Overview of SoC options


Although the support is still very basic and more user-space changes will be needed to make the full graphics stack run on top of it, we were able to successfully open channels and run simple pushbuffers with libdrm.
  I know almost nothing about graphics programming (and because of that maybe I'm terribly wrong), but to me, this looks promising. In fact integrating the know-how of nVidia into the noveau driver looks to me way better than releasing a new and divergent driver. As the article says, "This Tegra K1 Nouveau support is still proof-of-concept but is a huge step forward in coming from NVIDIA themselves.".


Without knowing much more than you (I think), this sounds to me like "We were able to successfully greet each other in a way that seemed not to cause long-lasting enmity on both sides" when trying to contact a new culture :) That is "Open channels and run simple pushbuffers" is the equivalent of "Hello world" or, in the language sense "Me Tarzan. Bring food". 

There's quite some way to go to get to Shakespeare. Or, back from the analogy, to something that approaches "open graphics driver" in any meaningful way.

Unless I'm gravely mistaken, that is.
 
I would be fine w/ that if he wasn't grossly misunderstanding why I made the comment about the USB ports.
After rereading your post, I guess I have to apologize for misreading you there, totally did not get what you where suggesting.But the end result is even worse:

" I was suggesting that EvilDragon could have said what he said to shoot down a request for more USB ports."

Why should he need to do so ? Was there ever a case where he did that ?. And speaking of putting intentions in other peoples mouth:

By using "I think", "I believe", or similar wording, you strengthen the fact that this is your opinion. Without it, the sentence reads as if you are informed, as if you are stating a fact rather than an assumption.
If I am informed, why should I use "probable" (and even relativize that by stating that it may be not be the case) ? It would only weaken my argument if I would do so.

Actually, EvilDragon did make comments about increasing ports. He specifically mentioned that he would have some sort of HDMI-out or similar, but that he was also keeping the EXT port for other purposes so that you could still reuse the TV-out cable for line in and out.
I know, but he also commented on a question wether the barrel jack will be kept or not, that he isn't sure wether there is space left. I could also start arguing wether the addition of one port could count as "massive", but I'd rather get back to the original question:
Where does the needed extra space wondrously come from ?
 
Last edited by a moderator:
but I'd rather get back to the original question:

Where does the needed extra space wondrously come from ?
I'm fine with that. We can drop the rest of the argument. (I have things I can say, but the comments people started making about marriage is starting to weird me out, so I'd rather not give them more fuel)

So, to address the question "Where does the needed extra space wondrously come from ?" there are a couple of points to make:

First, your point about the lack of space to place ports on the back of the device only proves that the back would be crowded. It doesn't say anything about the rest of the circuitboard. So it is possible that there could be extra space on the board somewhere else.

And most importantly, Grench has already explained 2 pages ago that there is no need for extra space in the case that EvilDragon includes a SIM card:

Well sheesh - if ED is going to put a SIM card slot into the thing, then microSD adds NO additional motherboard space requirement and removes the space used by the eMMC. 2 in 1 (SIM and microSD) slots have been around for years now.

http://www.smkusa.com/usa/featured_products/SIM_micro/

http://www.molex.com/molex/products/listview.jsp?query=&offset=0&filter=&fs=productname%3Aor(microSD||transflash)&npp=20&sType=z&autoNav=&path=cMEMORYCARDSOCKET%23%230%23%231j~~nf45||32&channel=Products&key=fb_fan
-God Ginrai
 
Last edited by a moderator:
First, your point about the lack of space to place ports on the back of the device only proves that the back would be crowded. It doesn't say anything about the rest of the circuitboard. So it is possible that there could be extra space on the board somewhere else.
No it doesn't as these ports don't exist for themselves, they need to be wired to other components. But that was only one part of the argument, look at the pcb again, and then tell me where to put it - keeping in mind, that you need additional space to fit in a xG module plus traces plus antenna, plus traces for the hdmi out, etc.
And most importantly, Grench has already explained 2 pages ago that there is no need for extra space in the case that EvilDragon includes a SIM card:
Which is (although that this specific component is discontinued) a valid possibility, but by using it, you are limiting your "tetris" possibilities again, as you need to put traces on the board for both which may get more complicated than having two seperate components where only one needs to be accessible by the user. But that was way after you stated your point of view, so I thought there was/is more.
 
Last edited by a moderator:
First, your point about the lack of space to place ports on the back of the device only proves that the back would be crowded. It doesn't say anything about the rest of the circuitboard. So it is possible that there could be extra space on the board somewhere else.
No it doesn't as these ports don't exist for themselves, they need to be wired to other components. But that was only one part of the argument, look at the pcb again, and then tell me where to put it - keeping in mind, that you need additional space to fit in a xG module plus traces plus antenna, plus traces for the hdmi out, etc.
And as I mentioned before: We're not putting these things onto the same PCB. A new PCB is being designed. It's not even the same person designing the PCB, so it could be very different. You can't go making these judgements off of the Pandora PCB.

