Pandora OMAP4 and POSITIVE FEEDBACK!!

[Elanzer]

OMAP4 isn't a large increase over what the Pandora has now, it isn't even a doubling. You can overclock the Pandora to roughly OMAP4 CPU performance levels (GPU is different, but that isn't an emulation bottleneck here). Rather than fragment the community so easily over next to nothing, it will require another couple generations before Pandora 2 should be bothered with.

 

[WizardStan]

Exactly. Yes, upgrading your PC is very easy, but how many people upgrade their 1.6Ghz processor to a 1.8Ghz just because they can, then to the 2.0Ghz a few months later, then the 2.4Ghz a few months after that? Nah, you buy the 1.6Ghz, it was good enough for a while until finally the 2.4Ghz came out and you decide it is time to upgrade.


Or maybe graphics card would have been a better example, since they're easier and more compatible. Normally you wouldn't upgrade to the latest video card as soon as it comes out, would you? Maybe upgrade every few revisions, skipping the ones in between. Unless you're insane

 

[Exophase]

OMAP4 isn't a large increase over what the Pandora has now, it isn't even a doubling. You can overclock the Pandora to roughly OMAP4 CPU performance levels (GPU is different, but that isn't an emulation bottleneck here). Rather than fragment the community so easily over next to nothing, it will require another couple generations before Pandora 2 should be bothered with.

OMAP4 has a dual-core Cortex-A9, which per-core should have substantially IPC performance than Cortex-A8. OMAP4440 is alleging 1.5GHz. An overclocked OMAP3530 isn't even in the same ballpark, and we don't know how far OMAP4 will overclock.


It'll be a while before we see ARM emulators get serious about multi-core, but it could be a big break for some like DS and Saturn.

 

[SNESFAN]

I have a hard set standard for new computers, the policy is actually borrowed a bit from the company I work for (it's a REALLY REALLY REALLY big company). But our lifecycle policy is as follows, I recommend it to all my clients as well when they ask for advice.


3 years for Desktops


2 years for Laptops


1 year for mobile phones (or upgrade eligibility with commercial company)


regardless of when the pandora is actually available, this IS a 2 year old device... not my problem it took so long to market, the specs that were once bleeding edge are now 2 years old. I'm sure you are very aware it's not just a MHz battle, there's lots of other factors involved, the motorola droid (same hardware as the pandora) uses the 600Mhz OMAP 3430 the motorola "droid 2" uses the 1ghz OMAP 3630 and it's an extremely noticeable difference in performance. Here we are talking about the "next gen" Omap processor. I think it's a valid request for an upgrade. If you don't agree then I can respect that.


Now I can see the point of view of open pandora and the community, the device doesn't have mature software on it. This is small beans compared to Motorola, funds aren't available yadda yadda... but from a consumer standpoint it's getting close to the time for an upgrade. (even though its not *fully* out yet)

 

[Exophase]

Personally, for the last several years I've only upgraded computers when some major component fails. But I'm pretty stingy with money and most things.


The last time I actually took the computer I built for my mom (a lowly 1.8GHz Core 2 duo level Pentium) and bought her some super cheap $300 set-top like computer. I may be getting even cheaper.

 

[SNESFAN]

It'll be a while before we see ARM emulators get serious about multi-core, but it could be a big break for some like DS and Saturn.

there's a ps2 and wii emulator that's caught my attention lately and can't wait until handhelds are able to emulate those as well. I know we're probably a long ways away from that though.


Fullspeed 60fps with no frameskip N64 (all titles) and psx (with the current available emulator) would be nice

 

[Prometheus]

but from a consumer standpoint it's getting close to the time for an upgrade. (even though its not *fully* out yet)

I guess I must be different to consumers, myself. I use stuff to death, generally, and only "upgrade" if something fails and it doesn't make financial sense to go for a repair, or there is some important benefit that can be gained from a change...

 

[WizardStan]

I guess I must be different to consumers, myself. I use stuff to death, generally, and only "upgrade" if something fails and it doesn't make financial sense to go for a repair, or there is some important benefit that can be gained from a change...

I upgrade when there is something that I want to do that can't by any means feasibly be done without an upgrade. This has meant playing some of the latest games on 4 and 5 year old hardware with settings turned way down, but it can be done.

 

[Prometheus]

^ That's what I meant by the important benefit part. (For example, I eventually purchased a DSi when there were a lot of titles that I wanted available on the DSiWare download service.)

 

[sebt3]

This has meant playing some of the latest games on 4 and 5 year old hardware with settings turned way down, but it can be done.

Oh god, playing with low poly-count, insanity....


/me is back to some xenon

 

[DaMummy]

im still enjoying my GeForce 8800 GTS 640MB that was bought in 2007, still runs everything i throw at it, with only an e6600 dual core cpu

 

[second exodous]

I think Elanzer is correct, I think it isn't a huge jump in speed from OMAP3 to OMAP4, maybe the OP team should go every other OMAP? In the end it doesn't really matter about the speed of the processor but the sales; is it really worth it for the OP team to upgrade already or are people still going to be buying the current model for at least a few more batches? I think it will be time to upgrade when mobile phones start to emulate the PS2 and not any sooner.


