• Akira Toriyama passed away

    Let's all commemorate together his legendary work and his impact here

Ampere Sell Through Estimates "Xbox Series S outselling Series X" PS5 - 12.8 mil, XSX - 6.7 mil, NSW - 89.7

S

Star Ocean

Banned
Student
User requested account closure



The data firm's Piers Harding-Rolls notes that PS5 supply improved significantly during the July to September window, which saw the PlayStation console outsell Nintendo Switch across Europe during that time. PS5 also managed to outsell Nintendo's console in September in the US, according to NPD.

However, he notes that Switch is likely to reclaim its place over Christmas thanks to the launch of the new OLED edition, improved supply and the fact Nintendo tends to perform very well over the Christmas period.
Click to expand...

www.gamesindustry.biz

Ampere: Xbox Series S outselling Series X in key markets

Data analysts Ampere says that Xbox Series S is out-selling Series X across several key markets.GamesIndustry.biz has a…
www.gamesindustry.biz www.gamesindustry.biz
 
Nintendo reported in their last quarterly reports that Switch sell-through surpassed 90m units in October, so the estimation is not far off for Switch total at least.
 
According to GSD data, over 90% of sales of Call of Duty: Vanguard on Xbox came as a digital download
Click to expand...

That's the only "tracking data" and not just analysts information in the article.
I also would like to mention and explain why that number differs form the math estimate done here, that puts Xbox digital share at 81%

This is GSD only and the math was done with GSD figures and GFK retail data in combination.
GSD and GFK retail tracking is quite a bit different.
 
Well, looks like Sony ordered FAR more chips for PS5 at TSMC than MS for Xbox. Like it was last year. By a double amount
 
The data firm's Piers Harding-Rolls notes that PS5 supply improved significantly during the July to September window, which saw the PlayStation console outsell Nintendo Switch across Europe during that time.
Click to expand...
I guess Switch will sell more during the holiday season.
 
SoonyXboneUhh said:
That's the only "tracking data" and not just analysts information in the article.
I also would like to mention and explain why that number differs form the math estimate done here, that puts Xbox digital share at 81%

This is GSD only and the math was done with GSD figures and GFK retail data in combination.
GSD and GFK retail tracking is quite a bit different.
Click to expand...
It is quite an interesting difference although there's lots of rounding involved too.

How reliable are Ampere's estimates? XBS is about a million lower than we expected here.
 
EDGalTBR said:
I'm not buying this. It looks like some random numbers like VG Chartz.
Click to expand...

Ampere has access to NPD, GSD and Media Create last I checked, so they're able to create estimates with integrity.

Worth noting that sell through and sell in numbers will always be hard to compare 1:1.
 
Numbers are sell thru but I’m guessing it’s missing a few markets/retailers since NSW is off by a bit (well I’m guessing since it’s again suffering from supply issues?)

Edit: like ZhugeEX said, not exactly 1:1 but close enough is good in my book.
 
ZhugeEX said:
Ampere has access to NPD, GSD and Media Create last I checked, so they're able to create estimates with integrity.

Worth noting that sell through and sell in numbers will always be hard to compare 1:1.
Click to expand...

Worldwide sales estimates are also notoriously challenging to compile. I would take any worldwide estimate with a grain of salt regardless of the analytics firm, unless their data source originates from the company they're tracking.

There's just too much margin of error to be definitive, even if they receive direct POS data from various regions.
 
Last edited:
The switch and PS figures make sense. And I suppose it is feasible that there's ~1.3m unsold series S and X, with most of that being the S which is in good supply in most countries and 1.3m is the about the normal gap we see for a console in good stock.
 
What exactly is the issue with the estimates? They seem reasonable and most likely based on actual sales data from most major markets. Two prominent sales analyst who's data and estimates are normally accepted here and elsewhere have also reported the data at face value. Just trying to figure out where the disconnect is when discounting this information.
 
gtj1982 said:
What exactly is the issue with the estimates? They seem reasonable and most likely based on actual sales data from most major markets. Two prominent sales analyst who's data and estimates are normally accepted here and elsewhere have also reported the data at face value. Just trying to figure out where the disconnect is when discounting this information.
Click to expand...
Nothing is wrong as long as you assume large error margins of up to +/- 1 million attached to each data point. It's when you treat these estimates as gospel that you get in trouble. Like I said, accept it, but also take it with a grain of salt.

