Steam Hardware Survey Results (Video cards)

[cbn]

Explain why the 1050 Ti is up 1% yet top on Amazon as well as a few other places while the 5% up 1060

These listings are for specific cards not a ranking for all cards with a specific GPU.

Also don't forget the GTX 1060 has been around longer.

But GTX 1060 vs. GTX 1050 Ti vs. GTX 1070 (or whatever) is not my main point.

My main point is that RX 470 ranks low on Steam and now I find out also on Amazon. (even the RX 570 doesn't rank that high in Amazon's top 100)

 

[tential]

So you are trying relate Steam's hardware survey to GPU sales?

Explain why the 1050 Ti is up 1% yet top on Amazon as well as a few other places while the 5% up 1060 is lower in sales? Doesn't seem to correlate that well imho.

I don't understand why you would ever use data in which you can't explain the testing methodology.

It's like if someone ran benchmarks of all games and it looked off slightly, then we find out he actually benched every game on Super ultra reduce your fps to 7 settings, and looked at a wall, and that's how he got 25-35 GPS range in his benches.

You need to know methodology to use data...

 

[cbn]

It's like if someone ran benchmarks of all games and it looked off slightly, then we find out he actually benched every game on Super ultra reduce your fps to 7 settings, and looked at a wall, and that's how he got 25-35 GPS range in his benches.

You don't think that analogy might be a bit extreme?

It's not believable that Nvidia is more popular than AMD?

 

[cbn]

Here is the best selling list from Newegg (GTX 1060, GTX 1050 Ti, RX 580, RX570 and RX 480. RX 470 is no longer sold by Newegg):

https://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709 601203793 601206485 8000 601205646 601273503 601296377 601296379 4814&IsNodeId=1&bop=And&order=BESTSELLING

GTX 1060 dominates here just like we see in the Steam Survey. (Bacon1 will also be happy because in Newegg's ranking GTX 1060 outdoes GTX 1050 Ti by a very strong margin occupying 14 of the top 20 spots compared to only 3 for GTX 1050 Ti)

Below is the top 20, but the link I provided actually has 110 entries.


(54)
EVGA GeForce GTX 1060 6GB SSC GAMING ACX 3.0, 6GB GDDR5, LED, DX12 OSD Support (PXOC), 06G-P4-6267-KR
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1607 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 06G-P4-6267-KR
• Item #: N82E16814487275
• Return Policy: Replacement Only Return Policy

• $249.99
• Sale Ends in 5 Days (Mon)
• $239.99 after $10.00 rebate card
• Free Shipping

Compare

(19)
EVGA GeForce GTX 1050 Ti GAMING, 04G-P4-6251-KR, 4GB GDDR5, DX12 OSD Support (PXOC)
Compact Size – 5.7in

• Core Clock: 1290 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 1 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 04G-P4-6251-KR
• Item #: N82E16814487290
• Return Policy: Standard Return Policy

• $134.99
• Sale Ends in 5 Days (Mon)
• $114.99 after $20.00 rebate card
• Free Shipping

Compare

(11)
EVGA GeForce GTX 1060 3GB SSC GAMING ACX 3.0, 03G-P4-6167-KR, 3GB GDDR5, LED, DX12 OSD Support (PXOC)
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1607 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 03G-P4-6167-KR
• Item #: N82E16814487284
• Return Policy: Replacement Only Return Policy

• $214.99
• $199.99 after $15.00 rebate card
• Free Shipping

Compare

(53)
EVGA GeForce GTX 1050 Ti SC GAMING, 04G-P4-6253-KR, 4GB GDDR5, DX12 OSD Support (PXOC)
SC Performance, Compact Size – 5.7in

• Core Clock: 1354 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 1 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 04G-P4-6253-KR
• Item #: N82E16814487291
• Return Policy: Replacement Only Return Policy

• $149.99
• $139.99
• Sale Ends in 5 Days (Mon) -Save: 7%
• $119.99 after $20.00 rebate card
• Free Shipping

