What if we back up a bit and make it more incremental?

[EssoAir]

I'd rather have the P2 be revolutionary, instead of evolutionary.

 

[TeDaDeS]

Until the full concept is published we can only speculate if the device will hurt the current open pandora.

You missed the point of the discussion. The issue was competition between the proposed upgraded pandora (called P1.5 by me) and Craig's proposed Pandora2 (P2) and NOT between either device and the current Pandora. Either P1.5 of P2 would replace the current openpandora completely.


The real motivation for the P1.5 may ultimately be financial: if P1 sales drop off too much around mid 2013, there may not be enough to clear the preorder queue and begin to generate profits. At this point it might be necessary to offer a new class of performance (including, for example, Dreamcast emulation) to boost sales.


The P1.5 proposal in the original post might require just a new motherboard, lcd and minimal changes to the lid. (also a piece of foam to hold the battery in place), possibly making development inexpensive enough to find funding.


(edit: it is possible that the changed connectors (hdmi, microusb) would not fit the existing holes in the case, making the whole idea unworkable)

I think you misread my comment. I was commenting on this post before me:

Whether we are surprised or not, Craig needs to be more specific if he wants to debunk misperceptions about his intentions. And if he doesn't really want to debunk anything then I don't know what to say, I will just stop here.

This was about the 'still unkown version of Craig' P2. We should not worry about this, as the "as you call it" P1.5 could probably be done next to the Craig-P2 device.


As I read it ED and CraigX are also not clear what the P2 will eventually be, as I personally think ED would also go for some P1.5-ish design.


What will happen eventually, who knows...


(btw. )

 

[Grench]

I would like to bring this back around for discussion.

The current Pandora has a lot of things going for it.

Amazing controls - nubs and d-pads are perfect, they keyboard has a pretty good feel, shoulder button bugs are pretty much squashed.

Sounds amazing - if you haven't connected a set of high quality head phones to your Pandora, do it now.  Listen to some very high bit rate recordings - you'll be amazed.

Case plastics and stylus are simple, but very functional now.

Dual full sized SD slots - no other device being sold has them.

It is my understanding that the 1Ghz chips cap out the capabilities of the current motherboard.

As a next step forward, what about keeping the case & motherboard form factor and the audio components and modernize the rest of the features?

Or maybe the right step is to build a super-compact P2 with microSD and the iCP2 keyboard then an, 'XL' version with the same motherboard & controls in a bigger case with full sized SD and a bigger screen & more battery?

Which path would get us 'there'?

 

[Galaxis]

Tbh, for me, the Pandora is almost perfect as is. For a small evolution (P 1.5) I'd consider just three main points: Slightly higher screen resolution (1024 x something would be great), an integrated USB hub to get rid of the problems with HID devices and such directly on the main USB port, and a more stable Wifi.

Apart from that, it's mainly the software that needs work. SuperZaxxon does the job, but it has too many rough edges in it's current state (I absolutely want this to stay a Linux device, though).

 

[EssoAir]

Tbh, for me, the Pandora is almost perfect as is. For a small evolution (P 1.5) I'd consider just three main points: Slightly higher screen resolution (1024 x something would be great), an integrated USB hub to get rid of the problems with HID devices and such directly on the main USB port, and a more stable Wifi.
 
Apart from that, it's mainly the software that needs work. SuperZaxxon does the job, but it has too many rough edges in it's current state (I absolutely want this to stay a Linux device, though).

My 1ghz is a bit slow and lacks ram capacity, but over all its good

I do hate the nubs, though

 

[bismuthdrummer]

I fully agree that the P1 is "nearly there"... a lot of the original issues have been solved. The external speakers sound great, build quality is better, etc... I also agree that the best improvements going forward would be screen resolution, USB compatibility (3.0 would be sexcellent), and internal wifi. On board storage will hopefully go up (no need to debate the details here though). Better SoC of course.

