REV C UPDATE

[Gerald Coley]

Fellow Beaglers,

 

We wanted to give you an update on the Rev C board. We have just finished up the latest spin of the board, so we are running behind schedule. At this point we are looking at the middle of Q1 to get the Rev C out of production.

 

The key changes are as follows:

 

1) Addition of the USB Host port....We are moving this to Port 2 on the OMAP3530 instead of Port1. This allows us to align from a SW perspective with other platforms out there such as Pandora.

 

2) We have added access to the native LCD signals...This however came at a price. We had to move the DVI-D connector to make room for a pair of headers to provide access to the signals. This was a very painful change, especially when coupled with the move to Port2 on the USB. It has taken us several weeks to get it worked in. We hated to move the DVI-D connector, but there wasn't a way around it.

 

I have attached a mockup of what the changes will look like in the area around the DVI-D connector to give you an idea of what the change will look like.

 

If you have any questions, please feel free to let us know.

 

Gerald

 

 

[Felipe Contreras]

Hmm, It looks like I won't be able to use both S-Video and DVI-D at
the same time, because the converter will cover that port.

--
Felipe Contreras

[Gerald Coley]

There is no conflict. As I said, this is a mock up to show the basic idea. This is not a picture of the board. You will be able to use both at the same time.

 

Gerald

[Adam Yergovich]

Gerald Coley wrote:

Changes look good. I think it will be generally better to have the
external connectors confined to just the three sides. I look forward to
playing with the LCD signals, is the LCD supported as the default
xterm/user console for any of the linux distros yet? I'm still digging
through Angstrom and Debian but its not immediately clear.

-Adam

--
Adam Yergovich
Engineer
JK Microsystems

1403 Fifth St. Suite D
Davis, CA 95616

Tel:(530) 297-6073

[Koen Kooi]

On 14 okt, 20:32, Adam Yergovich <aye...@jkmicro.com> wrote:

> Changes look good.  I think it will be generally better to have the
> external connectors confined to just the three sides.  I look forward to
> playing with the LCD signals, is the LCD supported as the default
> xterm/user console for any of the linux distros yet?  I'm still digging
> through Angstrom and Debian but its not immediately clear.

Angstrom has a console both on LCD and serial.

regards,

Koen

[Gerald Coley]

The same signals will drive the DVI-D framer. There is a signal that will allow you to disable the framer if you like form SW.

 

Gerald

[Doug Emes]

On Tue, Oct 14, 2008 at 11:01 AM, Gerald Coley <ger...@beagleboard.org> wrote:
> Fellow Beaglers,
> We wanted to give you an update on the Rev C board. We have just finished up
> the latest spin of the board, so we are running behind schedule. At this
> point we are looking at the middle of Q1 to get the Rev C out of production.

[snip]

For rev. d an onwards, have you considered having a 2nd not-expansion board?
That is to say, using another 10pin connector (like the serial port)
to place the
Svideo, Line-out, Mic-In, and other bulky connectors off the main board? Those
pins should be pretty tolerant of the added distance (by virtue of
their usage -
whats another couple inches). that should free up a bunch of space "on the
top end" of the board for signal paths that really need the short distances.

Just a thought off the cuff, so to speak.

[Gerald Coley]

That has been considered. we have a lot of ideas that we continue to bat around.

 

The decision we made early on was standard buses, which means standard connectors. We had not really planned to focus much at all on real expansion. The existing expansion header was just a few standard buses that in and of themselves did not have any sort of standard connectors, like I2C and I2S. This is why we did not make the LCD interface available because at 1.8V, it wouldn't drive any display anyway. We focused on keepng the cost down. Everything we add from now on must be with keeping the cost under control as it is our foremost concern. There are a lot of OMAP3 base boards out there and our goal is not to compete with them in the area of expansion and I/O. Having done 20 OMAP development board over the years, I have seen what happens when you make everyone happy. There is a lot of waste and everybody pays for it whether they use it or not.

 

As to Rev D, there are no solid plans in place for that. REV C could be the final revision or we could end up at Rev F. Who knows. We plan to listen, take input, and do what we can when we can, and when it still aligns with our overall goals.

 

Thank you for the input!

 

Gerald

[O D]

If I may ask this question here...
Digi-key Europe doesn't has any more Beagleboards on the stock. Does anybody knows when we can expect them again here?
Middle of Q1 seems a long time from now... :(

[Gerald Coley]

Middle of Q1 is for Rev C and not the current revision. We are switching to REV B6, which fixes the layout for the U9 and U11 issue to support a different package. This should solve the U9 and U11 issue we have been having. These boards will show up the first week in November. We should also be shipping another 200 Rev B5 boards this week.

 

Gerald

[Nathan Monson]

On Oct 15, 3:55 am, "Gerald Coley" <ger...@beagleboard.org> wrote:
> Middle of Q1 is for Rev C and not the current revision. We are switching to

> REV B6. These boards will show up the first week in November.

There is a rumor on the IRC channel that B6 will switch to ES3.0. Is
that going to happen before Q1?

- Nathan

[Gerald Coley]

No. The plan right now is that Rev B6 will be ES2.1. If it changes, I will let everyone know.

 

Gerald

[gregoir...@gmail.com]

Gerald,

Can you explain what is the process between "finishing the latest spin
of the board" and the "Rev is out of production". Why does this take 3
to 4 months?

Grégoire

>  Beagle_exp_LCD_RevC.JPG
> 87KViewDownload

[Gerald Coley]

Layout is under review.

 

2 weeks to get 20 protoypes in.

1 week to assemble the boards

4 weeks to test the boards and all the new SW

6 weeks to get production boards in.

2 weeks to get first 1000 assembled and tested

 

Total of  15 weeks, not counting the holidays. I think this makes it to the 1Q of 2009 as I stated.

 

Gerald

[Frantisek Dufka]

Gerald Coley wrote:
> The key changes are as follows:
>
> 1) Addition of the USB Host port....
>

> 2) We have added access to the native LCD signals...
>

> If you have any questions, please feel free to let us know.

Ther are some rumours about rev C having 256MB of RAM, see
http://elinux.org/BeagleBoard#Revision_C
http://www.beagleboard.org/irclogs/index.php?date=2008-10-01#T19:10:58

Are we getting 256MB RAM in rev C?

Regards,
Frantisek

[Gerald Coley]

As you say, that is a rumor. I have no plans to put 256MB on the Rev C board.

 

Gerald

[Johnny You 游文洲]

Hi,

        I’m a little bit curious about the beagle board, what type of application did you build with the beagle board?

We know from the datasheet that BB are suitable for developing handheld devices, home entertainment, etc.

But, what kind of application did you guys developed using this board?

 

 

Thanks &  Best regards !!  

---------------------------------------------------------------------------------------------------------------------

游文洲 / Yohannes Budiono Jinawi / Jonny You

National Central University

Room A305-1, No.300, Zhongda Rd., Zhongli City, Taoyuan County 320, Taiwan (R.O.C.)

9652...@cc.ncu.edu.tw

+886 910926117

Lab www : http://cilab.csie.ncu.edu.tw/

---------------------------------------------------------------------------------------------------------------------

 

 

[Gerald Coley]

As to 256MB I wanted to let everyone know what we are thinking and where we are.

 

We have been testing the 256MB memory to make sure there are no issues, so there are a few early REV C boards out there, 4 in fact. The concern of making 256MB standard on Rev C is the cost. The 256MB has a significant cost to it and we would need to raise the price of the board to cover the additional cost.

 

We are not sure if that additional cost will hurt us in allowing us to keep the Beagle affordable or if the additional cost is worth it. We also have the additional cost of the USB host on the REVC which we are planning to absorb, but when you add in the added memory cost, that makes it tougher.

 

So, right now, the plan is to keep the memory where it is with 128MB. Based on these discussions and other discussions, we may decide to add the 256MB as standard and raise the price of the Beagle. The question is how much increase in cost is acceptable. We may be able to pass on a portion of the cost and absorb the rest, but right now we are not sure if we can. We are due to discuss this tomorrow and I will update everyone with what we decide.

 

In the mean time, if we could get feedback as to what the added memory is worth, that will help us make the case for moving forward with 256MB as standard on Beagle.

