|
|
|
Hi everyone,
I have a Virgin Media M600 account and when I run speed tests they average around 290 mbps download. This is from my PC which is connected to the router by ethernet cable. However, I have a lower spec PC which is also connected to the router by ethernet but which gives download speeds that average 120 mbps. So my question is, is it reasonable to have lower download speeds on a lower spec computer?
Thanks in advance
|
|
|
|
Speed tests use the processor to run and so a less powerful processor may not be up to doing the speed tests if the specific speed test is processor heavy. Also, what sort of network card/wifi do they have as you may be limited by the maximum speed of the network adapter.
|
|
|
120mbit = 100mb ethernet.
Is the old PC being limited by the ethernet connection?
OPNSense
PiHole
Unifi for Wifi
|
|
Register (or login) on our website and you will not see this ad.
|
|
|
120mbit = 100mb ethernet.
Is the old PC being limited by the ethernet connection?
That's not correct if it was 100Mb ethernet you would see around 90Mb on speedtests.
If the PC is slow enough it will affect the speed shown on a speedtest.
Thanks
Dan
|
|
|
So my question is, is it reasonable to have lower download speeds on a lower spec computer?
Yes it is reasonable, and particularly apparent if the machine is older and perhaps has other 'bloat' like certain antivirus and antimalware software which will further hinder matters.
I would expect a relatively new machine, with a fresh build can easily maintain up to a gigabit in both directions, without much load on the CPU, but yes older machines may struggle.
|
|
|
|
I see this all the time on the Virgin Media forums and it's something ISP's should make people aware of. Especially when users are now getting the VMO2 boost to speeds. A lot of the new 1GB customers complain they are not getting the full speeds on, in technology terms, ancient kit.
A bit of PR around managing user expectations would not go amiss.
|
|
|
I see this all the time on the Virgin Media forums and it's something ISP's should make people aware of. Especially when users are now getting the VMO2 boost to speeds. A lot of the new 1GB customers complain they are not getting the full speeds on, in technology terms, ancient kit.
A bit of PR around managing user expectations would not go amiss.
I do believe Virgin try in a few pages on the site although it's too basic to explain why. I know they try and explain when you call them however the person who you normally calls assumes there is no fault and always blames the user, when a lot of the time isn't true.
I know they say that to me when i call about congestion issues, nothing to do with the router, but no its always lets swap the "Hub" as it magically solves all the lack of capacity everywhere from my house to the data centres.
Everyone says to me why don't you change provider if its slow WiFi instead of running ethernet and cables and the moment that phrase is said it deeply saddens me. They go I'm with provider "(BT)" with 36Meg down and 10 Up and i get signal at the end of the garden at full speed so why can't you get 100 meg in your whole house, i explain, they then go change to BT. I'm like, i know you get commission for every sign up but unless they do ultrafast then theres no way, especially if the prices are not competitive.
Its a big cycle of people not understanding the basics, trying to explain processors is a whole level of complexity on top. It would be nice if people didn't think virgin lied when they said WiFi is slow and unreliable. It would also be nice if people didn't think WiFi was their internet connection speed then we might get closer to average consumers understanding.
|
|
|
|
Many thanks to everyone. I think the solution is just an upgrade to the second PC.
Stay safe.
|
|
|
Many thanks to everyone. I think the solution is just an upgrade to the second PC.
Stay safe. Why upgrade for the sake of speed tests? If the old computer was or is satisfactory in real use you haven't got a problem.
The speed testing on the better computer tells you all is well with the service  .
Connections: OnePlus 8 Pro on Three 4+ (LTE)/5G and at home Three Mobile, with (Three)ZTE MF286D router giving about 113/20Mbps.
