W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

The Case for the Pi3A+

Wed Apr 04, 2018 2:42 am

This is to address what I think is the case to be made for the release of a Pi3A+. It's a result of a side-thread in the sticky thread about the Pi3B+.

The A+ is getting rather old, in Pi terms. It's two generations back with regard to both CPU and memory. There is comparatively little to distinguish it from the Pi0. When the Pi2B launched, we were told that a Pi2A was planned...later it was cancelled. When the Pi3B was launched, we were told that a Pi3A was planned and it was to follow the CM3. The CM3 and CM3L were launched nearly 15 months ago. Dr. Upton mentioned that there is a Pi3A+ board and indicated that a launch was being considered.

I think the time has come to launch a serious upgrade to the A+ in the form of a Pi3A+. It would be a significant upgrade for applications where A+ boards are used now. It would make the Pi3A+ a major step up from the Pi0/Pi0W without going fully to a Pi2B/Pi3B/Pi3B+.

And--most importantly, I think--a Pi3A+ would fulfill the original intent of the RPF in that it could be used as a stand-alone device with no extra hardware needed beyond keyboard, mouse, and monitor. The orignal Model A wa criticized, rightly, on the grounds that a single USB port and no network connectivity left it too restricted for general use. Built-in WiFi takes care of the network and BT can solve the keyboard and mouse needs, leaving the sole USB port free.

Even if the BoM and recent high (and getting higher) prices on memory meant the Pi3A+ had to go back to the original Model A price of $25 and/or have it's memory reduced to 512MB and even if it has to use SoC chips that don't test at the full 1.4GHz needed for a Pi3B+, I believe that a Pi3A+ would be a viable product.

ejolson
Posts: 1899
Joined: Tue Mar 18, 2014 11:47 am

Re: The Case for the Pi3A+

Wed Apr 04, 2018 3:27 am

W. H. Heydt wrote:
Wed Apr 04, 2018 2:42 am
go back to the original Model A price of $25
I think the price was $800 for the two seater and $900 for the four seater. Personally , I would prefer an original Model A to the newer one. But, for the same reasons the Model T outsold the Model A, so does the Zero. Maybe a new Model T will appear and have 4 cores in place of the 4 cylinders.

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Wed Apr 04, 2018 4:40 am

ejolson wrote:
Wed Apr 04, 2018 3:27 am
W. H. Heydt wrote:
Wed Apr 04, 2018 2:42 am
go back to the original Model A price of $25
I think the price was $800 for the two seater and $900 for the four seater. Personally , I would prefer an original Model A to the newer one. But, for the same reasons the Model T outsold the Model A, so does the Zero. Maybe a new Model T will appear and have 4 cores in place of the 4 cylinders.
Actually... Ford's Model T preceded their Model A.

ejolson
Posts: 1899
Joined: Tue Mar 18, 2014 11:47 am

Re: The Case for the Pi3A+

Wed Apr 04, 2018 4:46 am

W. H. Heydt wrote:
Wed Apr 04, 2018 4:40 am
ejolson wrote:
Wed Apr 04, 2018 3:27 am
W. H. Heydt wrote:
Wed Apr 04, 2018 2:42 am
go back to the original Model A price of $25
I think the price was $800 for the two seater and $900 for the four seater. Personally , I would prefer an original Model A to the newer one. But, for the same reasons the Model T outsold the Model A, so does the Zero. Maybe a new Model T will appear and have 4 cores in place of the 4 cylinders.
Actually... Ford's Model T preceded their Model A.
Wikipedia has 1903 for the original Model A and 1908 for the Model T.

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Wed Apr 04, 2018 3:46 pm

ejolson wrote:
Wed Apr 04, 2018 4:46 am
W. H. Heydt wrote:
Wed Apr 04, 2018 4:40 am
ejolson wrote:
Wed Apr 04, 2018 3:27 am
I think the price was $800 for the two seater and $900 for the four seater. Personally , I would prefer an original Model A to the newer one. But, for the same reasons the Model T outsold the Model A, so does the Zero. Maybe a new Model T will appear and have 4 cores in place of the 4 cylinders.
Actually... Ford's Model T preceded their Model A.
Wikipedia has 1903 for the original Model A and 1908 for the Model T.
Have it your way because it's irrelevant to this thread. Do you care to comment on the idea of there being a Raspberry Pi 3 Model A+? OR is all you want to d is derail the thread?

gkaiseril
Posts: 446
Joined: Mon Aug 08, 2016 9:27 pm
Location: Chicago, IL

Re: The Case for the Pi3A+

Wed Apr 04, 2018 3:53 pm

That would be a strip down of the Pi 3B/Pi 3B+. I do not see anyone else looking for this type of board.