 

Gerald

[Bill Gatliff]

Gerald Coley wrote:
> In the mean time, if we could get feedback as to what the added memory
> is worth, that will help us make the case for moving forward with 256MB
> as standard on Beagle.

I'd give you another US$30 per-board for the USB and memory updates, easy.

b.g.
--
Bill Gatliff
bg...@billgatliff.com

[Gerald Coley]

Any application that runs a high level OS such as Linux. Look at the Archos 5 paler that has just been released which has OMAP3530 in it. There is anew class of devices called MID (Mobile Internet Device) that are emerging onto the market and it is projected that in five years it will out sell laptops.

 

Here is a short  list of things that the OMAP is good for:

 

MID

NAS Servers

Medical Devices

Digital signage

Software Defined Radio

Telecomminucations

Audio

Appliances

 

I hope this helps!

 

Gerald

 

2008/10/16 Johnny You 游文洲 <9250...@cc.ncu.edu.tw>

[Gerald Coley]

I have $30. Do I hear $35!

 

Thank you!

 

Gerald

[Robert Kuhn]

2008/10/16 Gerald Coley <ger...@beagleboard.org>:

> I have $30. Do I hear $35!

$30-50 only for memory.No problem! We (for example) need it!

Robert

[Felipe Contreras]

On Tue, Oct 14, 2008 at 7:54 PM, Gerald Coley <ger...@beagleboard.org> wrote:
> There is no conflict. As I said, this is a mock up to show the basic idea.
> This is not a picture of the board. You will be able to use both at the same
> time.

Maybe I'm not explaining myself correctly.

If the DVI-D is right next to the S-Video, then when using this kind
of adapters will block the S-Video:
http://www.hometech.com/video/gc-1212321.jpg

--
Felipe Contreras

[Doug Emes]

Felipe,

They make adapters that have a cable in between the two heads that
would not block
the svideo port. I have one that came with my computer desktop
motherboard actually.
An example: http://www.nulime.com/Startech-Digital-Video-Cable/p75182

[Måns Rullgård]

Count me in. My experience with the prototype has been nothing but
good.

What about memory speed. My rev C prototype has faster memory than the
rev B, and it's showing.

You've mentioned a possibility of running two versions. Is that still
being considered as an option?

--
Måns Rullgård
ma...@mansr.com

[Måns Rullgård]

You can always use an HDMI-DVI cable, or put an HDMI-female to DVI-male
adaptor on the other end.

--
Måns Rullgård
ma...@mansr.com

[John Beetem]

Just for the sake of discussion, is it possible to increase RAM to
256MB and decrease NAND to 128MB with minimal change in cost? Any
serious Linux use seems to require an SD card for the file system, so
the actual NAND use is more or less limited to storing MLO, uBoot, and
a kernel image.

I can easily add more NAND using an SD card or USB Flash drive, but
adding high-bandwidth RAM is not possible.

[Bill Gatliff]

Ssssshhhhhhhhhhhhh!! I think we coulda had him for $30! :)

[Gerald Coley]

That could be the case. But, I recommend and always have to use the cable and not the adapter. You won't find the adapter in the reference manual at all.

 

There will always be the case that some adapter out ther may not work. We aren't in a postiotion to try and support every adapter out there.  This is just a price that will need to be paid in order to get the LCD signal access and I think it is worth it.

 

 

Gerald

 

On Thu, Oct 16, 2008 at 10:10 AM, Felipe Contreras <felipe.c...@gmail.com> wrote:

[Ilja Kamps]

On the other hand it would put me off, 128MB works for what I need to do
and the additional cost would be a shame.

[Gerald Coley]

Thank you for your inpu!

 

Gerald

[Matthias Klostermann]

I wouldn't mind to pay $30 for a 256MB RAM version of the Board. The idea to reduce the onchip NAND to balance out a bit of the cost for more RAM sounds good, too. Except for the bootloader, I don't have any use for it anyways, and more (and faster?) RAM would be more of an advantage.

Matthias

[Nathan Monson]

I'd hate to see the price go up. For me, the rock bottom price is the
main selling point of the BeagleBoard, not the amount of memory.
Running a swap partition on a newish SD card is faster than I
expected.

Ideas I do like:
Offering a $200 Power User Beagleboard without discontinuing the
$150 model
Reducing the NAND to pay for the additional memory without raising
the price

- Nathan

[Koen Kooi]

On 16 okt, 17:14, Måns Rullgård <m...@mansr.com> wrote:
> Robert Kuhn wrote:
>
> > 2008/10/16 Gerald Coley <ger...@beagleboard.org>:
>
> >> I have $30. Do I hear $35!
>
> > $30-50 only for memory.No problem! We (for example) need it!
>
> Count me in.  My experience with the prototype has been nothing but
> good.

$30-$50 more for extra ram would be acceptable for me as well.

regards,

Koen

> What about memory speed.  My rev C prototype has faster memory than the
> rev B, and it's showing.
>
> You've mentioned a possibility of running two versions.  Is that still
> being considered as an option?
>
> --
> Måns Rullgård

> m...@mansr.com

[Darcy Moen]

Mind if I toss my hat into the ring?

How about a 128 OR 256 ram, fully loaded SD card including preconfigured
OS (Angstrom or Ubuntu), power supply, USB, DVI, Serial, and Svideo
cables ready to go 'Beagle for Dummies' or 'Beagle of Non-geeks', or
'Beagle right out of the box'?

Fully loaded price: what ever you set. I'm sure there would be folks
willing to buy it, even if for the novelty of it.

I'd be in for one, cause I've been too busy to play. Would love to buy
one and be able to play right out of the box. Or would that offend the
purists with passion?

Darcy

[Robert Nelson]

I'd pay $30-$50 extra for 256MB, any chance at 512MB? ;)

--
Robert Nelson
http://www.rcn-ee.com/

[Gerald Coley]

So, let's take this to another angle. Is it feasible to have two versions, 128MB standard version at the $149 point, and a turbo version at a higher price point, say $175 as that is in the area most of you seem to be OK with? Does this make sense?

 

Now, from my perspective, this will be a nightmare trying to figure out how many of each to build and figuring out the correct mix. But, if this is what we need to do, then we will try to figure something out assuming that we could make it work.

 

What would everyone guess as to the mix, 50/50?

 

What say ye?

 

Gerald

[Paul Pignon]

Likewise. We really want that USB.

/Paul

Bill Gatliff wrote:
> Gerald Coley wrote:
>
>> In the mean time, if we could get feedback as to what the added memory
>> is worth, that will help us make the case for moving forward with 256MB
>> as standard on Beagle.
>>
>
> I'd give you another US$30 per-board for the USB and memory updates, easy.
>
>
> b.g.
>
>
>

--
Paul Pignon, CTO
pa...@bps.co.ee
Tel. +372 6803515
Mob. +372 53463942, +46 705508655
Skype: paulspignon

Borthwick-Pignon OÜ
Liivaoja 1-33
10155 Tallinn
Estonia
EU
http://bps.co.ee

[Bill Gatliff]

Gerald Coley wrote:
> So, let's take this to another angle. Is it feasible to have two
> versions, 128MB standard version at the $149 point, and a turbo version
> at a higher price point, say $175 as that is in the area most of you
> seem to be OK with? Does this make sense?
>
> Now, from my perspective, this will be a nightmare trying to figure out
> how many of each to build and figuring out the correct mix. But, if this
> is what we need to do, then we will try to figure something out assuming
> that we could make it work.
>
> What would everyone guess as to the mix, 50/50?

At US$175, I won't bother looking at the smaller unit--- I'll just standardize
on the turbo version as my "standard" version when I need actual hardware to
promote OMAP to a client.

As a project-builder, if I'm already "in it" for $149 then I'll spend the extra
money on the assumption that I'll find a way to need the extra resources on the
larger board (uniformly true for 15 years and counting!). I've made a pretty
serious investment already, the additional $30 is like insurance against losing
the whole $149 because the smaller board was too small (assuming the lack of
expansion options isn't a non-starter already).

