ayandon wrote on 2023-03-29, 06:42:

With the kind help and great work from this YouTube video and its author, I managed to build this awesome 41256 DRAM TESTER for so cheap and fully effective.

I tested all my 54 ICs (KM41256AP-12), 10 found to be dead/bad.
Now, I can test ICs ordered from Aliexpress/eBay also.

Wow, this looks like a great new tool. I will have to remember this for when I am next playing around with PC, XT and early AT RAM...

Got me wondering how fast it goes, not sure I'd start trusting it at 100ns and faster.... potentiallly could be ported to a bluepill board or similar for faster RAM testing.

Seems like you could also test 30 pin SIMM 4 bits at a time. Possibly the whole thing on an arduino mega with more pins... didn't really check how many pins got used up there.

What would be really exciting if we got it working on a ~160Mhz board, would be testing cache SRAMs maybe, need some careful layout by then though, lest your trace/wire lengths mess things up.

Right now I am somewhere between disappointed and happy. I bought this ginormous 286 motherboard back in 2020: Re: Bought these (retro) hardware today
I couldn't get it to do anything and it's been sitting under some shelves on carpet for 2 years, with parts being stolen from it occasionally.

I'm moving soon and going through these kinds of hidden things so they can be disposed of or stored properly. Spent like half an hour playing with this with the POST card and was getting no codes but occasional activity on the bus lines which implied the 286 was more working than I thought. Re-seated everything and still nothing, swapped the Odd/Even BIOS chips around and that made things worse, so maybe the BIOS is working.
On the thermal camera the 286 gets very toasty, it's an Intel 80286-12

Then I decided to try out just putting jumpers in places and switching the DIP switches back and forth so they make good contact. Only JP1 made it do something, it would give post code 11 and stop, huh. Gave it a speaker and it beeped twice, that's more than it's ever done so far.
So then I installed a riser and video card and it's posting, the keyboard works too. Wonderful!

I was genuinely considering removing some chips and putting it off for recycling but it's a whole working computer with a floppy controller and serial ports:

Checking the BIOS chips, it's a Tandon 286 built around 1991 and looking around it appears they used these in some big desktops with a riser. I can't use this really without the case but it's great that it runs and the CPU works.

Really got to get some of these added to the retro web, this board isn't even in TH99 and I have no idea what these jumpers do.

See that battery? somehow it's still working as a battery and has only *slightly* leaked into the negative terminal 😀

Woo, prize for the oldest Varta that's still got some Vart in it.

That's my first VGA card! Chips and Tech beastie. It was slow enough not to be called fast and fast enough not to be completely useless. Apparently they have some good compatibility modes, I didn't notice back in 1994 when I only wanted 800x600 in more than 16 colors 🤣 Anyhoo, mine I think got left behind in the intercontinental move... unless I tucked it in somewhere I forgot and it surprises me one day.

Jumpers probably don't amount to a whole lot, mono/color... FPU pres/absent ... core unlock 1/4 ... CMOS clear....

Got crazy enough and dropped a Turion ML-37 onto a ASUS K8N.

Oh, and destroyed a Sempron LE in the process, so I could get its IHS.

Now... to figure out why on earth it runs it at 800MHz.

Grrr. Setting up a dual-boot WinXP/DOS system on a machine without a CD drive (and without the possibility of adding one) is extremely infuriating. Especially when the computer is slow also (Via C7 CPU, 1 core at 1GHz). FWIW the following pages help:
https://www.jamesfmackenzie.com/2022/09/17/in … insetupfromusb/ <- using WinSetupFromUsb
https://web.archive.org/web/20080209122403/ht … mbr/bootini.htm <- using ntldr and bootsect.dos to boot DOS

Even so I ended up reinstalling XP twice. Several hours of work.

I've never managed to get XP to install from a bootable USB drive. It was never designed for it. Good job if you can figure it out.

You can setup the drive in a second machine, get DOS booting, set the second partition as primary, then... I'm not sure what to do then. It has to boot from the installer, format NTFS, copy files, and I think it still needs the CD after the next reboot wehb it is setting up devices.

If you manage to get it all working, it would be great if you could share a guide.

