[Anandtech] Discrete Q2 GPU Marketshare - AMD rises to 30%

[ShintaiDK]

Latest Mindfactory.de numbers taken from today.
RX460 545
RX470 2025
RX480 3415
GTX 1060 10090
GTX 1070 17535
GTX 1080 7505

 

[ShintaiDK]

I already did in this thread, twice!

Here is the quote again

I dont see any numbers.

 

[Keysplayr]

Now that the 4gb HBM Fury is memory starved, I guess there really is no reasonable defense for the 3gb "1060" right?


....right?

Oh brother. Is the 10603GB, or 6GB intended to run 4K? Come on man, think before you mock others.

 

[Keysplayr]

Except the Fury does handle 4k very well, it even performs nearly the same as the 1070.

https://www.techpowerup.com/reviews/ASUS/GTX_1060_STRIX_OC/26.html

Its performing closer to the 8GB 1070 than the 6GB 980 TI. Guess HBM does work better than GDDR5 after all.

Could be the great memory management they have going on with FuryX and the HBM memory.

 

[zinfamous]

Source? Because I quoted the CEO saying that sales were great.

lol. "anything but fantastic." First, official sales of dGPU from Q3 haven't been released. SO whatever Shintai is going on about those numbers, he is inventing information. 2nd, The first production of Polaris sold out. I guess that "isn't so great." Since he's on about how much more excellent nVidia is in comparison, I would think that Shintai is aware of the vendor reports--those people that actually order and sell these things--and how they were only seeing 1080/1070/1060 at roughly 1/10th to 1/20th the volume of Polaris cards.

But, we'll know in about a month. Until then, making claims about such numbers is pure speculation.

EDIT: lol--this guy still uses Steam numbers, knowing full well that this is a useless metric. Amazing. He just doesn't gas if his metrics are broken.

 

[AtenRa]

Please provide documentation for that.

Documentation for what ??? A higher MSRP card has a more expensive Chip.

R9 390X with 8GB ram and MSRP = $429

R9 390 with 8GB ram and MSRP = $329

R9 390X chip is more expensive (AIB pays more than the 390 Chip) thus R9 390X card has higher BOM and higher MSRP.



GTX 980 with 4GB ram and MSRP $549

GTX 970 with 4GB ram and MSRP $329

GTX 980 chip is more expensive (AIB pays more than the GTX 970 Chip) thus GTX 980 card has higher BOM and higher MSRP.


So again, AMD shipped some Polaris 10 chips in Q2 that had lower selling price than the Hawaii chips and thus ASPs got lower.

 

[ShintaiDK]

Documentation for what ??? A higher MSRP card has a more expensive Chip.

R9 390X with 8GB ram and MSRP = $429

R9 390 with 8GB ram and MSRP = $329

R9 390X chip is more expensive (AIB pays more than the 390 Chip) thus R9 390X card has higher BOM and higher MSRP.



GTX 980 with 4GB ram and MSRP $549

GTX 970 with 4GB ram and MSRP $329

GTX 980 chip is more expensive (AIB pays more than the GTX 970 Chip) thus GTX 980 card has higher BOM and higher MSRP.


So again, AMD shipped some Polaris 10 chips in Q2 that had lower selling price than the Hawaii chips and thus ASPs got lower.

You didn't document anything. Again, ASP isn't MSRP. If you claim the ASP change is due to Polaris, then document it. What does a Hawaii GPU cost, what does a Polaris GPU cost when AMD is selling them to AIBs. Not what AIB sells cards at.

 

[Krteq]

Well, nice thread crapping guys. Good job :/

 

[Gikaseixas]

Could be the great memory management they have going on with FuryX and the HBM memory.

It's has more to do with Fury's high bandwidth memory than the so called memory management

 

[AtenRa]

You didn't document anything. Again, ASP isn't MSRP. If you claim the ASP change is due to Polaris, then document it. What does a Hawaii GPU cost, what does a Polaris GPU cost when AMD is selling them to AIBs. Not what AIB sells cards at.

Same applies to NVIDIA then, higher ASPs doesnt mean they sold higher MSRP cards

 

[nurturedhate]

Steam hardware surveys, saying we can only compare certain cards under 4k, arguing illogically about ASP, 3gb versus 4gb, anything and everything to derail this thread.

 

[ShintaiDK]

Same applies to NVIDIA then, higher ASPs doesnt mean they sold higher MSRP cards

That's right. It means they earned more per chip sold. Irrelevant as such to the cards retail value.

 

[Gikaseixas]

Steam hardware surveys, saying we can only compare certain cards under 4k, arguing illogically about ASP, 3gb versus 4gb, anything and everything to derail this thread.

We could indeed use more moderation right now

 

[AtenRa]

That's right. It means they earned more per chip sold. Irrelevant as such to the cards retail value.

NO, higher ASP (Average Selling Price) doesnt mean they earned more per chip. It means the average price of the products sold got higher.

Higher margins means they earned more per chip. ASP and MSRP goes together because higher MSRP means higher selling price.

 

[DAPUNISHER]

We could indeed use more moderation right now

I say this as a member posting in this thread.

This is a AMD specific topic and should have been posted in the AMD forum. Posting it in the general area was a mistake. Or if done so because of the higher traffic, for exposure/attention, then be prepared for the negative attention that brings with it. Also as the sub forums have more specific posting guidelines,members are subject to stricter moderation as a consequence. Don't like the heat? Stay out of the kitchen, comes to mind.

 

[ShintaiDK]

NO, higher ASP (Average Selling Price) doesnt mean they earned more per chip. It means the average price of the products sold got higher.

