Pi vs BeagleBone-Black

These can provide waaaaaaay better hard realtime interfacing performance than any program running in Linux user-space,

There are also several real-time Linux kernel patch sets (Xenomia is one) that can provide way better real time (hard and soft) performance running in "RT" user space than anything running

in "normal" user space. They also supply full RT APIs to work with and hardware (when available) supported floating point if you need it. Any RT kernels avaialable for the PRUs ?

  If finding a cubieboard is a problem, you could take a look on Ebay at the Mele A1000 media player

I bought one off of Ebay several months ago which is one of the reasons I'm not as interested in the Cubieboard as I use to be. It does have wi-fi and Ethernet and although it is bit short on the

user I/O it does also have serial, I2C (have to add a header), IR remote control and a SATA 2 interface. There are several web sites that provide Linux for the A10 and it already comes with what

I think is an Android OS for the media player operation.

Gary Stewart wrote:

 

 

OK, unless you plan to make your own distro I don't understand why you would be recompiling 10000 packages. A majority of those packages have already been compiled for the existing distros and

there is nothing inherently ARM CPU version dependent in any applications that are not already compiled as long as you have all the proper tools: GCC, C libraries, and binary utilities which Raspian

does provide (not sure about Fedora but I assume they do to).

You're missing the point, Raspbian is a two man, unofficial, effort to port debian to an arch that the real Debian maintainers are unwilling to support. Raspbian may vanish tomorrow. At which point you will either need to start compiling them yourself, or move to a piece of hardware that someone else is willing to support. You could of course stick with the last Raspbian release if if you like, but that leaves you with other problems.

You're right in that nothing is inherently dependant on the CPU, the issue is simply that Debian doesn't want to support the older version CPU on the Pi, nothing more.

Morgaine Dinova wrote:

 

Spring greetings to you all ...  this thread has brought me out of my forum sabatical, naughty naughty.

 

Was wondering where you'd been hiding recently

• Farnell UK says "18584 will be available for delivery on 3 May, 2013", so they're clearly taking this seriously and expect BB Black to sell in much higher volumes than the first BeagleBone.  No surprise of course, price is king.
  

 

However after getting an 'in stock' notification from Newark (strange when I'm in the UK), Newark now show 0 in stock and more expected to ship on 28th June - no qty listed...

So I'm not holding my breath just yet.

Success has very little to do with engineering merit though, so I won't be making any guesses about how popular this new board will become.  However, for myself, the Pi Model B is now totally dead in the water since it lost its only advantage, which was price.

 

I'd hoped to see something like this where the price difference (in the UK anyway) is approaching zero. Regardless of the merits of each device, it'll be interesting to see what happens next.

Luc Cool wrote:

 

If finding a cubieboard is a problem, you could take a look on Ebay at the Mele A1000 media player. It's pretty much a cubieboard without the gpio lines.

At least for me, it's the GPIO that makes or breaks the device, without them I'm much less interested.

It will cost more than a Pi,

Approx 60 GBP, so around twice what the Pi costs. Sure it's not a straight comparison.

The A10 looks like it should be a good SoC and so far the cubieboard looks to be the closest to the sort of board I'm interested in. The danger, as always, is that something better and cheaper may come along by the time they start appearing in volume.

Either way, interesting times, and I'm really hoping we'll see more and more devices of this type.

You're missing the point, Raspbian is a two man, unofficial, effort to port debian to an arch that the real Debian maintainers are unwilling to support. Raspbian may vanish tomorrow. At which point you will either need to start compiling them yourself, or move to a piece of

  hardware that someone else is willing to support. You could of course stick with the last Raspbian release if if you like, but that leaves you with other problems.

No, I'm not. I think the worst case scenario you just described is not realistic for several reasons. Raspian is the official RPF distribution. There are now over 500,000 Raspberry Pi's out there, most using Raspian. There

is nothing to stop another group of people from taking over or forking Raspian if the current maintainers stopped, and I don't see the RPF not trying to find, or having a lot of trouble finding other people to maintain it if that

happened.

You're right in that nothing is inherently dependant on the CPU, the issue is simply that Debian doesn't want to support the older version CPU on the Pi, nothing more.

I don't think that official support from Debian is a prerequisite for a successful distribution. There are precedents for that thought.

selsinork wrote:

 

I'd hoped to see something like this where the price difference (in the UK anyway) is approaching zero. Regardless of the merits of each device, it'll be interesting to see what happens next.

With the price disincentive gone, it'll be great to see comparisons on engineering merit with Raspberry Pi Model B.  The Pi still has a technical advantage in media playback, but that is the only advantage that comes to mind at present, whereas at first glance it seems to be below par in every other technical respect compared to BeagleBone Black.  Detailed comparisons would really be useful.  Perhaps we can do that on the eLinux wiki.

My interest isn't in media playback but in computing, networking, home automation, embedding, and hardware interfacing, so Pi's media bullet points don't float my boat.  However, it's undoubtedly the case that the masses think otherwise, and good media handling for $35 is what really sold the Pi in vast numbers.  Plus the hype.

Interesting days ahead.  I hope many other manufacturers join in the fun at this price point.

Morgaine.

I've started adding information on BeagleBone Black to the eLinux wiki BeagleBone page.

