Wifi speed on old iPads

Found this https://support.apple.com/en-gb/guide/deployment/dep...

Does this mean 144Mbps is the only limit on Wifi speed?

Some are suggesting that an UL speed of 3Mbps on a 40/10 FTTP connection is down to the devices. How would this work when the linit looks to be over 45 times that?

Very briefly - 144Mbps is a headline marketing figure - only achievable in a perfect environment.

Since wi-fi relies on radio and radio is a shared medium more devices means less bandwidth available per device.

Add in interference from neighbours wi-fi, bluetooth, microwave ovens, passing buses with wi-fi on board.

Factor in the fact that wi-fi signals are attenuated by walls, furniture, bodies, etc.

Factor in that you can have reflections from all sorts of objects.

Good rule of thumb is half the theoretical “headline rate” in actual throughput and you’re doing quite well.

OK so not the advertised figure: I get that but only getting 1/45th is pretty dire, no?

So 1/45th inducates something is wrong?

In reply to a post by hk11:

How would this work when the linit looks to be over 45 times that?

If you are talking about uploads to the Internet, the maximum for your connection is 10Mbps. So, however you are connected to your router, that is the absolute maximum you could get; in practice you will never achieve that. You are getting about 1/3 of that speed, not 1/45.

If you Google "slow wifi iPad" you will find a number of suggestions to improve the speed. But, in my experience, old iPads are notoriously slow and inefficient nowadays. It's almost as if they have built-in obsolescence.

I'm on a 40/10 FTTP connection, so the 10 refers to the upload, not the download. Is the quoted 144 for downloads? If so what is the upload?

Let's say it is 30. So 3 is only a tenth of what is possible, no?

I understand old technology is going to be slower than modern, but I am taking advertised figures, do these change as the hardware ages?

Some are suggesting that an UL speed of 3Mbps on a 40/10 FTTP connection is down to the devices.

I read UL speed to mean "upload speed".

If you are talking about download speed then yo are getting about 1/13 of the theoretical maximum, not 1/45.

You read correctly"

Take your iPad and test it on another connection, ideally faster than your own domestic connection snd it should pinpoint whether it’s your iPad or broadband that’s at the limit / issue.

In reply to a post by Pheasant:

Take your iPad and test it on another connection, ideally faster than your own domestic connection snd it should pinpoint whether it’s your iPad or broadband that’s at the limit / issue.

It may give him an indication but not a cast iron guarantee. So many variables with this OP

I'm getting an 80/20 connection next week so we will see if the UL speed improves.

Then your upload speed is 10. (Your download speed is 40.)

In reply to a post by hk11:

I'm on a 40/10 FTTP connection, so the 10 refers to the upload, not the download. Is the quoted 144 for downloads? If so what is the upload?

WiFi is complicated, and the speeds quoted are not the same as fixed line ethernet. Also WiFi and Ethernet are designed as symmetric (same speed each way) unlike broadband DSL technology. However radio waves are not as predictable as fixed line and all sorts of interference can slow you down either router to tablet, or from tablet to router.

If you have an iPad Air 1st generation from 2013, this supports what was written as 802.11n and recently renamed to WiFi 4 (for version 4). Then there are two radio frequencies that WiFi can work on, the Air supports both 2.4 GHz and the 5 GHz frequencies. This technology was designed and first released in 2009.

Depending on the router you are using / provided by your ISP, the best you will see on a WiFi 4 network between two computers on the same network at home, is likely to be around 50 Mbps. As you are not the manufacturer of the iPad, you can't choose how many antennas it has or where they are physically. Depending how you hold the iPad you may be covering slightly one of the antennas, reducing throughput.

The router from your ISP may have one, two, three or four "streams" and the iPad Air gen1 supports what is called MIMO that means it can use two antenna at the same time to improve speeds. Each stream is on WiFi 4 is a maximum of 72 Mbps throughput, so with two streams you can get a theoretical 144 Mbps.

However a lot of that capacity is used for overheads, making the WiFi network work, so the amount you can actually get through from one device to another (e.g. two laptops in your home) is limited by these overheads. You can't test this stuff on an iPad as they are an appliance, fairly fixed function.

As long as the speed from router to iPad or laptop is faster than the speed from the router to the internet then the WiFi speeds are not restricting the connection. iPads are designed in the same way as iPhones and other Smartphones, to save battery. This makes them not the best tool to do speed tests with.

Air 1 technical specification
https://support.apple.com/kb/SP692?locale=en_GB

I am guessing you have an iPad that old, but if its even older then all bets are off. If you have an Air 2, or an entry level iPad that is newer than an Air 2 post back which one you have.

Newer WiFi standards such as WiFi 5 (AC) or WiFi 6 (AX) are designed to improve the total throughput for one or multiple devices. You may only have one device in use, but your neighbours are also using WiFi and causing interference...

Edited by jchamier (Wed 09-Aug-23 13:47:21)

A low upload could be caused by poor wifi - but if the download is much higher than the upload on the same device in the same location then it probably isn't the wifi. Do your other devices (wired or wireless) have a better upload speed?

It's actually 36/9 for some strange reason, but my question still stands: what is the max UL speed the ipads will do?

Thanks, so despite all the variables, in theory, 3mbps is very slow as the maximun *should be* 144Mbps?

Yes, I have an iPhone that generally gets better wifi upload, but I don't understand why downlad is much better than upload given wifi usually works the same in both directions. Or is that just a myth?

The theoretical maximum is 9 Mbs; in practice the maximum will be closer 6, or less.

That's the maximum to the router. The maximum upload speed to the Internet is almost always limited by the connection from the router to the Internet (absolute maximum 9Mbs in your case) rather than the connection from the device to the router.