And most importantly, Grench has already explained 2 pages ago that there is no need for extra space in the case that EvilDragon includes a SIM card:
Which is (although that this specific component is discontinued) a valid possibility, but by using it, you are limiting your "tetris" possibilities again, as you need to put traces on the board for both which may get more complicated than having two seperate components where only one needs to be accessible by the user. But that was way after you stated your point of view, so I thought there was/is more.
Actually, we're not really limiting the "tetris" possibilities, since we are actually creating positive space by removing the intended eMMC.

-God Ginrai
 
And as I mentioned before: We're not putting these things onto the same PCB. A new PCB is being designed. It's not even the same person designing the PCB, so it could be very different. You can't go making these judgements off of the Pandora PCB.
And as I already mentioned before, we got the same space, more features and there is no reason to believe that MWeston was very wasteful in placing things. Sure I can make judgements, thats what all people do all day long, taking their previous experiences/knowledge to predict the outcome of their next action. Why shouldn't that be working here.
Actually, we're not really limiting the "tetris" possibilities, since we are actually creating positive space by removing the intended eMMC.
Is the SIMSlot-MicroSD combo significantly smaller than eMMC + SIMSlot ?
 
And as I mentioned before: We're not putting these things onto the same PCB. A new PCB is being designed. It's not even the same person designing the PCB, so it could be very different. You can't go making these judgements off of the Pandora PCB.
And as I already mentioned before, we got the same space, more features and there is no reason to believe that MWeston was very wasteful in placing things. Sure I can make judgements, thats what all people do all day long, taking their previous experiences/knowledge to predict the outcome of their next action. Why shouldn't that be working here.
Yes, but there's also no reason to believe that MWeston was completely efficient. Also, while we have mostly the same things we are adding, we have no idea if any of those parts are the same size they were on the Pandora. (is the WiFi/BT module the same size, for example?)

Actually, we're not really limiting the "tetris" possibilities, since we are actually creating positive space by removing the intended eMMC.
Is the SIMSlot-MicroSD combo significantly smaller than eMMC + SIMSlot ?
That's what the conversation between Grench and slaeshjag seemed to point to.

-God Ginrai
 
Yes, but there's also no reason to believe that MWeston was completely efficient.
But the new designer will ? But I doubt that efficiency is of that importance in that case - not in general of course, but at this level of design competence, I see that as a "minor" problem.
Also, while we have mostly the same things we are adding, we have no idea if any of those parts are the same size they were on the Pandora. (is the WiFi/BT module the same size, for example?)
I doubt that you will see size decrease in a scale that matters, most of these components have a size that is mainly determined by the count of "pins" they need.
That's what the conversation between Grench and slaeshjag seemed to point to.
I thought it was about the clearance area, but maybe I'm wrong here. Another problem that could occur with the second solution grench presented is height and position: While "average joe/jane" may not have a big interest in replacing the interal memory, he/she will probably want to put in a Sim card, so you need it to be placed on a rather "exposed" position - preferebly under the battery compartment, but will it be placeable beneath it if it as thick as the Sim card+ microSD + housing ?
 
Yes, but there's also no reason to believe that MWeston was completely efficient.
But the new designer will ? But I doubt that efficiency is of that importance in that case - not in general of course, but at this level of design competence, I see that as a "minor" problem.
 So you won't doubt MWeston, but you're all for doubting the new guy? There's not "one" way to do things. A reorganization of the parts on the PCB, which is completely possible to happen with someone new behind the design, could potentially open up more space than the design that MWeston had created.

Also, while we have mostly the same things we are adding, we have no idea if any of those parts are the same size they were on the Pandora. (is the WiFi/BT module the same size, for example?)
I doubt that you will see size decrease in a scale that matters, most of these components have a size that is mainly determined by the count of "pins" they need.
 The dimensions of the TI wl1251 and the TI wl1271 differ a great degree. Sadly, the TI wl1271 is larger, but this supports my assertion that the parts we plan to include on the PCB could differ in size which would lead to different optimal organization patterns on the PCB, which is even more reason why we can't just look at a Pandora PCB and say "where is there room?".

That's what the conversation between Grench and slaeshjag seemed to point to.
I thought it was about the clearance area, but maybe I'm wrong here. Another problem that could occur with the second solution grench presented is height and position: While "average joe/jane" may not have a big interest in replacing the interal memory, he/she will probably want to put in a Sim card, so you need it to be placed on a rather "exposed" position - preferebly under the battery compartment, but will it be placeable beneath it if it as thick as the Sim card+ microSD + housing ?
Why wouldn't it be?

-God Ginrai
 
Is Linus getting more crazy every year or was he like that from the beginning?
 
