User avatar
ehem
Posts: 46
Joined: Mon Sep 09, 2019 1:17 am

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 1:52 am

lingon wrote:
Thu Nov 21, 2019 1:43 pm
This new Rev 1.2 string is reported by /proc/cpuinfo and by dmesg, so it has some implications for the software. If the Rev 1.2 changes would be fully transparent for the software, then it would hardly make any sense to report the revision in /proc/cpuinfo. Small component changes on the board might warrant a revision change of the silk screen text, but could be transparent to the software and would not warrant changes to /proc/cpuinfo.
Very good point. At which point the question is what changed that effects software? My understanding was there were some issues with fully powering down the ethernet interface due to the potential presence of PoE. Perhaps the hardware revision means it can now be fully powered down and still power the board via PoE (since the workaround required a firmware update, a better fix might require a revision).
DougieLawson wrote:
Thu Nov 21, 2019 7:57 am
The folks who don't have a clue assume that's going to mean a fix for the overblown and overegged USB-C non-problem.
I would like to note that I predicted such a fix in this time-frame before this thread was started. Even in September, well before I looked for historical dates for that post, I had been estimating 6 months for that to be fixed. This is the right time for that fix to show up. Until someone gets their hands on such a board and tests or the foundation comments though we won't know.
hippy wrote:
Fri Nov 22, 2019 12:48 pm
Conversely; if there wasn't a 1.2 it would be easy enough to come out and say that and put the entire thread to rest. I can see legitimate reasons to not admit 1.2 exists if it does, but I can't see any real reason not to admit it doesn't if it doesn't.
Actually I can think of one. If they make a habit of denying false rumors then everyone will learn that when they want to know they merely need to start a rumor. This is why many companies have a policy of never commenting on unreleased products.

I think it is quite legitimate to point out that all the claims of having seen such a board come from recently started accounts. The tactic of opening many accounts and trying to appear to be a zillion people who agree has been a favorite of Russian disinformation campaigns. On the flip side it could be that those people had been lurking and just discovered they had something important to report and had never bothered to have an account before.

Until there is very solid evidence (an announcement from the foundation or a corroboration from an account which has existed for some time) everything is speculation. I'm distinctly hopeful, but there is still plenty of room for doubt.

User avatar
DougieLawson
Posts: 37633
Joined: Sun Jun 16, 2013 11:19 pm
Location: A small cave in deepest darkest Basingstoke, UK
Contact: Website Twitter

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 7:45 am

ehem wrote:
Sun Nov 24, 2019 1:52 am
DougieLawson wrote:
Thu Nov 21, 2019 7:57 am
The folks who don't have a clue assume that's going to mean a fix for the overblown and overegged USB-C non-problem.
I would like to note that I predicted such a fix in this time-frame before this thread was started. Even in September, well before I looked for historical dates for that post, I had been estimating 6 months for that to be fixed. This is the right time for that fix to show up. Until someone gets their hands on such a board and tests or the foundation comments though we won't know.
Nonsense. I'd prefer the Raspberry Pi hardware and software engineers to spend their limited & valuable time doing things that have a benefit to the wider RPi4 user population rather than a fix for the tiny subset who think the USB-C non-problem is a massive issue.

As far as I'm concerned that USB-C non-problem can wait for Raspberry Pi "version next" in a couple of years time.
Note: Any requirement to use a crystal ball or mind reading will result in me ignoring your question.

Any DMs sent on Twitter will be answered next month.
All non-medical doctors are on my foes list.

User avatar
Gavinmc42
Posts: 4339
Joined: Wed Aug 28, 2013 3:31 am

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 7:52 am

fully powering down the Ethernet interface due to the potential presence of PoE.
Interesting, I have noticed when I shutdown, after the uSD has finished being written the Ethernet Leds are still flickering.
Could 1.2 have something to do with Ethernet booting and wake on Ethernet?

We know netbooting software is being worked on, perhaps network waking is/will be an option?
Does that Ethernet chip have that option, most do now?
The Industrial/embedded guys might like that feature.