The speed cannot exceed that of the slowest link in the chain from device to server.

WiFi works at the same speed in both directions, but that's not true for your Internet connection. There downloads are 4 times as fast as uploads (which makes sense as, in general, most traffic is from a server to your device rather than the other way round).

In reply to a post by hk11:

Thanks, so despite all the variables, in theory, 3mbps is very slow as the maximun *should be* 144Mbps?

If you had a speed test server in your home, connected by Ethernet to your router, and your iPad standing just in front of the router, and the iPad is running well, the WiFi speeds theoretically would max out at 144 Mbps in pure lab conditions. Very unlikely to be met on any WiFi version.

I have done this test myself using laptops on WiFi and a very fast server on 2.5 gigabit Ethernet, and I have seen WiFi 4 (N) manage around 80 Mbps in both directions. These were FAST laptops. iPads are not the same as laptops.

But the OP is talking about his 40/10 Internet connection.

I'm not asking about my Internet connection, just my wifi.

But if downloads are working at near the guarenteed speed why would uploads not follow?

I not expecting the full 144, given I am working over a 9Mbps connection, but why would there be a 3Mbps top speed given that the download is around the 36Mbps?

So you are testing the speed to a local server rather than using an Internet speed test?

I understand the maximum. It should be 10Mbps actually, but I think Vodafone say 9 as there is some sort of Ofcom rule saying they can't quote speeds above what they provide.

Given the max is 40, well 36 is quoted, download and I get near or even above this, I don't understand why I only get a third of the upload (9 or 10Mbps). It's obviously not an iPad limitation given the quoted 144Mbps, even if we allow for all sorts of veriables.

But why, given the download is actually obtained?

No, an Internet speed test.

Then you are testing the speed of your Internet connection.

I give up.

In reply to a post by hk11:

I understand the maximum. It should be 10Mbps actually, but I think Vodafone say 9 as there is some sort of Ofcom rule saying they can't quote speeds above what they provide.

Uh, its a bit complicated.

Given the max is 40, well 36 is quoted, download and I get near or even above this, I don't understand why I only get a third of the upload (9 or 10Mbps).

There are two products from Openreach that the ISP can purchase, the 40/10 service or the 80/20 service. They are both "up-to" these speeds, depending on the quality of the wire in the ground from the green box to your home, and the quality of the wiring in your home (number of extensions, etc).

The 40/10 service is the speed the router is connected to the green box at, but there are technical overheads (how computers work) that reduce the data throughput that you can get.The ASA and Ofcom agreed the current limits on ISP marketing.

With a laptop or desktop connected by Ethernet to a router on a 40/10 service, assuming a good and short copper line then the maximum speeds you can get is 36 download and 9 upload in megabits per second (Mbps)

Many copper lines in the street are not good quality and as you have a very expensive box at the Openreach end, and a £50 router at the home end, often the download speed is close to the quoted, but the upload speed is less good. This is why people want "full fibre" or "cable" instead of copper wires in the street, as there is less variability.

Edit - you are on an FTTP service, so you get the 40/10 speeds at ANY distance. The overheads of the TCP/IP packets and the PPPoE encapsulation is why the usable throughput of 36/9 is permitted to be advertised.

It's obviously not an iPad limitation given the quoted 144Mbps, even if we allow for all sorts of veriables.

This 144 is not guaranteed, if you don't live in the middle of the country there are other radio transmissions around you (coming through your walls) that can interfere with the WiFi speeds. This is why people upgrade WiFi technologies to get more total capacity so that with interference they get more usable throughput.

In your case, without an Ethernet adaptor for your iPad, or a laptop/desktop connected by Ethernet, you cannot reliably test this.

For example at home I have Virgin Media's service which is 250 down and 20 up. On my 3 year old desktop PC I can go to speedtest.net in my Firefox web browser and see speeds of 240 download and 19 upload. This is good. On a 5 year old Android phone connected by WiFi standing only 2 metres from my router I can only get 180 download and 6 upload.

Can you borrow a laptop or something with an Ethernet connection?

Which iPad do you have?

Edited by jchamier (Wed 09-Aug-23 19:58:58)

The OP is on a 40/10 FTTP service, so no copper line length involved. Just very poor Wi-Fi on an Ipad.

From everything the OP has said it is not a WiFi problem. If he gets 36Mbs download then he should expect near to 9 upload (although on my connection, with similar limits, I get 32 down and only 5 up; but that doesn’t cause me any problems). That points to an ISP problem.

Assuming that this upload speed is causing problems, and we’re not just obsessing about speed tests, the OP should contact his ISP’s customer support. Presumably he’s done the obvious and powered the router off and on again.

In reply to a post by Realalemadrid:

The OP is on a 40/10 FTTP service, so no copper line length involved. Just very poor Wi-Fi on an Ipad.

Ugh, I’d missed it was FTTP. Yes, an iPad before the Air 1 (which was 2007) is a pretty useless device to do speed testing on. The CPU can’t keep up with the throughput.

Thanks for the explanation of 80/20 and 40/10. I have found this all out recently as no one seems to offer the 40/10 FTTP service I am on now; hence my ordering the 80/20 one.

Surprised you mention copper as that is pretty much redundant round here, although some areas do still offer FTTC which is a mixture of copper and fibre, so ....

Back on topic: I have just done a couple of speed tests using an ancient iMac on both wifi and ethernet and got pretty much the 36/9 so my original question still remains. Why am I still only getting 36/3 on my iPads?

In reply to a post by hk11:

Surprised you mention copper as that is pretty much redundant round here, although some areas do still offer FTTC which is a mixture of copper and fibre, so ....