> With the price disincentive gone, it'll be great to see comparisons on engineering merit with Raspberry Pi Model B.

Such unbiased comparisons of engineering merit may be difficult to produce.

If you say the BB clock speed is greater than the RPi, you may be met with the

objection that the RPi can be overclocked to the same clock speed. 

If you note the RPI's overclocking potential, you may be met with the objection

that overclocking does not always succeed, and may result in damaged SD cards.

If you note the risks inherent in overclocking, you may be met with the objection

that the risks are overstated and that overclocking the RPi is not actually risky.

If you say the BB is faster, and therefore better for educational purposes

involving computationally intensive tasks such as IDE's and web browsing,

you may be told that programming should be learned on a slow machine

in order to teach the importance of optimization.

If you say the BB has better voltage regulation, you may be told that it should

be the responsibility of the power supply to regulate it's own voltage output.

If you say the BB works with a wider variety of keyboards without dropping

or repeating characters, you may be told that debugging hardware problems

builds character and is a valuable learning experience.

Hahaha. 

Indeed, the values system of fanbois is a glory to behold, and doesn't obey any known physical laws.

Fortunately as engineers we can just enumerate the facts that can't be disputed, except by fanbois of course.  And even engineering value judgements of comparative merits and demerits can be given in an NPoV manner as well, and reasonable people will often find them useful, fanbois excepted of course.

At the end of the day, we can only do our best using the engineer's normal methods of analysis.  Once we've solved world hunger and poverty, then maybe some brave soul will tackle much harder problems like blind fanboism.

coder27 wrote:

 

> With the price disincentive gone, it'll be great to see comparisons on engineering merit with Raspberry Pi Model B.

 

Such unbiased comparisons of engineering merit may be difficult to produce.

If you say the BB clock speed is greater than the RPi, you may be met with the

objection that the RPi can be overclocked to the same clock speed. 

If you note the RPI's overclocking potential, you may be met with the objection

that overclocking does not always succeed, and may result in damaged SD cards.

If you note the risks inherent in overclocking, you may be met with the objection

that the risks are overstated and that overclocking the RPi is not actually risky.

 

If you say the BB is faster, and therefore better for educational purposes

involving computationally intensive tasks such as IDE's and web browsing,

you may be told that programming should be learned on a slow machine

in order to teach the importance of optimization.

 

If you say the BB has better voltage regulation, you may be told that it should

be the responsibility of the power supply to regulate it's own voltage output.

 

If you say the BB works with a wider variety of keyboards without dropping

or repeating characters, you may be told that debugging hardware problems

builds character and is a valuable learning experience.

I'd say your bias is pretty clear too.  Why don't you wait for real responses instead of just making them up ? Here's a few for now.

16 bit memory data bus vs 32 bit for the Pi. Well at least it uses DDR3 vs. DDR2 for the Pi. That will not make up for the lack of bus width though.

The power management chip used by the BeagleBoard Black also uses low dropout voltage regulators for one of its 1.8V and one of its 3.3V supplies. Bad engineering eh Morgaine ? Built

right into the chip from TI no less.

The documentation for the real time PRUs is sorely lacking in the AM3359 Technical Reference Manual, 2 pages and no useful hardware description at all. Not even a block diagram. A search

on the TI web site produces only 5 hits all related to forum questions. At least the questions in the forum hint that there is access to better hardware documentation. The PRUs in the 3359 are

an upgrade to an existing device so the documentation may still be be "in the pipeline". They are mostly aimed at providing support for industrial real time communications protocols although

they apparently (no documentation) can be used for other purposes. The only software tools for it is an assembler which is not yet released for public use although mainly due to efforts at

BeagleBoard.org it is supposed to be by the end of this month.

The GPU is a proprietary design by PowerVR, which means at least for now, no open source drivers. It supports an OpenGL ES API same as the Pi so X11 hardware acceleration support is not

likely for it either. At least it is a few generations ahead of the Pi.

No built in camera interface.

Edited (again) to add:

It has been my experience that overclocking the CPU does not cause data corruption on the SD card, overclocking the memory does. I had one Pi (out of three) all running with 1 GHHz overclock

that did corrupt data until I changed memory timing from 600 MHz overclock to 500 MHz and have not had a problem with it since. After that I changed all of them to 500 MHz to be safe. Note

that I used the phrase data corruption which is an accurate description as opposed to damaged the SD card which is not.

On the plus side for the BB Black:

Lots more, and more useful I/O.

16 bit LCD interface.

More modern ARM architecture.

From Morgaine:

  Indeed, the values system of fanbois is a glory to behold, and doesn't obey any known physical laws.

Fortunately as engineers we can just enumerate the facts that can't be disputed, except by fanbois of course.  And even engineering value judgements of comparative merits and demerits can be given in an NPoV manner as well, and reasonable people will often find

   them useful, fanbois excepted of course.

At the end of the day, we can only do our best using the engineer's normal methods of analysis.  Once we've solved world hunger and poverty, then maybe some brave soul will tackle much harder problems like blind fanboism.

Yea, these are real good arguments from an normal engineering methods standpoint. Good to see that name calling hasn't fallen out of your favourite responses after your long absence.