That would be more important than Apple charger re-purposing?
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

andrum99
Posts: 1067
Joined: Fri Jul 20, 2012 2:41 pm

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 10:44 am

Gavinmc42 wrote:
Sun Nov 24, 2019 7:52 am
fully powering down the Ethernet interface due to the potential presence of PoE.
Interesting, I have noticed when I shutdown, after the uSD has finished being written the Ethernet Leds are still flickering.
Could 1.2 have something to do with Ethernet booting and wake on Ethernet?
Anything is possible. The fact that nobody has said what has changed in 1.2 would certainly lead one to suspect it is related to a future, as yet unannounced, feature.

User avatar
Gavinmc42
Posts: 4339
Joined: Wed Aug 28, 2013 3:31 am

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 11:28 am

would certainly lead one to suspect it is related to a future, as yet unannounced, feature.
Serious software involved?
Perhaps Arm cores and other stuff shutdown while GPU monitors the Lan.
That software could take some time and unknown weeks/months to get working.
It might have needed a minor PCB change.

All pure speculation.
I seem to remember shutdown mode is about 3ma.
But if lan chip is active it would be higher.
What is current in shutdown these days with beta eeprom code.
Do the lan leds blink with old critical code?
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

andrum99
Posts: 1067
Joined: Fri Jul 20, 2012 2:41 pm

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 11:50 am

Gavinmc42 wrote:
Sun Nov 24, 2019 11:28 am
would certainly lead one to suspect it is related to a future, as yet unannounced, feature.
Serious software involved?
Perhaps Arm cores and other stuff shutdown while GPU monitors the Lan.
That's not how wake on LAN works. The ethernet part, probably just the PHY, stays powered up and signals to the rest of the system that it should wake. The system can be completed powered down otherwise with just enough running to be able to respond to the wake on LAN signal. On PCs, everything can be off, and an embedded controller somewhere switches everything back on as if a soft power button was pressed.

W. H. Heydt
Posts: 11671
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 4:29 pm

Gavinmc42 wrote:
Sun Nov 24, 2019 11:28 am
All pure speculation.
I seem to remember shutdown mode is about 3ma.
But if lan chip is active it would be higher.
What is current in shutdown these days with beta eeprom code.
Do the lan leds blink with old critical code?
On the Pi4B (unlike previous B-series Pis), the LAN chip isn't involved with the Ethernet jack at all.

trejan
Posts: 1204
Joined: Tue Jul 02, 2019 2:28 pm

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 4:36 pm

W. H. Heydt wrote:
Sun Nov 24, 2019 4:29 pm
On the Pi4B (unlike previous B-series Pis), the LAN chip isn't involved with the Ethernet jack at all.
Uh what? Did you mean USB?

drgeoff
Posts: 10331
Joined: Wed Jan 25, 2012 6:39 pm

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 5:51 pm

W. H. Heydt wrote:
Sun Nov 24, 2019 4:29 pm
On the Pi4B (unlike previous B-series Pis), the LAN chip isn't involved with the Ethernet jack at all.
Let's start the pantomime season early with a loud "O yes it is!".

W. H. Heydt
Posts: 11671
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 6:23 pm

trejan wrote:
Sun Nov 24, 2019 4:36 pm
W. H. Heydt wrote:
Sun Nov 24, 2019 4:29 pm
On the Pi4B (unlike previous B-series Pis), the LAN chip isn't involved with the Ethernet jack at all.
Uh what? Did you mean USB?
Yes, I meant USB, though the USB hub on prior B-series was often referred to as the "LAN chip" as it included the Ethernet interface.

W. H. Heydt
Posts: 11671
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: New Pi 4B v1.2?

Sun Nov 24, 2019 6:23 pm

drgeoff wrote:
Sun Nov 24, 2019 5:51 pm
W. H. Heydt wrote:
Sun Nov 24, 2019 4:29 pm
On the Pi4B (unlike previous B-series Pis), the LAN chip isn't involved with the Ethernet jack at all.
Let's start the pantomime season early with a loud "O yes it is!".
See my just prior post.

User avatar
ehem
Posts: 46
Joined: Mon Sep 09, 2019 1:17 am

Re: New Pi 4B v1.2?