Thinkbroadband’s news pages show that Openreach has only reached about 45% of the UK built up area yet. A LOT of the country still has only copper from Openreach. Many cities and towns. Some are lucky to have an alternate network (such as Zzooomm or CityFibre) but not many yet.

Back on topic: I have just done a couple of speed tests using an ancient iMac on both wifi and ethernet and got pretty much the 36/9 so my original question still remains. Why am I still only getting 36/3 on my iPads?

The iPad is not capable of sending any faster then. That your iMac can get the full speed your ISP quotes means there is nothing wrong with your system.

Unlike a computer (iMac, MacBook, etc) the iPad is based on mobile phone technology, but made physically larger. For price and battery life reasons the choices made of the hardware inside are to a performance for price.

As the years go by the newer hardware goes faster for similar price. The original 2010 iPad was incredibly slow on the network, but people still loved it, even without a camera to do video calls. iPad 2 was a big seller with the added video camera. The Air models were then faster again, and now Apple has three ranges:
* Entry level iPad (ver 9 and ver 10 on sale)
* Mid range iPad - iPad Air gen 4 or gen 5
* High end iPad - iPad Pro 11” or iPad Pro 12.9”

The mid range and high end iPads include the CPU from the new Macbooks for speed reasons.

Then you don’t really have a problem. I can’t imagine that you need to upload large amounts of data from an ancient iPad.

45% eh. It's a pity folk have no choice. There's me on a tiny island that was quite happy with ADSL that is being forced onto full fibre and yet a friend in Liverpool that is stuck on ADSL but would love fibre, so falls into the other 55%.

I guess you are right about the iPad. Always been happy with mine, but wouldn't buy another due to Apple's continually making old hardware redundant. I'd love something with a bigger screen than my iPhone though. Are the £80 tablets worth looking at?

Still don't understand the difference between the advertised 144Mbps when the iPad is throttled at around 3Mbps, but we will see what happens when it gets offered 20Mbps next week. I am assuming that 3Mbps still won't be exceeded?

As an aside I have just found out the lack of an option for attachments when posting to Plusnet Community was that the iPads weren't using desktop mode. I always used to request desktop mode having been used to working on a PC, but I must admit to not having requested that for ages.

Edited by hk11 (Wed 09-Aug-23 20:37:47)

In reply to a post by hk11:

I have just done a couple of speed tests using an ancient iMac on both wifi and ethernet and got pretty much the 36/9

Hooray

9th December 2022 => 9th August 2023 a total of 9 months but at last you have the information that shows there is nothing wrong with your Vodafone FTTP service

I hope you will be calling Vodafone to apologise.

Edited by deleted (Wed 09-Aug-23 22:21:19)

😃

In reply to a post by dect:

In reply to a post by Pheasant:

Take your iPad and test it on another connection, ideally faster than your own domestic connection snd it should pinpoint whether it’s your iPad or broadband that’s at the limit / issue.

It may give him an indication but not a cast iron guarantee. So many variables with this OP

Got there in the end. Wow. So many posts and time devoted to this. I’m staggered.

There is no point in contacting Vodafone as they don't read what is sent to them and still insist BTPL-law are responsible for poor Openreach installation!

Customer service is not about the product supplied, it is about the way a company deals with complaints.

I'm still waiting for the apology from Apple that their product does not come anywhere near the 144Mbps wifi speed advertised.

Edited by hk11 (Thu 10-Aug-23 02:12:11)

In reply to a post by hk11:

I'm still waiting for the apology from Apple that their product does not come anywhere near the 144Mbps wifi speed advertised.

You’ll have to do a speed test with a server on your network before you can conclude that. All you can say so far is that there may be a problem with the combination of your iPad, your Internet connection and the particular speed test you are using.

Good luck having Apple admit a fault in your “old” iPad, which is presumably way past its warranty period. And you keep going on about 144Mbs when you have only tested a 9Mbs connection.

But if the device can't handle 9Mbps what chance is there that it could handle say 100Mbps?

I "keep" using the 144Mbps figure as I have yet to find an answer to the question I posted as to what dl speed the iPad could handle.

The wired service is fine - according to your own speed testing.

The issue is the wireless. As before wireless relies on radio, a shared medium - you are competing with your neighbours (up to 100m away), microwave ovens (when on - the magnetron/klystron emits milliwatt levels of interference in the same area of the spectrum), Bluetooth etc.

You want your wireless to work up to the spec - you need to move somewhere radio quiet like an isolated croft in Durness, not a terraced street in Sheerness.

You don't know that the device can't handle 9Mbs. You only know that (possibly) the combination of device, router and Internet connection is running slowly (in one direction). There are a host of factors that can affect that; the client device is just one of them. Your Internet connection, the router, the version of the firmware running on the router, the version of iOS running on the iPad, and the age of the iPad are all relevant factors, amongst others.

The comparison with the MacBook is meaningless unless you have ensured that both connections used the same frequency and the same channel. Even then it is not conclusive. There could be an incompatibility between the chipsets in your router and your iPad. You would need to do a lot more tests before you could conclude that the fault lies with the iPad - doing a speed test on your local network is the first as this would eliminate the Internet connection leaving the iPad/router combination as the important factor.

Bottom line - is the upload speed causing you problems? If not, don't obsess about it.

Please please don't repeat your errors of the last 9 months, everyone is trying to tell you the same thing but your continued pushing back is really not helping you.

Advice: Get yourself some better devices rather than keep using the dinosaurs stuff you currently have.

Strange that the wireless signals are only affected on the one (type) of device, as per the subject.

All fair comments. but it would be nice to have some answers.

In reply to a post by hk11:

it would be nice to have some answers.

First you have to ask the right questions.

It doesn't seem to matter what I ask, in the main, folk just answer what they *think* is the question.