Compare

(45)
EVGA GeForce GTX 1060 GAMING, ACX 2.0 (Single Fan), 06G-P4-6161-KR, 6GB GDDR5, DX12 OSD Support (PXOC), Only 6.8 Inches
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1506 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 06G-P4-6161-KR
• Item #: N82E16814487260
• Return Policy: Replacement Only Return Policy

• $239.99
• Free Shipping

Compare

(106)
GIGABYTE GeForce GTX 1060 DirectX 12 GV-N1060WF2OC-6GD 6GB 192-Bit GDDR5 PCI Express 3.0 x16 ATX Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: OC Mode: 1582 MHz Gaming Mode: 1556 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 1 x DisplayPort 1.4
• DVI: 2 x Dual-link DVI-D
• Model #: GV-N1060WF2OC-6GD
• Item #: N82E16814125901
• Return Policy: Replacement Only Return Policy

• $249.99
• Sale Ends in 5 Days (Mon)
• $229.99 after $20.00 rebate card
• Free Shipping

Compare

(45)
EVGA GeForce GTX 1060 GAMING, ACX 2.0 (Single Fan), 03G-P4-6160-KR, 3GB GDDR5, DX12 OSD Support (PXOC), Only 6.8 Inches
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1506 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 03G-P4-6160-KR
• Item #: N82E16814487263
• Return Policy: Replacement Only Return Policy

• $199.99
• $179.99 after $20.00 rebate card
• Free Shipping

Compare

(146)
EVGA GeForce GTX 1060 SC GAMING, ACX 2.0 (Single Fan), 06G-P4-6163-KR, 6GB GDDR5, DX12 OSD Support (PXOC), Only 6.8 Inches
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1607 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #:
• Item #: N82E16814487261
• Return Policy: Replacement Only Return Policy

• $249.99
• $239.99 after $10.00 rebate card
• Free Shipping

Compare

(26)
EVGA GeForce GTX 1060 DirectX 12 06G-P4-6368-KR 6GB 192-Bit GDDR5 PCI Express 3.0 FTW+ GAMING ACX 3.0 Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1632 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 06G-P4-6368-KR
• Item #: N82E16814487268
• Return Policy: Replacement Only Return Policy

• $299.99
• Free Shipping

Compare

(34)
EVGA GeForce GTX 1060 SC GAMING, ACX 2.0 (Single Fan), 03G-P4-6162-KR, 3GB GDDR5, DX12 OSD Support (PXOC)
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1607 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 03G-P4-6162-KR
• Item #: N82E16814487267
• Return Policy: Replacement Only Return Policy

• $179.99
• Sale Ends in 4 Hours
• $169.99 after $10.00 rebate card
• Free Shipping

Compare

(14)
EVGA GeForce GTX 1060 GAMING DirectX 12 06G-P4-6262-KR 6GB 192-Bit GDDR5 PCI Express 3.0 Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1506 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 06G-P4-6262-KR
• Item #: N82E16814487280
• Return Policy: Replacement Only Return Policy

• $269.99
• $249.99 after $20.00 rebate card
• Free Shipping

Compare

(150)
MSI GeForce GTX 1060 DirectX 12 GTX 1060 GAMING X 6G 6GB 192-Bit GDDR5 PCI Express 3.0 x16 HDCP Ready ATX Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1594 MHz (OC Mode) 1569 MHz (Gaming Mode) 1506 MHz (Silent Mode)
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x DL-DVI-D
• Model #: GTX 1060 GAMING X 6G
• Item #: N82E16814127963
• Return Policy: Replacement Only Return Policy

• $279.99
• $269.99
• $254.99 after $15.00 rebate card
• Free Shipping

Compare

(37)
PNY GeForce GTX 1060 DirectX 12 VCGGTX10606PB 6GB 192-Bit GDDR5 PCI Express 3.0 Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1506 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 3 x DisplayPort 1.4
• DVI: 1 x DL-DVI
• Model #: VCGGTX10606PB
• Item #: N82E16814133631
• Return Policy: Replacement Only Return Policy

