what this is supposed to be?

Apparently the second most popular AMD card is something that doesn't even exist.

 

[Shivansps]

Actually, it doesn't, and that is the real problem. It specifically does not reflect hardware accurately across all users, as in many cases it defines an IGP in systems over discreet GPU that are installed. It doesn't matter if that favors one brand over the other--it matters because it shows that the survey simply doesn't work. If it is broken in one direction, regardless of direction, it is not a useful metric.
http://forums.steampowered.com/foru...0da5e2ff1c856e49ef277&p=26196972&postcount=10

It doesn't properly distinguish between generations of GPUs
https://www.reddit.com/r/Amd/commen...ies_show_up_in_steam/?st=is3duap6&sh=b61ecd13

https://techreport.com/forums/viewtopic.php?t=117431

Look, it doesn't even properly reflect Linux/Steam's own OS
http://www.pcworld.com/article/3045...hier-than-steams-hardware-survey-implies.html


And, simply enough: Steam doesn't publish the methodology of how their survey works. I work in science, I read research articles all the time. I can tell you that any metrics used for any data in a report, where the methodology is not published and therefore verifiable by other investigators, is not acceptable. This wouldn't even pass approval to go into peer review. It would be rejected flat. So, it's simply an unacceptable method in its current use. About as reliable as the polygraph.

Look, this data and information has been out there long enough, widely available to the tech community-basically anyone that claims enough interest in these topics to post in these threads--that the unreliability of the steam hardware survey is a simple fact of the industry. It strikes me that anyone that posts in these pages using steam hardware survey data are either intentionally misleading, or guilty of the cardinal sin of any real enthusiasts discipline: failure to inform themselves of openly available, salient information.

To put a finer point on it: Using the steam hardware survey as a metric for hardware adaptation rates in the PC space is, quite literally, bringing the results of a phrenology exam into a diagnosis of a person's health and personality.

That mostly old news, but it does have problem with Optimus configurations as far i remember, the 300 and 200 series its the most intelligent thing they ever did, the 300 series DOES NOT EXIST, its 100% rebrand, its pointless to split it.

yes its a little glichy with multi gpu, but it now should properly detect the used GPU, instead of detecting the first device ID it finds, it also supports detection for SLI/CF but it does not seem to catalog it like that. And last time i think it detected both my 750TI and the Intel HD3000, when i had a monitor plugged on it. Its kinda hard to test becuase it only runs the "system information" once when you click on it, not sure how to enforce to re-check hardware after that.

Did you read the latter part of my post? It never accurately reflected multi-card gaming setups or even cases where you have IGP + discrete, like with some laptops using nVidia's Optimus implementation. So if you are trying to calculate market share (which has a specific definition) from the Steam hardware survey (which has a different, flawed methodology) you are comparing apples and kumquats.

market share is just not the same thing, still Steam is the primary gaming platform on PC, if we are gona talk about market share we need to include consoles, server and enterprise enviroments that does not matter at all for us.

 

[Shivansps]

Does anyone know what this is supposed to be?

Apparently the second most popular AMD card is something that doesn't even exist.

laptop graphic card.