Tell me what part of my original question I need to rephrase in order to get an answer appertaining to the subject line.

Does this mean 144Mbps is the only limit on Wifi speed?

I think that people have been more than patient in answering that question.

The question that you haven't asked is "Is my iPad's upload speed abnormally low?". And, until you do the appropriate tests no-one will be able to give you a reliable answer.

The other question that you haven’t asked is “Does it matter if my iPad’s upload speed is a bit slow?”. Without knowing what you need to upload there’ no answer to that one either.

Edited by TinyMongomery (Thu 10-Aug-23 15:08:23)

Thanks.

What are the "appropriate tests"?

At the risk of repeating what you have been told several times already, you need to measure the upload and download speeds to a server on your network. If you don’t know how to do that then it’s time to stop worrying about something unimportant that you can’t measure.

It’s certainly time that I stopped wasting time on this non-problem.

In reply to a post by hk11:

Strange that the wireless signals are only affected on the one (type) of device, as per the subject.

What wireless standard is the router, does it have 1, 2,3 or 4 antennas?
What wireless standard is your iPad, does it have multiple antennas?
What wireless standard is your iMac, does it have multiple antennas?


I have a Three branded pocket mobile broadband router that has only one antenna and supports WiFi 4 (802.11n) which means a Physical rate of maximum 72 Mbps. In some areas of the UK the Three network can easily provide 200+ Mbps over 4G. That device never managed to go any faster than 45 Mbps.

WiFi is seriously complicated, and if you want to know more read the small net builder site in the US….. you can’t say a 144 Mbps device (any WiFi 4 device with 2 antennas) is slow because it doesn’t get close to 100 Mbps.

You can only make that assertion with fixed line Ethernet. (Or Token Ring if you’re Pheasant)

The THG3000 does 2.4 and 5. Not sure how many aerials it has. https://m.youtube.com/watch?v=XBUXOMkLO6w May help?
The ipad uses 5 not sure about aerials. I asume from https://www.ifixit.com/Parts/iPad/Antennas it only has the one.
Not sure about the iMac.f

So what do you consider slow for a 144Mbps device?

I'd rather keep fixed line out of the equation ....... For now.

Edited by hk11 (Thu 10-Aug-23 20:35:15)

In reply to a post by hk11:

The THG3000 does 2.4 and 5. Not sure hiw many aerials it has.
The ipad uses 5 not sure about aerials.
Not sure about the iMac.

So what do you consider slow for a 144Mbps device?

I'd rather keep fixed line out of the equation ....... For now.

The frequencies (2.4 GHz and 5 GHz) are important when using Apple devices, as Apple limits the 2.4 GHz to only 20 MHz wide, whereas on 5 GHz you can have wider bands, 40 MHz on WiFi 4 (N), and 80 MHz on WiFi 5 (AC). The THG3000 was reviewed by TrustedReviews whom managed over 180 Mbps through the router. The unit is WiFi 5 (AC) capable as it is also provided for use with their full fibre services, so 500 or 900 Mbps. It most likely has 3 antennas, maybe 4. Unfortunately all internal means it is not visible unlike third party makes such as ASUS.

Due to physics, 5 GHz has less range. You might find if you force your iPad to the 5GHz connection and you are in the same room as the THG3000 you MAY get faster speeds.

The iPad Air 1 has the Apple A7 CPU which was their first 64bit chip. The unit went on sale 10 years ago, and Apple stopped supporting in 2016, it runs iOS 12.5.6 at the highest, and todays current release is 16.6

It was the first iPad to support two antennas in use at the same time, so the theoretical throughput speed increased from 72 Mbps to 144 Mbps. To get the highest speeds on an iPad with WiFi 4 (N) electronics your router needs to support 5GHz mode with two channels transmitting for 40 MHz.

Throughput across WiFi is very hard to test. WiFi 4 (N) is old and so when two devices transmit at the same time, even if one is in your house and one in your neighbours house, and they are talking to different routers, they will both get slower throughput. In busy homes with lots of family members all using the WiFi at the same time, even with an internet connection of 500 Mbps (e.g. Virgin Cable) a WiFi 4 (N) network would slow down and everyone on the network would get less good performance. WiFi 5 (AC) and WiFi 6 (AX) provide methods to solve this problem, but each end of the link (router, and device) need to be 5 or higher.

This guy (Tim Higgins) does WiFi throughput testing properly, he has a proper test lab and lots of routers. You can see from his reports that the speeds he gets in the lab are nowhere near the marketing numbers BUT the newer versions of WiFi are making huge jumps in the capacity and throughput. So you have to read the 144 as a guide number against other WiFi devices that support 72 or 300, 600 or even 3200. Not to compare against fixed line.

https://www.smallnetbuilder.com/wireless/wireless-fe...

Interesting, but it all seems to add extra questions rather than provide any real answers.

The average user isn't going to know any of this.

Figures and comparisons seem very limited. I can't even get basic power consumption figures for the THG3000 compared to Plusnet's Hub Two for example.

In reply to a post by hk11:

Interesting, but it all seems to add extra questions rather than provide any real answers. The average user isn't going to know any of this.

The average user works with what they’ve bought, or buys something newer.

The average IT professional with 25 years in the business, a BCS membership and Chartered status isn’t going to know how WiFi works. As I said it is insanely complicated and specialist.

Figures and comparisons seem very limited. I can't even get basic power consumption figures for the THG3000 compared to Plusnet's Hub Two for example.

ISP provided equipment is usually poorly documented. Third party products that sell in the USA often provide data to the US regulator (FCC). Sometimes you can look up information at WiFi Alliance’s pages.

Apple also have form in this regard: http://edition.cnn.com/2010/TECH/mobile/06/25/iphone...

