I've an SFF desktop (HP 8200) and i've a USB wireless mouse (unbranded) that occasionally becomes unresponsive and requires the unplugging and replugging of the dongle.
I decided to give the PC a boost with a GT710, recommended on many forums for its low power consumption.
When it's in the PC though the mouse is conking out constantly to the point of unusability. Logitech keyboard is fine
Any idea why that may be? I like the mouse, it's ergonomic, but one of them is going to have to go unless there's a potential solution.
As I type I wonder may it be worth trying the dongle in a hub so it's further from the card.....?
Try changing the video resolution. It's possible interference from the HDMI/Displayport signalling is interfering with the 2.4GHz mouse. Changing resolution (or frame rate) would change the harmonics of the signalling frequency.
That's worth a try too. Also USB 3 can interfere with cordless mice/keyboards. I don't suppose you have any USB 3 devices? If you're using a hub, try a USB 2 one.
From previous calculations, HDMI 1920x1080x50Hz gives a pixel clock of ~750MHz
I've not run the numbers for higher resolutions and frame rates, but I think you'd need something like 2560x1440 @70Hz using 10bpp to get to the 2.4MHz ballpark, then maybe dual-link would come into play and alter things ...
See if you can make a trivial adjustment downwards on the video card clocks, for fun.
*******
Emissions from DP or HDMI, would be a function of resolution, as already computed by others. The GT710, from the year 2014, may still have a VGA connector (the industry dumped VGA recently, which is why the only way to get it now is via "active" dongle adapters plugged to newer cards).
You could use the VGA connector on the faceplate, if the monitor has such a connection. VGA is only good for the lower resolution choices. Theoretically it might be driven at
2048 or so pixels, but the picture won't look good. That's why the digital format DP/HDMI are superior, in that they have no (apparent) issues with higher res. Maybe 1920x1080 is enough temptation on VGA. Or 1600x1200 might work. Anything lower will be fine.
The DP is 8B10B encoded, with sufficient transition density for clock recovery. The HDMI has a TMDS Clock+ and TMDS Clock- signal at up to 165, 340, 600MHz rate, and the clock would be the biggest emitter by about 20dB. The very last HDMI standard switched to 16b18b, so uses transitions for clocking. The GT710 isn't new enough for 16b18b.
Buggering with HDMI and DP resolution or refresh, to solve the problem, would be misery itself. Windows 10 recently removed support for some custom panel for making adjustments, as an example.
*******
I hope this is a household with "safety ground" on the wiring.
If you touch the metalwork on the chassis of the computer, do you get a slight tingling of a "shock-like" nature. The EMI prevention provided by a grounded chassis isn't worth a fig newton, but I have to ask, just to see if that is an aggravating factor. The house I was born in, has no safety ground, and the computer I installed there, had "shock potential" until I installed an auxiliary ground. The emissions control on the computer, is mainly provided by the PCB design of the motherboard (buried clock on inner layers).
I would guess the mouse is using 900MHz transmission and AM modulation, and is not a 2.4GHz Bluetooth and friends type of thing. There were even some HID devices operating for sure with AM modulation, at around 27MHz, because CB people noticed at the time. Over the years, there have been a few stupid ideas for such applications. But I understand some were done at 900MHz - I can't provide an example of such.
USB3 cabling emits at ~2.5GHz, and the broad sinx/x emission can impact Wifi at 2.4GHz. USB2 (it is claimed), has no RF emission of any importance... until someone finds an example where it does :-) But USB3 cabling (like to an external hard drive enclosure), that has wiped out wireless KB/Mouse before. Routing the air path of the RF signal for the HID devices on the other side of the computer, away from the USB3 cable, can provide some relief. It could be, while working on the computer recently, you moved an active USB3 cable closer to the air path used by the mouse signal. And the addition of the video card was purely coincidental and not the root cause.
*******
If the mouse worked at 2.4GHz, a cheap SDR (software defined radio) will not go to that frequency. If, on the other hand, the signal really is at 900MHz, that's within range of a RTL SDR (Realtek for the digital part, another company for the tuner up to 1700Mhz or so perhaps). It's too bad nobody makes one of those 6GHz front ends for peanuts, so an SDR could cover practically any computer problem you could come up with.
formatting link
I think your mouse is 900MHz - now, try and track down the mouse spec, then use Afterburner to detune the main video card clock. It could be the clock pins on the video card memory chips, which are emitting. FPGA memory instead of TSOP style (visible legs), might be slightly less polluting.
In the old days, we used to be concerned about chip emissions coupling into a heatsink. If the heatsink was electrically floating, there might be higher emissions. I don't know why nobody seems to worry about that any more.
That is the pixel rate, but the bit rate can be much higher, for up to
12 bits per colour and various colour schemes with 8/10bit encoding.
A local Wi-Fi is also likely to cause some issues, if Bluetooth and Wi-Fi hardware are co-located. Using a USB extension lead would be a cheap way of siting the dongle next to the mouse.
I might also use LatencyMon to check for OS and driver resilience to detect dropouts to Mouse Control:
That would be a steady clock on the clock pin of each video memory chip. Clocks are 20dB above random data patterns at the same rates.
We were taught by our Product Integrity, the important of burying clock signals on PCBs and how that pays off.
The question would be, why is the mouse that sensitive to other RF sources ? I thought the idea of using AM modulation went out yonks ago, with the 27MHz mice and keyboards. Modern wireless mice and keyboards use much more sophisticated modulation patterns.
I mean, think about it. When was the last time you saw an effect from plugging a video card into a computer ? I'm only guessing at the memory clock in this case, because it is a fixed clock signal, it's not suppressed in any way. Then it's a matter of the "loop size" as to how much coupling into the air occurs. If you bury the clock on an inner layer, the loop size is tiny.
What CPU & chipset is in the thing. It is a bit elderly but for 2D graphics and office work it should be fine on the Intel graphics.
Only when your are doing 3D rendering, bitcoin mining or gaming do graphics cards really matter. My i3770 2D graphics is faster than most dedicated graphics cards (3D graphics framerate is rubbish though).
If you want to give an office PC a boost the two most effective upgrades are double the RAM or add an SSD. Both make a significant difference to the speed of doing things - which one is better for you depends on the situation. Both are now quite affordable.
Worth a try. My fast laptop is quite capable of blinding Mifi dongles to mobile networks so I have a female USB socket on a 60cm extender lead. Something like this:
formatting link
Getting the RF receiver away from the hash of the PC might help.
I've got something similar though probably not of use to you.
Dual boot (Linux Mint and Windows 10), Logitech wireless combined keyboard and mouse, GT640 graphics. Works perfectly on Linux - same symptoms as you on Windows (used only to play Call of Duty).
Mine is on an extension cable and works a lot better for it, at least 4 metres from the computer so no screening or interference. I seldom use the mouse, but it was affecting the keyboard as well. the dongle is so tiny it cannot have much of an aerial after all and round the back of the pc which is mostly metal its hardly surprising. Brian
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.