So much for the U in USB

@wierd_w Ah. That makes unfortunate sense. Thanks.

@myne Exactly 😀

Oh, well: maybe there's a nano 1.1 floating on ebay or something.

The fix for the RTC problem is to remove R38 (next to the M6117D).

The time in the BIOS started ticking away as soon as I removed it and the year also defaults to 2001 now when reset. Left it off for 30 minutes and it kept the time fine!

Having an LBA problem; possibly someone can explain what I'm seeing before I start experimenting to qualify the problem.

Seemingly randomly the LBA status on the drives I'm using is set on or off. I've flashed the drives a few different ways and regardless of what flag is set on the partition it seems to have a random result.

I had assumed the BIOS being set on or off would trigger it for the device and then the flag setting on the actual device might be necessary to ensure it works. But apparently that isn't what's going on.

What size drives and what C/H/S do they detect as? It's probably those parameters that determine whether LBA is used or not.

BitWrangler wrote on 2024-05-06, 18:24:

What size drives and what C/H/S do they detect as? It's probably those parameters that determine whether LBA is used or not.

That's the weird thing. It seemed to be centering around how I had formatted and partitioned rather than the CF card itself. This is with 512MB, 2GB, 4GB, and 8GB cards formatted in a variety of different ways. I had assumed the card would be what was detected for that. The dual SDcard to CF adapter was a non-starter and resulted in a blank screen on boot btw.

I'll just go through and do some testing if there's nothing I'm missing about how this setup detects the CF as a drive (i.e. detecting the partition as a device rather than the physical CF card or something). Or that I'm just flat out not remembering from more than 20 years ago.

I did do some searches on LBA and how that works in practice to refresh my memory and came up with nothing that looked relevant.

With all my sympathy for those folks that have an outright dud battery, have "working battery" people noticed a short batt life, as in under two hours, as in when Fn-F5 monitoring of an idle machine, you can watch the gauge tick down in a hurry. I unearthed my old Tekkeon ( https://www.bhphotovideo.com/c/product/618847 … us_MP3450i.html ), which after 8 years of storage, STILL holds a better charge than the P386, and gives me plenty of juice, but a major departure from the form factor. The Tekkeon is almost as large as the computer.

jakfish wrote on 2024-05-06, 21:30:

have "working battery" people noticed a short batt life, as in under two hours

For the size of it I was thinking an hour or so was about right for the battery life. Size wise the thing seems similar to 1.5~2 18650 in capacity where as a "real" laptop would have a stack of six on the light side. I can do some practical draw and capacity tests whenever I get this thing disected on my bench.

The problem on mine is that the sense pin for detecting the charging state only gets pulled high (or low if that's correct) when the battery is completely dead causing an annoying shutdown warning while charging. Which ironically doesn't warn before shutoff. Otherwise I'd run it on my 20v capable USB-C battery pack.

I'm sorry to hear that you're wrestling with another charging issue. Are you getting anything close to a proper batt percentage with Fn-F5? I'm in no way certain that gauge is accurate regardless of batt state.

Your battery suppositions make sense to me, esp when I think back on a 3D-printed, tiny Banana Pi computer I picked up a while ago--its battery was 2000mAh and that lasted about the same time as the P386, but it had a smaller screen and no cf to thrash. I'll be interested to hear your results when you dissect.

For anyone who wants a closer look at this system and may be on the fence about getting one, I made a pretty in-depth unboxing and review video here -
https://www.youtube.com/watch?v=wVgITS8aLzc

Has anyone had any luck installing some better audio drivers in Windows 95 to get any WAV sounds working, or is this not possible due to the lack of a DAC?

Also, as to the battery concerns, my battery seems to last for quite a while. I haven't done a full test from 100 to 0 yet, but I'm estimating about 3 to 3.5 hours. It seems pretty good for such a small battery.

I very much enjoyed your video, from start to finish. Thanks for taking the time to put it together.

Glad that someone else at least is getting better battery life than I am. Can't help you on the sound, sorry. A couple of things I did pick up along the way:

1) For notepad users, Win95's notepad.exe is pretty rudimentary (no font choice, etc). Notepad++ is very slow, but NewPad is relatively quick to open/use, and has more to offer:

https://web.archive.org/web/19991109051128/ht … are/newpad.html

2) serial-to-wifi does work with oldnet's mtcp driver and the latest version of microweb browses well (white text on black screen is more readable, imo). Over at tindy store, both ISA and RS232 adapters are out of stock, however. I use retro-proxy on the desktop to get to https sites, though the heavier page can crash.