About the form factor. I think it's a stellar idea to have compact and XL versions, but I'm fairly certain that the primary argument against would be the reality of the costs involved. If it could be shown that the compact version could profit well, maybe this would be a viable compromise. Given this, I'd rather have the XL than the compact; I want it to feel good in my hands above being able to pocket it easily.

 

[Grench]

I think this could be done in stages.

P1 = CC 256

P1.25 = GC 512

P1.50 = 1Ghz Pandora

P2.0 could be all other hardware just as it is - screen, case, battery all the same.  New motherboard with updated SOC, wifi, mini hdmi port in place of the weird video out port.  Yes, the SD slots should be faster - but they'd be in the same position maybe even using the same component.  Yes, keep the existing audio system - it's damn good.  Yes, keep the now figured out shoulder buttons, keymat, nubs & pads.  The SoC can support screens and resolutions beyond the included one - that's OK.

P2.5 as P2.0 BUT replace the screen with the largest 16:9 that will fit in the lid.  Move the speakers down to the hinge area.  This revision is all about the screen, speaker changes and the changes to the plastics to support it.

Yes, it is a huge effort to make a new motherboard with updated components.  Is MW still around?

 

[Binky]

you mean:

P1.0 = CC 256

P1.1 = GC 512

P1.2 = 1Ghz Pandora

Surely a nerdy device such as the pandora should use proper version numbering?

 

[Grench]

We could call it Phred 2.0 if you're prefer.

I still think there may be some legs left in the current hardware design - what needs refreshed most is, well, the SoC + WiFi, which requires a new motherboard refresh/design to happen.  Start now and these could be ready for sale in 6-9+/- months?  How long does it take to design, test and productionalize a new motherboard?

Then once the dust settles and 5" 1080p displays are sold in candy stores, update the screen - maybe for a year from now?

It would really be a conversation for Ed and MW to have - my idea is to propose that they have that conversation and quantify a bit what it would take to get from here to there.

 

[Binky]

Rolling upgrades sound like a kludgy way to do things.

I'd rather have a much better toy in Two Years than a slightly better one in Two Months.

 

[zoranc]

To me Pandora is very good at filling the niche and I always wanted this niche filled. To me it is a smallest possible Linux PC on which you can do SOME software development even it is just for small/fun projects. I tried many (really many) alternative options. Added gaming controls are a huge bonus. In my opinion next revision should continue to fulfil this role.

To me the improvements should be in processor, maybe the ram, and definetelly the screen. Most important improvement for the screen would be readability outdoor since Pandoras usage and strong selling point is its mobility. Pixel Qi or similar solution would be ideal in my opinion. Beter backlight would be another alternative. I would not mind beter resolution either, but to me outdoor readability is much more important.

Next I would like an official linux distribution. Ubuntu would be ideal, but any other widespread one would do it. I have very small interest for Android since it lacks a C compiler.

Last point are noobs I did not like the original ones. Rebirth version vere beter I believe but I didn't test them extensively, now will have to check 1GHz version when arrives in the next days. So noobs might need improvement too.

To me evolutionary approach where Panda continues to fill a niche is the best approach.

 

[3_bit]

We should get everything sorted out then get the best CPU there is for the application. Do we really care if there's a new RAM chip that refreshes 50hz faster? no. The CPU is probably the biggest bottleneck, as emulation of systems would be using that the most. On the bright side, we might just bring the next generation of PC games into the P2 with the latest ARM. (as in 1998-2002ish). Of course, that's assuming ARM's power is growing as fast as x64's (annual rate)is.

 

[Grench]

Last point are noobs I did not like the original ones. Rebirth version vere beter I believe but I didn't test them extensively, now will have to check 1GHz version when arrives in the next days. So noobs might need improvement too.

Although some of our noobs need improvement, I usually let them get by with it since they are after all noobs & such.

I think you were referring to nubs though, not noobs.  No worries - but the typo/translation error was, to me, hilarious.

I think the nubs on my Pandoras (CC and 1Ghz) are much better than any game console I've used.  They're certainly more durable than the Wii nunchuck analog sticks.  They're more comfortable and controllable than Logitech's.  I'm not a fan of big floppy stick controllers though.  Which controller are you comparing the nubs to?