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Wed Apr 04, 2018 6:08 pm

gkaiseril wrote:
Wed Apr 04, 2018 3:53 pm
That would be a strip down of the Pi 3B/Pi 3B+. I do not see anyone else looking for this type of board.
Do you recall a couple of years ago when all the A+ boards...disappeared? It was because commercial display makers bought them up to make "smart" displays. While a Pi3A+ isn't going to challenge the dominance of the Pi3B/Pi3B+, I think it has a market. All the more so because there is otherwise little to choose from between the A+ and Pi0.

And, as I noted, you can make one into a functioning desktop without having to buy a hub and WiFi adapter.

hippy
Posts: 3786
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: The Case for the Pi3A+

Thu Apr 05, 2018 9:06 am

Deleted
Last edited by hippy on Sat Apr 07, 2018 11:36 am, edited 1 time in total.

fruitoftheloom
Posts: 17487
Joined: Tue Mar 25, 2014 12:40 pm

Re: The Case for the Pi3A+

Thu Apr 05, 2018 10:01 am

There is a case for a 2A / 2A+ / 3A / 3A+ as a drop in replacement for a 1A+.

In a previous discussion it was mentioned that was being explored.

I can not see WiFi / BT fits in with being a drop in replacement.
Adieu

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 20495
Joined: Sat Jul 30, 2011 7:41 pm

Re: The Case for the Pi3A+

Thu Apr 05, 2018 11:42 am

As you all know, we do not announce product in advance. However, it's not been forgotten, but we only have a small dev team, and we have a lot of work on with other products.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

hippy
Posts: 3786
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: The Case for the Pi3A+

Thu Apr 05, 2018 12:30 pm

Deleted
Last edited by hippy on Sat Apr 07, 2018 11:37 am, edited 1 time in total.

User avatar
bensimmo
Posts: 3222
Joined: Sun Dec 28, 2014 3:02 pm
Location: East Yorkshire

Re: The Case for the Pi3A+

Thu Apr 05, 2018 2:39 pm

For me there are two cases for it.

1) A stipped down B+ pulling the LAN/USB off it and leaving the WiFi/BT, that shouldn't be so hard as they squeezed it on the Zero. Running the 3B+ at whatever speed or an easily switchable speed setup (2B/3B style)

2) A stripped down B+ but leaving the Gigabit PoE lan, connection, trimming USB & HDMI to micro/mini and ridding it of the BT/WiFi (if it takes too much space)

Both still have the DSI for displays, though not sure how useful and used that connector actually is.

1) fits the general usage I can see, allows OTG, keeps the more robust camera connector and the ever useful WiFi (I don't find BT that useful as I have little that uses it, keyboards/mice cost more than a cheaper wireless kit, but others do I guess).
So pretty much what we bought them for.

2) for network endpoints that don't need the bulk the Pi3, saving a bit of cost too, which may be spent on the PoE instead ?

Cheaper and smaller than a customised Pi

Is it possible, I believe they could do either, seeing what they have managed before.

A need, I think that's mostly industrial, small Pi network demand.

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Thu Apr 05, 2018 6:53 pm

hippy wrote:
Thu Apr 05, 2018 9:06 am
But I am not convinced a 3A+ with WiFi and one USB is all that much better or preferable to a Zero WH.
Potentially, twice the RAM and perhaps as much as 10 times faster, plus the ability (at need) to run a 64-bit OS, whether Raspbian or not. Granted, size and power requirements are bigger, but I do think that it would provide an intermediate between the Pi0WH and the Pi3B/3B+ at a very small price increment (if the current price worked out), as the delta is only $5.
Also, given its design starting point would likely be the 3B+ with the LAN7515 and sockets removed, I am not sure it would all fit on an A+ sized board.
If you compare an A+ to any current B-series board, you can see that everything on the B-series that an A-series needs fits within the footprint of the A+, which--I suspect--was pretty much the point of the A+ form factor.
I believe there is less rationale for a 3A+ than there is for a 2B+ ( a 3B+ without WiFi ). I agree there is a place for a potential 3A+ but I think you would have to offer a compelling argument beyond there being that place for a 3A+ to see it filled.
I've been considering starting a similar thread to make the case for a Pi2B+. If you want to do that, be my guest. As it is, I think the case for a Pi2B+ is a much subtler one as the Pi2Bv1.2 leaves a much smaller "window of opportunity", though there are a number of features of the Pi3B+ that--if applied to a Pi2B+--could be seen as beneficial.

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Thu Apr 05, 2018 6:58 pm

jamesh wrote:
Thu Apr 05, 2018 11:42 am
As you all know, we do not announce product in advance. However, it's not been forgotten, but we only have a small dev team, and we have a lot of work on with other products.
*If* there is a decision to launch a Pi3A+ I certainly don't expect an announcement to that effect until it actually happens. Nor--if it gets planned--do I expect to see a launch for a minimum of 6 to 12 months. This thread is really just meant to lay out the case that this is a board that makes sense as a product. If the discussion helps persuade those that make such decisions, I''ll take that as a win. If not...nothing ventured, nothing gained.