This goes for all analytics firms that project out worldwide data. It is crucial to understand the scope and the limitations of the estimates that are provided. Also important to remember that all market research firms---even NPD---publish estimates. What varies is the degree of accuracy between them.
 
Nintendo Switch Sell-through as September 30 2021 :

Nintendo : >90M
Ampère : 89,7M
Vgchartz (as october 2) : 91,62M
 
Using numbers from the same source (unless adjustments have been made) overall 2021 sales as of September would be:
Switch - 13.96 million
PS5 - 8.22 million
XBS - 4.02 million

Aquamarine said:
Nothing is wrong as long as you assume large error margins of up to +/- 1 million attached to each data point. It's when you treat these estimates as gospel that you get in trouble. Like I said, accept it, but also take it with a grain of salt.

This goes for all analytics firms that project out worldwide data. It is crucial to understand the scope and the limitations of the estimates that are provided. Also important to remember that all market research firms---even NPD---publish estimates. What varies is the degree of accuracy between them.
Click to expand...
And yeah those numbers should be used to have a general idea of the current trend but of course when we talk about worldwide sales, even counting 3rd world markets, it's very easy to be slight off.
 
The sell through estimate for Q3 does track with my estimate of ~1.8M shipments since that's around when Series S started being more readily available in the US and shipments would also have some stock for October. Series S was already available in other parts of the world since Q2.
 
Aquamarine said:
Nothing is wrong as long as you assume large error margins of up to +/- 1 million attached to each data point. It's when you treat these estimates as gospel that you get in trouble. Like I said, accept it, but also take it with a grain of salt.

This goes for all analytics firms that project out worldwide data. It is crucial to understand the scope and the limitations of the estimates that are provided. Also important to remember that all market research firms---even NPD---publish estimates. What varies is the degree of accuracy between them.
Click to expand...

So don't trust any data is what I'm getting from this. We know the only official data is shipments and even how that is counted can vary from company to company and in the case of Sony there was a shift in reporting eternally between generations. If the idea is to take everything with a grain of salt why do we have NPD and Media Create threads each week and no one rushes in to say these are just estimates take with a grain of salt?
 
gtj1982 said:
So don't trust any data is what I'm getting from this. We know the only official data is shipments and even how that is counted can vary from company to company and in the case of Sony there was a shift in reporting eternally between generations. If the idea is to take everything with a grain of salt why do we have NPD and Media Create threads each week and no one rushes in to say these are just estimates take with a grain of salt?
Click to expand...
Because Media Create, NPD, GfK, GSD, Famitsu, etc. are established market research firms with known error margins (for example, NPD directly tracks 90% of the market and upweights the remaining 10%). Ampere Analysis is an opaque startup with no explanation about their methodology. Ampere is also projecting worldwide sales, which are notoriously challenging to compile.

Hence the initial doubt and the caveats behind the numbers. Everyone already knows the limitations of established firms.
 
gtj1982 said:
So don't trust any data is what I'm getting from this. We know the only official data is shipments and even how that is counted can vary from company to company and in the case of Sony there was a shift in reporting eternally between generations. If the idea is to take everything with a grain of salt why do we have NPD and Media Create threads each week and no one rushes in to say these are just estimates take with a grain of salt?
Click to expand...
It's very clear what is said, these numbers are based on actual data but also include estimates and have some error of margin. There are too many countries around the world without any tracking system.
 
master94ga said:
Keep in mind that some (don't know the numbers) of the Series X chips are used for xcloud
Click to expand...
That could very well explain why Series S is more available than Series X in demand-limited markets like the USA. Ampere is right on the nose with this.
 
