Pandora 1.5 by replacing ICs and chips with better, pin-compatible parts?


xyta

Active Member
Joined
Aug 28, 2009
Messages
886
I wonder what such a Pandora would be capable of...


If ED used the current board layout with the best pin compatible ICs available, what specs would be possible.


For example, the Pandora was planned with 128 MB of RAM. The first series came with 256 MB. Now it seems that future Pandoras will be delivered with 512 MB. How many MBs of RAM would be possible at maximum in theory?


Maybe we find a pin compatible wifi chip which supports higher data rates?


Or a better OMAP?


An IPS LCD display?


Gold plated headphone connectors?


etc...
 
I think something like this would only happen if there is still a demand for Pandoras when the current stock of parts have been used up. At that point P2 development will start so I don't know if it will be worth buying many more components.


I suspect that the current hardware is fairly balanced (i.e. upgrading one component would be only make sense if an other was upgraded to match, e.g no point in more ram unless accompanied by more processing power.


On a more positive note, I'm fairly sure there is a 1GHz OMAP that we could use. You'll have to catch one of the hardware guys if you want a fuller answer. :unsure:
 
The pandora uses pop memory which I believe breaks pin compatability the the AM37x series chips(I am guessing thats what you are looking at as far as an upgrade). it would be really hard to upgrade much without redesigning the pcb. Here is a link for a migration guide.


edit: nvm the pandora uses the cbc package, which does not have this issue. There could still be other issues though
 
Last edited by a moderator:
512MB RAM seems to be the maximum there is for the OMAP3 generation.


The only (mostly) pin compatible SoC out there that can be used out-of-the-box with our current PCB would be the DM3730 (AFAIK).


I don't thinkt there's a WiFi module that is pin compatible though.
 
OK... No public answers required, but something to consider...


What would it cost per unit for the smallest purchasable quantity of the DM3730?


What would it cost to do a 'short run' of that many DM3730 onto the new German boards?


What would it cost to throw an old CPU in the trash?


Add those up, multiply by 2, add it to the currently listed Pandora purchase price. Look at that number, look at the revenue & profit numbers it could generate to keep the overall project going.


In my humble opinion, If you want to do an upgrade program to further the project - there it is.


When can I pre-order one of THOSE?
 
This was discussed but quickly shot down when it was pointed out they still have thousands of the OMAP on hand they need to get rid of first. Maybe in 4 or 6 months when they've used them all up it'll be a consideration, but first they have fill existing pre-orders with existing stock.
 
I've said it before but why can't they try to resell the current stock to someone. The loss would probably quickly be offset by the increased sales. I know I'd pay more for one of those bad boys.
 
Last edited by a moderator:
OK... No public answers required, but something to consider...


What would it cost per unit for the smallest purchasable quantity of the DM3730?


What would it cost to do a 'short run' of that many DM3730 onto the new German boards?


What would it cost to throw an old CPU in the trash?


Add those up, multiply by 2, add it to the currently listed Pandora purchase price. Look at that number, look at the revenue & profit numbers it could generate to keep the overall project going.


In my humble opinion, If you want to do an upgrade program to further the project - there it is.


When can I pre-order one of THOSE?
This was discussed but quickly shot down when it was pointed out they still have thousands of the OMAP on hand they need to get rid of first. Maybe in 4 or 6 months when they've used them all up it'll be a consideration, but first they have fill existing pre-orders with existing stock.

You're ignoring this factor (see bold text). I'm not talking about skipping the old stock - it would get paid for twice as fast under this scenario.
 
I only skimmed the migration guide linked above because a lot of it made little sense to me. Please correct me if I'm wrong but I got the impression though that while the DM3730 may be used as a drop in replacement for the OMAP3530 with only a slight change in components, you would not be able to make the most of it in a board designed for an OMAP. - i.e. while the system is usable in that context, to take advantage of most of the additional features you would need a board designed to use them.


While there would certainly be some small advantages to switching SoCs at some point down the line, the question in my mind is whether it would be worth the cost of the upgrade, if that were the only change being made.


- Neelix
 
I only skimmed the migration guide linked above because a lot of it made little sense to me. Please correct me if I'm wrong but I got the impression though that while the DM3730 may be used as a drop in replacement for the OMAP3530 with only a slight change in components, you would not be able to make the most of it in a board designed for an OMAP. - i.e. while the system is usable in that context, to take advantage of most of the additional features you would need a board designed to use them.


While there would certainly be some small advantages to switching SoCs at some point down the line, the question in my mind is whether it would be worth the cost of the upgrade, if that were the only change being made.


- Neelix

The DM3730 has one really big advantage over OMAP3530: faster clock speeds. That goes for CPU, especially GPU, and to a much lesser extent memory. It also has twice as much L1 cache which can sometimes make a tangible difference. These are all performance benefits and are the bulk of what anyone cares about. Nothing that you need to redesign the board for.


Mind you, an OMAP3530 design isn't guaranteed to be DM3730 compatible, but the current revision of Pandora's PCB is. This was done as risk mitigation in case TI EOL'd OMAP3530 while there was still production demand for Pandora. If OPT didn't have a large supply of OMAP3530s already it wouldn't make a lot of sense to keep using it. Bear in mind that DM3730 is probably cheaper, on account of being on a smaller (but now quite mature) process.
 