Mon Nov 25, 2019 6:17 am

Gavinmc42 wrote:
Sun Nov 24, 2019 7:52 am
fully powering down the Ethernet interface due to the potential presence of PoE.
Interesting, I have noticed when I shutdown, after the uSD has finished being written the Ethernet Leds are still flickering.
Could 1.2 have something to do with Ethernet booting and wake on Ethernet?
This is the thing I was thinking of. There were those issues with the firmware where they had been trying to power the PI off completely, but restarts were occurring and they figured out the ethernet interface had to be left in a "warmer" state to avoid the restart. Could be when powered off the v1.2 will have the ethernet LEDs off.

Gavinmc42 wrote:
Sun Nov 24, 2019 7:52 am
We know netbooting software is being worked on, perhaps network waking is/will be an option?
Does that Ethernet chip have that option, most do now?
The Industrial/embedded guys might like that feature.

That would be more important than Apple charger re-purposing?
That depends upon which issues one is running into. My ordering of those priorities is USB booting, USB-C fix, wake-on-lan. The first is a firmware update which is almost certainly in the pipeline, but we don't yet have a time frame for. The second is a simple hardware fix which seems likely for v1.2 (due to being very simple to fix and having generated a lot of press). No idea of issues with the third, but right now I don't care. We'll have to wait for someone to identify things or the Raspberry PI Foundation to make an announcement.

andrum99 wrote:
Sun Nov 24, 2019 10:44 am
Anything is possible. The fact that nobody has said what has changed in 1.2 would certainly lead one to suspect it is related to a future, as yet unannounced, feature.
Or could simply be the pipeline isn't fully flushed and so their policy is to say nothing.

W. H. Heydt
Posts: 11671
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: New Pi 4B v1.2?

Mon Nov 25, 2019 6:22 am

andrum99 wrote:
Sat Nov 23, 2019 10:22 pm
W. H. Heydt wrote:
Sat Nov 23, 2019 10:01 pm
Which is rather odd, since the expectation that the Pi4B2 was going to be the bulk of sales.
They didn't actually say that, just that the relative initial sales of the 2 and 4GB models were different to what they expected.
I have a vague recollection of remarks indicating that the initial order (and, thus, the expected demand distribution) was something like 1:3:1 across the 1, 2, and 4GB models, respectively.

User avatar
ehem
Posts: 46
Joined: Mon Sep 09, 2019 1:17 am

Re: New Pi 4B v1.2?

Sun Dec 01, 2019 2:45 am

W. H. Heydt wrote:
Mon Nov 25, 2019 6:22 am
I have a vague recollection of remarks indicating that the initial order (and, thus, the expected demand distribution) was something like 1:3:1 across the 1, 2, and 4GB models, respectively.
Have you seen any numbers for actual demand? Guessing the 4GB model had massively greater demand than expected, but I've got no idea of the overall ratios. I've got one in hand, but am not going to power it on until I've got a case (gah, just missed getting the last piece today).

I'll have to take photos before and then see whether it matches the information for the rumored v1.2. Assume Tuesday at the earliest for a pro or con report.

W. H. Heydt
Posts: 11671
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: New Pi 4B v1.2?

Sun Dec 01, 2019 3:56 am

ehem wrote:
Sun Dec 01, 2019 2:45 am
W. H. Heydt wrote:
Mon Nov 25, 2019 6:22 am
I have a vague recollection of remarks indicating that the initial order (and, thus, the expected demand distribution) was something like 1:3:1 across the 1, 2, and 4GB models, respectively.
Have you seen any numbers for actual demand? Guessing the 4GB model had massively greater demand than expected, but I've got no idea of the overall ratios. I've got one in hand, but am not going to power it on until I've got a case (gah, just missed getting the last piece today).
The Pi4B4 certainly sold out quickly and it did appear that it was more popular than anticipated. I'm not at all sure that will be typical of longer term sales, though. It is certainly not surprising that the "early adopters" would spring heavily for the 4GB model. I actually started with 1 Pi4B1, 2 Pi4B2 and 1 Pi4B4. Future plans are (at minimum) 3 more Pi4B4 and 2 more Pi4B2.

As for a case...suit yourself. I generally try out new Pi models before I can get my hands on a case to fit. I normally do it on a convenient non-conducting surface. At present, I have a Pi4B1 running sitting on top of a copy of "The Naval War of 1812" by T. Roosevelt.