• $246.99
• Free Shipping

Compare

(7)
MSI GeForce GTX 1060 DirectX 12 GTX 1060 ARMOR 3G OCV1 3GB 192-Bit GDDR5 PCI Express 3.0 x16 HDCP Ready ATX Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1544 MHz
• DisplayPort: 2 x DisplayPort 1.4
• DVI: 1 x DL-DVI-D
• HDMI: 2 x HDMI 2.0
• Model #: GTX 1060 ARMOR 3G OCV1
• Item #: N82E16814137036
• Return Policy: Replacement Only Return Policy

• $204.99
• $184.99 after $20.00 rebate card
• $3.99 Shipping

Compare

(37)
MSI GeForce GTX 1060 DirectX 12 GTX 1060 GAMING X 3G 3GB 192-Bit GDDR5 PCI Express 3.0 x16 HDCP Ready ATX Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1594 MHz (OC) 1569 MHz (Gaming) 1506 MHz (Silent)
• DisplayPort: 3 x DisplayPort
• DVI: 1 x DL-DVI-D
• HDMI: 1 x HDMI
• Model #: GTX 1060 GAMING X 3G
• Item #: N82E16814137033
• Return Policy: Replacement Only Return Policy

• $199.99
• Sale Ends in 7 Days (Wed)
• $179.99 after $20.00 rebate card
• Free Shipping

Compare

(8)
MSI GeForce GTX 1060 DirectX 12 GTX 1060 ARMOR 6G OCV1 6GB 192-Bit GDDR5 PCI Express 3.0 x16 HDCP Ready ATX Video Card
Pick For Honor or Ghost Recon: Wildlands game w/purchase, limited offer

• Core Clock: 1544 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 2 x DisplayPort 1.4
• DVI: 1 x DL-DVI-D
• Model #: GTX 1060 ARMOR 6G OCV1
• Item #: N82E16814137040
• Return Policy: Replacement Only Return Policy

• $245.99
• $225.99 after $20.00 rebate card
• Free Shipping

Compare

(29)
EVGA GeForce GTX 1050 Ti SSC GAMING ACX 3.0, 04G-P4-6255-KR, 4GB GDDR5, DX12 OSD Support (PXOC)
EVGA ACX 3.0 Cooling for improved performance

• Core Clock: 1366 MHz
• Max Resolution: 7680 x 4320
• DisplayPort: 1 x DisplayPort 1.4
• DVI: 1 x Dual-Link DVI-D
• Model #: 04G-P4-6255-KR
• Item #: N82E16814487292
• Return Policy: Standard Return Policy

• $159.99
• $149.99 after $10.00 rebate card
• Free Shipping

Compare

(5)
PowerColor RED DRAGON Radeon RX 480 DirectX 12 AXRX 480 8GBD5-3DHD 8GB GDDR5 PCI Express 3.0 CrossFireX Support ATX Video Card

• Max Resolution: 4096 x 2160
• DisplayPort: 3 x DisplayPort
• DVI: 1 x DL-DVI-D
• HDMI: 1 x HDMI
• Model #: AXRX 480 8GBD5-3DHD
• Item #: N82E16814131704
• Return Policy: Replacement Only Return Policy

• $249.99
• $229.99 after $20.00 rebate card
• Free Shipping

Compare


ASUS ROG Strix Radeon RX 570 O4G Gaming OC Edition GDDR5 DP HDMI DVI VR Ready AMD Graphics Card (ROG-STRIX-RX570-O4G-GAMING)

• Max Resolution: 5120 x 2880
• DisplayPort: 1 x DisplayPort 1.4
• DVI: 2 x DVI-D
• HDMI: 1 x HDMI 2.0
• Model #: ROG-STRIX-RX570-O4G-GAMING
• Item #: N82E16814126189
• Return Policy: Replacement Only Return Policy