Last edited by a moderator:
I'll go ahead and just ask here, in case someone else knowledgeable and with more time than Ed knows, would there be any tangible gains in switching the headphone port over to using "Gold plated headphone connectors?"


I mean, I do recall at one point quite a bit of discussion about the Pandora as a suitable mp3 player for it's audiophile aspects (bulk aside) and figure if there was notable benefits it would either have been included as a feature (and overlooked by the TC) or forgotten (there seems to perpetually be a lot of ideas for how to improve/customize the Pandora).
 
I don't think gold plated connectors would make a huge difference here. Imho the sound is as good as it can get. Very clean and powerful.
 
"Gold plated headphone connectors?"

As a result of the whole "Monster Cable" marketing thing a few years back, many people are convinced that they need OO gauge wires with gold plated connectors to transfer a home audio signal 20'. Blind testing has proven that they can't tell the difference between an 18ga Radio Shack generic speaker wire, giant $20/foot Monster Cable wire and a pair of coat hangers.


Anyone who knows anything about electricity understands how silly it is, but there are people who -think- they can tell the difference between a gold and alloy connector on their headphone jack. Just nod and laugh at them on the inside - they don't respond well to people crushing their delusions.


I really like the sound from my Pandora. I keep the volume wheel set 3/4 turn down from maximum then control the end volume via the wheel on my headphone cord.


I'm using Sennheiser PX200-II headphones on it.


I have found that the Pandora responds well to higher bit-rate recordings - but that is for another thread.
 
"Gold plated headphone connectors?"

As a result of the whole "Monster Cable" marketing thing a few years back, many people are convinced that they need OO gauge wires with gold plated connectors to transfer a home audio signal 20'. Blind testing has proven that they can't tell the difference between an 18ga Radio Shack generic speaker wire, giant $20/foot Monster Cable wire and a pair of coat hangers.


Anyone who knows anything about electricity understands how silly it is, but there are people who -think- they can tell the difference between a gold and alloy connector on their headphone jack. Just nod and laugh at them on the inside - they don't respond well to people crushing their delusions.


I really like the sound from my Pandora. I keep the volume wheel set 3/4 turn down from maximum then control the end volume via the wheel on my headphone cord.


I'm using Sennheiser PX200-II headphones on it.


I have found that the Pandora responds well to higher bit-rate recordings - but that is for another thread.


Only true benefit of gold is it's resistant to corrosion..


But I understand the Monster Cable ripp offs, but even worse is people that get suckered by these products..
 
Last edited by a moderator:
Well, I don't care how much of a rip-off it is, electronics have nothing on stuff like this. Marketing a stick of wax and tiny (imperceptible) traces of a flower and a known carcinogen, and people believe it actually works to relieve headaches. "it doesnt cure ur headache, its teh cooling zensation it helps teh pain noting else works c all these othr ppl who say it works that means IT MUST WERK!!!"


I mean, at least with these electronics, you're still getting ripped off, but you're actually getting something that does something. With crap like HeadOn (the common term is "homeopathic remedies"), you're getting a tube of wax, or a sugar pill, or a tiny bit of water, i.e. something that does literally nothing.
 
Last edited by a moderator:
Then there is the line of HDMI cables which best buy sells that can have prices over 1k dollars


Oh and BTW, HeadOn DOES work, they said so on TV
 
Last edited by a moderator:
Then there is the line of HDMI cables which best buy sells that can have prices over 1k dollars

I wonder how long HeadOn usually lasts for people? It's $20, so after 50 sticks you'd have spent $1,000. Assuming it lasts for a month, that means $1,000 would be spent after a little more than 4 years.


So which is worse for your pockets really depends on how long these HDMI cables last. If they last 5 years or more, assuming HeadOn sticks last a month, the HDMI cables are easily not as bad as HeadOn.


We can definitely agree on one thing: all scams like these (and scams in general) are terrible. It's too bad people are so easily fooled.

Oh and BTW, HeadOn DOES work, they said so on TV

Oh, of course. Everything on TV is true. Like the fact that Blendy Pens are totally l33t because they have THOUSANDS of color possibilities! And of course, a gold-plated fake "American Buffalo" should be sold for $50 because it's so valuable, so the people selling it are so nice to be selling it for the like, totally cheap price of $20! Oh, and the Potty Putter! Using this amazing product will absolutely definitely improve your Golf skillz!!!11 The list of these aw3some things goes on and on!
 
OK... No public answers required, but something to consider...


What would it cost per unit for the smallest purchasable quantity of the DM3730?


What would it cost to do a 'short run' of that many DM3730 onto the new German boards?


What would it cost to throw an old CPU in the trash?


Add those up, multiply by 2, add it to the currently listed Pandora purchase price. Look at that number, look at the revenue & profit numbers it could generate to keep the overall project going.


In my humble opinion, If you want to do an upgrade program to further the project - there it is.


When can I pre-order one of THOSE?
This was discussed but quickly shot down when it was pointed out they still have thousands of the OMAP on hand they need to get rid of first. Maybe in 4 or 6 months when they've used them all up it'll be a consideration, but first they have fill existing pre-orders with existing stock.

You're ignoring this factor (see bold text). I'm not talking about skipping the old stock - it would get paid for twice as fast under this scenario.

This would increase the Pandora price another 50 - 60 EUR approximately,


The SoC is one of the most expensive parts.
 
Back
Top