===========================================================================
The price of liberty, and even of common humanity, is eternal vigilance. (Aldous Huxley version of the well-known saying)
Were there none who were discontented with what they have, the world would never reach anything better. Florence Nightingale (Cassandra: an Essay (1860 edition?)
|
|
|
|
True, if the computer is doing what they want satisfactorily then upgrading is unnecessary.
|
|
|
Good excuse though 
Bill
|
|
|
|
Pluralist, the reason I did speedtests on my second PC was that recently it had been buffering videos. I know that the processor etc would have some say in this but I just wondered why the speedtest was so much lower. Now I know that the PC specs do have some say on the speedtests thanks to you guys. I had just thought; computer, same thinkbroadband speed test, ethernet cable (both cat5e), same router, why not same result as it's only testing a speed along cables? Still seems a bit counter intuitive.
|
|
|
That's fine . The machine was not performing satisfactorily and you have found the probable reason .
The point I and those after me were making was that chasing speed for its own sake is a choice, not always a necessity. Many people don't think of that.
Connections: OnePlus 8 Pro on Three 4+ (LTE)/5G and at home Three Mobile, with (Three)ZTE MF286D router giving about 113/20Mbps.
===========================================================================
The price of liberty, and even of common humanity, is eternal vigilance. (Aldous Huxley version of the well-known saying)
Were there none who were discontented with what they have, the world would never reach anything better. Florence Nightingale (Cassandra: an Essay (1860 edition?)
|
|
|
May I just ask OP what CPU's are in both machines if possible?
Now is not a time to decide on upgrading as the market for components and prebuilt machines is all over the place at the minute, some components are ok and others are 1-3 times their RRP due to C*vid disruption.
Would love to look for the cheapest bang for buck alternative is all.
|
|
|
|
Redayejedi
Main computer has an Intel i7 4790K at 4GHz with 16 GB of memory
Secondary has an Intel i5 3470 at 3.2 GHz with 8 GB of memory
Both processors are getting long in the tooth now.
|
|
|
Both of those should easily cope with gigabit.
However third party security software (assuming you are running windows) can break everything.
22 years of broadband connectivity since 1999 trial - Live BQM
|
|
|
|
I do go a bit overboard with security software and backup software; AVG Internet Security, AVG TuneUp and Malwarebytes with Acronis True Image backing up to five drives - I once lost a whole computer when drive C: became corrupted so now I have 9 drives of various sizes either storing data or backing up. The secondary computer only has two drives but the same software. Geek or what?
|
|
|
Redayejedi
Main computer has an Intel i7 4790K at 4GHz with 16 GB of memory
Secondary has an Intel i5 3470 at 3.2 GHz with 8 GB of memory
Both processors are getting long in the tooth now.
I have the same CPU as your main computer in my desktop in the bedroom. It handles a gigabit without any issues.
I got rid of an i5 3470 about a year ago. It also handled a gigabit throughput.
It might be your Ethernet NIC or more likely some kind of security software or anti virus.
|
|
|
|
Thanks for that info. I'll disable my security software and test it then. I'll also check my NIC but I'm pretty sure they are both GB.
|
|
|
You could always run a speed test just between the 2 machines across the LAN. Before and after.
OpenSpeedTest
|
|
|
Like others have said, your hardware is capable of more speed than you're getting.
I'm willing to bet both these system are running mechanical hard drives? I'm asking this because I have upgraded many older systems to SSD's giving them a new lease of life, my game server has a lower spec to your main CPU wise but is capable of downloading at over 60MB/s when downloading from Steam. I think full gigabit speed is pushing it though imho.
Your CPU/RAM/NIC are all capable of higher speed.
|
|
|
|
Having multiple backups is a good thing, especially so if you have important data. Myself, I have drives in a RAID setup to help prevent data loss if a drive fails and I also backup regularly overnight to an off-site server using Duplicacy.
As others have said, it could be your 'security software' that's causing a slowdown. I can't comment on AVG Internet Security as I use something else, although primarily that looks for viruses and malware instead of also acting as a firewall or whatever else some of the security software out there does in addition.
|
|
|
|
On the main page of TTB there is a link to download files - try downloading one of these and monitoring the data download rate using task manager and see if it is higher (this takes out the main processor intensive tasks of speed tests). Be interesting to see if you can get over 120Mb/s doing that.
|
|
|
|
Well:
Pheasant, I used the TB Speedtest and the Open Speed Test within a minute of each other and got 267 Mbps and 334 Mbps a big difference.