• $199.99
• Free Shipping

Compare

(19)
PowerColor RED DRAGON Radeon RX 480 DirectX 12 AXRX 480 4GBD5-3DHD 4GB 256-Bit GDDR5 PCI Express 3.0 CrossFireX Support ATX Video Card

• Max Resolution: 4096 x 2160
• DisplayPort: 3 x DisplayPort
• DVI: 1 x DL DVI-D
• HDMI: 1 x HDMI
• Model #: AXRX 480 4GBD5-3DHD
• Item #: N82E16814131708
• Return Policy: Replacement Only Return Policy

• $199.99
• $4.99 Shipping

 

[Headfoot]

"Top Seller" is even more worthless than Steam data for determining market share. It's entirely unclear how something makes the "best seller list" and there's an enormously huge chance they bump things to that list IN ORDER for it to sell more, not based on marketshare.

 

[tential]

"Top Seller" is even more worthless than Steam data for determining market share. It's entirely unclear how something makes the "best seller list" and there's an enormously huge chance they bump things to that list IN ORDER for it to sell more, not based on marketshare.

I really think people don't understand which data is significant "hard data" compared to "soft data".

 

[Cuular]

Steam data is only relevant if you consider that they only collect it from people who haven't turned off letting them collect it.

Any smart person, has had that turned off for years.

 

[cbn]

It's entirely unclear how something makes the "best seller list" and there's an enormously huge chance they bump things to that list IN ORDER for it to sell more, not based on marketshare.

I doubt either Newegg or Amazon is faking their list.

 

[VirtualLarry]

but GTX 750 Ti is such an old card (without HEVC or HDMI 2.0) and lower performance per dollar than RX460.

The Steam masses are much less educated about video cards, than the enthusiasts on this site.

There was such a buzz about the 750ti when it first came out... it appears that the legend of the 750ti won't die. Plus, well, NVidia. "The way it's meant to be played." They hear that coming out of every other game they play, they think that it must be true. Plus, all of their other (less-informed, non-enthusiast) buddies all have NVidia too.

 

[cbn]

There was such a buzz about the 750ti when it first came out... it appears that the legend of the 750ti won't die. Plus, well, NVidia. "The way it's meant to be played." They hear that coming out of every other game they play, they think that it must be true. Plus, all of their other (less-informed, non-enthusiast) buddies all have NVidia too.

Yep, that card had great performance considering lack of PCIe connector (much faster than the R7 250 GDDR5 and GTX 650/GT 740). Digital foundry even branded it a console killer (both Xbox One and PS4) when used with a Core i3-4130.

Some examples below:

https://www.youtube.com/watch?v=iipDWbd6HNg

https://www.youtube.com/watch?v=gGf4SVWEw2g

https://www.youtube.com/watch?v=9OE2iI7OLh8

And even today I can even imagine that card still doing fairly well with the exception of some recent (or recent-ish) AAA titles (eg, Deus Ex Mankind divided).

So I'll bet what is happening is folks are keeping that one or handing down to someone else because it still allows a good level of performance in a pre-built without needing to swap the power supply.

 

[cbn]

AMD needs a big hit with Vega.

It would be good to see that GPU (and variations of that GPU) be able to make a good showing against the GTX 1060 and GTX 1070 which appear to be the bread and butter of Nvidia's consumer cards at this time.

EDIT: Fixed quote.

 

[OatisCampbell]

Steam data is only relevant if you consider that they only collect it from people who haven't turned off letting them collect it.

Any smart person, has had that turned off for years.

Its kind of mysterious how the STEAM, Etailer, and Jon Peddie Research numbers all seem consistent.

WHAT could be behind this strange coincidence?

Is it some kind of attempt at mind control, to makes us all think NVIDIA sells the most cards these days?!