In reply to a post by hk11:

but it all seems to add extra questions rather than provide any real answers.

You are getting something of a reputation with these threads

In reply to a post by hunnymonster:

Apple also have form in this regard:

Yeah they screwed up, but this was mobile connection on 2G and 3G with the iPhone 4, and they solved the problem with the 4S.

Third parties worked out they had not allowed for people holding the phone without a case, so touching the antenna, and with only one antenna this caused a major degredation in signal reception. With the 4S the phone has two antennas and chooses which one to use ("antenna diversity") as the competition were already using. This was a learning experience for apple.

Didn't affect WiFi

In reply to a post by jchamier:

You can only make that assertion with fixed line Ethernet. (Or Token Ring if you’re Pheasant)

Oi. Leave me out of this😜

[Token Ring rocks btw. As you know 😂]

In reply to a post by Pheasant:

[Token Ring rocks btw. As you know 😂]

Until you leave a cable unplugged and the token falls out and rolls under a cabinet.

Hehehe. 😂 Please sir, have you seen my token? I’d like it back.

Anyone seen a small bag of BNC terminators for my Thinnet? The gremlins have nicked them and now nothing works anymore.

In reply to a post by Pheasant:

The gremlins have nicked them and now nothing works anymore.

The kids/students were always doing this to avoid lectures....

In reply to a post by Pheasant:

Anyone seen a small bag of BNC terminators for my Thinnet? The gremlins have nicked them and now nothing works anymore.

Thick ethernet and vampire connectors, much safer

Apple can't give you the answer you want, it seemingly being "how fast should my iPad's wifi be, in my house, connected to my router, with my neighbouring radio frequency interference".

All they can do is tell you the maximum the hardware can physically achieve, which they do. I have no doubt that the iPad's wifi can achieve the quoted 144 Mbps when tested on the right hardware with the right radio frequency environment.

What answer do you expect them to give you ?

If really you want to know how fast your iPads's wifi is in practice, go to the effort of measuring it properly.

In reply to a post by hk11:

I guess you are right about the iPad. Always been happy with mine, but wouldn't buy another due to Apple's continually making old hardware redundant. I'd love something with a bigger screen than my iPhone though. Are the £80 tablets worth looking at?

In what way do you imagine that Apple is making old iPad hardware redundant ? Is it that they stop releasing new OS updates after a few years ?

If so, I've got some bad news for you if you venture into the Android tablet market...

We have no idea what the speed of your wifi connection actully is, you've never to my knowledge tested it.

To that you need an iPerf3 server on an Ethernet connected machine:

https://formulae.brew.sh/formula/iperf3 - Mac via homebrew
https://iperf.fr/iperf-download.php - Windows

and an iPerf3 app on the iPad:

https://github.com/ndfred/iperf-ios

Then you'll be able to test how fast the wifi actually is.

Edited by andynormancx (Sat 12-Aug-23 12:42:56)

You don't know that the wifi is performing at 1/45th of the headline speed. All you know is that the combination of your iPad, your router, your neighbours radio emissions and your ISP are only allowing that speed.

Unless you test the speed of the wifi independently, you are just guessing at where the bottleneck is.

Very rare to get actual realistic power consumption figures published for many devices. Basically because of the variability of doing so on different usage scenarios, so you may get a worst case figure or look at the power supply wattage or VA rating, again worst case and usually hugely over what the device will consume long term in steady state operation.

I’m tempted to dig out an old iPad 2 from the cupboard and test it to see what it will do on WiFi throughput - as mentioned several times though, that won’t necessarily mean you will achieve similar rates due to the wide factors / variability of WiFi based on the specific user setup and locale etc.

In reply to a post by Pheasant:

Very rare to get actual realistic power consumption figures published for many devices. Basically because of the variability of doing so on different usage scenarios, so you may get a worst case figure or look at the power supply wattage or VA rating, again worst case

For someone prepared/determined to change the kit if the usage is unnecessarily high then its worth them knowing but for most folk they are still going to use the kit regardless so what is the point. The router from both these ISPs is going to be basic so I wouldn't expect a massive different.

If people really want to know then one of those plug in display things that shows the usage is a good investment. I have no doubt you already have one of these

It's strange that electrical appliances don't have figures, even if they vary. Car manufacturers publish figures and they have similar wide variables, but manage to constrain them to @56mph, for example.

Please do get yer iPad out; if only to throw yet more variables into the mix. One of mine has decided it won't do anything but try and contact Apple now to register itself, despite this having been done when new years ago!

ISP change Tues, so that will eliminate one variable, then another when I swap the router over. Wifi interference and my iPad will remain unchanged, unless of course another neigbour changes their setup at the same time.

In reply to a post by hk11:

Please do get yer iPad out

A few tbb tests for interest... I'm on a (nominal) 80/20 FTTC connection, but too far from the (crowded) cabinet to achieve those speeds.

iMac (wired) for reference: 64.4 down, 16.4 up.

iPhone 12: 63.6 down, 15.8 up.

iPad 2 mini: 56.0 down, 16.7 up. (Better than I expected, it always "feels" a lot slower than the other two)

All tests HTTPx6, IPv6, taken within about 10 minutes of each other with iPad and iPhone on 5GHz less than a metre from the router. iMac uses Chrome, other two use Safari.

Edited by billford (Sun 13-Aug-23 09:50:45)

In reply to a post by hk11:

Please do get yer iPad out; if only to throw yet more variables into the mix.

Ok, if you have more modern hardware this is what you can get.

