SoC: Back and forth!


Without actually looking up anything about the SoC itself I can already guess that ED won't be interested: if the A80 is effectively off the table because it would take too long to get to market something that has only just been announced will definitely not be viable.
 
Or do you just want the Pyra out faster even if it's weaker?
 
Or do you just want the Pyra out faster even if it's weaker?
@Natsu


Or do you just want the latest or greatest thing without considering the many factors that could make or break the project.. Like lack of Linux A80 drivers, cloudy release/delivery schedule.. etc
 
Last edited by a moderator:
Remember, we have a daughter board to swap.

Even if the OMAP 5 is pretty sure used in the first Pyra it still makes sense to watch new processors as we might be able to use them later.
 
Without actually looking up anything about the SoC itself I can already guess that ED won't be interested: if the A80 is effectively off the table because it would take too long to get to market something that has only just been announced will definitely not be viable.
Actually it's not a SoC, but only a processor.Also it's not made for mobile use, but for performance, so it has higher power requirements.
 
Or do you just want the Pyra out faster even if it's weaker?
There will always be something better. If ED waits for the A80 there will, without doubt, be something even better "coming soon". At that point will you say that the A80 is weaker, that anyone that wants the A80 just wants it out faster?Your appeal to emotion has no effect on me, and it has no effect on ED: he's doing his best fueled entirely by hard facts.
 
Last edited by a moderator:
Or do you just want the Pyra out faster even if it's weaker?
@Natsu
Or do you just want the latest or greatest thing without considering the many factors that could make or break the project.. Like lack of Linux A80 drivers, cloudy release/delivery schedule.. etc
Why did you say "@Natsu" and then quote me anyway? :lol: A80 isn't really the latest or greatest thing. A80 Linux drivers seem to already have an answer. I already know about the time restraint, but he seems to be against the A80 or any change really, from the get go. That is why questioned him.
Or do you just want the Pyra out faster even if it's weaker?
There will always be something better. If ED waits for the A80 there will, without doubt, be something even better "coming soon". At that point will you say that the A80 is weaker, that anyone that wants the A80 just wants it out faster?Your appeal to emotion has no effect on me, and it has no effect on ED: he's doing his best fueled entirely by hard facts.
Of course there will always be something better, but at least a80 is closer to now and more capably. OMAP5 is so far behind.
Oh, didn't give me that it's facts only when this forum is filled with opinions that affect decisions.
 
Last edited by a moderator:
I think the best method is the safest on this issue. 

We can have more features than Pandora, but having too many things going on at the same time is spreading it too thin. 

There's no point in giving Pyra the most powerful soc possible if that does not increase the sale. Yes, it can make pyra somewhat future proof, but pyra by itself is future proof enough. 

It's not worth it to risk losing money for a bit more performance. 

Even if the A80 is not much more expensive than Ti's offer, we still loss time. By the time A80 is ready for everyone, who knows what the next hardware hype is. 

As long as Pyra can be an alternative to my laptop, it's good enough. 800 bucks is my top bid, anything more than that, it would be a tough purchase. It's also true that the earlier Pyra's coming out, the easier my purchase decision would be. I don't see the demand of more performance being any sincere, no one's willing to pay for that. 

This being said, Pyra 1 should be a good foundation for anything coming after it. We may even have a x86 pyra if pyra 1 did well enough. Those promises are all for nothing until ED puts something saleable on the table. 
 
Or do you just want the Pyra out faster even if it's weaker?
There will always be something better. If ED waits for the A80 there will, without doubt, be something even better "coming soon". At that point will you say that the A80 is weaker, that anyone that wants the A80 just wants it out faster?
Your appeal to emotion has no effect on me, and it has no effect on ED: he's doing his best fueled entirely by hard facts.
my personal feelings on this matter is that the a80 is worth waiting for regardless of how long the wait ends up being, just like how the pyra will be worth the wait regardless of how long the wait ends up being and without knowing quality of the end product. But only the a80 specifically, not just because it's bleeding edge. But just the fact that it's true to spec big.little and will see lots of community support development support which I'm guessing the omap will hardly see, if it sees any at all outside our ring. A big.little a15/a7 is literally designed for mobile handhelds, where the omap looks to be made for other applications and is being shoehorned into a mobile. I strongly feel the wait will be worth it.

Unofficial or sketchy linux support is a less than ideal card that the a80 holds, but there is contenders to make it a reality even if unofficially, but official looks promising as well.

I understand ED's position that he doesn't want to rest the fate of his company in a gut feeling, or a unreliable thing that looks good on paper too however.

This isn't about catching moore's law, this is way more than that, even if something better came out after the a80 which is sure to happen in 6 months, the a80 would still remain a solid choice. Where I'm not so sure about the omap. Without seeing thermal and power consumption data from the omap compared to a unreleased a80 it's impossible to say for certain. But I have a very strong feeling that the a80 is going to perform very well in both arenas, a feeling I don't get at all with the omap. The thing that keeps leaving a really bad taste in my mouth with the omap is the m4 cores that I haven't really even seen a good suggested use for them. Where the a7 cores I've seen plenty suggested and implemented in other devices and are proven to be good at what they advertise at being good at.

