Hardware/Software Survey needs overhaul.
Does this thing even update anymore? I have yet to see one pop up in like 4-5 years.
So if people get to update their HW for steam 2 times in a decade, can this statistics shown there even be taken seriously to reflect any reality? It's like at least 3-4 years old information, no?

How about allowing us to actually update manually (or at least trigger manually the update) our HW survey once per month? And/or add it to our profile and when it's not anymore up to date, refresh it?


Also to the survey itself. Don't you think it's here and there a bit outdated and insufficient?
----------------------------------------------------------------------------------------------------------

For example "Network speed"?:
Unspecified 99.97%+0.01%
10000.0 Kbps 0.01%-0.01%
33.6 Kbps 0.00%0.00%
1024.0 Kbps 0.00%0.00%
768.0 Kbps 0.00%0.00%
2048.0 Kbps 0.01%0.00%
112.0 Kbps 0.00%0.00%
56.0 Kbps 0.00%0.00%
256.0 Kbps 0.00%0.00%

Who even has 56kbps bandwidth in 2019? I mean like really? People have fiber internet in their homes today with speeds up to several gigabits per second. I have had 300M internet for 7 years now with upgrade option up to 1G for example.

Also you have a typo there, it's supposed to be 'kbps' not 'Kbps' as K would represent Kelvin, which has nothing to do with network bandwidth nor speed. https://en.wikipedia.org/wiki/International_System_of_Units#Prefixes

Also don't confuse "speed" with "bandwidth". Network speed is measured between my computer and lets say game server in milliseconds, "bandwidth" is the amount of data I can upload/download through my internet "pipe" in amount of time (second), which reflects in download speeds, but download speed (so you can say "I downloaded this 10 gigabit game in 1 minute") is not network speed. So either call it "Download speed" or "Network bandwidth". Network/connection speed/lag you measure in milliseconds, not bits per second.

----------------------------------------------------------------------------------------------------------

Why are linux distributions limited to top 5? Give us 10 or 15 :-)
And why do we even care about distribution subversions? It's one line for Ubuntu 18 enough and maybe add another for 19?

Ubuntu 18.04.2 LTS 64 bit 24.57% +1.28%
Linux 64 bit 8.96% +8.96%
Ubuntu 18.10 64 bit 8.22% -0.35%
"Manjaro Linux" 64 bit 7.56% +0.68%
Linux Mint 19.1 Tessa 64 bit 6.70% +6.70%
Other 43.99% -8.20%

This "Other" really needs to be divided so there isn't more than 5% others. And different versions or same distro should be grouped together, like Ubuntu v18.4 and 18.10 into one "Ubuntu Linux", exactly like there is one "Manjaro Linux". What exactly is this "Linux 64bit" - sounds pretty much same as "Other".

I really would like to see in this list also other major flavours, like Fedora, Debian, maybe Pop_OS!, Arch etc. The current short list of Ubuntu, Manjaro and Mint is slightly underwhelming. Top 10 please! :-)

----------------------------------------------------------------------------------------------------------

"Physical CPUs" : 4 CPU's 54.96%

Really? I bet there is 99.99% 1 (ONE) physical cpu per computer. With several cores - so lets start counting cores instead of cpus.

----------------------------------------------------------------------------------------------------------

VR headsets - how are those even counted/measured? (never had the honor to answer this survey question before, because after I got Rift, the survey has never shown up) Do it only see devices currently connected to computer or what? I mean I physically disconnect OR disable the service for my Rift every time I don't use it, just because by default it is always active and heats up the device and takes power even when not using it. Does it even count it's presence then? How about when the survey pops in a computer where I have not used it (from my 3 different computers where I have Steam installed) - BUT I STILL OWN the unit and can theoretically use it on whichever computer I want to; but because of the wall mounted stations I prefer not to move it around) ?
Legutóbb szerkesztette: deemon; 2019. máj. 17., 5:24
< >
111/11 megjegyzés mutatása
Ten percent of people get queried each month. You not getting asked for that period of time is well within the range of statistical probability.

As for the network speeds, outdated but also irrelevant these days. The lowest common denominator is already more than good enough for multiplayer.
Washell eredeti hozzászólása:
As for the network speeds, outdated but also irrelevant these days. The lowest common denominator is already more than good enough for multiplayer.

