Not perfect, but development just started.
SoC Poll - What's most important for you?
- Thread starter EvilDragon
- Start date
Not perfect, but development just started.
God Ginrai
Godmaster
Umm... If you've noticed Exophase's comments, he's pointed out numerous times (because people seem to keep ignoring it) that the x86 SoC doesn't seem to have an EVM available, whereas the OMAP has its EVM and the A80 has the OptimusBoard. EvilDragon has already been testing the OMAP EVM, the OptimusBoard comes out in March or April, and the intel SoC doesn't have an apparent EVM for him to test with. How do you propose EvilDragon does what you are asking him to do?
-God Ginrai
hollowshell18
Still Fresh
- Joined
- Jun 20, 2013
- Messages
- 9
In response to WizardStan, that's a fair assessment sure. I may not be a frequent poster, but I am long time fan of the project and I'm quite familiar with the current situation. I completely agree that priority number one should be keeping EvilDragon in business, and I agree that he should avoid as much risk as possible. I just felt after reading this discussion that the question of "what will we do with the Pyra once it's finished" was being drown out by people's attachment to one architecture over another.
I've cast my vote for the "I don't care about the architecture" option; I'm confident that we'll be gaining quite a bit either way.
...Though I'd still love to have Oblivion on a handheld someday.
This eases a lot of doubts. Sure, its a long ways off from being full speed on most games, but it's still very impressive progress. This inspired me to look up what was available for the Gamecube, and I was pleasantly surprised. I'd forgotten just how many great games came out for it, and how many great ones were ported to it as well. Lots of gems.
I've cast my vote for the "I don't care about the architecture" option; I'm confident that we'll be gaining quite a bit either way.
...Though I'd still love to have Oblivion on a handheld someday.
SNESFAN
Retro game fanatic
There are a few exceptions, he specifically along with some others I won't name (as to not exclude or put them on the podium) are the valued opinions that we as a community need to listen to. Not these arguments about it possibly inviting populations of people that may not be welcome and things of that nature. If something isn't possible its obviously not an option so why bother drilling the issue into the ground.
Edit: But to say xy or z option should or shouldn't be used because of a preconceived idea that it will be hotter, cooler, faster, more capable or efficient from a hardware perspective without actually testing those and providing those tests seem counterproductive. Having an official poll on the mater before info is given seems to just call out to see which direction the bias in the community leans. Not which would actually be the best option.
Edit: But to say xy or z option should or shouldn't be used because of a preconceived idea that it will be hotter, cooler, faster, more capable or efficient from a hardware perspective without actually testing those and providing those tests seem counterproductive. Having an official poll on the mater before info is given seems to just call out to see which direction the bias in the community leans. Not which would actually be the best option.
Last edited by a moderator:
God Ginrai
Godmaster
Well, if this topic was locked like I asked, then people wouldn't be able to attempt to use this topic to sway people's votes.
-God Ginrai
They would tried to do that in the other thread. Stop trying to stop discussion.
Last edited by a moderator:
Hồng Thất Công
Đả Cẩu Bổng Pháp
I'd say lock this thread up, scrap the poll. Let ED and his team make the decision behind closed door.
Null
Snug
I vote for D-Wave Two.
Neelix
Insecticidal Maniac
Is this thread the new Absurdism Corner then?
- Neelix
- Neelix
Null
Snug
Last edited by a moderator:
I'd say lock this thread up, scrap the poll. Let ED and his team make the decision behind closed door.
God Ginrai
Godmaster
They may try, but they can't directly effect them like discussion in this topic can. The vote is always at the top of the page in this topic, so someone could say something to sway your vote, and OH, there's the vote button.They would tried to do that in the other thread. Stop trying to stop discussion.
-God Ginrai
Current voting trend show that people don't care about long term support.
Well nothing is unusual here compared to the rest of the world as people usually prefer fast processor when buying phones or tablets nowadays (some prefer battery life or better both) and starting to complain about (the lack of) support AFTER they bought it. So they have what they want - manufacturers produce faster toys every year and provide little to no support (especially when comparing to usual computers).
What's so special about our community if it have basically the same values as others but just pretending to be different?
Or maybe there is a difference but it's not shown because there is no 'what's your second most important question' question in this poll?