My reverse-angle argument goes like this. At $150 just to play with
beagleboard, if $30 is a deal-breaker then maybe the buyer needs to reconsider
the whole effort. Beagleboard is unique in some ways, not in others.

Don't bother flaming, you don't have to like or adopt my opinions. :)

[Gerald Coley]

Do you think that difference in price will scare away students for example? We are looking to work Beagle into the universities at some point. It would be very nice as I mentioned to build only one version.

 

Gerald

[Bill Gatliff]

Gerald Coley wrote:
> Do you think that difference in price will scare away students for
> example? We are looking to work Beagle into the universities at some
> point. It would be very nice as I mentioned to build only one version.

Can't say for sure (who can?).

The biggest "scary point" is $100, and I don't see a way to get below that.

I'm mostly familiar with Extension and corporate students, and in that crowd
anything sub-$200 seems palatable.

For undergrads, I guess it's whatever their parents will spend :), and sub-$200
still seems ok--- especially if the curriculum re-uses the board for multiple
classes.

[Frantisek Dufka]

Gerald Coley wrote:
> So, let's take this to another angle. Is it feasible to have two
> versions, 128MB standard version at the $149 point, and a turbo version
> at a higher price point, say $175 as that is in the area most of you
> seem to be OK with? Does this make sense?

That would be great. Makes perfect sense. Please go for it if you can.

Also maybe in such case it even makes sense to bump the price up to 189
or 199 and add some other low hanging fruit goodies you may have on the
'removed to keep the cost down' list. Rationale is that the price
difference is small and the target group for the turbo version is less
sensitive to price. But if this feel like opening can of worms then
please forget what I just said. I'm perfectly happy with $179 256MB
variant of REV C board :-)

>
> Now, from my perspective, this will be a nightmare trying to figure out
> how many of each to build and figuring out the correct mix. But, if this
> is what we need to do, then we will try to figure something out assuming
> that we could make it work.

Maybe some web based voting on beagleboard.org could help?

Frantisek

[Malina, Jim]

I'll second that.

$150 (or lower) is an extremely compelling beagleboard feature for us.

If you were to offer a turbo product (at higher cost) are there any other integrations that might be considered? Memory+, Ethernet? Battery? Wifi, Bluetooth...

Jim

[V. Elishah Frey]

It seems absolutely silly to me to integrate a battery or wifi or bluetooth.  What atracts me to the beagleboard is it's costamizability, and the instant you build something in you close the option to the developer to use a variation on that built in device.  What is built into the board, what I think is succesfull, are devices that everybody wants to use, and, more importantly, devices that would be much much much more expensive to be put together by a hobbiest.

 

-E.L.F.

[Geof Cohler]

On the Special Adapter versus Special Cable issue... The advantage of
the Adapter is that it costs about $4 and then you can plug in the
cable you already have that came with your monitor. So I think you're
right that the Rev C design will make it so that you must use a
special $15 cable (HDMI-M to DVI-D-M) rather than a Special $4 Adapter
plus a Standard Cable. It's not the end of the world.

Geof

On Oct 16, 11:13 am, "Doug Emes" <kyo...@gmail.com> wrote:
> On Thu, Oct 16, 2008 at 10:10 AM, Felipe Contreras
>

[Gerald Coley]

Adding additional functions to the board adds cost, which at least to me, seems to be going in the wrong direction. As E.L.F. points out, the ability to work with work with different suppliers of these devices. Instead of having OMAP3 drivers for every single supplier of BT and WiFi devices, you would only have one. Another factor is that anything added other board will cost us more than having those devices being made my the thousands in the commercial market. 

 

Gerald

[John Beetem]

Here's my 1.49 cents...

Personally, what really drew me to BeagleBoard was the $149 price
point (versus 10x for Mistral or >2x for other ARM development
boards). I figured that not only would it save me money, but it would
foster a large user community which is critical for open source
software of this complexity. [As an aside, I did have to match the
$149 with another $150 for a DVI-D monitor, so I really didn't save
that much up front. Except that I do have a really nice monitor now.]

128MB DRAM is plenty for my current and immediate applications. I was
looking for a development board that had capabilities similar to a
smart phone or netbook, and BeagleBoard fits that nicely.

There is always a danger with adding more DRAM: software has a
tendency to bloat to fill available resources. One of the beauties of
the embedded world is that limited resources have encouraged more
efficient use of processor cycles and memory. It would be a shame to
see software bloat take over the embedded world as it did the desktop.

My recommendation: please make sure there is still a $149 version
available. It's OK to have an extra-cost version for high-end users,
but that $149 sure is compelling. Whenever I show it to other
engineers, it's the $149 that causes their jaws to drop.

[DOUAT Laurent]

Hello,

My 2 cents.
I plan to make rich ui + web browsing + tv services. for such classic project, I think 128MB is feasible but will need more integration effort.
Going up to 256MB (with faster speed: Mans remark) is making sense and will create a breaking difference for the beagleboard.

The Pandora box costs 330 $ with 128MB
Another point : Maybe mid/end 2009 the 256MB SDRAM will cost the current price of the 128MB ?

 

Voting time :

175$ for 256MB : All in.

 

Laurent

 

 

----- Original Message -----

From: Gerald Coley

To: beagl...@googlegroups.com

Sent: Thursday, October 16, 2008 8:11 PM

Subject: [beagleboard] Re: REV C UPDATE

[Frantisek Dufka]

V. Elishah Frey wrote:
> It seems absolutely silly to me to integrate a battery or wifi or
> bluetooth. What atracts me to the beagleboard is it's costamizability,

Yes, I was thinking about perhaps additonal MMC slot or any other
existing OMAP3 I/O currently missing for cost reasons. I guess $175 or
$199 may not matter much for those going for the turbo version. >200 may
be too much though. If there is nothing easy to add I am happy with just
256MB vesrion of existing board.

The mmc slot would be handy for permanent root file system while still
having sdio or removable card option. True that you can add usb card
reader but this may increase power consumption when comparing to already
present mmc interface and having removable usb device for root
filesystem (for keeping sdio) is not ideal. Still it is not a must. OTOH
RAM is really something you cannot add easily and for desktop-like usage
or advanced multimedia usage where DSP or maybe 3d chip grabs big chunks
of memory for itself 128MB is not enough but 256MB may really improve
situation.

Frantisek

[Bill Gatliff]

Frantisek Dufka wrote:
> V. Elishah Frey wrote:
>> It seems absolutely silly to me to integrate a battery or wifi or
>> bluetooth. What atracts me to the beagleboard is it's costamizability,
>
> Yes, I was thinking about perhaps additonal MMC slot or any other
> existing OMAP3 I/O currently missing for cost reasons. I guess $175 or
> $199 may not matter much for those going for the turbo version. >200 may
> be too much though. If there is nothing easy to add I am happy with just
> 256MB vesrion of existing board.

Heh, swap the SD slot for a micro-SD and I'm happier. That's a lot of
real-estate gained, who knows what someone could use the room for? :)

[Frantisek Dufka]

Bill Gatliff wrote:
> Heh, swap the SD slot for a micro-SD and I'm happier. That's a lot of
> real-estate gained, who knows what someone could use the room for? :)

Well, one full SD slot is nice if there is any ready made slot at all.
You can't buy 32GB in microSD now and 16GB is just released with higher
price and lower speed (class 2). Perhaps microSD will always lag in
price/capacity/speed behing full SD.

The alternative is just ready to use SD/MMC pins without any socket.
That would be enough for me for second OMAP MMC interface. There are
pins on expansion socket that can be configured via mux registers as MMC
but I guess these are raw signals I cannot use for soldering sd card
directly. As I understand it there should be some voltage regulator chip
to set corect SD/MMC voltage (in 1.6-1.8V or 2.8-3.4 range) like the
Menelaus one [1] in N8x0. Such voltage regulator chip could fit the
$175-$199 space. Please correct me if I am wrong with this and I can in
fact attach second memory card directly.