Higher margins means they earned more per chip. ASP and MSRP goes together because higher MSRP means higher selling price.

Higher selling price, but for who? AMD isn't the AIB. AMD could for the sake of argument sell 50$ Polaris chips and 50$ Hawaii chips to their AIBs. Remember all the discounted Hawaii cards now and then? What about those? The GPU that AMD sells is just 1 part of the cards.

The production cost for a Polaris chip and a Hawaii chip should also be relatively close.

 

[Gikaseixas]

I say this as a member posting in this thread.

This is a AMD specific topic and should have been posted in the AMD forum. Posting it in the general area was a mistake. Or if done so because of the higher traffic, for exposure/attention, then be prepared for the negative attention that brings with it. Also as the sub forums have more specific posting guidelines,members are subject to stricter moderation as a consequence. Don't like the heat? Stay out of the kitchen, comes to mind.

When the heat is solely from a single source one has to wonder why don't someone put it away after so many similar events. But I get your point and its completely fair

Sent from my HUAWEI MT7-L09 using Tapatalk

 

[DAPUNISHER]

When the heat is solely from a single source one has to wonder why don't someone put it away after so many similar events. But I get your point and its completely fair

Sent from my HUAWEI MT7-L09 using Tapatalk

I will say this then stop dragging this OT. I personally do not see anyone violating the guidelines. And I also like it when people challenge my stance or opinion on something. Instead of digging my heels in, and being even more steadfast and intractable, I try to see things from their POV. It keeps my thinking dynamic, and keeps me from falling into dogmatic, burned in neural pathways. It does, also, require checking the ego at the door, and accepting I may be too emotionally invested in the topic to maintain the necessary objectivity.

And when the facts are on my side, but they are the ones being intractable? I desist, because we all know the saying about arguing on the internet.

 

[Sweepr]

When the heat is solely from a single source one has to wonder why don't someone put it away after so many similar events.

Maybe you should be thinking about adding something useful to the discussion instead of seeking infractions for that particular source only because you don't agree with them.

 

[Gikaseixas]

Maybe you should be thinking about adding something useful to the discussion instead of seeking infractions that particular source only because you don't agree with them.

Like your post added anything to the discussion
AMD is back on track, making inroads as planned by its management. Now let's wait and see if Vega can compete too.

Sent from my HUAWEI MT7-L09 using Tapatalk

 

[Madpacket]

Steam, Mindfactory.
http://store.steampowered.com/hwsurvey/directx/
August 2016:
GTX 1080 0.94% (+0.38%)
GTX 1070 1.48% (+0.72%)
GTX 1060 0.48% (+0.42%)
RX 480 0.26% (+0.14%)
RX 470 0.01% (+0.01%)
RX 460 0.02% (+0.02%)

https://www.reddit.com/r/Amd/comments/4xo5ip/amd_vs_nvidia_sales_numbers_one_outlet/
RX 480 - 3,560
RX 470 - 225
RX 460 - 50
GTX 1080 - 5,830
GTX 1070 - 11,230
GTX 1060 - 4,265

Great compared to what? What numbers? Show me!

Some advice Shintai, please don't take this the wrong way but every time you quote Steam for sales metrics you lose credibility.

 

[ShintaiDK]

Some advice Shintai, please don't take this the wrong way but every time you quote Steam for sales metrics you lose credibility.

Why? I know some people dont like the steam numbers because it shows something they dont like. That doesn't make it less accurate.

I also posted the latest mindfactory.de numbers. And they are in line with steam.

RX460 545
RX470 2025
RX480 3415
GTX 1060 10090
GTX 1070 17535
GTX 1080 7505

 

[zinfamous]

Why? I know some people dont like the steam numbers because it shows something they dont like.
I also posted the latest mindfactory.de numbers. And they are in line with steam.

You still don't get it. it is a metric that simply doesn't work. We went over this with you in two previous threads and you have simply ignored the piles of data that show you how broken this is.

Until steam fixes the way that their software identifies actual hardware, and restructures the survey to capture an accurate sample of users, reading your posts will continue to be like bringing a phrenology chart into a lecture on human physiology. No one cares if you take yourself seriously--just understand that no one else does because you simply do not understand bad data when it whips you in the face.

 

[Arachnotronic]

You still don't get it. it is a metric that simply doesn't work. We went over this with you in two previous threads and you have simply ignored the piles of data that show you how broken this is.

Until steam fixes the way that their software identifies actual hardware, and restructures the survey to capture an accurate sample of users, reading your posts will continue to be like bringing a phrenology chart into a lecture on human physiology. No one cares if you take yourself seriously--just understand that no one else does because you simply do not understand bad data when it whips you in the face.

It does work. See the performance of NVIDIA's GeForce GTX business and the performance of AMD's high end GPU biz. AMD's "share gains" like I keep telling you are mainly in low-end OEM systems, markets from which NVIDIA has pretty much pulled out.

 

[Dribble]

You still don't get it. it is a metric that simply doesn't work. We went over this with you in two previous threads and you have simply ignored the piles of data that show you how broken this is.

Until steam fixes the way that their software identifies actual hardware, and restructures the survey to capture an accurate sample of users, reading your posts will continue to be like bringing a phrenology chart into a lecture on human physiology. No one cares if you take yourself seriously--just understand that no one else does because you simply do not understand bad data when it whips you in the face.

He's not alone, anyone not too invested in what the figures show could see that steam is a reasonable statistical tool - huge data set, random sampling, regular surveys. Not perfect, no survey every is, but still very useful. No - I can't be bothered arguing with you over it because you don't want to listen because it isn't saying what you want to hear. Just don't be under the assumption that banging the drum louder then anyone else has made you any more correct.