Basically in theory what will the OMAP4 bring us specifically? Is it just a little faster with 1080p output or are there systems just out of reach for the OMAP3?


Having said that I think it would be wise for the OP team to get a Pandaboard and be preparing for a possible upgrade in case sales slow down but I think they have at least a few more batches in them.

 

[Exophase]

craig has expressed the desire to follow an iPhone-like revision schedule with a new major architectural redesign being released at two year intervals. There's no way you'll be seeing a new OMAP every year - even following that every two years will be tight, it might end up closer to 3 years.


OMAP4 might not seem like a big jump to you guys but it's really very substantial. Even getting off of a 65nm part will help since it'll lower power consumption. Dual core will help you not have to worry about all those low CPU background processes suddenly spiking out of nowhere and interrupting your emulators like they do now. The L1 and L2 cache bumps will help a lot for some programs that are more memory starved, and the memory latency looks like it may have been improved somewhat. The SGX540 will have several times (like 6x) the compute power so it'll be able to push much better shaders and may eventually be useful for some OpenCL stuff. And regardless of what the real benefits are, tech sells.


Right now everything but filling quotas is moot. Demand still vastly outstrips supply, and if OP can't get to the point where they can start shipping thousands a month somehow then there's no real need to talk about doing anything different for a long time, because they'll be riding the current Pandora until absolutely no one wants one anymore. The only thing that'd really kick them off is a major competitor doing the same kind of thing better, but I'm not sure OP could very easily keep up with such a force no matter what they did.


But if they did manage to really turn production around and could start handling big volumes in the tens of thousands then I think a new model is needed sooner than later. Pandora is just looking too far behind the millions of cell phones out there, and public perception still suggests that a clam shell with a keyboard be more powerful than a cell phone (if it isn't a DS). Pandora clearly missed its opportunity on that but it could more easily get it the next time.


At the same time, I'm not sure how I feel about them reusing the case verbatim. It'd save a lot of time and headache, but I can't help but think that there are some major things that would help them here. An extra latching position and a better shoulder design at the very least. I'd also honestly go for a smaller battery.. I know the very long battery life is a big factor but how many people really feel they need 10+ hours and aren't content with something over 7 or so? The huge battery makes Pandora comparatively thick/heavy and adds to the price, and precludes the possibility of using off the shelf parts.


I'd also want to see if there were any options for a capacitive screen, because stylus based input is viewed as being quite passe now. But that'd beg a separate touch-optimized GUI design. We'll see more and more Linux based options here (not just Android, there's Moblin/Meego, Chrome, etc) and I'm sure it'll be possible to get a popular, slick GUI on top of a custom kernel and userland libraries that'll be easy to port existing Pandora stuff to.

 

[gadgetoid]

All this talk of a Pandora 2 isn't going to help sales of the current iteration. We all know why devices like the iPad 2 are kept hush-hush with little but unsubstantiated rumours to hint at their arrival. Who'd want an already obsolete device?


Fair enough, the Pandora has been proven to perform, and what ED has shown it doing is seriously impressive. It'd be more impressive if I had one in my hands now. But at the end of the day, it's a stupendously ambitious and noble project which has most likely given my my monies worth of entertainment and community fun already.


Anyway, wistful digressions about community giggles and video entertainment aside, the Pandora 2 isn't the best topic to be discussing right now, we should be focussing on what performance can be squeezed out of the existing handheld. Console generations have proven time and time again that unimaginable things can be done late it the lifecycle of a device and, in fact, only now has the Pandora really crystallized and become a device with an impressive feature-set. It's got a few years in it yet, to be sure.


All that said, if a Pandora 2 went up for pre-order tomorrow, I'd be the first in line. Can't change a geek.


If you folks wrote half as many lines of code as you did forum posts, we'd be in an even better position! (yes, I'm aware of the hypocrisy of this statement)

 

[Exophase]

All this talk of a Pandora 2 isn't going to help sales of the current iteration. We all know why devices like the iPad 2 are kept hush-hush with little but unsubstantiated rumours to hint at their arrival. Who'd want an already obsolete device?

What difference does it make? OP hasn't even come close to fulfilling their current preorders. It doesn't really matter if no one else wants it, so long as everyone doesn't up and cancel now.

Anyway, wistful digressions about community giggles and video entertainment aside, the Pandora 2 isn't the best topic to be discussing right now, we should be focussing on what performance can be squeezed out of the existing handheld. Console generations have proven time and time again that unimaginable things can be done late it the lifecycle of a device and, in fact, only now has the Pandora really crystallized and become a device with an impressive feature-set. It's got a few years in it yet, to be sure.