Anyway, all this means changes so it is not REV C material. The first
baby step is to have second rev C board with just 256MB RAM in Q1/2009.
I definitely plan to get this one if Digi-Key allows me to buy it :-)

Frantisek

1.
http://mxr.maemo.org/diablo/source/kernel-source-diablo-2.6.21/kernel-source/arch/arm/mach-omap2/board-n800-mmc.c#48

[Koen Kooi]

On 17 okt, 01:19, Geof Cohler <g.coh...@computer.org> wrote:
> On the Special Adapter versus Special Cable issue...  The advantage of
> the Adapter is that it costs about $4 and then you can plug in the
> cable you already have that came with your monitor.  So I think you're
> right that the Rev C design will make it so that you must use a
> special $15 cable (HDMI-M to DVI-D-M) rather than a Special $4 Adapter
> plus a Standard Cable.  It's not the end of the world.

the 1.5 meter hdmi -> dvi cable cost me €3.50, which is about $5.

regards,

Koen

[Bala Gajen]

Beagleboard is a development board and its not the final product. I think it's a good idea to provide only the IO pins accessible (for additional devices such as Bluetooth, wifi) for the developers. Otherwise developers will be limited to use their ideas.

 

Developers can then add additional hardware to the board, either by incorporating a readily available hardware (devices) or by designing their own bit of circuits.  Developers should suggest what pins should be accessible and what not (pins which are not important). So pins can be made available on new Rev Boards.

 

Making the LCD related pins accessible is a good move on Rev C.

 

In the case of developing their own circuits; This IO pins will helps developers to make their own expansion cards, for example Ethernet card. Developers should publish the expansion card designs to other developers so they can use it without re inventing the wheels. So we will end up having a set of expansion cards (Bluetooth card, wifi card, touch screen controller card etc). This will help us pick and choose what we need for our own project.

 

In this way developers can make their own working system. If its successful they could make their own single board beagle, incorporating all the expansion-card designs init.

 

Even though beagle is design to be used on multimedia applications nothing stops us from using it on other applications. So I would even remove Audio, Mic and S-Video connectors and use pin headers instead to save space. This space can be used to make additional omap pins accessible.

 

Requesting additional memory is good thing as we can not add these by ourselves. only a fraction of Nand is used so NAND memory can be smaller than 128mb. Apart form that I suggest the new board should have more IO pins, and headers for Audio and Video. SD card slot should be left as it can take different cards. As I said once a designs is finalized this slot can be changed to microSD, if that's what needed.

[Johnny You 游文洲]

I’ve install angstrom in the BB, it takes about 90MB of RAM to boot up. This means that I have only 30MB free memory. I think there is necessary to raise the RAM size and reduce the NAND for cost-down.

 

Btw, is there any other version of linux that takes less memory?

 

From: beagl...@googlegroups.com [mailto:beagl...@googlegroups.com] On Behalf Of Bala Gajen
Sent: Friday, October 17, 2008 5:22 PM
To: beagl...@googlegroups.com
Subject: [beagleboard] Re: REV C UPDATE

 

Beagleboard is a development board and its not the final product. I think it's a good idea to provide only the IO pins accessible (for additional devices such as Bluetooth, wifi) for the developers. Otherwise developers will be limited to use their ideas.

[Koen Kooi]

On 17 okt, 11:21, "Bala Gajen" <balaga...@gmail.com> wrote:
> Beagleboard is a development board and its not the final product. I think
> it's a good idea to provide only the IO pins accessible (for additional
> devices such as Bluetooth, wifi) for the developers. Otherwise developers
> will be limited to use their ideas.

USB provides support for BT, wifi and wired ethernet.

regards,

Koen

[Bala Gajen]

Yes Koen you are right.

 

BT, Wifi and Ethernet can use USB.

 

I think I have given the wrong examples there; I haven't studied the board and the omap datasheet properly. Someone mentioned that there are unimplemented OMAP pins in the beagle board (correct me if I am wrong). I can't say much without knowing more about these unimplemented pins. In the future a Developer may want to use these pins in their design.

 

If i am to add USB based devices to the project. I would probably develop my own USB hub circuit (If possible) and will use module based devices (can be directly mounted on the board, e.g BT module) to save space.

 

Not sure about a USB based touch screen controller, and infrared port. But if there facilities to implement these in OMAP directly (without using USB) then I would prefer that.

 

regards

 

Bala

[Nishanth Menon]

Bala Gajen said the following on 10/17/2008 06:27 AM:

> Yes Koen you are right.
>
> BT, Wifi and Ethernet can use USB.
>
> I think I have given the wrong examples there; I haven't studied the
> board and the omap datasheet properly. Someone mentioned that there
> are unimplemented OMAP pins in the beagle board (correct me if I am
> wrong). I can't say much without knowing more about these
> unimplemented pins. In the future a Developer may want to use these
> pins in their design.
>
> If i am to add USB based devices to the project. I would probably
> develop my own USB hub circuit (If possible) and will use module based
> devices (can be directly mounted on the board, e.g BT module) to save
> space.
>
> Not sure about a USB based touch screen controller, and infrared port.
> But if there facilities to implement these in OMAP directly (without
> using USB) then I would prefer that.
>

Beagle, Zoom MDK, EVM, SDP3430, Pandora. each of these provide various
benefits for the price point. For folks who need a board with
LCD,touchscreen(thru OMAP), etc.. there are zoom mdk, evm options.. for
folks who would like to play around with changing cameras and lcds every
so often, need serial/parallel and other exotic display, cameras etc..
there is the SDP.. omap is a 400odd pin chip, there are lots one can
play with, but one thing even on a development platform like SDP you
dont get is: every single feature OMAP has to offer. IMHO, 149$ is folks
like me can pay for a hobby project where i spend a few hours a week..

Beagle looses it's attractiveness if the cost crosses the 149$.. for the
idea of creating a "automobile like high end trim" of beagle with
features, it does not make much sense to me.. I gather the input from
this discussion could be: there is a need to have a bare board (IMHO
some other name other than beagle - call it bull dog or Dallas
//Chihuahua// or something) giving just expansion ports to plug in
custom displays, custom cameras, touchscreen, probably a non POP memory
in the form of a module and other stuff this chain has wishlisted....

Regards,
Nishanth Menon

[Laurent Desnogues]

I am all in favour of 256 MB of faster DRAM and less NAND at < $200.

I think there are two main categories of people: those who see it as
a low-priced
embedded board and those who see it as a nice small computer. For the second
category, to which I belong, 256 MB would be extremely nice (I would
even say it's
a requirement).

BTW it's not because you have plenty of RAM that you start doing
bloated software,
it's an education problem, and usually the need for more RAM arises
from software
you need/use rather than the other way around.

Laurent

[Bill Gatliff]

Oooh. Could we get some solder points for USBH, so that I can attach stuff to
the bus with something other than an ordinary cable?

[kapare]

Don't go high our some will take your place ;)

http://www.hackszine.com/blog/archive/2008/10/linuxstamp_embedded_linux_syst.html?CMP=OTC-7G2N43923558

But the 256MB of RAM sound nice!

kap

[David Goodenough]

The other interface missing on the current BeagleBoard is the Camera ISP.
I had hoped to be able to use the BeagleBoard as either just a development
board or a short run production board, but I need the Camera interface, so
I have had to look elsewhere.

David

[David Goodenough]

On Friday 17 October 2008, Johnny You 游文洲 wrote:
> I’ve install angstrom in the BB, it takes about 90MB of RAM to boot up.
> This means that I have only 30MB free memory. I think there is necessary to
> raise the RAM size and reduce the NAND for cost-down.
>
>
>
> Btw, is there any other version of linux that takes less memory?

Have you looked at OpenWrt? It can be run on machines with only a few
MB of both memory and NAND. There is even discussion on their forums
about doing the port, but the current thought is to await Rev C with its
USB port. Given that much of the OMAP 3530 support is already in the
kernel it should not be difficult.

David

[Koen Kooi]

On 17 okt, 12:16, Johnny You 游文洲 <92502...@cc.ncu.edu.tw> wrote:
> I've install angstrom in the BB, it takes about 90MB of RAM to boot up.
> This means that I have only 30MB free memory. I think there is necessary to
> raise the RAM size and reduce the NAND for cost-down.
>
> Btw, is there any other version of linux that takes less memory?

Linux only takes a few megabytes of memory, the rest is occupied by
userspace. So, no, there is no other version of linux that takes less
memory.
You can remove (and add) packages to the angstrom demo images using
opkg to suit your needs. Or you can build a custom image using
OpenEmbedded.