Redayejedi. Both my drive C:'s are SSDs
ian72 I downloaded the very large test file and monitored it with Task Manager as you suggested. Briefly, it touched 302 Mbps
I am no wiser!
|
|
|
I'm on Virgin Media 1 gigabit. With any web based speedtester it maxes out my CPU so I never see anywhere near 1 gig throughput.
If I use a native Windows app or a command line app, such as the ookla ones on speedtest.net, then I see 900+ mbps.
https://www.speedtest.net/my-result/d/8d87ddd2-1379-...
|
|
|
|
Good point.
The OP started out with a simple question that has now evolved into a bit of a forensic diagnostic search for the cause (as these things inevitably tend to go). There are a couple of facets to this, to my mind anyway:
- speed test results are not a consistent beast. Rather than focussing on individual figures, it is preferable to look at a 'basket' of speed tests from various servers and also various speed test tools and at various times to get a broader view of performance. The law of averages.
- speed testing is all rather pointless and 'artificial' - it is rare to be running a high-speed domestic broadband connection consistently flat out. There are other factors that are just as if not more important, such as performance variation at different times/days, how much latency and jitter is present at 'idle' as well as when loaded, are there buffer-bloat issues, etc.
- to your point, the nature of the speed test itself, stresses not only the network but the host hardware - something perhaps OP may not have appreciated previously. As you say web-based speed tests are far more resource hungry than other forms of tests such as native or command line. Therefore how closely does the testing tool match reality? Answer, not always very close at all.
- content comes from all over, and isn't always readily available from the nearest CDN or cache, therefore the real world performace may be just a fraction of what a headline Speedtest suggests is possible.
- no matter how much 'we' collectively deny it, we are in some thrall of our speed test results (and ping times!)
|
|
|
|
If your laptop is old, then the ethernet will be Fast Ethernet: 100 Mbps max, on the laptop.
Also wifi will always be less than optimal.
|
|
|
If your laptop is old, then the ethernet will be Fast Ethernet: 100 Mbps max, on the laptop.
Also wifi will always be less than optimal.
How does one get 267 and 334 Mbps from a Fast Ethernet port? 🙃
|
|
|
If your laptop is old, then the ethernet will be Fast Ethernet: 100 Mbps max, on the laptop. Also wifi will always be less than optimal.
Not always the case, my old Thinkpad T40 had a gigabit Ethernet port. The WiFi was only 802.11b so back at the start of WiFi around 2003.
Gigabit Ethernet has been around a LONG time. Some retail brands fitted 100 Mbps to save money, not for any other reason.
22 years of broadband connectivity since 1999 trial - Live BQM
|
|
|
How does one get 267 and 334 Mbps from a Fast Ethernet port? 🙃
Broken AV software causing caching effects and breaking a web based speed test? I've seen some AV product convince a person was getting 8 Gigabit from their ADSL internet
22 years of broadband connectivity since 1999 trial - Live BQM
|
|
|
|
Yeah. That’s just pure [censored] isn’t it. I mean fast.com just told me I’m getting 1.2 Gbps. Utter tosh.
|
|
|
I only use fast.com in a private / incognito browser; on computers without third party AV/security suite. I consider those suits “snake oil” from an industry that knows its heading for irrelevance.
22 years of broadband connectivity since 1999 trial - Live BQM
|
|
|
|
They did a test via a standard test download rather than a speed test that hit 302Mbps - this wouldn't be impacted by AV as task manager is specifically looking at throughput on the interface which suggests they do have a gigabit NIC.
|
|
|
So my question is, is it reasonable to have lower download speeds on a lower spec computer?
Late, but the simple answer is "maybe". 
In short, it comes down to how much resources your hardware has, and that's not necessarily down to 'fast' or 'slow' hardware, so much as system config too.
Run a Userbench benchmark and see what aspects of your system are slow (if any). High levels of background CPU will be a telling sign of your system resources being leeched from apps, programs etc. It might be that it'll come down to faster hardware, but it might come down to servicing the software running on said system and making it less bloaty.
|
Pages in this thread: 
Print Thread