Well nothing is unusual here compared to the rest of the world as people usually prefer fast processor when buying phones or tablets nowadays (some prefer battery life or better both) and starting to complain about (the lack of) support AFTER they bought it. So they have what they want - manufacturers produce faster toys every year and provide little to no support (especially when comparing to usual computers).
What's so special about our community if it have basically the same values as others but just pretending to be different?
Or maybe there is a difference but it's not shown because there is no 'what's your second most important question' question in this poll?
Last edited by a moderator:
It seems while it may not be their most important,it still shows that they prefersome support.
hmc
Active Member
Hi guys,
in 1992, which, IIRC, was the era when most people's PCs had 80386 processors and 80486 were the top of the line, Intel produced an x86 SoC that was compatible with the 80186 processor and had some of the features of the 80286 on-board, too. The "Hornet" chip.
It was so low-power that the device (HP 100LX / HP 200LX) could be used for WEEKS on one set of 2 AA alkaline batteries without recharging. Okay, it did not have a color backlit TFT screen of course, but a 640x200 monochrome LCD, but - doesn't this prove, that x86 and low-power is no contradiction in general?
It's a mystery to me, why it was not possible to build another low-power x86 SoC for so many years in between. Maybe simply because OSes (or people) are so demanding that it would not be possible to satisfy their needs with a processor that is 1-2 generations back in time.
But if you also use an OS that's a few years old, if you use optimized code, if you don't expect the pocketable computer to have similar performance specs as your high-end desktop PC, you can be very happy with an older processor, too!
I understand that using software from some years ago will be harder and harder, because development cycles become shorter and shorter, and to keep compatibility with other systems, or to secure your system against attacks from the Internet (also not an issue 20 years ago!), you NEED to have the latest software versions.
Going the x86 way could mean attracting a group of potential customers, that are now completely blocked out:
Business customers.
Imagine you can run Windows (if we go x86, we need to take care that the device is Windows compatible IMHO, even if most of us - me included - don't care about Windows at all), install Outlook, sync with the Exchange server, be connected to your company's network and work on the go, using the Pyra. I think this could really be a breakthrough, a new market. A market that I always wanted to serve with my mobile computers "business", and which I could serve well with the HP 200LX back then, but no other pocketable device was really capable of it so far (which results in some people still using their 200LX daily for business-related stuff. I know, because they are my customers ). Even the Atom-based Viliv and UMID Palmtops failed, just because of their bad quality, and then the companies even shut down. I offered the Pandora to some business users, but it failed on them (no: they failed on it ;-), mainly due to lack of relevant software.
The question is of course: Do we want to serve that market at all? Do we want to have a lot of Windows "Spam" in our forums. The least would be, to make a clear separation. Call the x86 Pyra differently (maybe Pyrus? ;-) No, there is an ebook reader with that name already! But you get the idea), have a different forum for it...
I still don't know what's better
My favorite way would still be to go the ARM route first, to make Pyra famous as the "Pandora successor", including all the backwards compatibility to the many hundreds of existing PNDs, and a few months later offer an x86 Pyra. Then we will see, which device gets more attention and which device has most potential.
in 1992, which, IIRC, was the era when most people's PCs had 80386 processors and 80486 were the top of the line, Intel produced an x86 SoC that was compatible with the 80186 processor and had some of the features of the 80286 on-board, too. The "Hornet" chip.
It was so low-power that the device (HP 100LX / HP 200LX) could be used for WEEKS on one set of 2 AA alkaline batteries without recharging. Okay, it did not have a color backlit TFT screen of course, but a 640x200 monochrome LCD, but - doesn't this prove, that x86 and low-power is no contradiction in general?
It's a mystery to me, why it was not possible to build another low-power x86 SoC for so many years in between. Maybe simply because OSes (or people) are so demanding that it would not be possible to satisfy their needs with a processor that is 1-2 generations back in time.
But if you also use an OS that's a few years old, if you use optimized code, if you don't expect the pocketable computer to have similar performance specs as your high-end desktop PC, you can be very happy with an older processor, too!
I understand that using software from some years ago will be harder and harder, because development cycles become shorter and shorter, and to keep compatibility with other systems, or to secure your system against attacks from the Internet (also not an issue 20 years ago!), you NEED to have the latest software versions.