Linus is finnish, from the swedish contingent. There's a particular streak of dark humour, weirdness and anti-authoritarianism that goes with the endless forests and the dark winters. I find Linus to perfectly sensible, especially compared to other known linux personalities (like ESR or, god forbid, RMS), but that might come with the shared worldview :) This is the same kind of humour that causes a Swedish game manufacturer to name their otherwise completely normal card game "Spank the Monkey", just to laugh at the silly americans who for some reason seems to be unable to stock the game :D

Or that makes us think this is a fitting entertainment for a collection of world leaders and prominent politicians, to explain the nordic countries:

http://www.youtube.com/watch?v=M2ez11LkUwM

Yes, you can see the then swedish prime minister and royal family in there, laughing their butts off. The foreign dignitaries seems a bit more...confused :)
 
So you won't doubt MWeston, but you're all for doubting the new guy? There's not "one" way to do things. A reorganization of the parts on the PCB, which is completely possible to happen with someone new behind the design, could potentially open up more space than the design that MWeston had created.
No, I don't doubt the new guy in general. I doubt that he will do a much(!) better job than MWeston did. I think it is quite possible that he will create a more space efficient layout. He has already experience in designing at least two phone pcbs, which gives him the credit of bringing up workable designs within tight spaces, and also the "judgement" from the outside, that he is doing a fine job (otherwise he may not have had a chance to design the neo900 pcb or Ed choosing him for the Pyra). I just doubt that he can work wonders.
The dimensions of the TI wl1251 and the TI wl1271 differ a great degree. Sadly, the TI wl1271 is larger, but this supports my assertion that the parts we plan to include on the PCB could differ in size which would lead to different optimal organization patterns on the PCB, which is even more reason why we can't just look at a Pandora PCB and say "where is there room?".
And the Omap5 is also significantly larger than the Omap3, but even that size difference looks rather small in comparison to what the possible xG Module is eating up (I hope I got the right one). And there is still the RAM, which will too eat space (but I do not know how much, the pop package size may differ greatly from the size of single modules). And if I remember Eds FOSDEM speech correctly we will also need a heatsink which may limit the placement of parts too.

Why wouldn't it be?
Because Ed mentioned one of the design goals was to not get the Pyra thicker than the Pandora. So with a latch mechanism like the part that Grench found, you will need access from the top. Which either needs a seperate door/lid/hatch to the outside if you want to place it anywhere on the board (which I guess is not a preferable option as its a possible weak point for the case), or you place it under the battery as you won't need a door/latch then, with the battery protecting the whole thing. But if you place it under the battery thickness will (or at least could) be a problem.
 
Last edited by a moderator:
So you won't doubt MWeston, but you're all for doubting the new guy? There's not "one" way to do things. A reorganization of the parts on the PCB, which is completely possible to happen with someone new behind the design, could potentially open up more space than the design that MWeston had created.
No, I don't doubt the new guy in general. I doubt that he will do a much(!) better job than MWeston did. I think it is quite possible that he will create a more space efficient layout. He has already experience in designing at least two phone pcbs, which gives him the credit of bringing up workable designs within tight spaces, and also the "judgement" from the outside, that he is doing a fine job (otherwise he may not have had a chance to design the neo900 pcb or Ed choosing him for the Pyra). I just doubt that he can work wonders.
 I don't think finding space for one part is "working wonders". If I were wanting him to add a laundry list of things, then I'd totally agree with you.

The dimensions of the TI wl1251 and the TI wl1271 differ a great degree. Sadly, the TI wl1271 is larger, but this supports my assertion that the parts we plan to include on the PCB could differ in size which would lead to different optimal organization patterns on the PCB, which is even more reason why we can't just look at a Pandora PCB and say "where is there room?".
And the Omap5 is also significantly larger than the Omap3, but even that size difference looks rather small in comparison to what the possible xG Module is eating up (I hope I got the right one). And there is still the RAM, which will too eat space (but I do not know how much, the pop package size may differ greatly from the size of single modules). And if I remember Eds FOSDEM speech correctly we will also need a heatsink which may limit the placement of parts too.
Yes, ED did mention a heatsink. I have no idea if this will limit it in any way. :\

Why wouldn't it be?
Because Ed mentioned one of the design goals was to not get the Pyra thicker than the Pandora. So with a latch mechanism like the part that Grench found, you will need access from the top. Which either needs a seperate door/lid/hatch to the outside if you want to place it anywhere on the board (which I guess is not a preferable option as its a possible weak point for the case), or you place it under the battery as you won't need a door/latch then, with the battery protecting the whole thing. But if you place it under the battery thickness will (or at least could) be a problem.
I was under the impression that there was already stuff under the battery on the Pandora PCB. (Just not exposed) Am I wrong? If I'm not, then how would something being under the battery on the Pyra PCB increase its thickness beyond what the parts under the Pandora battery did?

-God Ginrai
 
Back
Top