A Pixel Qi screen with daylight readability would be awesome in this device.  A year or two ago when I/we had asked about it, Craig had stated that he had talked to them, but that the screens would be prohibitively expensive.  As far as I know they have never gone into production with a 5" +/- screen.  Maybe that has changed?

I would like to see the next version - whether it is a whole new machine or an upgrade to the existing system - we need to consider updating the OS to something Joe public has heard of and can properly spell on a QWERTY keyboard without looking up extended ASCII code sequences.  This would likely require more onboard flash or moving from onboard flash to an on-motherboard microSD card.

In my very humble opinion, the weak points in the current units are the WiFi, SoC being outdated and the screen being smaller and lower resolution than my phone.  The overall design, layout, buttons, ergonomics, card slots, audio - those are all downright awesome as is.

 

[Grench]

We should get everything sorted out then get the best CPU there is for the application. Do we really care if there's a new RAM chip that refreshes 50hz faster? no. The CPU is probably the biggest bottleneck, as emulation of systems would be using that the most. On the bright side, we might just bring the next generation of PC games into the P2 with the latest ARM. (as in 1998-2002ish). Of course, that's assuming ARM's power is growing as fast as x64's (annual rate)is.

Mmm.... I don't think it's quite that simple.  We're not talking about a unit with a discrete CPU + everything else as components.  The CPU in the Pandora is a System on Chip.  I.e. the majority of the 'everything else' is actually taken care of on the chip itself.

MWeston is a bit of a genius for creating the design on the original Pandora motherboards - when was it?  3.5 - 4 years ago?  I'm really curious what he could come up with while using the same physical board size and the same external features (changing the weird video port on back to something more standard too) but with otherwise modern 2013 off the shelf components.

 

[3_bit]

I agree with this. I would buy a Pandora now, but I really want to get it when it's "cutting edge" again like my iPod 3rd gen was when I got it. (of course I got it before I was into retro computing and open source at all, which is why it is what I have right now and not a first batch)

 

[Grench]

I don't want this to get lost in the shuffle.

With a heroic effort from ED and MWeston and DaveC, we could have a Pandora II in the same form factor as the Pandora I.

It would be like a 'sleeper dragster'.  It might look like the same old box, but it would have a whole new lease on life.

 

[xiongxioi]

We could call it Phred 2.0 if you're prefer.

I still think there may be some legs left in the current hardware design - what needs refreshed most is, well, the SoC + WiFi, which requires a new motherboard refresh/design to happen.  Start now and these could be ready for sale in 6-9+/- months?  How long does it take to design, test and productionalize a new motherboard?

Then once the dust settles and 5" 1080p displays are sold in candy stores, update the screen - maybe for a year from now?

It would really be a conversation for Ed and MW to have - my idea is to propose that they have that conversation and quantify a bit what it would take to get from here to there.

I'm totally agree with you on every front. Even through the OPT has to get rid off all its old chips first, Time is still important. 

I hope P2 will have a futuristic feel to it to justify the price-tag. P1 looks so utilitarian; it's like something Hitler would order to make for his annual tax return. 

My opinion is that having a more unique and luxury design would help boosting P2's sale.

 

[Grench]

Bringing this back up for thoughts.

It's kind of like taking an old IBM XT case, gutting it and putting in modern components.

What's more retro-gamer than that?

 

[3_bit]

We really should do this, but make sure to update ports (SEE: TV OUT AND GPIO) first.

Bringing this back up for thoughts.

It's kind of like taking an old IBM XT case, gutting it and putting in modern components.

What's more retro-gamer than that?

I think that's just called "Anti-Theft". (What theif would steal that, assuming they don't know the high value of vintage hardware?)

 

[TrashyMG]

Bringing this back up for thoughts.

It's kind of like taking an old IBM XT case, gutting it and putting in modern components.

What's more retro-gamer than that?