Since I try to use the internal mouse, I do wish there was a way to drag windows. Has anybody found a way?

Hi all,

New here. Just received my Pocket 386 and really enjoying it. I agree with a lot of the impressions others have on it so far—battery life somewhat on the short end, slow USB file transfer. But all around a fun machine to tinker with.

I saw a question on the first page about display resolution and I’m really curious if we can find a way to force the graphics chip in Windows 95 to display 800x480. The Toshiba Libretto 100c output 800x480 in windows 9x, and I think the CL graphics chip that our machine has built in can output up to 1024x768. Anyone got any solutions there?

Also, total newbie question but how does one install the Freddy V driver in windows 95? What device in device manager should be selected?

I love the idea of people building 3d-printed “cartridge” expansion boxes for use pin headers like this guy did with the Hand. A sound card is a natural choice. Maybe a Wi-Fi modem. Graphics card?? What else can we dream up? https://youtu.be/wWUtMPxmKD8?si=bFwKVwswb_eaDiLi

The device does not have a DAC. There is a PC speaker WAV driver for Windows 3.x and probably 9x too, but it hangs up the system while playing sound and fidelity will be terrible over the built in piezo. Wish the designers routed PC speaker audio through the built in amp like the early IBM PS/1s did. I desoldered my speaker, bet someone could patch output into the amp + add a hardware audio switch.

Seems like the device is best navigated using Windows keyboard hot keys, but I recognize most shareware and later commercial software will need a mouse, ditto for most 3 x games.

Would in general recommend early 90s software, e.g Word 2.0 or *maybe* Word 6.0 A 386SX/40 just won't have the scoot to run something 32-biy like Office 95.

I agree--an Office suite, no matter the year, would be taxing. I'm able to run Word 97 standalone and despite its lengthy install, its load time is manageable. MS quietly released a .docx file converter for Word 97, both open/save, and that helps keep things a bit more modern.

To experiment, I just installed Adobe Reader 4.0 and the load times are untenable. Actually, Adobe Reader 2.1 is fairly snappy, but can't open today's .pdf

The FreddyV driver is a DOS driver. Load it in config.sys. Using it or the original USB driver will force Windows into slower 16-bit disk access as they are real mode DOS device drivers.

You must be patient to consider Word 97 acceptable. I wouldn't even want to run Word 6.0 under 3.x, prefer it quick and snappy. Word 97 was really made for a fast 486 or Pentium and 16MB of RAM.

RTF should allow interchange between legacy WPs and newer ones. Might be a legacy Word filter or conversion program on the modern PC side too.

On my P386, right after boot, the stock WordPad and Word 97 both take 16 seconds to bring up a blank document. Maybe a 16-bit Word 6.0 loads faster, I don't know. After closing them out, WordPad and 97 leave enough of a memory trail to go down to 9 seconds upon reloading. It seems relative compared to 1:30+ cold boot to a no-hour-glass desktop 😀

I hadn't realized that WordPad's default file format is Word 6.0, but I need professional syntax (smart quotes, em dashes, etc) and I try to run Word 97 on all my kit (even in WINE) to preclude conversion problems.

When I was considering WordPad, I wasted a couple of hours trying to find the abandoned Spell Catcher, a spell checker/thesaurus. But I couldn't track down a Win95 version for sale.

Just found another issue with my unit. Apparently it can read data from the USB fine, but trying to copy to USB leads to it hanging. The USB stick can be written fine when using with the Book 8088, so at least it's not a problem with the stick itself. Changing to a different ch375 driver does not solve this issue.

Do you have a smaller-sized usb drive to try? I use an old Cruzer 1GB and read/write is fine. Weird, though, that your current drive works in the 8088.

I don't have any < 2GB USB drive at the moment. I will look for one tomorrow.

for those wishing to experiment with word processing: it might be a good idea to look into ClarisWorks/Appleworks, Microsoft Works, or Office 4.X or something... I imagine Office 95 would technically work but unless you want to simulate using a school computer back in 1995 (minus stolen keyboard keys and the elastic band stolen from the CD drive...) I imagine it wouldn't be all that great of an experience.

that said, if you're ONLY using it as a word processor and not multitasking then once it is loaded its a perfectly cromulent experience. doubly so with the CH375 driver unloaded (i imagine because it probably desperately thrashes the virtual memory and the driver makes things slow in windows)

honestly i would probably just put windows 3.x on it and then use calmira or something similar.