When Vega parts launch those numbers will start to shift back toward the 40 - 50% they used to have. I just saw a 144hz 34" 3440 x 1440 LG monitor for $699 MSRP- with Freesync. AMD comes out with parts that crush that resolution and they will sell. That's an enthusiast dream set up for $1200 if Vega is $499.- the price of just the monitor going Gsync.

 

[Stuka87]

I doubt either Newegg or Amazon is faking their list.

There have been other threads here that show that Amazon's best seller list is in no way a list to use what is currently the best selling product. Which is why old products show up, such as the nVidia GT210. I will see if I can find it. But the algorithm they use pretty much makes it impossible to see what is currently the top selling product. And whats worse, is different product types use different algorithms.

BTW: Not saying nVidia would not still be at the top, just that the list should not be trusted as a whole between what is actually at the top. IE: A GT210 should not be that high up. But its been out for ages, so its results are skewed.

 

[Blue_Max]

How does a GTX 750 Ti have a higher adoption % gain? (Particularly when going from March 2017 to April 2017)

GTX 750 Ti March 2017 = 5.13%
GTX 750 Ti April 2017 = 5.34%

That "just drop one in" factor is huge. We'll be seeing 1050/TI take that spot over time. Between used cards never dying, used and new being picked up cheap... I can see it staying strong for a while.

I'll bet nVidia didn't even realize how big of a hit the 750TI would be.

 

[Dribble]

I'll bet nVidia didn't even realize how big of a hit the 750TI would be.

For quite a while the cheapest card that could actually game properly and didn't require a huge case and psu was a 750Ti. Hence when someone said - "I need a card for gaming, don't want to spend a fortune" they all bought 750Ti's.

On this forum and everyone would say "spend more and buy a used 7970/290 off ebay" and then argue for 10 pages about the size, power and it being second hand.

 

[Qwertilot]

That tendency was really, really odd yes I guess the 1060 can also pretty much live in the 'just drop in bracket'. More power of course but it isn't remotely crazy.

 

[T1beriu]

Please excuse the offtopic. But I thought you might find this interesting.

Jon Peddie Research (JPR) has published their new GPU data on the market share.

 

[OatisCampbell]

Yeah...BUT....that only shows us the market as a whole and we all know Intel owns that because 90%+ of PCs come with an Intel cpu and only a small percent have a dedicated GPU.
So those charts make it look like "Gosh NVIDIA and AMD are basically tied in the graphics market".
On an enthusiast forum those figures are meaningless because we all know the above and none of us (save Intel employees) have any reason to care about integrated graphics.
We're discussing discrete cards here. Theoretically AMD will be back in the 30s to 40s by 1/1/2018, but they sure aren't now.

 

[Headfoot]

I doubt either Newegg or Amazon is faking their list.

No one is saying the list is fake. What I'm saying is no one knows what criteria puts you on the list. You're assuming it's number of sales over some undetermined period. I say it's much, much more complicated than that.

It doesn't matter who's right, what matters is you can't rely on data that you have no idea how it was collected/generated. Basic scientific method.

 

[cbn]

"Top Seller" is even more worthless than Steam data for determining market share. It's entirely unclear how something makes the "best seller list" and there's an enormously huge chance they bump things to that list IN ORDER for it to sell more, not based on marketshare.

I doubt either Newegg or Amazon is faking their list.

No one is saying the list is fake. What I'm saying is no one knows what criteria puts you on the list. You're assuming it's number of sales over some undetermined period. I say it's much, much more complicated than that.

It doesn't matter who's right, what matters is you can't rely on data that you have no idea how it was collected/generated. Basic scientific method.

Yes, I do think it is just sales.

I don't think Amazon or Newegg is altering the ranking to make a particular brand and model of video card appear to more popular than it is.

 

[tential]

Yes, I do think it is just sales.

I don't think Amazon or Newegg is altering the ranking to make a particular brand and model of video card appear to more popular than it is.

Did you even read the post you quoted?

It doesn't matter what you think it is. You don't know. You don't know the time period or how amazon or newegg calculates top sellers so you can't use that as data.