Well using the guide in that first page I linked to above, I made a bootable XP USB drive. That was the easy part. And installing XP from it works fine. But getting DOS installed and dual-booting is where I get bogged down. Especially since this machine refuses to boot from a FreeDOS USB drive (which BTW works on every other computer I tried it in).

The software used (WinSetupFromUsb) does all the work for you. Copies the install files to USB drive, and sets up a dual stage bootloader. You use the first stage to do the initial setup, and after the first reboot it goes into the second stage which completes the setup. Even on this slow machine, it takes about 15 minutes or less to install a fully bootable, standalone XP.
The guide is all in the page I linked. I just followed that guide.

FWIW I got Rufus to burn me a copy of XP x64 install disc onto a 4GB microSD card, so I'd give Rufus a spin.

I thought it was easiest if you install DOS first, then XP. Haven't set up a dual boot in almost 20 years though (I'll be doing more retro stuff to test everything in my collection once I get my new house organised, so I have room to work on stuff without storing half of it in a damp garage)

Yeah I want to install DOS first but I can't get it done on this machine. It refuses to boot from any USB DOS media. I'll have to pull the drive (which is not easy) and get that done on a separate machine, then put it back and install XP.

PcBytes wrote on 2023-03-29, 20:08:

Got crazy enough and dropped a Turion ML-37 onto a ASUS K8N.

Oh, and destroyed a Sempron LE in the process, so I could get its IHS.

Now... to figure out why on earth it runs it at 800MHz.

That's it's default cool and quiet speed, you may need a util to make it go to max multi on a desktop board. Some desktop boards boot CPU at min mobile multi, some boot them at max. Being an Nforce board it might be that Nvidia caved to user pressure from criticism on Nforce 2 that it made mobile modded chips unusable by not booting them at desktop or min mobile multi, because nothing would do 24x 200 or 166, or even 133 without volts. So went for min multi default.

Yeah... it won't boot any kind of way on that K8N. Flashing DFI's Lanparty UT nF3 250 BIOS didn't help either, despite being listed working w/o CnQ (not that I'd have a issue with that).

With the original BIOS it hangs after testing for USB devices, with the Lanparty BIOS I get boot loop - POST shows up, then the RAID BIOS is invoked and as soon as it tries to move on to the device table it resets.

Both BIOS work fine with standard Athlon 64 3000+.

Reportedly the K8N4-E should work, but the only one I have needs a SuperIO replacement as the one that's on the board currently is shorted.

So anyway, I managed to install the dual boot setup of FreeDOS+XP. It turns out that my FreeDOS install USB drive had gone bad.
I made another one, created a 2GB partition on the SSD, installed FreeDOS 1.3 there, left the remaining space unpartitioned.

Next booted from the XP USB install drive that I made per the tutorial at https://www.jamesfmackenzie.com/2022/09/17/in … insetupfromusb/ and installed XP following those instructions.

After it was installed, XP put automatically the files needed for dual-boot on drive C: (where FreeDOS is): ntldr, boot.ini and bootsect.dos. I did not have to do anything special to setup the dual boot, it was basically setup automatically.
The only change I had to make was cosmetic: in c:\boot.ini, the FreeDOS installation was labeled as "Unknown operating system on drive C:" and I changed this label to "FreeDOS". Nothing else was needed. The dual boot system works as intended.

I agree. Always install DOS first, and this can be done on any other machine. You can always fix DOS not booting, but it's a lot harder if you blow away what XP needs.

My preference is to have a dedicated FreeDOS partition that boots into GRUB4DOS. This isn't always possible, depending on your setup, since it steals one of the four available primary partitions.

I'll have to revisit booting XP from a USB drive again. Thanks for the advice.

Thermalwrong wrote on 2023-03-29, 18:09:

See that battery? somehow it's still working as a battery and has only *slightly* leaked into the negative terminal 😀

Yeah, I noticed that right away! But since it is leaking, it needs to go though ... 😉

H3nrik V! wrote on 2023-03-30, 03:40:

Thermalwrong wrote on 2023-03-29, 18:09:

See that battery? somehow it's still working as a battery and has only *slightly* leaked into the negative terminal 😀

Yeah, I noticed that right away! But since it is leaking, it needs to go though ... 😉

Fair point, it really should come off now that I've charged it up again through yesterday's use. I was looking up this board and Deksor apparently has the computer this came from, but the board was destroyed by this battery.