Mayby they want it for other statistics? How long it will take for people to download those new terabyte games with several gigabytes of updates weekly?
Legutóbb szerkesztette: deemon; 2019. máj. 11., 4:37
deemon eredeti hozzászólása:
Does this thing even update anymore? I have yet to see one pop up in like 4-5 years.
So if people get to update their HW for steam 2 times in a decade, can this statistics shown there even be taken seriously to reflect any reality? It's like at least 3-4 years old information, no?
Yes it does, but its at random.


deemon eredeti hozzászólása:
How about allowing us to actually update manually (or at least trigger manually the update) our HW survey once per month? And/or add it to our profile and when it's not anymore up to date, refresh it?
I'm not sure if allowing people to activate it would mess with their numbers or not. But if it wouldn't I don't see an issue with allowing people to activate it once a year on their own.


deemon eredeti hozzászólása:
Also to the survey itself. Don't you think it's here and there a bit outdated and insufficient?
----------------------------------------------------------------------------------------------------------

For example "Network speed"?:
Unspecified 99.97%+0.01%
10000.0 Kbps 0.01%-0.01%
33.6 Kbps 0.00%0.00%
1024.0 Kbps 0.00%0.00%
768.0 Kbps 0.00%0.00%
2048.0 Kbps 0.01%0.00%
112.0 Kbps 0.00%0.00%
56.0 Kbps 0.00%0.00%
256.0 Kbps 0.00%0.00%

Who even has 56kbps bandwidth in 2019? I mean like really? People have fiber internet in their homes today with speeds up to several gigabits per second. I have had 300M internet for 7 years now with upgrade option up to 1G for example.

You would actually be surprised to know that even in the US there are many who can't get anything but dial up because high speed internet companies don't want to do their jobs and put 10s of thousands of dollars worth of equipment out in the middle of no where to service 1 to 5 people over a 10 or so mile area.

You should be thankful you have the option of an upgrade, many people don't have any choice over what they get, and when that happens, they get stuck with expensive, super crappy "high speed" internet that doesn't even break 1 megabit download and their upload is almost not even worth it. And in some cases this is in the middle of a city like New York. I'd link to a recent video by a guy on you tube but there is swearing in it. Search for Louis Rossmann, and look for one of his latest posts where he rants about his crappy internet in the middle of New York.


deemon eredeti hozzászólása:
Also you have a typo there, it's supposed to be 'kbps' not 'Kbps' as K would represent Kelvin, which has nothing to do with network bandwidth nor speed. https://en.wikipedia.org/wiki/International_System_of_Units#Prefixes
Actually Kbps is correct. This is usually why I spell things out to people though cause its less confusing to people, many who don't follow tech that much don't see the difference between Kbsp, KBps, mbps, MBps and so on.
https://www.google.ca/search?hl=en&q=kbps
https://searchnetworking.techtarget.com/definition/Kbps
https://techterms.com/definition/kbps


deemon eredeti hozzászólása:
Also don't confuse "speed" with "bandwidth". Network speed is measured between my computer and lets say game server in milliseconds, "bandwidth" is the amount of data I can upload/download through my internet "pipe" in amount of time (second), which reflects in download speeds, but download speed (so you can say "I downloaded this 10 gigabit game in 1 minute") is not network speed. So either call it "Download speed" or "Network bandwidth". Network speed you measure in milliseconds, not bytes per second.
Many people look at network speed as how fast you can download, on most if not all high speed sale pages it usually says speed. I use speed for how fast, and bandwidth with how much you can download in a month. I ask people how fast their ping in between servers. That helps keep the confusion down.


deemon eredeti hozzászólása:
Why are linux distributions limited to top 5? Give us 10 or 15 :-)
And why do we even care about distribution subversions? It's one line for Ubuntu 18 enough and maybe add another for 19?

Ubuntu 18.04.2 LTS 64 bit 24.57% +1.28%
Linux 64 bit 8.96% +8.96%
Ubuntu 18.10 64 bit 8.22% -0.35%
"Manjaro Linux" 64 bit 7.56% +0.68%
Linux Mint 19.1 Tessa 64 bit 6.70% +6.70%
Other 43.99% -8.20%

This "Other" really needs to be divided so there isn't more than 5% others. And different versions or same distro should be grouped together, like Ubuntu v18.4 and 18.10 into one "Ubuntu Linux", exactly like there is one "Manjaro Linux". What exactly is this "Linux 64bit" - sounds pretty much same as "Other".
Because there are like 100+ different versions and they have to make the cut off somewhere. Also the majority of them are not great for gaming so why bother naming them. And then you start getting into the different version numbers of each one... really its kind of a mess. They should just lump all one kind of linux into one, and maybe make a pull down to see each version number break down.