Going the x86 way could mean attracting a group of potential customers, that are now completely blocked out:
Business customers.
Imagine you can run Windows (if we go x86, we need to take care that the device is Windows compatible IMHO, even if most of us - me included - don't care about Windows at all), install Outlook, sync with the Exchange server, be connected to your company's network and work on the go, using the Pyra. I think this could really be a breakthrough, a new market. A market that I always wanted to serve with my mobile computers "business", and which I could serve well with the HP 200LX back then, but no other pocketable device was really capable of it so far (which results in some people still using their 200LX daily for business-related stuff. I know, because they are my customers
). Even the Atom-based Viliv and UMID Palmtops failed, just because of their bad quality, and then the companies even shut down. I offered the Pandora to some business users, but it failed on them (no: they failed on it ;-), mainly due to lack of relevant software.The question is of course: Do we want to serve that market at all? Do we want to have a lot of Windows "Spam" in our forums. The least would be, to make a clear separation. Call the x86 Pyra differently (maybe Pyrus? ;-) No, there is an ebook reader with that name already! But you get the idea), have a different forum for it...
I still don't know what's better
My favorite way would still be to go the ARM route first, to make Pyra famous as the "Pandora successor", including all the backwards compatibility to the many hundreds of existing PNDs, and a few months later offer an x86 Pyra. Then we will see, which device gets more attention and which device has most potential.
I fully agree. But maybe it will be helpful for them to have our input from this thread.I'd say lock this thread up, scrap the poll. Let ED and his team make the decision behind closed door.
Last edited by a moderator:
@hmc In business world money matters and anything else does not and it's not a good business practice for ED to go wintel expecting that business customers would just pop out suddenly.
A business customer can pay more money than non-business customer but it requires real support in return, not some geeky community spending some of their free time on pyra software. Can ED satisfy such requirements?
Maybe you can do it instead if the need arises?
But basically business prefers large enough companies just for that reason (ability to provide phone support and maybe even some engineering support).
You probably know how it works: a company makes a contract with some company (called system intergrator) to provide some product and support for it. That contractor company then buys hardware from manufacturing company (in our case it's ED's DragonBox shop), installs some software and provides support for that software and solves hardware-software integration problems by whatever means necessary.
In the end it's all about money. If you can find enough customers willing to pay for wintel pyra AND could provide support for such configuration (or this can be done by ED's company if he's willing to do so for that money) then we'll see wintel pyra.
No money = no wintel pyra. Spread the word, try to interest business customers. That's the only way.
My guess is that business customers could show more interest in such device after pyra starts producing because it will be evident by then that ED's company can successfully design and produce such device. We believe in ED but business people don't like additional risks without big revenue for them attached to these risks, just look at ED
A business customer can pay more money than non-business customer but it requires real support in return, not some geeky community spending some of their free time on pyra software. Can ED satisfy such requirements?
Maybe you can do it instead if the need arises?
But basically business prefers large enough companies just for that reason (ability to provide phone support and maybe even some engineering support).
You probably know how it works: a company makes a contract with some company (called system intergrator) to provide some product and support for it. That contractor company then buys hardware from manufacturing company (in our case it's ED's DragonBox shop), installs some software and provides support for that software and solves hardware-software integration problems by whatever means necessary.
In the end it's all about money. If you can find enough customers willing to pay for wintel pyra AND could provide support for such configuration (or this can be done by ED's company if he's willing to do so for that money) then we'll see wintel pyra.
No money = no wintel pyra. Spread the word, try to interest business customers. That's the only way.
My guess is that business customers could show more interest in such device after pyra starts producing because it will be evident by then that ED's company can successfully design and produce such device. We believe in ED but business people don't like additional risks without big revenue for them attached to these risks, just look at ED
Last edited by a moderator:
I don't think that x86 vs ARM makes a big difference. But Windows vs GNU/Linux makes a huge difference. If the Pyra would run Windows, I think a big part of the current community (and most of the devs) would immediately say "meh" and lose interest. I am not interested in a Windows device at all. I hate Windows, I hate it even more than iOS, which I also hate. Having to use Windows makes me want to punch and strangle things. It is a extremely frustrating experience to me. I try to stay as far away as possible from anything that even remotely smells like Windows.
Similar threads