Internet = Virgin Media 250 Mbps / 20 Mbps.
PC = desktop PC (12th Gen Intel i7) with 1 Gigabit Ethernet connection to router
Router = ASUS WiFi 6 with 4 physical antennas for 4 stream MIMO, 160 MHz channel on 5 GHz, 20 MHz on 2.4 as Apple devices won't use any more than 20MHz on 2.4 GHz

iPerf running on the PC, using iPad Pro 11" (about 3 years old) I get a speed range between 480 Mbps and 590 Mbps for both upload and download. Theoretical speed not stated on Apple tech specs page. At sale was "over 1 Gigabit"

iCloud Private Relay disabled to avoid any redirection.

Thinkbroadband (in Safari) and speedtest.net (app) tests on the iPad show the same WAN speeds as the desktop PC, 270 Mbps download and 25 Mbps upload.

When I can find my USB-C adaptor I will try my personal laptop on an Ethernet link with iPerf just to confirm the switch can handle the full gigabit speed (940 expected).

Edited by jchamier (Sun 13-Aug-23 10:07:15)

In reply to a post by hk11:

It's strange that electrical appliances don't have figures, even if they vary. Car manufacturers publish figures and they have similar wide variables, but manage to constrain them to @56mph, for example.

Please do get yer iPad out; if only to throw yet more variables into the mix. One of mine has decided it won't do anything but try and contact Apple now to register itself, despite this having been done when new years ago!

It’ll be a couple of weeks before I get near that iPad again, and I’ve no idea if the battery has completely crapped out…so other folks will doubtless post up their own experiences if they have one handy and running.

In any event, if it’s going to run modern apps and you really want to maximise performance then running an ancient (relatively speaking tablet) is not the best way of going about it. For a start the operating system on these things is about seven versions behind the current release of iOS (well, well beyond support) and the hardware was out of support about a decade ago.

For all intents they are now museum pieces, and you can’t expect them to be performant like a device that is up to 5 years old, so within the current support envelope of hardware and iOS versions. It’s well and truly a sweated asset. But I’m sure you know that.

In reply to a post by jchamier:

When I can find my USB-C adaptor ....

So I found my two USB-C ethernet adaptors, both claim to be Realtek chips. Same desktop PC as the iPerf server, using the laptop as the client with either USB-C to Ethernet adaptor, both PCs connected directly to the ASUS router with its internal Gigabit switch, iPerf reports 940 Mbit/sec download and upload. As expected, and showed the laptop and desktop have no problems keeping up with 1Gig ethernet as you'd expect for machines with 12th Gen intel CPUs.

The laptop is capable of WiFi 6E (ax) with 160 MHz channels, and the router is WiFi 6 with 160 MHz support, so after a couple of reboots I was able to get the laptop to connect at 160 MHz on 5 GHz. iPerf managed a throughput over the 160 MHz channel of between 820 to 700 Mbits/sec.

Note the claimed throughput of 160 MHz WiFi 6 is 2400 Mbit, but even this couldn't reach the limit of the server's 1 Gig connection (940 Mbit/sec).

I then attached the USB adaptors to the iPad Pro and disabled its built in WiFi, and the iPad was able to achieve 940 Mbits/sec.

Edited by jchamier (Sun 13-Aug-23 13:25:44)

In reply to a post by Pheasant:

For all intents they are now museum pieces

People don't realise how bad their legacy kit is until they upgrade, I've just gone from an Intel G2030 processor to an Intel 13 Gen i7 processor and I can't believe the difference.

In reply to a post by jchamier:

The laptop is capable of WiFi 6E (ax) with 160 MHz channels, and the router is WiFi 6 with 160 MHz support, so after a couple of reboots I was able to get the laptop to connect at 160 MHz on 5 GHz. iPerf managed a throughput over the 160 MHz channel of between 820 to 700 Mbits/sec.

Note the claimed throughput of 160 MHz WiFi 6 is 2400 Mbit, but even this couldn't reach the limit of the server's 1 Gig connection (940 Mbit/sec).

I then attached the USB adaptors to the iPad Pro and disabled its built in WiFi, and the iPad was able to achieve 940 Mbits/sec.

The best I’ve seen on an Ookla-based test on my phone running 2x2 mimo on an AX-based access point using 80 MHz channels is about 780 Mbps down and a smidge over 800 Mbps up. That’s all in the same room about 5 metres from the AP with a bit of 18mm thick timber between the AP and phone / that is the cupboard door! Near enough line of site 😅

On AC-based access points that throughout drops to around 480-500 Mbps both ways with no obstructions.

In reply to a post by dect:

In reply to a post by Pheasant:

For all intents they are now museum pieces

People don't realise how bad their legacy kit is until they upgrade, I've just gone from an Intel G2030 processor to an Intel 13 Gen i7 processor and I can't believe the difference.

Indeed. The “bang per watt” astounds me most these days. Either battery-supping Apple Silicon or all out AMD sheer power is pretty phenomenal step change from only a few years ago.

In reply to a post by dect:

People don't realise how bad their legacy kit is until they upgrade, I've just gone from an Intel G2030 processor to an Intel 13 Gen i7 processor and I can't believe the difference.

That's around a 10 year technology jump, I'd hope you would see a difference. My corporate tries to upgrade every 4 to 5 years, and on laptops that is usually very noticable. On desktops it depends if you've gone over one of those performance cliffs

In reply to a post by Pheasant:

The best I’ve seen on an Ookla-based test on my phone running 2x2 mimo on an AX-based access point using 80 MHz channels is about 780 Mbps down and a smidge over 800 Mbps up. That’s all in the same room about 5 metres from the AP with a bit of 18mm thick timber between the AP and phone / that is the cupboard door! Near enough line of site 😅 On AC-based access points that throughout drops to around 480-500 Mbps both ways with no obstructions.

2x2 AX 80 MHz achiving 800 is pretty impressive anyway, I guess any nearby networks are nicely far away, and to get 480-500 on AC is great!