deemon eredeti hozzászólása:
"Physical CPUs" : 4 CPU's 54.96%

Really? I bet there is 99.99% 1 (ONE) physical cpu per computer. With several cores - so lets start counting cores instead of cpus.
You're just nit picking now.



deemon eredeti hozzászólása:
VR headsets - how are those even counted/measured? (never had the honor to answer this survey question before, because after I got Rift, the survey has never shown up) Do it only see devices currently connected to computer or what? I mean I physically disconnect OR disable the service for my Rift every time I don't use it, just because by default it is always active and heats up the device and takes power even when not using it. Does it even count it's presence then? How about when the survey pops in a computer where I have not used it (from my 3 different computers where I have Steam installed) - BUT I STILL OWN the unit and can theoretically use it on whichever computer I want to; but because of the wall mounted stations I prefer not to move it around) ?

Yes it only shows headsets that are connect to the system at the time the survey is done. I don't remember if it does this for controllers too. But its the same for hard drives and everything else. I have 5 hard drives, only 1 is usually connected to my system all the time.

As for VR headsets, I could see if there was some sort of switch that was flipped when you have the hardware and software needed for it installed. Then after say 6 months if you have not plugged it in once since or you uninstall the software then the switch gets turned off.

Maybe even have the hardware survey ask if you have VR headset and game controllers and asks you to plug them in for the survey.

But there is also the issue with what you pointed out, you have 3 different computers, what if someone plugs in the single VR headset for all 3 system and then trigger the survey or plug it in when the survey asks. I guess they could provide a unique serial number for the headsets and if it detects that same headset on more than 1 system it ignores it for those other systems beyond the first one.
Gwarsbane eredeti hozzászólása:

deemon eredeti hozzászólása:
Also you have a typo there, it's supposed to be 'kbps' not 'Kbps' as K would represent Kelvin, which has nothing to do with network bandwidth nor speed. https://en.wikipedia.org/wiki/International_System_of_Units#Prefixes
Actually Kbps is correct. This is usually why I spell things out to people though cause its less confusing to people, many who don't follow tech that much don't see the difference between Kbsp, KBps, mbps, MBps and so on.
https://www.google.ca/search?hl=en&q=kbps
https://searchnetworking.techtarget.com/definition/Kbps
https://techterms.com/definition/kbps

Nope, you are wrong. And so are all those links you you gave and it's mindboggling how much such misinformation is spread in the Internet.

https://en.wikipedia.org/wiki/Kilo-

quote: "It is used in the International System of Units where it has the unit symbol k, in lower case." Just because you 'muricans like to capitalize everything doesn't make it right.

https://en.wikipedia.org/wiki/Data-rate_units#Decimal_multiples_of_bits

Gwarsbane eredeti hozzászólása:
Because there are like 100+ different versions and they have to make the cut off somewhere. Also the majority of them are not great for gaming so why bother naming them. And then you start getting into the different version numbers of each one... really its kind of a mess. They should just lump all one kind of linux into one, and maybe make a pull down to see each version number break down.

I didn't ask them to post all. Just instead of top 5 a little bit more. Like 10 for example. Would be much more helpful than this list for Mac:
MacOS 10.14.3 64 bit 24.09% -14.62%
MacOS 10.14.4 64 bit 23.29% +23.29%
MacOS 10.13.6 64 bit 19.36% -0.58%
MacOS 10.12.6 64 bit 7.69% -0.28%
MacOS 10.14.2 64 bit 5.67% -4.03%
MacOS 10.14.0 64 bit 4.93% -1.13%
MacOS 10.11.6 64 bit 4.87% -0.24%
MacOS 10.14.1 64 bit 2.59% -0.99%
MacOS 10.10.5 64 bit 1.95% -0.07%
Other 5.57% -1.33%

Where they actually have 10 (9+Other). I mean, does anybody actually care how many people use MacOS 10.14.2 compared to 10.14.1? It's like entirely and utterly pointless information (as most probably you can't even downgrade there but just upgrade), compared to different linux distributions to see which ones are more popular and trending.

Gwarsbane eredeti hozzászólása:
As for VR headsets, I could see if there was some sort of switch that was flipped when you have the hardware and software needed for it installed. Then after say 6 months if you have not plugged it in once since or you uninstall the software then the switch gets turned off.

Maybe even have the hardware survey ask if you have VR headset and game controllers and asks you to plug them in for the survey.

