Comments on: A Big Batch of Quickies

This contest is great, BUT I'm pretty sure cliesource isn't run by Sony.

cliesource is independently owned and operated. Sony even requested that they take down the NX specs PDF that was leaked....

----[self-mutilation or self-empowerment?]----

This is awesome. Finally, they can open up that sleeping CF slot. However, what I fear is that the drivers which will be developed, will probably cost a lot of money, and will not be sold included with the NX series in the box. Therefore, Im not sure how many "average Joe's" would actually know about these drivers which would be included, and im sure these average people might turn down the NX series seeing this big bulk on the back of the PDA, and understanding it wont be able to expand.

"Now thats just PRIME!"

Well if someone writes a driver for the ibm microdrive, then suddenly I might become a Clie user!

Well if someone writes a driver for the ibm microdrive, then suddenly I might become a Clie user!

Don't get too excited - The MicroDrive has been long supported by HandEra/TRG, and the big problem with it is that it eats batteries by the fi****ll - It cuts the life of the TRGPro's battery to 1/4 it's normal run time. I can't imagine the NX is going to do much better.

There's plenty of 512K CF cards around now - By the time the Sony CF driver is 'ready for prime time', there may even be 1gb models available, without the high current draw of the MicroDrive.

Sony could easily satisfy everybody's needs by releasing the bigger Memory Sticks. I have no CF cards, so either way I'd end up buy new cards. I'd rather them be Memory Sticks.

Why? Memory Sticks are slow - At least CF is a real interface - There are more peripherals available for CF currently than are planned for Memory Stick, and (If the peripheral end pans out), you can probably use both at once.

That said, I'm sure that if somebody COULD make bigger memory sticks economically, they WOULD. Sony opening up it's licencing agreement has pretty much obsoleted the 'Sony is holding back!' line of thinking.

I've seen occasional posts that the CF slot allows one to use certain models of dial-up modem in it already - Any confirmation on that?

[i]Pretec has 3GB CF cards now.[/i]

Isn't that an expensive card...I mean, [b]really[/b] expensive? What if it gets lost? That would stink to pay so much and then lose it. I guess you'd have to be careful.

[i]That's a more than 4.5 CDs worth of information.[/i]

You could probably get 4.5 CDs for under $5. Wouldn't a 3GB card be like almost 100 times more expensive?

http://www.palminfocenter.com/view_story.asp?ID=4378

I tried McPhling on a CUSA TT demo model. Works great.

PopUpNote has a bad bug -- only shows the first 9 Memos -- nothing more! I've emailed Benc to ask about OS5ing it -- no reply... Maybe if others joined in??

PopUp Clock & Calc work fine under OS5 -- oh, with TEALMASTER, I mean. I used PopUpNote with Tealmaster too. They don't work w/o it, of course.

A couple of weeks ago I tried TapPad on a T|T with no joy; there doesn't seem to have been a new releae since then :(

>Palm IIIx, Palm OS 3.5, NS Basic/Palm 2.1 63 seconds
> Palm Tungsten, Palm OS 5.0, NS Basic/Palm 3.0 15 seconds

Note that he compared against a 16 MHz IIIx. If he had compared against a 66 MHz NR70, the results would have been close to even.

Admittedly, that's still a decent result for their virtual machine. HotPaw Basic on a Tungsten T is almost as fast as on an NR70 when doing math intensive calculations, but is around 35% slower doing simple loops.

But native ARM virtual machines (several are under development) should be at least 10x faster on an OS 5 model than the equivalents on 66 MHz Dragonball based handhelds (or more than 40 times faster than a Palm IIIx).

The DSP is a TMS320 variant, which is a non-general-purpose 16-bit integer media stream (audio, video, communications codec, etc.) processor. I don't see anything in the Basic (or C) language specification that deals with these data types directly. The DSP far more useful to developers who want to do mp3, MPEG2, speech recognition, MIDI synthesis, 2d graphics setup, or software modem libraries which other applications (in Basic, C, LispMe, etc.) can then call. For data types other than 8-bit and 16-bit vectors, the ARM925 CPU is probably faster, even at its slower clock rate.

Those are for OS compatibility, I was referring to resolution compatibility. When it says that an application "supports Palm Hi-Res (320x320)", all it basically means is that it won't crash or misbehave with that resolution, not that it actually takes it advantage of it. It may do so, but not necessarily. I feel that not crashing or misbehaving should be inherent to the OS 5 compatibility predicate which would then render the resolution compatibility statement superfluous. I'd much rather have an indication of whether or not the application actually does something useful with those higher resolutions.