I do have plans to modify cases to use a fan. (The black cases are being very difficult to get, at least from Newark/Element14.)
I'll have to take photos before and then see whether it matches the information for the rumored v1.2. Assume Tuesday at the earliest for a pro or con report.
I will look forward to that.

User avatar
PeterO
Posts: 5623
Joined: Sun Jul 22, 2012 4:14 pm

Re: New Pi 4B v1.2?

Sun Dec 01, 2019 4:49 pm

spooker wrote:
Sun Dec 01, 2019 4:40 pm
is the foundation already aware of the wifi jamming issue on the rpi4.
https://hackaday.com/2019/11/28/raspber ... -own-wifi/
There is mention of quotes from Eben on the subject in the comments, so it's pretty clear they are aware (and it's mentioned in other threads on here as well).

PeterO
Last edited by PeterO on Sun Dec 01, 2019 5:46 pm, edited 1 time in total.
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

andrum99
Posts: 1067
Joined: Fri Jul 20, 2012 2:41 pm

Re: New Pi 4B v1.2?

Sun Dec 01, 2019 5:37 pm

PeterO wrote:
Sun Dec 01, 2019 4:49 pm
spooker wrote:
Sun Dec 01, 2019 4:40 pm
is the foundation already aware of the wifi jamming issue on the rpi4.
https://hackaday.com/2019/11/28/raspber ... -own-wifi/
There is mention of quotes from Ebon on the subject in the comments, so it's pretty clear they are aware (and it's mentioned in other threads on here as well).

PeterO
That particular quote has Eben theorising that the cause is a badly shielded HDMI cable.

Heater
Posts: 14721
Joined: Tue Jul 17, 2012 3:02 pm

Re: New Pi 4B v1.2?

Sun Dec 01, 2019 8:06 pm

andrum99
That particular quote has Eben theorising that the cause is a badly shielded HDMI cable.
I have no idea if this is related but it sounds familiar...

A few years ago I got a Microsoft Surface Pro 4.

Everything was fine except when I plugged it into my big Samsung monitor. Then the WiFi stopped working.

Searching around I found this was a common problem. I never found that MS had a response to this issue.

A year or so later the monitor started showing green speckles everywhere. They changed as I moved the HDMI cable.

So I got a new HDMI cable. No more green speckles. And, as if by magic, WiFi was working just fine.

The moral of the story?

Certainly whatever HDMI does it can radiate EMI that cripples WiFi.

Do not trust cheap shitty HDMI cables.
Memory in C++ is a leaky abstraction .

User avatar
ehem
Posts: 46
Joined: Mon Sep 09, 2019 1:17 am

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 2:02 am

Heater wrote:
Sun Dec 01, 2019 8:06 pm
The moral of the story?

Certainly whatever HDMI does it can radiate EMI that cripples WiFi.

Do not trust cheap shitty HDMI cables.
This is true. I've got this one long and really thin HDMI cable. I'm actually astonished it manages to handle 1920x1080, but that it does. For the moment it serves until it eventually gets replaced. This though is waay off-topic.

W. H. Heydt wrote:
Sun Dec 01, 2019 3:56 am
As for a case...suit yourself. I generally try out new Pi models before I can get my hands on a case to fit. I normally do it on a convenient non-conducting surface. At present, I have a Pi4B1 running sitting on top of a copy of "The Naval War of 1812" by T. Roosevelt.

I do have plans to modify cases to use a fan. (The black cases are being very difficult to get, at least from Newark/Element14.)
There is a reason my hardware lasts, I'm very nice and as a result it keeps running. This is a blessing since I don't need to replace things too often, but on the flip side this is a curse since I don't need to replace things for a long time. That dual-P3-550MHz was still 100% operational 2 years ago and still handled its key tasks, so it didn't require replacement. It was only when Spam Assassin started exhausting the 512MB of memory and 2GB of swap that it required replacement (finally got rid of that snail, and very carefully installed Xen when it was time to replace).

I was thinking of the Flirc case, but lack of access to the pins seemed too much of a downside. Kept searching until a mention of the Argon40 Neo showed up and that was what I was looking for. Though given how far the RPF has pushed the temperatures down, perhaps such cases are now a bit overkill.