But there is also the issue with what you pointed out, you have 3 different computers, what if someone plugs in the single VR headset for all 3 system and then trigger the survey or plug it in when the survey asks. I guess they could provide a unique serial number for the headsets and if it detects that same headset on more than 1 system it ignores it for those other systems beyond the first one.

Also not good. Maybe you sold or gifted your headset and new owner can't fill then HW survey correctly because it's attached to your old steam? Where it's already overwritten by your new Valve Index maybe?

Anyway I use same steam account in all of those 3 computers, so having the option to set one(or several) VR headset per account would solve the problem. If the same serial headset is associated with the new account, it's removed from previous one...

Same thing with controllers. They are not glued to one computer.


In entirely differnet topic (or account vs computers topic) - I do not play at all in all 3 of them, just 2 (Windows and Manjaro). In one old laptop I just have it installed to buy games and read discussions etc. So survey popping up in this one would be entirely wasted, because in all reality I would even decline to participate there because it would not reflect the reality of my "gaming PC" status.
Legutóbb szerkesztette: deemon; 2019. máj. 11., 8:41
deemon eredeti hozzászólása:
Nope, you are wrong. And so are all those links you you gave and it's mindboggling how much such misinformation is spread in the Internet.

https://en.wikipedia.org/wiki/Kilo-

quote: "It is used in the International System of Units where it has the unit symbol k, in lower case." Just because you 'muricans like to capitalize everything doesn't make it right.

https://en.wikipedia.org/wiki/Data-rate_units#Decimal_multiples_of_bits

From the link you provided:

For the kilobyte, a second definition has been in common use in some fields of computer science and information technology. It uses kilobyte to mean 210 bytes (= 1024 bytes), because of the mathematical coincidence that 210 is approximately 103. The reason for this application is that digital hardware and architectures natively use base 2 exponentiation, and not decimal systems. JEDEC memory standards still permit this definition, but acknowledge the correct SI usage.

https://en.wikipedia.org/wiki/JEDEC_memory_standards
The specification cites three prefixes as follows:

kilo (K): A multiplier equal to 1024 (210).
mega (M): A multiplier equal to 1,048,576 (220 or K2, where K = 1024).
giga (G): A multiplier equal to 1,073,741,824 (230 or K3, where K = 1024).

When dealing with computers, the standards dictate a capital K, not a lower case one.

This has become common practice in IT and with computer when writing in regards to data, such as internet speeds, computer RAM, computer storage, ect.
Legutóbb szerkesztette: Spawn of Totoro; 2019. máj. 11., 9:42
Spawn of Totoro eredeti hozzászólása:
https://en.wikipedia.org/wiki/JEDEC_memory_standards
When dealing with computers, the standards dictate a capital K, not a lower case one.
This has become common practice in IT and with computer when writing in regards to data, such as internet speeds, computer RAM, computer storage, ect.

No, it's not "common practice in IT" - rather a niche usage only for JEDEC and memory. And we're not talking about any memory standards and size here, but network bandwidth. Let's stick to this. I link once more:
https://en.wikipedia.org/wiki/Data-rate_units#Kilobit_per_second

We don't even use the "base 2 exponentiation" here, but kilobit used for data rate measurement is 1000 bits = 125 bytes, not some 1024 bits.

read those:
https://en.wikipedia.org/wiki/Bit_rate
The bit rate is quantified using the bits per second unit (symbol: "bit/s"), often in conjunction with an SI prefix such as "kilo" (1 kbit/s = 1,000 bit/s), "mega" (1 Mbit/s = 1,000 kbit/s), "giga" (1 Gbit/s = 1,000 Mbit/s) or "tera" (1 Tbit/s = 1000 Gbit/s).[2] The non-standard abbreviation "bps" is often used to replace the standard symbol "bit/s", so that, for example, "1 Mbps" is used to mean one million bits per second.

https://en.wikipedia.org/wiki/Kilobit

The kilobit is a multiple of the unit bit for digital information or computer storage. The prefix kilo- (symbol k) is defined in the International System of Units (SI) as a multiplier of 103 (1 thousand),[1] and therefore,

1 kilobit = 103bits = 1000 bits.

The kilobit has the unit symbol kbit or kb.

Using the common byte size of 8 bits, 1 kbit is equal to 125 bytes.

