I have two Identical HP DC7800 SFF desktops running XP Pro connected to a 10MB broadband cable link if I connect one at a time to the 10MB link and use speedtest.net it tells me that one downloads at 5.5MB the other at 9.7MB. I have checked the settings on the onboard gigabit NIC and they are identical. I cant see anything obvious thats different between the two PC's. Its driving my crazy does anyone have any thoughts on why this is happening?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations derfloh on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
To identical PC's one downloads at 5.5MB the other at 9.7MB
- Thread starter SJO34fl
- Start date
Presume they have identical NICs?
Make sure they both have the same (and latest) NIC driver.
Ping test both to the router to see that they give identical results.
Try assigning each the same IP address (in turn) and test.
Swap the network RJ45 cables; swap the ports they are using testing each time to see if the d/l changes.
Regards: Terry
Make sure they both have the same (and latest) NIC driver.
Ping test both to the router to see that they give identical results.
Try assigning each the same IP address (in turn) and test.
Swap the network RJ45 cables; swap the ports they are using testing each time to see if the d/l changes.
Regards: Terry
get tcp optimizer
start it and compare/optimize settings
M. Knorr
MCSE, MCTS, MCSA, CCNA
start it and compare/optimize settings
M. Knorr
MCSE, MCTS, MCSA, CCNA
goombawaho
MIS
Follow tf1's suggestions. That is SOLID troubleshooting thinking there.
There are so many variables that could affect the speeds, even with identical hardware, that it's almost not worth the time to investigate it.
Are they using the same cables? The same port on the routers? Do the cables fit snugly? Are they running the exact same software with the same configuration and CPU load at the time of testing? Are the speed tests being taken at exactly the same time, or different times of days? Are you running the speed test using the same web browser, the same revision number, the same plugins, the same versions of flash/shockwave/java/etc?
The biggest determining factor in how fast you can download from the internet is the speed and congestion of your internet connection. Your internal network and PC hardware plays some small role, but the most important factors are all well outside of your control and highly variable from moment to moment.
It's really a bit like saying "I had a 2008 Accord and it took me 20 minutes to get to work on Monday, but when my car was in the shop on Tuesday the Honda dealer gave me a loaner that was an identical 2008 Accord and it took me 22 minutes that day."
________________________________________
CompTIA A+, Network+, Server+, Security+
MCTS:Windows 7
MCTS:Hyper-V
MCTS:System Center Virtual Machine Manager
MCTS:Windows Server 2008 R2, Server Virtualization
MCSE:Security 2003
MCITP:Enterprise Administrator
Are they using the same cables? The same port on the routers? Do the cables fit snugly? Are they running the exact same software with the same configuration and CPU load at the time of testing? Are the speed tests being taken at exactly the same time, or different times of days? Are you running the speed test using the same web browser, the same revision number, the same plugins, the same versions of flash/shockwave/java/etc?
The biggest determining factor in how fast you can download from the internet is the speed and congestion of your internet connection. Your internal network and PC hardware plays some small role, but the most important factors are all well outside of your control and highly variable from moment to moment.
It's really a bit like saying "I had a 2008 Accord and it took me 20 minutes to get to work on Monday, but when my car was in the shop on Tuesday the Honda dealer gave me a loaner that was an identical 2008 Accord and it took me 22 minutes that day."
________________________________________
CompTIA A+, Network+, Server+, Security+
MCTS:Windows 7
MCTS:Hyper-V
MCTS:System Center Virtual Machine Manager
MCTS:Windows Server 2008 R2, Server Virtualization
MCSE:Security 2003
MCITP:Enterprise Administrator
goombawaho
MIS
Again: Follow tf1's suggestions. That is SOLID troubleshooting thinking there.
Identical machines SHOULD perform nearly identically. That difference is too far off to be non-statistically meaningful.
In other words - something be wrong (or different).
Identical machines SHOULD perform nearly identically. That difference is too far off to be non-statistically meaningful.
In other words - something be wrong (or different).
Identical doesn't mean absolutely identical. While they are the same brand/model, that doesn't mean that the software/drivers are the same. For that matter, it doesn't mean that they even have the same hardware. If you look here you can see that there are several NIC options on the DC7800 SFF:
________________________________________
CompTIA A+, Network+, Server+, Security+
MCTS:Windows 7
MCTS:Hyper-V
MCTS:System Center Virtual Machine Manager
MCTS:Windows Server 2008 R2, Server Virtualization
MCSE:Security 2003
MCITP:Enterprise Administrator
________________________________________
CompTIA A+, Network+, Server+, Security+
MCTS:Windows 7
MCTS:Hyper-V
MCTS:System Center Virtual Machine Manager
MCTS:Windows Server 2008 R2, Server Virtualization
MCSE:Security 2003
MCITP:Enterprise Administrator
Well I agree that there are a host of possible differences down to the firmware on the NICs, but if these were purchased at the same time, they should be pretty much identical. The difference between them is too high to put down to just the difference in one thing.
I would suspect an Ethernet port, bad physical network connection, a bad NIC or a software problem for the sort of differences experienced here.
Regards: Terry
I would suspect an Ethernet port, bad physical network connection, a bad NIC or a software problem for the sort of differences experienced here.