W. H. Heydt wrote:
Sun Dec 01, 2019 3:56 am
ehem wrote:
Sun Dec 01, 2019 2:45 am
I'll have to take photos before and then see whether it matches the information for the rumored v1.2. Assume Tuesday at the earliest for a pro or con report.
I will look forward to that.
This has been pushed to Wednesday or Thursday. Even though getting to the US wasn't too bad, apparently the USPS is going to be slow at delivering. Still have no idea whether I've got a rumored one or not (distinct chance I do, but distinct chance I don't).

User avatar
mahjongg
Forum Moderator
Forum Moderator
Posts: 12663
Joined: Sun Mar 11, 2012 12:19 am
Location: South Holland, The Netherlands

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 12:42 pm

spooker wrote:
Sun Dec 01, 2019 4:40 pm
is the foundation already aware of the wifi jamming issue on the rpi4.
https://hackaday.com/2019/11/28/raspber ... -own-wifi/
Yes they are, according to this post:
PhilE wrote:
Mon Dec 02, 2019 6:58 am
Friday's rpi-update firmware release includes a patch to choose an HDMI mode for [email protected] with alternate timing (reduced blanking) to avoid harmonics in the 2.4GHz band.

lingon
Posts: 130
Joined: Fri Aug 26, 2011 7:31 am

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 1:42 pm

This link contains images that are claimed to be of the underside of a Raspberry Pi Model 4 B Rev 1.1 and a Rev 1.2.
It is shown that there are some changes on the Rev 1.2 above the letter "I" in the text "MICRO SD CARD".
http://akkiesoft.hatenablog.jp/entry/20 ... 1574911998
Any ideas what these new traces would be for?

lingon
Posts: 130
Joined: Fri Aug 26, 2011 7:31 am

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 2:22 pm

Comparing the pictures there is 7 soldered pins where the uSD-card is inserted into the reader on the Rev 1.1. and 8 soldered pins on the Rev 1.2. Could this be related to an enhancement of the SD-card speed?

trejan
Posts: 1204
Joined: Tue Jul 02, 2019 2:28 pm

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 2:32 pm

lingon wrote:
Mon Dec 02, 2019 1:42 pm
This link contains images that are claimed to be of the underside of a Raspberry Pi Model 4 B Rev 1.1 and a Rev 1.2.
It is shown that there are some changes on the Rev 1.2 above the letter "I" in the text "MICRO SD CARD".
http://akkiesoft.hatenablog.jp/entry/20 ... 1574911998
Any ideas what these new traces would be for?
Interesting. Finally some evidence of this new revision. Thanks for finding that. Odd that the PCB artwork has changed but the silkscreen was never updated to add a v1.2.
lingon wrote:
Mon Dec 02, 2019 2:22 pm
Comparing the pictures there is 7 soldered pins where the uSD-card is inserted into the reader on the Rev 1.1. and 8 soldered pins on the Rev 1.2. Could this be related to an enhancement of the SD-card speed?
No. All of the connections on the micro SD are used. You only see 7 pads in the top Pi because of the angle of the photo. The socket is a Molex 1040310811.

hippy
Posts: 6851
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 3:01 pm

lingon wrote:
Mon Dec 02, 2019 1:42 pm
It is shown that there are some changes on the Rev 1.2 above the letter "I" in the text "MICRO SD CARD".
http://akkiesoft.hatenablog.jp/entry/20 ... 1574911998
Any ideas what these new traces would be for?
My guess would be that it's a switch to control power to the SD Card, to allow forced power-cycling or to turn power off when the board is reset or shutdown.

trejan
Posts: 1204
Joined: Tue Jul 02, 2019 2:28 pm

Re: New Pi 4B v1.2?

Mon Dec 02, 2019 3:18 pm

hippy wrote:
Mon Dec 02, 2019 3:01 pm
My guess would be that it's a switch to control power to the SD Card, to allow forced power-cycling or to turn power off when the board is reset or shutdown.
That does sound plausible and https://github.com/raspberrypi/linux/is ... -529212585 mentions a problem with the current arrangement.

Return to “General discussion”