Ryng™ said:
Using numbers from the same source (unless adjustments have been made) overall 2021 sales as of September would be:
Switch - 13.96 million
PS5 - 8.22 million
XBS - 4.02 million


And yeah those numbers should be used to have a general idea of the current trend but of course when we talk about worldwide sales, even counting 3rd world markets, it's very easy to be slight off.
Click to expand...

That Switch number is hefty. 2M/4M/7M for Oct/Nov/Dec could get sell through up to ~27M for the year. Do we have a good non-VGC estimate of what it was last year as a comp?

2021 Oct 28 day period was around 1.15M from Famitsu/NPD alone I believe?
 
Last edited:
master94ga said:
Keep in mind that some (don't know the numbers) of the Series X chips are used for xcloud
Click to expand...

Also keep in mind that was few months ago. MS surely filled their servers with XSX for Xcloud, just for that purpose . They are not doing it endlessly. There are not milions and millions XSX for that purpose, couple ten of thousands surely

@Aquamarine

That could very well explain why Series S is more available than Series X in demand-limited markets like the USA. Ampere is right on the nose with this.
Click to expand...

XSS chip is smaller, IMO, MS can produce more XSS chips per wafer than for XSX.
 
Last edited:
The gap between sell in and sell through is 600k for PS5 and ~1.3m for Xbox Series.

Even considering Ampere data and Xbox Series shipments aren't 100% accurate, it's fair to say that Xbox is losing sales by not having more Series X available.
 
Salesman said:
Also keep in mind that was few months ago. MS surely filled their servers with XSX for Xcloud. They are not doing it endlessly. There are not milions and millions XSX for that purpose, couple ten of thousands surely.
Click to expand...
xCloud is part of Microsoft's core growth strategy. They're planning on making Xbox ubiquitous to all consumers through streaming high-fidelity Series X performance on any smartphone and any device with a web browser worldwide.

Microsoft has been upgrading ALL of their old Xbox One S server blades worldwide with Xbox Series X (eight in one blade) to prep for significant xCloud growth within the next five years.

I have no idea how many of Microsoft's 4 million Azure servers were prepped with Series X, but it doesn't make much sense to only be a couple of ten thousands.

Rather, I can see them temporarily de-emphasizing their initial consumer production lines for Series X because of this and the component shortages unique to Series X. They then restarted the line for the Halo bundle now that this comprehensive upgrade is complete.
 
Aquamarine said:
I have no idea how many of Microsoft's 4 million Azure servers were prepped with Series X, but it doesn't make much sense to only be a couple of ten thousands.
Click to expand...

It makes sense if Xcloud is just for gaming using the XSX as server blades. Also, vast majority of Azure servers are for Virtual Computing, Storage, Database, Networking and what not. XSX as server blades surely aren't used for these purposes.
 
Last edited:
Salesman said:
It makes sense if Xcloud is just for gaming using the XSX as server blades. Also, vast majority of Azure servers are for Virtual Computing, Storage, Database, Networking and what not. XSX as server blades surely aren't used for these purposes
Click to expand...
The compute cloud architecture of their xCloud streaming service is dependent on the server blades they add to the chassis, which are then integrated with the rest of Azure (Storage / Computer / App Service / etc.) It's an integrated add-on that works in tandem with the other services, essentially. Very smart design and allows them to scale easily.

It doesn't make sense to launch / replace all the old Xbox One S blades with that small of a capacity but only Microsoft knows the answer. Anything we say here is just pure speculation.
 
Salesman said:
Xss is selling a bit slower. It is easy to find, really
Click to expand...
It honestly blows my mind how easy the XSS is to find. The XSX is still so hard and anytime you try to buy it, it’s gone within seconds. It’s nice to see gamers constantly going for the better models, even if they’re pricier!
 
Salesman said:
Also keep in mind that was few months ago. MS surely filled their servers with XSX for Xcloud, just for that purpose . They are not doing it endlessly.
Click to expand...
They are expanding xCloud to new contries month after month. However as you say I don't think console used for xCloud are millions, probably just thousands.
 