[Jason Kridner]

On Oct 14, 11:54 am, "Gerald Coley" <ger...@beagleboard.org> wrote:
> There is no conflict. As I said, this is a mock up to show the basic idea.
> This is not a picture of the board. You will be able to use both at the same
> time.

The converter spoken about is fixed and fat, such as
http://www.nextag.com/Phoenix-Gold-DVI-D-78293610/prices-html, not the
HDMI-to-DVI-D cables that we normally use. I think the concern is
valid and we need to understand if we can live with it.

>
> Gerald
>
> On Tue, Oct 14, 2008 at 11:32 AM, Felipe Contreras <
>
> felipe.contre...@gmail.com> wrote:
>
> > On Tue, Oct 14, 2008 at 7:01 PM, Gerald Coley <ger...@beagleboard.org>
> > wrote:
> > > Fellow Beaglers,
>
> > > We wanted to give you an update on the Rev C board. We have just finished
> > up
> > > the latest spin of the board, so we are running behind schedule. At this
> > > point we are looking at the middle of Q1 to get the Rev C out of
> > production.
>
> > > The key changes are as follows:
>
> > > 1) Addition of the USB Host port....We are moving this to Port 2 on the
> > > OMAP3530 instead of Port1. This allows us to align from a SW perspective
> > > with other platforms out there such as Pandora.
>
> > > 2) We have added access to the native LCD signals...This however came at
> > a
> > > price. We had to move the DVI-D connector to make room for a pair of
> > headers
> > > to provide access to the signals. This was a very painful change,
> > especially
> > > when coupled with the move to Port2 on the USB. It has taken us several
> > > weeks to get it worked in. We hated to move the DVI-D connector, but
> > there
> > > wasn't a way around it.
>
> > > I have attached a mockup of what the changes will look like in the area
> > > around the DVI-D connector to give you an idea of what the change will
> > look
> > > like.
>
> > > If you have any questions, please feel free to let us know.
>
> > Hmm, It looks like I won't be able to use both S-Video and DVI-D at
> > the same time, because the converter will cover that port.
>
> > --
> > Felipe Contreras
>
>

[Jason Kridner]

On Oct 16, 10:14 am, Måns Rullgård <m...@mansr.com> wrote:
> Robert Kuhn wrote:
>
> > 2008/10/16 Gerald Coley <ger...@beagleboard.org>:
>
> >> I have $30. Do I hear $35!
>
> > $30-50 only for memory.No problem! We (for example) need it!
>
> Count me in.  My experience with the prototype has been nothing but
> good.
>
> What about memory speed.  My rev C prototype has faster memory than the
> rev B, and it's showing.
>
> You've mentioned a possibility of running two versions.  Is that still
> being considered as an option?
>

We are committed to maintaining a $149 version, so if we need to raise
the price of the extra-memory and USB-host version, I'm fairly
confident we'd continue to run both versions. This causes a lot of
heartache for some people dealing with the planning, which is one
reason we really need to understand the demand and value for the
version with these features.

[Gerald Coley]

If we must support this DIV-D adapter, then we will either need to live with no S-video at the same time or remove the LCD signals.

 

Is everyone OK with removing the LCD signals?

 

Is S-Video and DVI-D at the same time that critical?

 

Gerald

[Robert Kuhn]

Hello,
yes, for me it is. I need both s-video and DVI or the BB is useless for me.

Robert

2008/10/17 Gerald Coley <ger...@beagleboard.org>:

[c.dicaprio]

Hello Gerald,
I don't agree with remowing the LCD signals. I think is an important
feature that make the beagle board suitable for portable devices.
And... I'm waiting rev C to start my project! :-)

Best Regards,

Clemente.

On 17 Ott, 18:08, "Gerald Coley" <ger...@beagleboard.org> wrote:
> If we must support this DIV-D adapter, then we will either need to live with
> no S-video at the same time or remove the LCD signals.
>
> Is everyone OK with removing the LCD signals?
>
> Is S-Video and DVI-D at the same time that critical?
>
> Gerald
>

[Doug Emes]

On Fri, Oct 17, 2008 at 11:08 AM, Gerald Coley <ger...@beagleboard.org> wrote:
> If we must support this DIV-D adapter, then we will either need to live with
> no S-video at the same time or remove the LCD signals.

> Is everyone OK with removing the LCD signals?
> Is S-Video and DVI-D at the same time that critical?
> Gerald

I feel the need to comment: S-Video is not nearly as strict as DVI-D
for cabling
parameters... If it comes down to it, put a 4 pin header where the
s-video is and
folks can direct wire the lum/ground and chrome/ground connections to it, or
make a connection facing straight up, or some other solution. LCD
direct signals
are way more "exciting" than a NTSC cablehead connector.

Having both DVI-D and SVideo is critical, but having a $0.50 connector on board
causing heartache should not be. The rev. C may or may not need a board layout
change to replace the s-video connector with a 4 pin header.

[Keith Williams]

Gerald Coley wrote:
> If we must support this DIV-D adapter, then we will either need to
> live with no S-video at the same time or remove the LCD signals.
>
> Is everyone OK with removing the LCD signals?
>
> Is S-Video and DVI-D at the same time that critical?
>

Please leave the planned LCD signals intact. That would be a deal
breaker for just about anything that I do embedded.

Keith

[Geof Cohler]

I think we can live without the adapter. So my vote is: LCD signals
+ S socket + HDMI-A Socket -- although I've never used the S socket
yet. I would vote to add the USB Host (Std-A) Socket be added too (in
addition to the USB OTG Mini-A/B socket).

Geof


On Oct 17, 12:52 pm, Keith Williams <esp...@linuxinstruments.com>
wrote:

[spec]

On Oct 17, 2008, at 9:08 AM, Gerald Coley wrote:

> If we must support this DIV-D adapter, then we will either need to
> live with no S-video at the same time or remove the LCD signals.
>
> Is everyone OK with removing the LCD signals?

No! Removing features from the BB to satisfy a minority of users
peripherals makes no sense. I have two projects in mind which would
benefit greatly from the lower cost of driving an LCD directly.

While making my BB kit purchases, I chose the slightly more expensive
HDMI-->DVI cable over the adaptor for the very reason of not covering
other ports. Plus the extra weight of those adaptors really puts
unnecessary wear and tear of the connectors.

Inexpensive cables from China and elsewhere are cheap and plentiful.
I can't see limiting the design, and therefore the utility of the BB
to placate a few developers with inadequate tools.

> Is S-Video and DVI-D at the same time that critical?

I'm not using S-video, but that's not to say I won't find an
application in the future. I suppose those signals could be moved to
an expanded J3. Eliminating P4 and stacking the audio connectors
might create enough header space to provide the extra signals others
are requesting.

-Glen

[John Beetem]

My vote is to keep the LCD signals in Rev C. I've seen lots of
interest for LCD on this list and the IRC. It's unfortunate that this
blocks the use of a DVI-D adapter, but HDMI to DVI-D cables are
available and inexpensive, while converting DVI-D back to LCD is a
pain. Another advantage to keeping the LCD signals is that (I
believe) this makes it possible to build an inexpensive SVGA expansion
board consisting of a 3-channel DAC and an SVGA connector. That would
be a lot cheaper than a DVI-D monitor or a DVI-D to VGA converter box.

[rt1k]

Hi Gerald,
please do not remove LCD driving signal from rev. c board.
I think that this is the most important feature to make the beagle board the definitive solution for my embedded projects.

Best regards.
Rt1k.