Interesting: thanks.

I'm upgrading from 40/10 FTTP to 80/20 FTTP Tues, so ...

Nice!

I'm waiting for Linux on the iPad.

Just installed Debian 12 on one of the iMacs and I can now view the Flickr website and actually see what is going on. The only device working up to now is an iPhone.

In reply to a post by jchamier:

My corporate tries to upgrade every 4 to 5 years, and on laptops that is usually very noticeable

I did use to get a new laptop every year or so before I retired. The last one was in 2016 I think it was a Lenovo, but once the company image has been PXE'ed on and all the build layers have been added and security applied they never seem as fast

I didn't realise you could get an external wifi connector for the iPad!

In reply to a post by hk11:

I didn't realise you could get an external wifi connector for the iPad!

The more modern iPads (Air4/Air5, the 2018 11" pro's and later) have USB-C which supports lots more accessories.

The lightning port equipped older ones can work with the Camera Connection Kit which provides a USB-A connector, and there are some reports that USB ethernet adaptors can work, but much slower as many iPads only support USB 2.0

Interesting: thanks.

New (meaningless?) speedtest on Plusnet 80/20 FTTP v Wi-Fi on iPad - https://www.thinkbroadband.com/speedtest/16919933852...

V Wi-Fi on iPhone - https://www.thinkbroadband.com/speedtest/16919939505...

Edited by hk11 (Mon 14-Aug-23 07:26:04)

In reply to a post by hk11:

New (meaningless?) speedtest on Plusnet 80/20 FTTP v Wi-Fi on iPad - https://www.thinkbroadband.com/speedtest/16919933852...

V Wi-Fi on iPhone - https://www.thinkbroadband.com/speedtest/16919939505...

I think the iPhone is newer then the iPad. The iPhone can keep up with the new speed service, 75 download and 20 upload is pretty good. This shows the iPad is struggling.

Which model iPad and iPhone are you using?

Well that is pretty conclusive evidence that the Ipad wi-fi is rubbish. End of story.

It’s only evidence that this particular combination of iPad, router, frequency, channel, and external influences is rubbish.

Tests with another router might produce completely different results.

I’m still intrigued to know what anyone does with an ancient iPad that requires a higher upload speed. In general iPads consume data rather than producing it interactively.

In reply to a post by TinyMongomery:

It’s only evidence that this particular combination of iPad, router, frequency, channel, and external influences is rubbish. Tests with another router might produce completely different results.

Very true, but the use of the iPhone as well showing quite different speeds at basically the same time, means the likelihood is the age of the iPad, with a slow CPU, at a guess an iPad 1 or 2, is just not remotely sensible for doing speed tests.

and the weekend testing shows that WiFi speeds of 144 or even 2400 Mbit/s don't translate to that in the real world. Unlike Ethernet.

Not really. The chipsets in the two devices are different, and they could be using different frequencies. Testing with a different router is an obvious step to test if it’s the particular device/router combination.

If you have a variety of variables you can only draw meaningful conclusions if you try varying each of those variables to test how they affect the result. And you need to first identify what those variables are and ensure everything else is kept constant.

Yup, my thoughts as well, hence my post here.

All that talk about interference and which way the wind was blowing might be valid, but not really the issue here.

I think they iPads are iPad2, but no model numbers on them as far as I can see.

The iPhone is an SE.

In reply to a post by hk11:

All that talk about interference and which way the wind was blowing might be valid, but not really the issue here.

With WiFi sadly interference is a fact of life.

I think they iPads are iPad2, but no model numbers on them as far as I can see.
The iPhone is an SE.

See this apple page to identify the iPad:
https://support.apple.com/en-us/HT201471

Is the SE a version 1 (looks like iPhone 5/5s) or a version 2 or 3 (looks like the iPhone 8). The original iPad from 2010 supports both 2.4 GHz and 5 GHz for connectivity, but it has no cameras at all. The iPad 2 has cameras, and introduced FaceTime.

In reply to a post by hk11:

Interesting: thanks.

New (meaningless?) speedtest on Plusnet 80/20 FTTP v Wi-Fi on iPad - https://www.thinkbroadband.com/speedtest/16919933852...

V Wi-Fi on iPhone - https://www.thinkbroadband.com/speedtest/16919939505...

So your new broadband service with PlusNet has gone live this morning, are you now using their router rather than the Vodafone one?

Just found an old iPad 3 (A1416) and run the TBB HTTP speed test (the iPad is so old it isn't even loading the normal speed checker - it's only running iOS 9 and no updates available).

On the iPad I got 28.4 down and 11.6 up (on a 500Mb OpenReach FTTP package).

On my iPhone in the same location I got 143 down and 70.2 up.

So, the iPad not running as well as the iPhone but it is getting more upstream than your ISP link would be capable of so if yours has the same WiFi hardware as the iPad 3 (can't remember if you said what model iPad you had) then it seems it is capable of more than you are getting at the base WiFi level.

The Apple page doesn't load on my iPad which is rather ironic! I'll look later on another device.

No idea on the iPhone, but is still getting updates"

Yes, the iPads both have cameras.

Yes and yes.

I couldn't find where to change my username and password on the THG.

All very odd.

I just think I'll get a basic tablet. Certainly not worth getting another iPad. They have been good, but with things more than two years old being old relics I can probably get several tablets for the price of an iPad.

In reply to a post by hk11:

They have been good, but with things more than two years old being old relics I can probably get several tablets for the price of an iPad.

I'd love to hear why you think a two year old iPad is an "old relic" ?

In reply to a post by hk11:

They have been good, but with things more than two years old being old relics I can probably get several tablets for the price of an iPad.