I guess the question remains is it worth the money and investment to pursue the omap, and have the a80 on the back burner for an upgrade, then invest in the a80 in the future if that ends up being a choice. Also splintering first adopters from later adopters with such a radically different chipset. ED will have to do something with the omap chips if he buys them and they don't sell off before the a80's become available which could push a release date for the upgrade even further and even if appealing now and in the future will be less apealing when he finally gets the lesser ideal chips offhand. I remember the upgrade discussions on the pandora when they were talking about having to burn off their onhand supply of omap 3530's before going to the DM3730 chips.
 
Of course there will always be something better, but at least a80 is closer to now and more capably. OMAP5 is so far behind.
Which is exactly the situation that will be when the A80 is available and the "something better" is closer: the A80 will be "so far behind" and anyone unwilling to wait for whatever that may be obviously "just want(s) the Pyra out faster even if it's weaker". With any luck it will be an infinite and neverending cycle, I am afraid of the time when there isn't something new and exciting just around the corner. In the meantime we have to settle on what is available now.
Oh, didn't give me that it's facts only when this forum is filled with opinions that affect decisions.
Opinions alone haven't driven a single decision; if you have a counter point I would love to see it. Every decision that has been made on an opinion has been reinforced with facts. ED isn't in the habit of adding something just for the hell of it, it needs to be justified.Calling one thing "weaker" is a blatant appeal to emotion: you could have used a lot of different phrases but you chose the one that tries to make someone feel bad for going against it: it is a bad argument and it won't work; if you genuinely think waiting for the A80 is worth it try doing some work, get the facts together, and present your case. If you think the risk of the 6 months now will pay off in the long run then you should have no problem proving it.
 
But only the a80 specifically, not just because it's bleeding edge. But just the fact that it's true to spec big.little and will see lots of community support development support which I'm guessing the omap will hardly see
@Natsu, take notes, this is how you present an argument, an excellent point on why the A80 is worth waiting for that doesn't recurse into waiting forever.
 
A80 isn't really the latest or greatest thing. A80 Linux drivers seem to already have an answer.
And that answer is where...?

Allwinner doesn't offer Linux drivers, so do you have a source for Linux drivers, and kernel mainline support?

I already know about the time restraint, but he seems to be against the A80 or any change really, from the get go. That is why questioned him.
If I was against a change, I wouldn't have contacted Allwinner, would I?

Getting the A80 would push us back at least 6 months, as we need to learn about the new SoC, read the documentation, implement the hardware and get the software (drivers, etc.) to work.

OMAP5 stuff is mostly implemented in Linux Mainline Kernel and that doesn't seem to be the case for the A80.

So besides getting the hardware to work, software would be the next big thing.

And as long as you don't have the software side working, you can't fully test the hardware...
 
I meant WizardStan was against change. I look more at this later.
 

@Natsu, take notes, this is how you present an argument, an excellent point on why the A80 is worth waiting for that doesn't recurse into waiting forever.
Pretty much the same point, just greatly expanded. Tho, I wouldn't call it bleeding edge.
Besides, can you really type that? :p Pot calling the kettle black.
 
Last edited by a moderator:
A80 isn't really the latest or greatest thing. A80 Linux drivers seem to already have an answer.
And that answer is where...?
Allwinner doesn't offer Linux drivers, so do you have a source for Linux drivers, and kernel mainline support?

I already know about the time restraint, but he seems to be against the A80 or any change really, from the get go. That is why questioned him.
If I was against a change, I wouldn't have contacted Allwinner, would I?
Getting the A80 would push us back at least 6 months, as we need to learn about the new SoC, read the documentation, implement the hardware and get the software (drivers, etc.) to work.


OMAP5 stuff is mostly implemented in Linux Mainline Kernel and that doesn't seem to be the case for the A80.


So besides getting the hardware to work, software would be the next big thing.


And as long as you don't have the software side working, you can't fully test the hardware...
If you are able, how many omap chips do you think you will be ordering if any in your first bulk order if that ends up being the first iteration pyra?


I have heard they can be ordered in smaller quantities, but even if you could in theory order them in batches less than 100 you wouldn't want to only populate that many daughter boards.


Knowing if you are going to order 10k pieces would give a ballpark idea on how long choices made now will have an effect in the future.
 
I meant WizardStan was against change
Lies and slander. I love change, I'm against bad arguments.
Pretty much the same point, just greatly expanded
Not the same point. Your entire argument, the one and only thing you said, was that the OMAP5 was weaker. After I pointed out the fault in just using that as an argument you effectively re-iterated the same flawed argument. SNESFAN saw the flaw in the argument and made additional points that make the A80 specifically worth waiting for, not just because it is better but because of features it and it alone has.
 
But waiting until the A80 is out mean that the dragon box will die... I dont understand how that many person cant graps that :(
 
Back
Top