[Steve Franks]

Buying a $30 cable to connect a $200 monitor to a $150 pcb is a touch
silly IMHO. The DVI is great for debug, but if many of us are going
to use this for something other than "it works, stick it on the
shelf", then we need cheap display, and that means svideo or LCD.
Moreover, what I saw on the wiki a week or so ago looked like a really
expensive driver board, connected to those "lcd signals" we keep
talking about, so I question seriously if exposing them actually would
make an inexpensive display possible, or if it would just be one more
expensive display option. My ideal board would be a $200-$250 board
that I can pretty much stick a LVDS and/or TTL LCD straight into (no
backlight controller, obviously, don't see that fitting on the pcb).
Some of the real low-end stuff is going into a 320x200 or so OLED
display. I want cheap graphics, and I don't care how! Maybe the $99
15" DVI lcd is rally the cheapest option, who knows...and no, I
haven't actually found a $99 15" lcd, so don't ask ;)

Steve

[Jakov af Wallby]

On Fri, Oct 17, 2008 at 04:27:31PM -0700, Steve Franks wrote:

[snip]

> silly IMHO. The DVI is great for debug, but if many of us are going
> to use this for something other than "it works, stick it on the
> shelf", then we need cheap display, and that means svideo or LCD.

DVI or HDMI is essential for digital signage.

--
regards,
Jakob

It's is not, it isn't ain't, and it's it's, not its, if you mean it
is. If you don't, it's its. Then too, it's hers. It isn't her's. It
isn't our's either. It's ours, and likewise yours and theirs.
-- Oxford University Press, Edpress News

[kapare]

Reasons why I'm interested on the beagleboard is:

1-cheap and all in one board. I don't like to add multiple converter
and USB TO etc....
2-OMAP
3-LCD -> OK I could leave with USB LCD but this add $$
4-S-VIDEO was to me important for what I was interested to do. If
their is another solution to convert to s-video. (ex: USB 2.0 to tv-
tuner). Again it ADD extra $$
5-I didn't buy revision B because I was interested to have a connector
for DVI and not a HDMI to DVI and also the fix for USB HOST (EHCI)
fix.
But I don't think that my patience will wait until earliest at late
December 2008 or early January 2009 (http://elinux.org/
BeagleBoard#Revision_C). I WANT MY CHRISTMAS GIFT ;)

kap

On Oct 17, 12:08 pm, "Gerald Coley" <ger...@beagleboard.org> wrote:
> If we must support this DIV-D adapter, then we will either need to live with
> no S-video at the same time or remove the LCD signals.
>
> Is everyone OK with removing the LCD signals?
>
> Is S-Video and DVI-D at the same time that critical?
>
> Gerald
>

[Gerald Coley]

There are no plans for HDMI on BeagleBoard. We will only be supporting DVI-D.

 

Gerald

[Måns Rullgård]

"Gerald Coley" <ger...@beagleboard.org> writes:

> There are no plans for HDMI on BeagleBoard. We will only be supporting
> DVI-D.

HDMI video is a subset (restricted modes) of DVI-D. What you probably
mean is that the Beagle will not support HDMI audio or HDCP. DVI-D
allows HDCP but not audio.

--
Måns Rullgård
ma...@mansr.com

[Gerald Coley]

Nope. I mean we won't support HDMI. That is our standard message as all we support is DVI-D and we use the HDMI connecotr. Yes, DVI-D is a subset, but we don't want to give anybody the impression that we even support a subset, even though as you say, DVI-D is indeed the digital portion of HDMI.

 

Gerald

[Måns Rullgård]

"Gerald Coley" <ger...@beagleboard.org> writes:

> On Mon, Oct 20, 2008 at 3:56 PM, Måns Rullgård <ma...@mansr.com> wrote:
>
>>
>> "Gerald Coley" <ger...@beagleboard.org> writes:
>>
>> > There are no plans for HDMI on BeagleBoard. We will only be supporting
>> > DVI-D.
>>
>> HDMI video is a subset (restricted modes) of DVI-D. What you probably
>> mean is that the Beagle will not support HDMI audio or HDCP. DVI-D
>> allows HDCP but not audio.
>

> Nope. I mean we won't support HDMI. That is our standard message as
> all we support is DVI-D and we use the HDMI connecotr. Yes, DVI-D is
> a subset, but we don't want to give anybody the impression that we
> even support a subset, even though as you say, DVI-D is indeed the
> digital portion of HDMI.

No, that's still not accurate. Perhaps a table will help.

Feature DVI-D HDMI Beagle
--------------------------------------
1024x768 60Hz X * X * optional
1280x720 60Hz X X X
1920x1080 60Hz X X
2560x1600 X
HDCP X X
Audio X

Strictly speaking, DVI-D only specifies the data link. Interpretation
of the data is up to connected devices. The spec includes an EDID
channel over which supported formats can be communicated.

The HDMI spec allows for any timing mode, and includes a list of
required modes as well as specific optional modes. As we've all seen,
many HDMI monitors choose to only support the listed modes. A generic
DVI-D sink (monitor) is not required to support any specific formats.

The DVI specification predates HDCP, so there is no mention of it
there. Numerous devices do, however, support HDCP over DVI-D
connectors.

The Beagle supports any video mode within its clocking capabilities,
and provides a DVI-D compatible signal through an HDMI connector.

--
Måns Rullgård
ma...@mansr.com

[Gerald Coley]

I agree with what you have down. The Beagle supports any video mode within its clocking capabilities,

and provides a DVI-D compatible signal through an HDMI connector.

 

Gerald

[NovaSmart Technology R&D]

I vote for LCD. With the best size, performance and power consumption,
i believe beagle board is the best choice for portable product. With
built-in LCD features... it will be perfect.

On Oct 20, 2:03 pm, "Gerald Coley" <ger...@beagleboard.org> wrote:
> Nope. I mean we won't support HDMI. That is our standard message as all we
> support is DVI-D and we use the HDMI connecotr. Yes, DVI-D is a subset, but
> we don't want to give anybody the impression that we even support a subset,
> even though as you say, DVI-D is indeed the digital portion of HDMI.
>
> Gerald
>

> On Mon, Oct 20, 2008 at 3:56 PM, Måns Rullgård <m...@mansr.com> wrote:
>
> > "Gerald Coley" <ger...@beagleboard.org> writes:
>
> > > There are no plans for HDMI on BeagleBoard. We will only be supporting
> > > DVI-D.
>
> > HDMI video is a subset (restricted modes) of DVI-D.  What you probably
> > mean is that the Beagle will not support HDMI audio or HDCP.  DVI-D
> > allows HDCP but not audio.
>
> > --
> > Måns Rullgård

> > m...@mansr.com

[doug....@gmail.com]

On Oct 16, 11:49 am, Bill Gatliff <b...@billgatliff.com> wrote:
<snip>
>
> My reverse-angle argument goes like this.  At $150 just to play with
> beagleboard, if $30 is a deal-breaker then maybe the buyer needs to reconsider
> the whole effort.  Beagleboard is unique in some ways, not in others.
>

That would be fine if everything needed was in the $149 kit but it is
not.
Most have to purchase all these extra parts at more than $50:

Power supply, HDMI-DVI cable, RS232 cable/conn, USB HUB.
And most currently must get a USB-B(?) cable/adapter to use the HUB.

The RevC USB Host connector is going to help eliminate the USB-B
adapter
needed for the USB hub use. It may also eliminate the need for a power
supply since 5v from the USB-OTG may power the beagle and still let
the
USB Host port work.

IMO, the breaking point for more widespread use is to keep the
experience
under $200. It's already just over $200 with all the extra's required
but when
the base kit( Beagleboard ) closes in on $200, those extras are going
to
throw off a good number of experimenters.

I like the idea of keeping the $149 kit and having another "PLUS" kit
with the
extra memory.

SInce we are dreaming, how about this:
A software kit which would reduce the cost to the experimenter out of
the box.
It would be an Angrom or Ubuntu image for the Beagleboard which had
built-in USB network forwarding and X keyboard/mouse redirection from
the
desktop computer. This eliminates the power supply, the need for a
hub,
USB ethernet device, USB kbd, USB mouse. And if the BeagleBoard was
shipped to boot from such an SD image, they wouldn't even need the
RS232
cable unless problems occurred. Couple that with a virtual machine
image
of the development environment and a good 2 weeks of messing around
with
package installations, git and bitbake configuring, etc. Not to
mention having
a non-broken development image pre-configured. Atleast 3 times I had
git
update from the dev tree and was broken by new updates added to dev.

my .02

[Jakov af Wallby]

On Tue, Oct 21, 2008 at 02:16:08PM -0700, doug....@gmail.com wrote:

> SInce we are dreaming, how about this:
> A software kit which would reduce the cost to the experimenter out of
> the box.
> It would be an Angrom or Ubuntu image for the Beagleboard which had
> built-in USB network forwarding and X keyboard/mouse redirection from
> the

[...]