Salesman said:
Maybe per wafer you can get more S consoles, i think. I dunno
Click to expand...
Definitely this. The die size of Series X is ~360mm² which is ~164 dies per wafer in the best case. The die size of Series S is ~197mm² which is around ~305 dies per wafer in the best case. One would need to know the exact dimensions as well as the current yield to know the exakt number of actually usable dies per wafer, but in any case that's definitely way way more for Series S.
 
If accurate, this means the vast majority of Game Pass subscribers are still on last gen Xbox and PC/cloud.

In Microsoft's earnings call on Jan 26 of this year, they confirmed 18 mil subs. It's probably safe to assume they're above 18 mil at the end of September of this year.

Best case and impossible scenario is ALL 6.7 mil of those Series owners all have Game Pass. Even in this scenario, Series owners would only account for 37.2% of all Game Pass subs. Factor in that they have to be above 18 mil subs at the end of September of this year, and that 37.2% will be even lower.
 
I thought Xbox was closer to 10 million going off a recent article though I would not surprised if this number is more accurate.
 
Xbox really needs to step up production. This holiday should be huge, but in all likelihood it wont be that spectacular cause of supply constraints.
 
MysticGon said:
I thought Xbox was closer to 10 million going off a recent article though I would not surprised if this number is more accurate.
Click to expand...

Then someone were way off with the numbers. LOL
 
XSS is really popular here in Brazil. Its value with Gamepass is almost unbeatable. X360 was also very popular but that was due to piracy. Things are different now.

But I guess Playstation 4 is still the king and the most desired hardware is Playstation 5, the problem is the expensive price and stock issues.

Out of the big 3, the only irrelevant gaming company around here is Nintendo due to a mix of factors like no game localization, reaaaally expensive prices (especially games), and lack of official representation (they are back now but in a very superficial way).
 
We could just wait until end of January for the next earning release and see if Xbox Series is still the fastest selling Xbox.

Xbox One approached 10m (sold in) in mid November, so 1 year after launch
 
Last edited:
Hi guys, good to see the discussion here.

I’d be surprised if Xbox Series sell-in was at 8m at end of Sept considering our sell-through estimates but I guess it’s possible.

Switch sell-through passed 90m in October according to Nintendo (btw the console companies face the same challenges analysts do compiling global sell-through estimates :)). I think our end of Sept global estimate is pretty bang on. We aggregate our data across 33 country markets, so use a bottom-up approach. We do adjust historical numbers if additional information comes to light that is different to our estimates.

Ampere is a new name in games research but was founded in 2015 by ex. Screen Digest execs. Screen Digest started tracking the console market in the mid 1990s and I led the games research team there from the early 2000s.
 
PiersHR said:
Hi guys, good to see the discussion here.

I’d be surprised if Xbox Series sell-in was at 8m at end of Sept considering our sell-through estimates but I guess it’s possible.

Switch sell-through passed 90m in October according to Nintendo (btw the console companies face the same challenges analysts do compiling global sell-through estimates :)). I think our end of Sept global estimate is pretty bang on. We aggregate our data across 33 country markets, so use a bottom-up approach. We do adjust historical numbers if additional information comes to light that is different to our estimates.

Ampere is a new name in games research but was founded in 2015 by ex. Screen Digest execs. Screen Digest started tracking the console market in the mid 1990s and I led the games research team there from the early 2000s.
Click to expand...
That's cool. So based on your tracking how is Series doing compared to One? Is the "fastest selling Xbox" claim still valid?
 
PiersHR said:
Hi guys, good to see the discussion here.

I’d be surprised if Xbox Series sell-in was at 8m at end of Sept considering our sell-through estimates but I guess it’s possible.

Switch sell-through passed 90m in October according to Nintendo (btw the console companies face the same challenges analysts do compiling global sell-through estimates :)). I think our end of Sept global estimate is pretty bang on. We aggregate our data across 33 country markets, so use a bottom-up approach. We do adjust historical numbers if additional information comes to light that is different to our estimates.

Ampere is a new name in games research but was founded in 2015 by ex. Screen Digest execs. Screen Digest started tracking the console market in the mid 1990s and I led the games research team there from the early 2000s.
Click to expand...
always great to have some insight
 
You must log in or register to reply here.
Back
Top Bottom