Or as W. S. Gilbert put it...
"You must stir it and stump it
And blow your own trumpet
Or trust me, you haven't a chance."

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Thu Apr 05, 2018 7:07 pm

hippy wrote:
Thu Apr 05, 2018 12:30 pm
Trying to argue for a 3A+, without WiFi or Bluetooth, without a LAN7515, a closer A+ replacement, when the A+ is probably the least popular of the Pi range doesn't have much going for it. One has to therefore argue for something a bit better; a 3A+ with WiFi and Bluetooth.

I personally think including WiFi and Bluetooth is mostly a means to make a 3A+ sound more viable, more useful, and potentially more popular, than I believe it would be to justify production of the 3A+. But that's just my opinion.
I would only argue in favor of WiFi/BT because of the general popularity...after reading years of posts pleading for it's inclusion. I use WiFi very little (and not at all at home) and am content with WiFi dongles, at least when they work. BT I don't use at all...but it does appear to be popular.
Currently, excluding CM, it is really a choice of Zero WH, 2B or 3B+. There is a gap for an 'A+' product, but I am not convinced there is a need for it, with or without WiFi and Bluetooth. Again; just my opinion.
Yup. That gap is there and I think putting something there has significant potential.
I accept and appreciate W. H. Heydt's argument and case. I just don't agree with it. I would hope we can agree to disagree because this is after all simply a discussion.
Agreed. This is a simple discussion and opinions can differ in good humor and friendship.

W. H. Heydt
Posts: 8892
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: The Case for the Pi3A+

Thu Apr 05, 2018 7:13 pm

bensimmo wrote:
Thu Apr 05, 2018 2:39 pm
2) A stripped down B+ but leaving the Gigabit PoE lan, connection, trimming USB & HDMI to micro/mini and ridding it of the BT/WiFi (if it takes too much space)
Interesting point. PoE header would seem to be without a use on a Pi3A+ as there would be no Ethernet connector to tap for power. From a design perspective, it might make sense to leave the header holes in place and not populate it. Alternatively, populate the header and find a spot to provide a way to connect the proper voltage supply and the enterprising could use the PoE hat to power the Pi3A+. This could work in those cases where the Pi3A+ is to be installed at what would otherwise be at the far end of a cable that is too long to use 5v effectively.

drgeoff
Posts: 8516
Joined: Wed Jan 25, 2012 6:39 pm

Re: The Case for the Pi3A+

Thu Apr 05, 2018 8:47 pm

bensimmo wrote:
Thu Apr 05, 2018 2:39 pm
1) A stipped down B+ pulling the LAN/USB off it and leaving the WiFi/BT,
@bensimmo
Please do not use the term B+ when you mean a 3B+. There was a B+ model long before Wi-Fi and BT were present on any RPi model.

User avatar
bensimmo
Posts: 3222
Joined: Sun Dec 28, 2014 3:02 pm
Location: East Yorkshire

Re: The Case for the Pi3A+

Fri Apr 06, 2018 6:58 am

Ok #B+, B+ Series...

I'm sure context works perfectly fine here.
Last edited by bensimmo on Fri Apr 06, 2018 3:28 pm, edited 1 time in total.

hippy
Posts: 3786
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: The Case for the Pi3A+

Fri Apr 06, 2018 10:03 am

Deleted
Last edited by hippy on Sat Apr 07, 2018 11:37 am, edited 2 times in total.

fruitoftheloom
Posts: 17487
Joined: Tue Mar 25, 2014 12:40 pm

Re: The Case for the Pi3A+

Fri Apr 06, 2018 10:31 am

hippy wrote:
Fri Apr 06, 2018 10:03 am
W. H. Heydt wrote:
Thu Apr 05, 2018 6:53 pm
I've been considering starting a similar thread to make the case for a Pi2B+. If you want to do that, be my guest.
Not wanting to swing this thread off-topic beyond this: I'll give that a miss as I don't see much rationale for a 2B+. I see that as mostly a desire of makers to save a few pennies in having the WiFi module left off. That's a "meh" for me; just disable it.

The RPT have a road map, and these types of Posts are just "pi in the sky thinking".