Regards: Terry
goombawaho
MIS
Yes tf1, the other poster is WRONG (kmcferrin). IF, IF, IF both PCs have the same motherboard, same BIOS, same hard drive, same network card, same NIC driver, same internet connection, same cable, same router, same modem, etc. the setups should yield EXTREMELY similar performance benchmarks. They are clones.
Now, if any of those things is not the same, you have a point Mr. kmcferrin
Now, if any of those things is not the same, you have a point Mr. kmcferrin
The user says he is connecting them one at at time to the "10MB link". My guess is that he/she is already using the same cable and swapping it between the two PC's one at a time.
Even if the hardware is "exactly" identical, kmcferrin's point is that there are other factors that come into play:
1. NIC driver - different versions can show different performances
2. Network congestion - this can vary at any given instant, as well as the path in which a data packet may travel
3. Speedboosting - some ISP's (like Time Warner's Turboboost) have an option that temporarily increases the speed of a file download for the first few seconds before dropping down to the normal speeds. Performance here is highly inconsistent.
4. Windows performance - A PC bogged down with spyware, virus's, trojans, etc. might be hijacking some of the bandwidth and/or tying up the CPU
The first three are highly unlikely since the test results are very different from each other, but they are points to always consider. However, #4 might very well apply in this situation assuming both PC's are using the same ethernet cable and switch port. And it doesn't stop there - a lot more could be added to the short list above. That's just the tip of the iceberg. It would be easier to narrow it down if we had more info from the OP.
~cdogg
"Insanity: doing the same thing over and over again and expecting different results." - Einstein
[tab][navy]For posting policies, click [/navy]here.
Even if the hardware is "exactly" identical, kmcferrin's point is that there are other factors that come into play:
1. NIC driver - different versions can show different performances
2. Network congestion - this can vary at any given instant, as well as the path in which a data packet may travel
3. Speedboosting - some ISP's (like Time Warner's Turboboost) have an option that temporarily increases the speed of a file download for the first few seconds before dropping down to the normal speeds. Performance here is highly inconsistent.
4. Windows performance - A PC bogged down with spyware, virus's, trojans, etc. might be hijacking some of the bandwidth and/or tying up the CPU
The first three are highly unlikely since the test results are very different from each other, but they are points to always consider. However, #4 might very well apply in this situation assuming both PC's are using the same ethernet cable and switch port. And it doesn't stop there - a lot more could be added to the short list above. That's just the tip of the iceberg. It would be easier to narrow it down if we had more info from the OP.
~cdogg
"Insanity: doing the same thing over and over again and expecting different results." - Einstein
[tab][navy]For posting policies, click [/navy]here.
If after running through the tests mentioned previously (swapping cables, ports, checking drivers and firmware, and ensuring tests are done at similar times of the day with the same speed test site), then number 4 becomes a significant possibility.
But I'd run all the network tests first because a bogged down PC should have symptoms other than just a tardy connection speed.
Regards: Terry
But I'd run all the network tests first because a bogged down PC should have symptoms other than just a tardy connection speed.
Regards: Terry
goombawaho
MIS
CDOGG - I already discounted different hardware and drivers as a reason. Nobody was talking about spyware. Let's not LOOK for reasons for machines to differ. We should focus on reasons that they would be the same if they are truly clones.
1. NIC driver - ASSUMING same - DISCOUNTED
2. Network congestion - ASSUMING testing from same download site
3. Speedboosting - Assuming same ISP, thus similar performance on two tests done back to back.
4. Windows performance - This one I didn't think about as I was thinking these were NEW PCs. Definitely would want to compare PCs that are clean and have similar items running on them when windows boots.
1. NIC driver - ASSUMING same - DISCOUNTED
2. Network congestion - ASSUMING testing from same download site
3. Speedboosting - Assuming same ISP, thus similar performance on two tests done back to back.
4. Windows performance - This one I didn't think about as I was thinking these were NEW PCs. Definitely would want to compare PCs that are clean and have similar items running on them when windows boots.
Yet the tests runs clearly do differ in some way or else the results would be more consistent. The OP was curious as to why they would differ, and since the largest variable that changes between tests is the PC itself, it is not unreasonable to expect that there is some difference between them.
We're making the point that two PCs that are the same make, model, and OS still have an enormous amount of potential variability between them. That may or may not be the cause, but it's certainly worth looking into. With the limited amount of information that the OP has provided it is impossible to make a definitive determination, so we're simply pointing out options for him to investigate.
________________________________________
CompTIA A+, Network+, Server+, Security+
MCTS:Windows 7
MCTS:Hyper-V
MCTS:System Center Virtual Machine Manager
MCTS:Windows Server 2008 R2, Server Virtualization
MCSE:Security 2003
MCITP:Enterprise Administrator
goombawaho,
That is not necessarily true, which is why I listed it as a possible cause. I have Time Warner's Turboboost. If I run a speedtest test twice on the same PC back to back, I can get very different results. Sometimes it is over 20 Mbps. Other times it may only be 13-14 Mbps. I always had consistent results when I had the standard 7 mbps plan, but when I bumped up to the elite Turboboost package, I started noticing the difference. This is expected though as the technology is not perfect.
~cdogg
"Insanity: doing the same thing over and over again and expecting different results." - Einstein
[tab][navy]For posting policies, click [/navy]here.
- Status
Similar threads
- Locked
- Question
