1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading ... Loading ...

A Call to Hacking the Treo 650’s Memory Problems

Posted by Levi on Feb 15th, 2005
2005
Feb 15

My Treo 650 is due to arrive on Friday, maybe Saturday, or if I’m extraordinarily lucky, maybe even Thursday! With that in mind I have been researching upgrade procedures from the Treo 600. It seems the prevailing wisdom right now is not to do the standard, backup and restore procedure, but rather to start from scratch and install applications one by one. Part of this is to avoid potential incompatibility issues. The other is due to the memory issue on Treo 650’s.

As you probably know, the Treo 650 has a different KIND of memory than the Treo 600. This memory is “non-volatile flash” memory which means that it can retain information without power, unlike the Treo 600’s. Unfortunately, the minimum allocation size of this memory is much bigger than the Treo 600’s at 512 bytes. So a file that only has a few bytes of information still takes up 512 bytes. The effective result of this is that the applications that took up only 50K on your Treo 600 may take up 80K on your Treo 650, and so memory gets eaten up much more quickly.

Pa1mOne in their ultimate wisdom for some reason decided that most users did not need any more memory on the Treo 650 than on the Treo 600. Even if we believe that Pa1mOne was unaware of the less efficient memory usage on the Treo 650 (and if this is true I’m not sure whether this is a harsher statement against Pa1mOne than them realizing it but simply ignoring it), the phone still deserved an increase in memory. Pa1mOne’s competition is PocketPC Smartphones that typically come with 64MB of memory (double what’s in Treo’s), and sometimes more. It seems rudimentary that in this day of cheaper and cheaper memory that this was a very cheap method to compete better. But as it stands, the memory has become the chief bottleneck for performance of such phones for power users.

Speaking of power users, it seems that Pa1mOne doesn’t really value them as much as the mass of other users. While this may be a good decision based on standard marketing practices, the problem with this for a company peddling advanced electronics is that those driving the development, hardware and software innovation, and even a lot of the buzz and referred sales for these devices happen to be those power users. I have heard countless users on discussion forums and other websites complaining about this memory issue and about Pa1mOne’s seeming disregard for the issue. There are some who have decided (and are I’m sure advising their friends) that the Treo 650 is not the phone to upgrade to, and that they plan to either wait for a future edition that gets some of these things – especially the memory – right, or even that they are going to vote with their feet and buy a PocketPC phone.

This is truly unfortunate and I hope for Pa1mOne’s sake that they get their act together, but what are we all to do in the mean time - if there even is a mean time? Pa1mOne’s answer to the firestorm that erupted when the memory issue became public was to offer a free 128MB SD card to users who requested it. Since then it seems that even those requesting it (the offer is not really advertised anywhere) haven’t been able to get one. This is not really any compensation either because a 128MB card is probably worth $15 on the open market and really doesn’t go towards fixing the issue. Most power users will want at least a 512MB for their phone, so the 128MB will just be used as an added card for storing some temporary files like MP3’s.

So, my thought this morning was towards fulfilling this need from someone other than Pa1mOne, since I can’t expect them to offer a real solution any time soon. TiVo has worked with a dedicated community of hardware and software hackers and developers since its inception. The result of this is a few companies (the best know of which are Weaknees and pvtupgrade) that offer kits or full installation services that will take your TiVo and expand its storage capacity by leaps and bounds. What I want to know is why aren’t there some enterprising electrical engineers looking into if not starting a company to do this for the Treo, at least trying it out on their own? Is there something inherent in the Treo 650’s design that makes it too difficult to get under the hood and change things? So I would just call on all people who have this kind of technical ability to do this. Unfortunately Pa1mOne made a big mistake, but that doesn’t mean that it can’t be fixed by it’s most enterprising and knowledgeable “power users,” despite the lack of regard we seem to garner from Pa1mOne.

So come on people, let’s create our own solution to this problem! I’m not about to try myself because that would require at least four more years of schooling, but I know some of you have the expertise to do this right now!