 

[cbn]

Did you even read the post you quoted?

Yes, I did. Here is what Headfoot wrote:

No one is saying the list is fake. What I'm saying is no one knows what criteria puts you on the list. You're assuming it's number of sales over some undetermined period. I say it's much, much more complicated than that.

It doesn't matter who's right, what matters is you can't rely on data that you have no idea how it was collected/generated. Basic scientific method.

And yes I am assuming it is sales over some undetermined period.

You don't know the time period or how amazon or newegg calculates top sellers so you can't use that as data.

I don't know what time span is used for calculating best sellers. Maybe they go back as far as three months (or whatever) in order to keep very old sales from affecting the rankings of newly introduced cards too much?

So I agree I don't know the exact methodology.

But with that noted, I certainly don't believe the following about Amazon or Newegg:

"Top Seller" is even more worthless than Steam data for determining market share. It's entirely unclear how something makes the "best seller list" and there's an enormously huge chance they bump things to that list IN ORDER for it to sell more, not based on marketshare.

 

[tential]

I don't know what time span is used for calculating best sellers. Maybe they go back as far as three months (or whatever) in order to keep very old sales from affecting the rankings of newly introduced cards too much?

So I agree I don't know the exact methodology.

But with that noted, I certainly don't believe the following about Amazon or Newegg:

Well on the note of the bolded, that's speculation. I don't believe they would do that to boost a brand. They would do it if they believed it would boost sales.

Having created my own algorithm to do this, the primary reason I did this was... To boost sales!

Unless this algorithm is done well, it's a self fulfill prophecy. You could potentially create a top selling list that people then buy what's on it giving you further aim to manipulate it (adding price floors for products that can be included, etc.).

I'm speaking from experience when I say that you could purposely or accidently use your top sales list to promote sales.

But there isnt much of a benefit in pushing people to products they don't want. The point of using a top seller algorithm is to hopefully convert more sales.

So to sum my ramble up,
Top seller lists are used as a way to boost conversion ratios, and they are successful at this.
Doesn't mean they are gamed purposely, but no matter what the intention is there is human motivation behind it and its not perfect.

 

[cbn]

Tential,

The customer buys an item.....then the algorithm ranks the item based on whatever time span is set (eg, items bought in the last three months)

A person at Amazon or Newegg is not artifically changing the best seller rank in order to sell whatever item they want to sell the most of.

Now regarding the idea of "best selling lists" boosting sales, I don't doubt that occurs to some degree.

If a person looks at the best selling list and bases decisions to buy video cards based on the best selling list.....then yes, it could further boost sales of those video cards on the best selling list.

But I just don't think there is any artificial manipulation happening here with those best selling lists. If an item is the 7th best selling video card on a list.....it really is the 7th best selling video card on the list.

 

[tential]

Tential,

The customer buys an item.....then the algorithm ranks the item based on whatever time span is set (eg, items bought in the last three months)

A person at Amazon or Newegg is not artifically changing the best seller rank in order to sell whatever item they want to sell the most of.

Now regarding the idea of "best selling lists" boosting sales, I don't doubt that occurs to some degree.

If a person looks at the best selling list and bases decisions to buy video cards based on the best selling list.....then yes, it could further boost sales of those video cards on the best selling list.

But I just don't think there is any artificial manipulation happening here with those best selling lists. If an item is the 7th best selling video card on a list.....it really is the 7th best selling video card on the list.

I think we're both agreeing. I'm just giving ways the algorithm can be altered and how it does effect sales.

I don't think putting an item in the top sales list would have a an effect. I don't have the data though since that was the next thing I was testing before I left that job.

To be clear, my list was called "featured" and not top sellers, but I was in process of testing to see if customers would favor the top sellers over random products when put into a "featured" list.

Honestly the largest reason I doubt any product is specifically placed into the top seller list is its hard work.... I could barely get development team to keep up with all my ideas, let alone breaking code to insert my favorite product then fixing it when you're done. Too much work....