This is the best idea ever! So easy.

[Wouter Bijlsma]

I just tried ordering a BB from Europe (Netherlands) before I found
out a rev. C with LCD signals was coming, so I cancelled the order. I
was told the supplier for NL wasn't getting any boards for the coming
7 weeks anyway, and did not receive any information on the rev. C
boards yet. Too bad, because I'm dying to get my hands on a BB,
they're really, really hard to get hold of here.

Anyway, I'd like to add that just like the previous posters I'd
_gladly_ put down another $30-$35 for an extra 128MB of RAM, maybe
even more. I'd even want to pay around $200 to get the board within
the coming weeks if I could. I think the impact of constrained RAM
should not be underestimated, doubling to 256 would enlarge the range
of viable applications enormously and add a lot to the BB lifetime.
Especially now that Gumstix has the Overo board, which isn't as
complete as the BB without expansion boards, but is similarly priced
and has 256MB. My experience with open embedded systems is that before
you know people start making stuff no-one ever imagined possible, just
to hit a brick wall because of constrained memory, and 128 isn't much.

I'll be keeping a close eye on any BB developments, hopefully all goes
well with the development of the rev. C board and I'll have something
to burn my spare time on soon :-)

On 16 okt, 16:14, Måns Rullgård <m...@mansr.com> wrote:
> Robert Kuhn wrote:
>
> > 2008/10/16 Gerald Coley <ger...@beagleboard.org>:
>
> >> I have $30. Do I hear $35!
>
> > $30-50 only for memory.No problem! We (for example) need it!
>
> Count me in.  My experience with the prototype has been nothing but
> good.
>
> What about memory speed.  My rev C prototype has faster memory than the
> rev B, and it's showing.
>
> You've mentioned a possibility of running two versions.  Is that still
> being considered as an option?
>

> --
> Måns Rullgård
> m...@mansr.com

[Gerald Coley]

D/K will be Rev B getting boards in late next week, so the 7 week thing is not true. The Rev C won't be out until mid to late Q1 of next year.

 

 

Gerald

[Wouter Bijlsma]

Hi Gerald, thanks for the reply!

Any word on the amount of RAM and/or if the LCD signals will stay?
Because if I have to wait until Q1 next year _and_ the board will
still only have 128MB ram _or_ no LCD signals, I might want to either
try to buy a rev B. board anyway, or look around for a completely
different board that satisfies my hunger for embedded power. What
exactly is the status of the EHCI support on the rev B. board ATM? Any
chance it will be usable through software updates?

Wouter

On 6 nov, 16:28, "Gerald Coley" <ger...@beagleboard.org> wrote:
> D/K will be Rev B getting boards in late next week, so the 7 week thing is
> not true. The Rev C won't be out until mid to late Q1 of next year.
>
> Gerald
>

[Gerald Coley]

The LCD signals will definitely stay. It looks like we may end up with two versions where one is 128MB and the other is 256MB. We are not sure how to handle all of that. The responses we got were that some wanted to keep the price and the memory while others wanted more memory no matter what! So, we are trying to find a way to make everyone happy.

 

The status of the EHCI port on Rev B is that there isn't an EHCI port on Rev B. Rev B does not have the port populated. This is not a SW issue.

 

Gerald

[Wouter Bijlsma]

Ok so the 'high-speed USB 2.0 OTG' feature advertised on
beagleboard.org is in fact not present on the current board at all?...
That kind of rules out the rev. B board for me, because hooking up an
USB LCD panel + an ethernet dongle + possible more USB devices through
a 12 Mbit USB1 port is going to suck big time.

The rev C. board will have USB2 right?

If you need a volunteer to test the prototype rev. C boards, I might
know someone heheh ;-)

Wouter

On 6 nov, 16:51, "Gerald Coley" <ger...@beagleboard.org> wrote:
> The LCD signals will definitely stay. It looks like we may end up with two
> versions where one is 128MB and the other is 256MB. We are not sure how to
> handle all of that. The responses we got were that some wanted to keep the
> price and the memory while others wanted more memory no matter what! So, we
> are trying to find a way to make everyone happy.
>
> The status of the EHCI port on Rev B is that there isn't an EHCI port on Rev
> B. Rev B does not have the port populated. This is not a SW issue.
>
> Gerald
>

[Jason Kridner]

On Nov 6, 10:02 am, Wouter Bijlsma <wouter.bijl...@gmail.com> wrote:
> Ok so the 'high-speed USB 2.0 OTG' feature advertised on
> beagleboard.org is in fact not present on the current board at all?...

Incorrect. The high-speed USB 2.0 OTG port is certainly there and
functional. The USB port being discussed is a *second* USB 2.0 host-
only (EHCI) port.

> That kind of rules out the rev. B board for me, because hooking up an
> USB LCD panel + an ethernet dongle + possible more USB devices through
> a 12 Mbit USB1 port is going to suck big time.

The OTG port is 480Mbit/s. People are using it successfully for
connecting USB-to-VGA adapters.

>
> The rev C. board will have USB2 right?

Rev B also has this.

[Robert Kuhn]

2008/11/6 Gerald Coley <ger...@beagleboard.org>:

> The status of the EHCI port on Rev B is that there isn't an EHCI port on Rev
> B. Rev B does not have the port populated. This is not a SW issue.

Sorry if this was already answered: will Rev. C has this port?

R.

[Gerald Coley]

Yes. Rev C will have the EHCI port on it.

 

Gerald

[Robert Kuhn]

2008/11/6 Gerald Coley <ger...@beagleboard.org>:

> Yes. Rev C will have the EHCI port on it.

I will love it.
Thanks for the answer.

R.

[Wouter Bijlsma]

On Nov 6, 5:43 pm, Jason Kridner <jkrid...@beagleboard.org> wrote:
> On Nov 6, 10:02 am, Wouter  Bijlsma <wouter.bijl...@gmail.com> wrote:
>
> > Ok so the 'high-speed USB 2.0 OTG' feature advertised on
> > beagleboard.org is in fact not present on the current board at all?...
>
> Incorrect.  The high-speed USB 2.0 OTG port is certainly there and
> functional.  The USB port being discussed is a *second* USB 2.0 host-
> only (EHCI) port.

My bad, I misunderstood that from the stuff I've been reading about
the current
board, I thought that the port had issues with USB2.0 in general.

>
> > That kind of rules out the rev. B board for me, because hooking up an
> > USB LCD panel + an ethernet dongle + possible more USB devices through
> > a 12 Mbit USB1 port is going to suck big time.
>
> The OTG port is 480Mbit/s.  People are using it successfully for
> connecting USB-to-VGA adapters.

Ok, if the port works fine at full speed I might have to reconsider,
no problem
driving a USB LCD with that... Eventually I'd like to interface a
small graphic
LCD (5" 320x240 or something similar), LCD signals would be fun to
play
with (though I have no idea how & what you'd need even with a rev C
BB).

OTOH, an extra full-speed USB port and double the RAM would seriously
increase the potential of the board, so I'm still not sure if It
wouldn't make
more sense waiting for the new board... I want to create a multi-
purpose
in-car computer with (ideally) mp3 player + ipod itf/ engine sensor
readout & display (there's a custom engine in it already)/ GPS
tracking/
wifi and/or ethernet/ UMTS via USB/ + ??? capabilities. I think you
can do
all of that with a rev B but seeing that this pet project might very
well end
up as an epic failure, the extra specs might at least make a nice
simple
X terminal or cute server out of it.

