@jazzmonger
[OT] For 199.00 € my son happily housed HASS on a blazing fast refurbished laptop w/ I7 + 8GB RAM + USB3 + 256GB Internal_SSD. Display, keyboard and UPS are free addons 
Today paying 50% more one can find a Intel_I7 + 16GB RAM + 500 GB SSD refurbished notebook.
All true, but do not forget the power consumption considerations. Has anyone a clear view on this. E.g. raspi 4 4GB + SSD versus laptop of Intel NUC?
2 Likes
Depends on the NUC.
I am powering a mini PC with Pentium N5000 CPU with 4GB of RAM and 2 SSDs through a POE splitter (12W) which is lower power than a rPi4 can run on (it requires a 3A-5V to boot so 15W) and it performs 4-6X faster than a rPi4. The rPi4 is not power efficient at all. The idea that it consumes less power is just false perception. It is just cheap. If you want the same performance you can get yourself a compute stick with a cherry trail CPU which is powered with less than 5W, 1/4th of the rPi4. Cost is another story…
2 Likes
And you’ve downloaded the latest firmware from github and put it on the boot partition?
The pi 4 does not require 3A/5V it is advised.
It’s a assumption that your setup or a compute stick uses more equal or less than a pi4.
Unless there is proof of all of this(except the speed of a laptop or nuc)
Advising someone on just assumption isn’t really advise.
It is more than assumptions. It is based on what I have observed and tested. Thanks for your inputs though. The rPi4 has problems under load with 2A/5V and is prone to rebooting. Under load it benchmarks about the same as a compute stick which never consumes more than 7W. Likewise my mini PC is the N5000 never consumes more than 9W under stress with 2 SSDs, averaging 8W which the rPi4 easily hits with nothing attached with less than 1/4th the performance. The temperature from all the testing is also another data point. Under stress, my passively cooled N5000 never exceeds 65C. The same stress program on the rPi4 with an active cooler reaches 75C. The TDP of the intel chips are 3W for the Z and 7W for the N, the rPi’s is also 7W but power regulation on the rPi seems to be poorer causing power spikes/burst. At idle the bare rPi4 consumes ~3W, the mini PC with the SSD consumes 0.7W. Essentially the rPi consumes the same or in some situations a lot more power running much slower and generating a lot more heat, causing throttling which my mini PC never does. The rPi4 is cheaper by 25-30% when considering the case, cooling, power supply and storage to match the mini PC.
References which matches my observations:
https://www.raspberrypi.org/documentation/hardware/raspberrypi/power/README.md
https://androidpctv.com/review-beelink-s2/
And some benchmarks with links I collected when I had to make this decision, the N4100 is the same chip as the N5000, just a lower clock:
2 Likes
Considering the documentation states 600ma (3 Watt) this is for the whole board.
A tdp of a processor cannot be considered the whole use of a system.
That is also stated in the documentation of raspberry hdmi WiFi Bluetooth etc etc also use power.
Also the the pi 4 has more integrated in the broadcom chip (Lan) witch also contribute to a more power consuming chip.
Temperature wise I have my pi 4 passively cooled.
The highest I have ever seen in my history was around 55 degree’s. Do not have a clue what the load was at that time but it idles at 43 degree’s passively.
That is all with a ssd, usb powered usb hub with 3 devices plugged in and a rfxcom plugged in the pi directly.
I agree, the mini PC is also for the whole PC with wifi, ethernet and 2 SSDs… it consumes 0.6W vs ~3W for the rPi at idle. Under load they are much closer but the mini PC is 4-10x faster. The point is that the rPi4 consuming less power than the NUC under any circumstance is a myth. In general I noticed a magic number of 4. It is 4x slower at the same power or consumes 4x more power running at the same speed and the same applications than a current equivalent PC. If you look at the benchmark table, the rPi4 is really closer in performance to the atom Z8350 (although arguably still slower in various benchmarks) compute stick or board which consumes 3-4W under max load, ~4x less that the rPi4. This was in response to the power question… it’s not even close. It’s 4x lower. I did try by the way to power the rPi4 with the same POE splitter (rated for 5V 2A) and it would crash and reboot randomly even with nothing attached under various load. My mini PC running (12V 1A) never does even under stress test. (there is a linux package called stress). I am measuring power with a smart meter.
I would venture to say that if you run exactly the same programs (home assistant for example) and equivalent OS (rasbian/debian) with the equivalent RAM(4GB) and peripherals SATA SSD on the board the same generation NUC (gemini lake) will consume significantly less power than the rPi4, likely 4x less. (ie 1W vs 4W)
1 Like
Now showing actual data as I went to pull a spare rPi3B+ which supposedly consumes a little less power than the rPi4 and connected it right next to my mini PC with the exact same POE splitter (Trendnet TPE-104GS) on the exact same POE switch (unifi 24 pro) and displaying the power going out.