Lately I've been trying to remove the charging circuits from older motherboards to allow for use of lithium cells and while the 486 era all use fairly reasonable power (~6 to 8 uA). Which will run for two to four years on a CR2032 coin cell, these older boards draw enough power that external battery packs are probably the way instead.

edit: just wanna share this because it turned out looking cool. This is put together in Krita since it does good skew/scale, with a colour adjustment layer to change the tint of the reverse layer:

I've been running Windows XP SP3 on my P4 machine, but I've lately started to feel like SP2 would be more appropriate for the system even though I've had no issues whatsoever with SP3. I wanted to test things first before nuking the SP3 install I had on SSD, so I whipped out my test HDD to make sure SP2 installs fine and gets along with the machine, which it did, and just for fun I also tried Windows 2000 SP4 which also worked fine.

Since all worked well I formatted the SSD and proceeded to install WinXP SP2, the first part where it copies needed files after selecting the install partition went as normal but then all of a sudden it failed to boot from the SSD after the first reboot, gave some error message which I already forgot, something about not being able to boot the system right after the "press a button to boot from CD" timed out. After some troubleshooting I found WinXP SP3 installs just fine but WinXP SP2 does not no matter what I've tried, it always hangs at the first boot from SSD.

But the same WinXP SP2 installation CD works just fine on the HDD! Which is SATA II by the way, I'm running both the SSD and HDD on IDE compatibility mode so it shouldn't be a SATA issue.

So whats with WinXP SP2 and SSD's? Anything I could try to get it working? I've already tried to clean the drive and create new partition with Windows 7's diskpart (with correct partition alignment), no go. Also tested a small 10GB partition size in case it doesn't like large partition sizes, didn't work either.

Testing out a Logitech Momo driving wheel I picked up recently.

Everything works except for the brake pedal. For whatever reason, it shows as fully depressed (100%) in the Logitech software. Pressing the brake pedal doesn't change anything. Meanwhile, pressing the accelerator causes both the accelerator to move to 100% and the brake to reduce to ~5%.

Tested this on two different computers with two different operating systems (Windows 98SE and Windows XP) to rule out any system shenanigans.

I popped open the pedals themselves to test some things. Continuity seems to be fine through the attached cable. And the pot itself seems functional, although I don't know at this stage how it is calibrated and what the expected values are.

I'll need to do some more testing with the unit plugged in to see if I can get any response from the brake pedal's pot. Then maybe need to pop open the wheel itself, since everything runs through that.

Sombrero wrote on 2023-03-30, 15:31:

So whats with WinXP SP2 and SSD's? Anything I could try to get it working? I've already tried to clean the drive and create new partition with Windows 7's diskpart (with correct partition alignment), no go.

I vaguely remember partition alignment being a problem for one of my retro systems. The OS wouldn't install on a correctly aligned partition, but if I let it format the disk on its own, it worked fine.

I think it may have been Win2K in my case, not sure. It was a few years back.

Shponglefan wrote on 2023-03-30, 15:40:

Testing out a Logitech Momo driving wheel I picked up recently. […]
Show full quote

Testing out a Logitech Momo driving wheel I picked up recently.

Everything works except for the brake pedal. For whatever reason, it shows as fully depressed (100%) in the Logitech software. Pressing the brake pedal doesn't change anything. Meanwhile, pressing the accelerator causes both the accelerator to move to 100% and the brake to reduce to ~5%.

Tested this on two different computers with two different operating systems (Windows 98SE and Windows XP) to rule out any system shenanigans.

I popped open the pedals themselves to test some things. Continuity seems to be fine through the attached cable. And the pot itself seems functional, although I don't know at this stage how it is calibrated and what the expected values are.

I'll need to do some more testing with the unit plugged in to see if I can get any response from the brake pedal's pot. Then maybe need to pop open the wheel itself, since everything runs through that.

Are they not wired as top and bottom of same axis? If you were using a joystick, you'd be pushing forward or pulling back on the Y. ... though true simulation you should be able to do both at once, to destabilise vehicle on accel and brakes for throwing it sideways in turns.