I don't mean to be especially negative, but Pandora isn't really like a console and probably won't ever be driven by the same sort of motivation to push the hardware. You're more likely to see things happen like they did on the GP2X than like on a console. No squeeze towards better results late in its life, just a quiet abandonment and replacement by the making company, with that replacement itself replaced not even two years later.


I'd hardly say that Pandora has "only now crystalized" when people have had them for some 7 months.


Companies like OP can only really survive on hardware sales, not software sales. It's their prerogative to have a release cycle more like Apple's than Nintendo's.

If you folks wrote half as many lines of code as you did forum posts, we'd be in an even better position! (yes, I'm aware of the hypocrisy of this statement)

Oh god, don't start Torporing us. That's just petty complaining about someone not doing enough of what you want, and drawing in something that's not even tangentially related. Or at least your belief that they're not doing it; you don't actually know what people are working on behind the scenes. I know you didn't mean anything bad by that comment, but please just take my advice - any people who are working hard on code for Pandora or meant to eventually be on it will not be happy with you telling them to post less and code more.

 

[second exodous]

Hmm, every two years, I haven't gotten my Pandora yet but I think I could go every other one as I use PCs until doing anything is painfully slow which is like 6 or so years. Game wise I'll use it until it dies as that is what I do with my current handhelds.


As for battery size I think they should always aim for 10 or 12 hours at minimum. I don't know how much power the OMAP4 uses but adjust the battery size accordingly. Also batteries might get better around the time Pandora2 comes out.

 

[Exophase]

Why do you insist on 10-12 hours? How often are you away from a charger after that many hours of continual use? Seems to me that 8 hours is the high end sweet spot, with anything over 6 being acceptable.

 

[second exodous]

Why do you insist on 10-12 hours? How often are you away from a charger after that many hours of continual use? Seems to me that 8 hours is the high end sweet spot, with anything over 6 being acceptable.

Long plane flights. I would like it if I could take one device to play video, listen to music, and play games on. Right now I take my DS and a laptop and combined these die in under 8 hours so a flight longer than that I need a third device. With 10-12 hours I wouldn't need nearly as much junk to take on a flight. I have a nook but I get sick reading in a cramped pressurized environment. I'd also sleep but it makes my jet lag four times as bad.


Also, did the big battery really add that much to the device? It doesn't look that large considering what it can do. I just like long battery life is all.

 

[SNESFAN]

What difference does it make? OP hasn't even come close to fulfilling their current preorders. It doesn't really matter if no one else wants it, so long as everyone doesn't up and cancel now.

that would be the only thing that OP could do to kill their own project.... even the most devoted "till the end" preorders would cancel and order a pandora 2 if it was possible to be had in any reasonable time frame. Unless you could afford both like some.


OP doesn't have a back fill of hardware they would have to sell at a discounted rate like microsoft and sony and apple etc, nor could they afford to supposedly. They can't even make the hardware fast enough to keep up with orders they haven't fulfilled yet. They would really not have a whole lot to lose once the preorders are supplied. That is their business model up until now. They sell the pandora as a love for the project rather than actual hardware sales, so if they are selling these at cost or close to it, there is no return on investment from the whole pandora idea.


It might be a slap in the face to release a new version so soon after finally getting the first one out to customers. But time is their enemy, and the potential sales over the past 2 years is what they have lost. To stay ahead they need to start thinking now on what they want for the future, once customers realize the specs compared to what else is currently available, or currently unreleased but may become available at the drop of the hat. Their customer base might fall out from under them. I think alot are here because nothing else is available, craig himself said before that might change. OP needs to maintain the edge on the niche to stay on top with so much teetering on the balance.


A year from now would be, in my belief, a good time to have a new model. The current technology thats available today and based off the age of the specs of what is being sold right now... But if like ED said they need money from the ICP to fund the development of the pandora 2, then we're probably looking at 2 years from now, if not more... If I had to take a running leap at an estimate. 2 years is not bad, 3 might be fatal for this company though if I had to be honest.

 

[Exophase]

Long plane flights. I would like it if I could take one device to play video, listen to music, and play games on. Right now I take my DS and a laptop and combined these die in under 8 hours so a flight longer than that I need a third device. With 10-12 hours I wouldn't need nearly as much junk to take on a flight. I have a nook but I get sick reading in a cramped pressurized environment. I'd also sleep but it makes my jet lag four times as bad.


Also, did the big battery really add that much to the device? It doesn't look that large considering what it can do. I just like long battery life is all.

I do feel that the device is bulkier than it needs to be, and I know popular public perception falls at least somewhat along these lines. Size plays a big roll in image these days.


It'd probably be worthwhile to carry a second battery for cases like this. It'd be cool if Pandora had an internal battery that could let it last for a few seconds without the primary battery, so you could hot swap it. No matter what, I really hope that the next Pandora has better support for transitioning from DC in to battery power. Then you could have external supply batteries that let you hot swap.