The mini PC with 4GB of RAM is on the port called “Spare2” and runs proxmox with a VM in which an instance of Home Assistant is running. Wifi and BT are on but not used, gigabit ethernet is connected, 2 SSDs connected, one through the internal SATA 60GB Crucial C300 and the other through a USB-SATA cable also a 60GB Crucial C300. It shows 3W.
The rPi3B+ has 1GB of RAM, runs raspbian booted from the USB SSD also with an instance of HA with the same configuration (pretty much virgin) with no wifi or BT and only one SATA SSD (Crucial C300 60GB again) connected through an exact same USB-SATA cable from Sabrent. It is connected to port7 and consumes 7.1W.
I let them settle for 5min before collecting the data and I am pretty sure the splitter itself has an efficiency factor (can likely shave 1 or 2W for each to get the actual device consumption) but this should give you the idea. We are talking only a couple of watts here but it illustrates the false perception that the rPi is energy efficient. It really isn’t. It is actually much less so than a NUC.
3 Likes
sorry all, long thread and just want to double check, before i start…
which process do i follow to boot an RPI4 (8GB) from an SSD, using HASS OS
i got the following board and M.2
Pi 4B only takes 1A so 5W when it’s running.
Not quite, from the reviews link above:
The rPi4B idle eats 3.4W while the mini PC with a few more peripherals takes 0.6W.
Under load, the rPi4B uses 8W while the mini PC peaks at 10W. This corresponds to my observation as well. Under any equivalent usage case (running the same programs), the rPi consumes 4x more power than the intel equivalent. When pushed to full load, they consume about the same power but the mini PC runs ~4x faster and can do ~4x more tasks simultaneously. Notice from the review that not a single of the tested mini PCs consume more than 1W at idle. Not AMD, not the intel NUCi7.
1 Like
See this thread
Its should work as long you run the procedure correctly
Hi,
since this is a bit off topic, but rather fundamental info, could it be you moved this to a separate thread, and continue there? I really think community members should easily find the power consumption comparisons easily, and not like I just did by chance…
especially since
snippem s observations dont come close to what I notice locally, and this is a calm moment in the life of my Rpi4 4gig:
very often nearing the 75-80 C point…especially in the summer 
so please dedicate a thread to the topic?
Feel free to start a new one. I am not a mod here so I can’t split the topic.
sure, but it was almost completely copied from another thread which I cant find at the moment… my apologies to the original coder.
edit found it: Remove (border) space of cards - #14 by hajo62 it was @hajo62 
type: picture-elements
image: /local/homeassistant/homeassistant.png
style:
'#root > bar-card':
$: |
ha-card {
background: none;
box-shadow: none;
}
bar-card-name {
color: white;
font-size: 100%;
font-weight: normal;
margin: 4px;
}
bar-card-currentbar {
border-radius: 7pt;
}
bar-card-backgroundbar {
border-shadow: 2pt;
margin-top: -2pt;
margin-left: -2pt;
border-radius: 7pt;
border-style: solid;
background: rgba(255, 0, 0, 0.8);
}
bar-card-value {
font-size: 125%;
color: white;
font-weight: bold;
margin: 4px;
}
elements:
- entity: sensor.rpi_last_boot_macro
prefix: 'Last boot: '
style:
color: white
font-size: 90%
left: 65%
top: 25%
title: text
type: state-label
- type: custom:bar-card
animation:
state: 'on'
columns: 5
direction: up
width: 10px
height: 90px
positions:
icon: 'off'
indicator: 'off'
minmax: 'off'
title: outside
value: outside
name: outside
style:
left: 50%
top: 68%
width: 100%
scale: 100%
entities:
- entity: sensor.processor_use
max: 100
min: 0
name: Cpu
severity:
- color: '#40bf40'
from: 0
to: 40
- color: '#ffde00'
from: 41
to: 80
- color: '#fd0000'
from: 81
to: 100
- entity: sensor.cpu_temperature
max: 100
min: 0
name: Temperature
decimal: 0
severity:
- color: '#40bf40'
from: 0
to: 40
- color: '#ffde00'
from: 41
to: 75
- color: '#fd0000'
from: 75
to: 100
- entity: sensor.memory_use
max: 4096
min: 0
name: Memory
decimal: 0
severity:
- color: '#40bf40'
from: 0
to: 1280
- color: '#ffde00'
from: 1280
to: 2500
- color: '#fd0000'
from: 2500
to: 4096
- entity: sensor.disk_use_home
max: 64
min: 0
name: Disk
decimal: 0
severity:
- color: '#40bf40'
from: 0
to: 25
- color: '#ffde00'
from: 25
to: 50
- color: '#fd0000'
from: 50
- entity: sensor.swap_use
max: 1024
min: 0
name: Swap
decimal: 0
severity:
- color: '#40bf40'
from: 0
to: 640
- color: '#ffde00'
from: 640
to: 768
- color: '#fd0000'
from: 768
2 Likes
thank you !!
Really interested in the details of your mini-pc if you could provide them ?