With , one expects, predictable consequences !
Correct.
Hello, I am unable to flash this firmware onto my pi 4!PhilE wrote: ↑Tue Jul 02, 2019 8:09 amHi all,
Download the archive from here: https://drive.google.com/file/d/1PXwrnh ... sp=sharing
With the zip file on the Pi:Code: Select all
$ unzip vl805_update_0137a8.zip $ chmod a+x vl805 $ sudo ./vl805 -w vl805_fw_0137a8.bin $ sudo reboot
Code: Select all
pi@raspberrypi:~/Downloads $ sudo ./vl805 -w vl805_fw_0137a8.bin
VL805 FW version: 00013701
* failed to open 'vl805_fw_0137a8.bin'
Code: Select all
pi@raspberrypi:~/Downloads $ cat /etc/debian_version
10.1
pi@raspberrypi:~/Downloads $ cat /proc/device-tree/model
Raspberry Pi 4 Model B Rev 1.1
Code: Select all
pi % ~ md5sum vl805_update_0137a8.zip
431938eb649fde38ee00c3ec87912344 vl805_update_0137a8.zip
pi % ~ unzip -l vl805_update_0137a8.zip
Archive: vl805_update_0137a8.zip
Length Date Time Name
--------- ---------- ----- ----
423012 2019-09-12 12:40 vl805
96984 2019-05-03 10:09 vl805_fw_013701.bin
98712 2019-09-05 11:38 vl805_fw_0137ab.bin
--------- -------
618708 3 files
RFBomb wrote: ↑Sat Sep 21, 2019 1:43 pmHello, I am unable to flash this firmware onto my pi 4!
Following the steps above, I get a message showing this:When I tried to use the -v to verify my current firmware to the 13701 file, that declared it was correct. I was also able to reflash the pi using the 13701 file. But it fails 100% of the time attempting to open the 137a8 file.Code: Select all
pi@raspberrypi:~/Downloads $ sudo ./vl805 -w vl805_fw_0137a8.bin VL805 FW version: 00013701 * failed to open 'vl805_fw_0137a8.bin'
Not sure what is going on, any ideas?
Here is my board infoCode: Select all
pi@raspberrypi:~/Downloads $ cat /etc/debian_version 10.1 pi@raspberrypi:~/Downloads $ cat /proc/device-tree/model Raspberry Pi 4 Model B Rev 1.1
jamesh wrote: ↑Fri Sep 13, 2019 6:37 amBeing inaccurate is ok (just) , blatant lying is unacceptable.
You clearly know very little about the Pi.
1. Not been around for 8 years.
2. Always had HDMI.
3. Unless you work for either Pi or Broadcom, your suggestions were probably not even noticed. If indeed you actually made them. And I'm pretty sure you don't work for either of those given your subject knowledge level. That said, how interesting would it be to know what you suggested, who you suggested it to, and when.
4. There are other avenues to reduce heat than reducing die size.
So, since it's clear you are now just making stuff up, I thinks it's time for a holiday.
You appear to have flashed the flashing utility to the EEPROM. No wonder your USB devices don't work any more.czarship wrote: ↑Mon Sep 23, 2019 4:01 amI've tried to update the firmware on my Pi 4 from within Raspbian via PhilE's instructions on Page 4.
However, I continually received the error ""failed to open 'vl805_fw_0137a8.bin'" when running the command "sudo ./vl805 -w vl805_fw_0137a8.bin"
Instead, it accepted the command "sudo ./vl805 -w vl805" and told me to reboot to activate the update.
Afterwards, no matter which OS I boot into, none of my USB devices (simple keyboard & mouse or USB controllers) are registering. All OSes boot but don't recognize any input from USB devices and my keyboard doesn't appear to be receiving power (caps lock key doesn't light up LED on keyboard).
I'm not sure what happened, but is there any way to troubleshoot or revert these recent changes given I can't interact with the OS via USB devices? I can ping the Raspbian OS from my Windows desktop but SSH is disabled so I can't remote in.
Any help would be greatly appreciated, thanks!
Unless you're ordering dozens of cases or more in bulk, $15 is only a premium price tag when compared to the official case. And the official case, with zero provision for cooling or ventilation, causes the Pi4 to throttle so badly that it really ought to be recalled. The FLIRC case absolutely does what it is claimed to do, lowering the temperature by 20C or more compared to no case and preventing throttling under real-world workloads, and I imagine you verified that for yourself when you 'saw what kind of difference it makes.'
The interface between the processor and the case doesn't have to achieve the same low thermal resistance (which is measured in degrees C per watt) as a 100W CPU's cooling system does, because the Pi4's CPU uses just ~6W max. Rather, the difficulty is making sure that heat is well dissipated from the heatsink to the surrounding environment. So a mirror finish and a very thin layer of thermal paste, which you expect for a 100W CPU's heatsink, simply aren't relevant here.Imperf3kt wrote: ↑Thu Sep 26, 2019 3:47 amThe walls are thin, the top is made of a removable rubberised plastic, the block that touches the SOC actually sits about 1.5mm above it and relies on some thick thermal pad to make contact and the surface isn't even flat. The block itself is not solid either, it's a mere cavity... There's less aluminium here than my current Pi3b cooling method - a 25*25mm block of aluminium.
You only need to show 1 value. The GPU and CPU is the same chip with the metal heatspreader. The value from /sys/devices/virtual/thermal/thermal_zone0/temp and vcgencmd measure_temp is from the same temperature sensor so there is no way to determine GPU or CPU. There is a difference in the implementation for how the value from the internal sensor is converted to a temperature reading which is why they slightly differ.
You **might** get lower temperature without the heatsink on the SoC. The issue is that the heatsink is stops the cooler air reaching the heat spreader on the SoC and depending on the quality of the thermal pad some people have found it better to remove the heatsink.genodesiful wrote: ↑Sat Sep 28, 2019 2:16 amI came up with this makeshift solution, just so I could play with it.
What does the "eeprom" do? I did update/upgrade, rpi-update (upgrade? You know what I mean, firmware!), but I haven't heard of that one.... Also, what would the advantage be of angling the fan? I'll give it a try and see if I get better cooling. One of the long ends of the box is almost completely cut away, and there's a slit on the cables side, as well as a small window on the SD card end, so the air goes in the top and out of the ends. You can see the alterations below, I've (somewhat) sealed the fan to stop blowback and to push the air out of the ends of the box, which is the path of least resistance. I've stressed it several ways now with various Python codes, and it hasn't risen much above 50C, which is pretty awesome, esp. considering the only reason I hesitated buying a 4B was the talk of heat and throttling problems!And update Raspbian and install the boot eeprom updates (if it doesn't do it automatically).
Thank you very much for that info. Very very useful IMHO.jamesh wrote: ↑Sat Sep 28, 2019 1:08 pmBoot eeprom information
https://www.raspberrypi.org/documentati ... teeprom.md
Replace the sticky pads on the Armor fan case with 0.5mm copper shim and proper thermal paste and you can knock another 10 degrees off your temperatures.