The RPT are well aware of what is "feasible" and after 6 years now employ a team of very competent people, the OP making a case (in several posts) is just hot air and insulting to RPT/RPF IMO
Adieu

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 20495
Joined: Sat Jul 30, 2011 7:41 pm

Re: The Case for the Pi3A+

Fri Apr 06, 2018 11:10 am

fruitoftheloom wrote:
Fri Apr 06, 2018 10:31 am
hippy wrote:
Fri Apr 06, 2018 10:03 am
W. H. Heydt wrote:
Thu Apr 05, 2018 6:53 pm
I've been considering starting a similar thread to make the case for a Pi2B+. If you want to do that, be my guest.
Not wanting to swing this thread off-topic beyond this: I'll give that a miss as I don't see much rationale for a 2B+. I see that as mostly a desire of makers to save a few pennies in having the WiFi module left off. That's a "meh" for me; just disable it.

The RPT have a road map, and these types of Posts are just "pi in the sky thinking".

The RPT are well aware of what is "feasible" and after 6 years now employ a team of very competent people, the OP making a case (in several posts) is just hot air and insulting to RPT/RPF IMO
Yes we do have a road map for the main product line (1,2,3,4,5 etc), and a lot of development is going in to that.

However, sub products are not so well defined. So we could decide to build a 3A+, or a 2A or a sandwich maker. It's all down to what engineering capability is available, and whether it would actually sell enough to compensate for the development time and the time taken out from the mainline development.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

gordon77
Posts: 3365
Joined: Sun Aug 05, 2012 3:12 pm

Re: The Case for the Pi3A+

Fri Apr 06, 2018 2:39 pm

fruitoftheloom wrote:
Fri Apr 06, 2018 10:31 am
hippy wrote:
Fri Apr 06, 2018 10:03 am
W. H. Heydt wrote:
Thu Apr 05, 2018 6:53 pm
I've been considering starting a similar thread to make the case for a Pi2B+. If you want to do that, be my guest.
Not wanting to swing this thread off-topic beyond this: I'll give that a miss as I don't see much rationale for a 2B+. I see that as mostly a desire of makers to save a few pennies in having the WiFi module left off. That's a "meh" for me; just disable it.

The RPT have a road map, and these types of Posts are just "pi in the sky thinking".

The RPT are well aware of what is "feasible" and after 6 years now employ a team of very competent people, the OP making a case (in several posts) is just hot air and insulting to RPT/RPF IMO
I was interested to hear in the 3B+ launch video that Eben said they listen to users likes and dislikes, so it can't all be hot air.

User avatar
davidcoton
Posts: 3163
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: The Case for the Pi3A+

Fri Apr 06, 2018 10:29 pm

jamesh wrote:
Fri Apr 06, 2018 11:10 am
So we could decide to build a 3A+, or a 2A or a sandwich maker.
How about a Pie Maker? They could be cooked using recycled Forum Hot Air.
Of course the official version would make Raspberry Pie. No doubt others will attempt to take over with Apple, Banoffee, ..., Pork, ....
"Thanks for saving my life." See https://www.raspberrypi.org/forums/viewtopic.php?p=1327656#p1327656
“Raspberry Pi is a trademark of the Raspberry Pi Foundation”

alphanumeric
Posts: 1535
Joined: Tue Jan 19, 2016 2:17 pm
Location: Sydney, Nova Scotia, Canada

Re: The Case for the Pi3A+

Sun Apr 08, 2018 12:52 pm

I would buy at least two 3A's. I have two A+'s I'd happily replace with 3A+'s. They have Sense Hats on them and run headless. The A+ form factor is perfect for mounting Hats on top and keeping a small footprint, IMHO anyway. One normal size HDMI and one full size USB is all I need to do the setup. I find that the four mounting holes lining up with those in the Hat makes mounting them in a case easier than say a Zero. A 3A or 3A+ with the same layout as the A+ would work for me. I have a few other setups where I would happily swap out a Zero for a 3A. Rovers etc where the extra USB and Ethernet aren't needed.

User avatar
DavidS
Posts: 3096
Joined: Thu Dec 15, 2011 6:39 am
Location: USA
Contact: Website

Re: The Case for the Pi3A+

Tue Apr 10, 2018 7:28 pm

Personally I would rather see a RPi B+W, that is a BCM2835 SoC based B+ with the addition of Bluetooth and WiFi.

The BCM2835 has the advantage of being an ARMv6 CPU that is more compatible with more older binary only software, that will not run on the BCM2835 SoC with its ARMv8 CPU cores and the deprecated swp instruction (a heavily used instruction in a lot of software that is closed source, still usable, and no longer maintained).

Also having the same Bluetooth and WiFi on three different RPi boards would increase the chances of us getting WiFi on systems other than n*x on our RPi (EG RISC OS, Baremetal projects, etc).
26-Bit R15 to 32-bit. 16-bit addressing to 24-bit. ARM and 65xx two CPU's that continue on, and are better than ever. Assembly Language forever :) .

Return to “Off topic discussion”

Who is online

Users browsing this forum: No registered users and 2 guests