Well, you can always arm wrestle to see who gets the new computer!
We need to figure out what else we have in common. I’ll list everything I can think of which may be a factor (this isn’t a brag post, honestly!)
Hardware
My PC is essentially this one, but with an nVidia video card:
The 6 cores are overclocked to 4.19Mhz (just checked) (anyone else overclocking? if so, we should check USB/PCI speed etc. settings)
Gigabyte X58 UD9 Motherboard
nVidia video card
12GB RAM
SSD
Other Hardware
All MCUs and similar devices are connected to a pair of powered hubs.
JTAGICE-III compatible AVR programmer/debugger
Software
Windows 7 64 bit, ultimate
Visual Studio 2010 Ultimate (via an MSDN subscription)
Gadgeteer and FEZ SDKs, drivers
Gadgeteer module builder templates
Netduino SDK and drivers
AVR Studio 5.1
Network
Home network, gigabit, my own domain controller
That’s probably enough for the first round.
Pete
My assumption is that it is related toe the USB host controller, which is usually built into your motherboard’s chipsets.
Another thing is, can someone try to trace USB somehow? We have some expensive analyzers here but the problem is that non one our PCs show this, not even the super machine we just got.
Lenovo M58 - Intel Core Duo E8400 3.0GHZ - 8 GB memory
Window 7 Enterprise 64 Bit
Visual Studio 2010 Premium SP1
I have VMWare installed on my Lenovo, and I am using WMWare Fusion on my IMac.
It seems like a long shot, but is it possible that having a virtual machine support program installed, with its drivers inserted in the USB chain, is causing the problem? This could/would effect the timing.
@ Pete: Do you have a virtual machine support program installed?
@ Mike
Nothing on this machine. I don’t do anything with VMs on this box.
@ Gus
USB trace: I haven’t the faintest idea where to start or what to look for. What do I need to do one?
Pete
Per my email on the previous page I’ve tested with a sequence of VMs. That’s how I got it to work.
I’ll do a windows ETF trace on the Hydra today - I offered last week.
Phil
How about the Power Supply??? could that be your in common component…
maybe the type and the wattage of the PS could help.
and last not least if you have nvidia USB EHCI chipset go here and get the hotfix…
I hope the above helps.
Jay.
I doubt it is the power supply. A fully loaded spider works fine while a Hydra with nothing except the USB client module has problems.
My power supply is more than enough 
Corsair AX850 850W efficient power supply
I wouldn’t expect that to be the issue in any case, as my hubs are all powered.
USB chipset: Onboard USB 2 and NEC USB 3 chipset Intel X58 + ICH10R. nVidia is only for my video card.
I’m not going to discount something incompatible on the board; the UD9 had a number of “early adopter” issues. Like I said, though, it works fine with the Spider, Netduino, Panda, and all my MicroElektronika boards. Same cables, same hubs (also tried different).
Pete
FLASH…
To test my latest theory, on the possible impact of VM software USB drivers, I did a little experiment.
I have the latest version of VMWare running on my machine.
On my Lenovo, I first inserted the Hydra and pinged it with MFDeploy. Nothing happened.
I then went to the Windows 7 Service Manager and stopped the VMWare USB Arbitration Service. This caused the VMWare Server Service to also stop.
I then inserted the Hydra, and tried to ping it. I got back a response from the Hydra TinyBooter. I do not have firmware installed on my Hydra, only the latest TinyBooter.
When I tried to deploy the latest firmware, deployment started but I immediately got a message back saying “Error: Unable to deploy to device”.
I was then still able to ping the TinyBooter.
When I restarted the VMWare USB Arbitration Service I was no longer able to ping the TinyBooter.
I think this is an interest data point!
I know at least one other person with the problem had virtualization software loaded.
Hi Mike,
that’s interesting indeed, can you please try the following:
after starting the VM in this mode as you mentioned you are no longer able to ping the Hydra now go to the Device Manager and right click on the parent node and choose scan for hardware changes…
when it all done please try and ping the Hydra see if that works…
Cheers,
Jay,
Jay:
I did not start a VM in my experiment. VMWare automatically starts a bunch of services at boot. I was turning off the service that did the USB Arbitration.
Steven / Gus,
I’ve got USB traces for you. I used the Windows Logman tools and Netmon so there is no need for expensive analysers … all free.
How should I get the files to you? I can email the info & tools info and give you download links for the files.
Phil
Those do not show everything like expensive analyzers do 
Please email to me gusi@ GHI domain
I have tested the ping with MFDeploy on 4 PC (Real one, not VM). On those 4 PC only 2 can Ping the Fez Hydra. These two have less than 3 Ghz in CPU.
jango
My new observation was about a machine with VM software installed, not running within a VM.
But I still believe it is speed related.
[quote]Gus
Support
Those do not show everything like expensive analyzers do Smiley
Please email to me gusi@ GHI domain[/quote]
The Microsoft tools show everything but bit timing. The tracing is built right into the HUB driver, so they’re also non-invasive.
The email should be sitting in your inbox.
If you need something more than the Microsoft traces I have access to both a logic analyser and a storage oscilloscope,but the scope is only a 25MHz dual trace (I’m away from my office on business, and my good tools are back there). The logic analyser will give you bit timing to nanosecond resolution but they’re a b*stard to decode the USB (it will decode to the byte level).
Phil
My fear! The logs didn’t show anything we can work with. We would need to step int eh USB drivers to see what is going on. Anyone with a spare PC we can borrow? If you do, you know we will have the 80W laser soon and I am sure we can run few freebies for you 
@ Gus:
Last night I realized that the Hydra I tested with yesterday had the special TinyBooter that Aron had sent me last week installed on it. The one he said not to try to deploy the firmware. That could explain why I was unable to deploy the firmware.
I will be trying again at work today, with the latest released Hydra firmware, to see if shutting off the VMWare services cures the problem on a 3.0GHZ machine.
It might be productive to download a trail version of VMWare, or the free VMWare Player, install it on your new machine, and see if it introduces a problem. You just have to install it.
I think there might be two issues; one with pure processor speed, and another with VM USB arbitration.