Wouter

[Jason Kridner]

On Nov 6, 2:34 pm, Wouter Bijlsma <wouter.bijl...@gmail.com> wrote:
> On Nov 6, 5:43 pm, Jason Kridner <jkrid...@beagleboard.org> wrote:
>
> > On Nov 6, 10:02 am, Wouter  Bijlsma <wouter.bijl...@gmail.com> wrote:
>
> > > Ok so the 'high-speed USB 2.0 OTG' feature advertised on
> > > beagleboard.org is in fact not present on the current board at all?...
>
> > Incorrect.  The high-speed USB 2.0 OTG port is certainly there and
> > functional.  The USB port being discussed is a *second* USB 2.0 host-
> > only (EHCI) port.
>
> My bad, I misunderstood that from the stuff I've been reading about
> the current
> board, I thought that the port had issues with USB2.0 in general.
>
> > > That kind of rules out the rev. B board for me, because hooking up an
> > > USB LCD panel + an ethernet dongle + possible more USB devices through
> > > a 12 Mbit USB1 port is going to suck big time.
>
> > The OTG port is 480Mbit/s.  People are using it successfully for
> > connecting USB-to-VGA adapters.
>
> Ok, if the port works fine at full speed I might have to reconsider,
> no problem
> driving a USB LCD with that... Eventually I'd like to interface a
> small graphic
> LCD (5" 320x240 or something similar), LCD signals would be fun to
> play
> with (though I have no idea how & what you'd need even with a rev C
> BB).

It works at *high* speed, in addition to full speed and low speed. In
USB terminology, full speed is 12Mbits/s. The OTG port supports all 3
speeds. The EHCI port only supports high speed.

>
> OTOH, an extra full-speed USB port and double the RAM would seriously
> increase the potential of the board, so I'm still not sure if It
> wouldn't make
> more sense waiting for the new board... I want to create a multi-
> purpose
> in-car computer with (ideally) mp3 player + ipod itf/ engine sensor
> readout & display (there's a custom engine in it already)/ GPS
> tracking/
> wifi and/or ethernet/ UMTS via USB/ + ??? capabilities. I think you
> can do
> all of that with a rev B but seeing that this pet project might very
> well end
> up as an epic failure, the extra specs might at least make a nice
> simple
> X terminal or cute server out of it.

Current specs are sufficient for X terminal functionality and the
other things you mention.

[Hanimal]

Hi Gerald,

I read something about "Traced to a bad package on TXS0102
(Delamination issue)" but i don't understand.

What will it change on BB RevC about U9 and U11?
Which new package o new reference component ?

Thank you,

Agustí

On 15 oct, 11:55, "Gerald Coley" <ger...@beagleboard.org> wrote:
> Middle of Q1 is for Rev C and not the current revision. We are switching to
> REV B6, which fixes the layout for the U9 and U11 issue to support a
> different package. This should solve the U9 and U11 issue we have been
> having. These boards will show up the first week in November. We should also
> be shipping another 200 Rev B5 boards this week.
>
> Gerald
>
> On Wed, Oct 15, 2008 at 4:55 AM, O D <ognis...@gmail.com> wrote:
> > If I may ask this question here...
> > Digi-key Europe doesn't has any more Beagleboards on the stock. Does
> > anybody knows when we can expect them again here?
> > Middle of Q1 seems a long time from now... :(
>
> > On Wed, Oct 15, 2008 at 1:37 AM, Gerald Coley <ger...@beagleboard.org>wrote:
>
> >> That has been considered. we have a lot of ideas that we continue to bat
> >> around.
>
> >> The decision we made early on was standard buses, which means standard
> >> connectors. We had not really planned to focus much at all on real
> >> expansion. The existing expansion header was just a few standard buses that
> >> in and of themselves did not have any sort of standard connectors, like I2C
> >> and I2S. This is why we did not make the LCD interface available because at
> >> 1.8V, it wouldn't drive any display anyway. We focused on keepng the cost
> >> down. Everything we add from now on must be with keeping the cost under
> >> control as it is our foremost concern. There are a lot of OMAP3 base boards
> >> out there and our goal is not to compete with them in the area of expansion
> >> and I/O. Having done 20 OMAP development board over the years, I have seen
> >> what happens when you make everyone happy. There is a lot of waste and
> >> everybody pays for it whether they use it or not.
>
> >> As to Rev D, there are no solid plans in place for that. REV C could be
> >> the final revision or we could end up at Rev F. Who knows. We plan to
> >> listen, take input, and do what we can when we can, and when it still aligns
> >> with our overall goals.
>
> >> Thank you for the input!
>
> >> Gerald
>
> >> On Tue, Oct 14, 2008 at 5:42 PM, Doug Emes <kyo...@gmail.com> wrote:
>
> >>> On Tue, Oct 14, 2008 at 11:01 AM, Gerald Coley <ger...@beagleboard.org>
> >>> wrote:
> >>> > Fellow Beaglers,
> >>> > We wanted to give you an update on the Rev C board. We have just
> >>> finished up
> >>> > the latest spin of the board, so we are running behind schedule. At
> >>> this
> >>> > point we are looking at the middle of Q1 to get the Rev C out of
> >>> production.
> >>> [snip]
>
> >>> For rev. d an onwards, have you considered having a 2nd not-expansion
> >>> board?
> >>> That is to say, using another 10pin connector (like the serial port)
> >>> to place the
> >>> Svideo, Line-out, Mic-In, and other bulky connectors off the main board?
> >>> Those
> >>> pins should be pretty tolerant of the added distance (by virtue of
> >>> their usage -
> >>> whats another couple inches). that should free up a bunch of space "on
> >>> the
> >>> top end" of the board for signal paths that really need the short
> >>> distances.
>
> >>> Just a thought off the cuff, so to speak.

[Gerald Coley]

This has actually been fixed on Rev B6 which is in production now. We switched from a BGA (YSP) package to a SOIC (DCT) package.  Basically it looks like the BGA package could not stand the heat of the lead-free soldering process. We have had no failures of these devices in the latest production run.

 

Gerald

[Gene]

I agree with the previous posters who suggest decreasing NAND to 128MB
to offset the increase of RAM to 256MB. NAND memory can be easily
offset by using an SD Card, which probably 99.9% of the users here
utilize; whereas there is no easy option to increase the RAM.

Please, keep the price at $149, increase RAM to 256 and decrease nand
to 128mb. You will have a better product that will appeal to more
users.

On Oct 16, 6:51 am, "Gerald Coley" <ger...@beagleboard.org> wrote:
> As to 256MB I wanted to let everyone know what we are thinking and where we
> are.
>
> We have been testing the 256MB memory to make sure there are no issues, so
> there are a few early REV C boards out there, 4 in fact. The concern of
> making 256MB standard on Rev C is the cost. The 256MB has a significant cost
> to it and we would need to raise the price of the board to cover the
> additional cost.
>
> We are not sure if that additional cost will hurt us in allowing us to keep
> the Beagle affordable or if the additional cost is worth it. We also have
> the additional cost of the USB host on the REVC which we are planning to
> absorb, but when you add in the added memory cost, that makes it tougher.
>
> So, right now, the plan is to keep the memory where it is with 128MB. Based
> on these discussions and other discussions, we may decide to add the 256MB
> as standard and raise the price of the Beagle. The question is how much
> increase in cost is acceptable. We may be able to pass on a portion of the
> cost and absorb the rest, but right now we are not sure if we can. We are
> due to discuss this tomorrow and I will update everyone with what we decide.
>
> In the mean time, if we could get feedback as to what the added memory is
> worth, that will help us make the case for moving forward with 256MB
> as standard on Beagle.
>
> Gerald
>
> On Thu, Oct 16, 2008 at 9:00 AM, Gerald Coley <ger...@beagleboard.org>wrote:
>
> >  As you say, that is a rumor. I have no plans to put 256MB on the Rev C
> > board.
>
> > Gerald
>
> >   On Thu, Oct 16, 2008 at 8:19 AM, Frantisek Dufka <duf...@gmail.com>wrote:
>
> >> Gerald Coley wrote:
> >>  > The key changes are as follows:
>
> >> > 1) Addition of the USB Host port....
>
> >> > 2) We have added access to the native LCD signals...
>
> >> > If you have any questions, please feel free to let us know.
>
> >> Ther are some rumours about rev C having 256MB of RAM, see
> >>http://elinux.org/BeagleBoard#Revision_C
> >>http://www.beagleboard.org/irclogs/index.php?date=2008-10-01#T19:10:58
>
> >> Are we getting 256MB RAM in rev C?
>
> >> Regards,
> >> Frantisek