Unless you get an Amazon Fire tablet, they're all pretty expensive now. Check out the OnePlus tablet, or the Pixel tablets, both released in the last 6 months. (The YouTuber MKBHD reviewed them recently).

My iPad Pro 11" is 3 years old by the way. I bought mine as a second hand from CEX when it was 8 months old. Worth looking in CEX stores near you as they give warranty.

I gave up on Android when an update killed my (expensive) Asus Nexus.

Thanks: I'll check them out.

Technology moves so fast!

Anything over two years old is, it would appear, out of date.

I think my iPad Pro is 6 or 7 years old and still receiving the latest iPadOS updates. Apple are one of the best for the time they support new OS versions on phone and tablet devices, and all devices eligible get updates at the same time whereas on Android you generally have to wait to see if the manufacturer will support it and it could take a year or more.

In reply to a post by hk11:

Technology moves so fast!

Anything over two years old is, it would appear, out of date.

I think like everything you have to know when its the right time to upgrade, the average time people keep a mobile phones in the UK is 4 years and I think thats about right.

Devices may be superseded but in my opinion it doesn't mean they are obsolete, If you have a legacy device and you're not bothered by the throughput of the wifi then no need to worry but once you are doing speed tests and complaining to your ISP about slowness then you are clearly not happy with the wifi of the device and should upgrade it.

Edited by deleted (Mon 14-Aug-23 17:45:18)

In reply to a post by hk11:

Anything over two years old is, it would appear, out of date.

Depends how you define out of date. My company says we can use any iPhone/iPad still getting security updates. Whereas they have had to set a 2 year rule for Android as so many devices don't get updates. The directors don't want a client to sue the company because of an unpatched security hole.

What you may not have realised is a 10 year old laptop with 1st Gen Core i3 CPU can easily keep up with higher WiFi speeds, but tablets (and phones) have been dramatically changing in technology for the last 10 years. This has probably ended now, just incremental change, not dramatic jumps.

The 2018 iPad Pro 11" 1st edition with the A12 CPU will be pretty close in WiFi speeds to the end of 2022 model iPad Pro 11" 4th edition, but the original iPad Pro 9.7" from 2016 won't keep up at all.

How long people keep things and when they become out of date is two different things.

All the while something does the job there is no need to upgrade, but when it starts to give you problems ....

So, like a lot of things, it varies.

Trouble with something like speed is you don't notice a gradual reduction. When you are used to 1Mbps (ADSL) then you are forced to pay for a ten fold increase (fibre) and don't get it, you question why.

Edited by hk11 (Mon 14-Aug-23 23:37:43)

You keep saying that, with no justification whatsoever. Exactly what about a two year old iPad is out of date ?

In reply to a post by hk11:

When you are used to 1Gbps (ADSL)

Did you mean this?

Edit: You are the first person I can ever recall who clearly prefers ADSL (ADSL2+ max 24Mbps) over full fibre, you may want to think about why you stand alone with that opinion. I also know one person on here who prefers FTTC over full fibre and he also stands out from the crowd with all his opinions.

Edited by deleted (Mon 14-Aug-23 22:08:59)

No, clearly a typo. Upload on ADSL was in Mbps not Gbps. Have corrected.

It’s not a case of preferring ADSL per say; just that it did the job. But I suppose you could say that about most older communication? I wouldn’t want to go back to dialup, but on the other hand what is being gained by pushing all this data around at even greater speeds? Emails of a few lines used to do the job, now it’s a couple of pages of html most of which is advertising!

If your uploaded emails are being affected because of the amount of advertising you include in them, you certainly need to do something about it.

In reply to a post by hk11:

Emails of a few lines used to do the job, now it’s a couple of pages of html most of which is advertising!

You can with many companies these days elect to receive plain text mail, and opt out of marketing communications.

In reply to a post by hk11:

Emails of a few lines used to do the job, now it’s a couple of pages of html most of which is advertising!

Yes online marketing of any kind can be frustrating at times but lets be honest it isn't really causing speed issues for your setup is it?

"Car manufacturers publish figures and they have similar wide variables, but manage to constrain them to @56mph, for example."

And these figures are not based on real driving. No wind resistance so of no use in the real world. Bit like theoretical Wi - Fi speeds.

In the real world two Ford Fiestas (for exaple) driven by same person will have slightly different figures from each other.

At least the figures are published. If you tie them down enough they wil have some relevence: if only against each other.

In reply to a post by hk11:

At least the figures are published. If you tie them down enough they wil have some relevence: if only against each other.

And the WiFi numbers are calculated and scientifically determined, for example if you have a WiFi 4 (N) device with single antenna it does 72 Mbps, and if you have two antennas it is 72+72 = 144 Mbps. That has no real relationship with throughput on fixed Ethernet however.

Compare the WiFi 4/N 144 against a WiFi 6 1200 or 1400 and you can see the relative improvement. You can’t compare WiFi 6 1200 with Ethernet 1000.

Car efficiency figures may be published but are of no use as they do their best to shown the most efficient value they can achieve.

Bears no resemblance to real world figures.

Some PHEV show figures of over 300 mpg. Funny that! Then when the battery has run down the fuel consumption is worse than a non PHEV and more pollutants out of the tail pipe.

In reply to a post by broadband66:

Some PHEV show figures of over 300 mpg. Funny that! Then when the battery has run down the fuel consumption is worse than a non PHEV and more pollutants out of the tail pipe.

The research group Emissions Analytics found that, and others, which along with the DieselGate stuff, caused the likes of EU to change the method they test cars... still not sure its great.

The analogy with WiFi is that the published number is only comparable to another published number, not to actual throughput achieved. UNLIKE with Ethernet where 1000 Mbit/s ethernet will give 940 Mbit/s throughput on a TCP/IP connection.