The kilobit is commonly used in the expression of data rates of digital communication circuits as kilobits per second (kbit/s or kb/s), or abbreviated as kbps,[2] as in, for example, a 56 kbps PSTN circuit, or a 512 kbit/s broadband Internet connection.

https://en.wikipedia.org/wiki/Measuring_network_throughput
The throughput of communications links is measured in bits per second (bit/s), kilobits per second (kbit/s), megabits per second (Mbit/s) and gigabits per second (Gbit/s). In this application, kilo, mega and giga are the standard S.I. prefixes indicating multiplication by 1,000 (kilo), 1,000,000 (mega), and 1,000,000,000 (giga).
Legutóbb szerkesztette: deemon; 2019. máj. 15., 0:39
deemon eredeti hozzászólása:
Spawn of Totoro eredeti hozzászólása:
https://en.wikipedia.org/wiki/JEDEC_memory_standards
When dealing with computers, the standards dictate a capital K, not a lower case one.
This has become common practice in IT and with computer when writing in regards to data, such as internet speeds, computer RAM, computer storage, ect.

No, it's not "common practice in IT" - rather a niche usage. And we're not talking about any memory standards here, but network bandwidth here. Let's stick to this.

Is there a difference between Kbps and KBps?

Hint there is an 8-fold difference between those. Capitalisation is important in IT measurements.
deemon eredeti hozzászólása:
Spawn of Totoro eredeti hozzászólása:
https://en.wikipedia.org/wiki/JEDEC_memory_standards
When dealing with computers, the standards dictate a capital K, not a lower case one.
This has become common practice in IT and with computer when writing in regards to data, such as internet speeds, computer RAM, computer storage, ect.

No, it's not "common practice in IT" - rather a niche usage only for JEDEC and memory. And we're not talking about any memory standards and size here, but network bandwidth. Let's stick to this. I link once more:

With over 20+ year experience myself, not to mention the combined 300+ years experience of those I work with and those I know who work in IT, yes, it is common practice to use a capital K.

That is far more then a "niche" usage. No one any one us have know has ever used a lowercase K. One reason being it looks out of place when next to the uppercase M and G.

So, again, it is common practice in IT to use an uppercase K over a lowercase k.

Darren eredeti hozzászólása:
deemon eredeti hozzászólása:

No, it's not "common practice in IT" - rather a niche usage. And we're not talking about any memory standards here, but network bandwidth here. Let's stick to this.

Is there a difference between Kbps and KBps?

Hint there is an 8-fold difference between those. Capitalisation is important in IT measurements.

His argument is that it should be kbps and kBps, with a lowercase k. Nothing to do with the B being upper or lowercase.
How about allowing us to actually update manually (or at least trigger manually the update) our HW survey once per month? And/or add it to our profile and when it's not anymore up to date, refresh it?

Manual updates aren't a great idea. When you're gathering statistics, if you pick your sample at random from the population then it gives you an unbiased estimate of the population, and will tend towards being perfectly accurate as the sample size increases. Unless they've changed methodology, Steam does the random sampling by splitting the user-base into 12ths at random, and asking each 12th to do the survey on a particular month, so you should get asked once a year.

If you let people self-select by manually updating, your sample is no longer random, and will produce results which are biased in some way. It's bad enough for the statistics that you have to say yes to your information being included in the survey (although that is obviously necessary for privacy reasons), never mind manually submitting to the survey.

As regards the k/K debate; to throw a spanner into the works, what we should be using to denote a unit which has 1024 bits is 1 Kibit (kibibit) and 1024 bytes is 1 KiB (kibibyte). I suppose alternatively, we could use shannons instead of bits (as 1 bit = 1 Sh) or perhaps nats, where 1 nat is approximately 1.44 bits. I'm downloading at 5.545*10^5 nats/s, we might say.
Spawn of Totoro eredeti hozzászólása:
With over 20+ year experience myself, not to mention the combined 300+ years experience of those I work with and those I know who work in IT, yes, it is common practice to use a capital K.

That is far more then a "niche" usage. No one any one us have know has ever used a lowercase K. One reason being it looks out of place when next to the uppercase M and G.

So, again, it is common practice in IT to use an uppercase K over a lowercase k.

Not in my 35+ years of IT experience (I don't even bother including here "additional other people to prove my point and try to make myself more right this way"). I have NEVER used uppercase K over a lowercase k and never seen anyone around here use it either. Must be then regional thing.

However I quite often run into amateur articles in the interwebs that use K over k, written by journalist, not by IT people.
Legutóbb szerkesztette: deemon; 2019. máj. 16., 1:22
WOW so good
< >
111/11 megjegyzés mutatása
Laponként: 1530 50

Közzétéve: 2019. máj. 11., 3:25
Hozzászólások: 11