bofh.fromhell wrote on 2020-04-20, 02:13:

Very nice!
What printer is that?

A Canon Pixma MX922 which I got used for CAD$50. Keep in mind you also need ink jet printable media, plenty of discs would just smear the ink off.

Recapped my BBC Micro's power supply today. Using a desoldering station definitely makes it an easier job. Seems like someone had been in the PSU before, filter cap C1 was gone, and both the pads for filter cap C1 and C2 had rather ugly solder rework done to them. I did all of the caps, as I like just doing everything in one go and not having to worry about the PSU for at least a few decades, at least when it comes to aging capacitors.

All electrolytics are either Panasonics or Nichicons from Mouser, so good quality, legitimate caps, and they're also 105°C rated instead of the originals' 85°C. The new RIFA filter caps are also from Mouser. It is funny to see how much smaller these new electrolytics are in comparison to the older ones. After a small screw up with one of the positive and negative wires being the wrong way around, the BBC Micro thankfully still works, I could hear the PSU chirping when I had it the wrong way around, likely going into protection mode. D'oh! At least I can rest easy knowing I won't get any magic smoke out of it any time soon! (Yes, that TV is a bit dusty, it's in a spare bedroom that's almost never used, just the quickest display I could hook up!)

Swappee out my P166mmx test mobo for an AUSUS socket 7 mobo rocking an AMD K5-266. More pci slots, slightly quicker, 2 16-bit ISA slots, 1 AGP slot and more available memory. I can test a few more cards now.

Repaired three PAL PSone consoles (SCPH-102)... Replaced bad SMD caps in the A/V output of two of them, repaired the third one's weak/cracked modchip solders, and replaced its dead CD drive, its broken DC in jack and its broken CD lid from a fourth one with more serious mainboard issues... Also started the recap of a '93 SEGA Game Gear, got the audio to work. Purposely mismatched two caps' capacitance rating due to lack of the appropriate parts, display mostly works apart from random "flashes" (sudden brightness increase/decrease) - will have to find the exact parts to complete the repair. Wanna be done with it and play some OutRun Europa!

Just a heads up, if you guys keep a Game Gear around, recap it ASAP! If it's dead (dim display, wrong contrast, no audio, general weirdness), the caps are 99.9% to blame. These damned ones are SMD, barrel in a box-like plastic enclosure, a bit hard to remove, they are glued to the board and the leakage might not be apparent, but they corrode the PCB badly to the point that it's hard to apply fresh solder without thoroughly cleaning/scratching the old pads. Also be careful with the positioning of the new caps so that you can close it back, it's pretty tight in there...

Caluser2000 wrote on 2020-04-20, 19:37:

Swappee out my P166mmx test mobo for an AUSUS socket 7 mobo rocking an AMD K5-266. More pci slots, slightly quicker, 2 16-bit ISA slots, 1 AGP slot and more available memory. I can test a few more cards now.

K5-266?..

Attempted to connect a CD drive to the secondary channel on my Packard Bell and doesn't get detected because the HDD on the primary channel is set to single mode. If there was a 4-6 footer IDE cable, I would put the HDD as master and the CD drive as slave and it'll get detected, or even connect the CF-IDE adapter to the primary channel as a primary slave and then get the CD drive to work.

Today, well (yesterday actually) I have been looking at my collection of Retro Universal Power Adaptors for my laptops! Currently editing all the ums and erms out of the video I did - not particularly exciting but I'm still learning to video edit so doing videos on all sorts of things so I've got some real footage to practice on! 😉

imi wrote on 2020-04-20, 14:21:

it's pretty much just showing the obvious... that at higher resolutions, in the case of Voodoo 2 SLI 1024x768, you are GPU limited beyond a certain point and gains are minimal, but at lower resolutions it scales very well up to 1000Mhz.

but yeah a K6-2 400 is going to be limiting to a Voodoo 3 in any case.

Functionality and stability > bottlenecks. Screw that implied "slow cpu need older video card because voodoo scales to the moon" newthink. There's absolutely nothing wrong with having a V3 with a K6, and on a K6 there's still practical advantages of V3 over a V2 SLI by the means of not wasting 2/3 slots for video (a K6 main PC then could have a TV card, modem, or something else useful there) and thrashing less / fillrate (as you know 3d games didn't stop at Forsaken / Incoming / Expendable).

Nobody said otherwise.

Oj0 wrote on 2020-04-20, 13:15:

appiah4 wrote on 2020-04-20, 12:51:

Oj0 wrote on 2020-04-20, 12:26:

That seems a bit high? I was running my V2 SLI on a Celeron 300A @ 504 and there was a definite gain from single to dual card.

At 800x600 it would run virtually at the same speed.

I only ran 800x600 or sometimes even 640x480. At the time I was actually using them they were connected to a 15" CRT which looked quite garbage at 1024x768.

We're going back a long, long time, so I can't give you exact numbers, but I remember there was definitely a performance difference.

I have been benching most of the Voodoo3 series this past couple of weeks (another post in this topic is due). Wizmark, 3DMark 99/00, 3DWinbench98/99/2000, Final Reality, Quake I/II/III and Unreal/UT99. I also have my Voodoo2 SLI results for comparison Will throw these somewhere onto the forum if someone is interested? Just let me know the appropriate place.

are you sure? ^^ got a photo of that CPU?

e: and the post vanished x3

So after successfully making disk images of a copy of Deluxe Paint II Enhanced that I bought a few months back which required spending a bunch of time cleaning the disks with Q-tips and alcohol, making an attempt, cleaning some more, making another attempt, etc until both disks finally read completely (woot!) ... well... that process had clearly left a bit of the gunk (mold, dirt, whatever) that was on the Deluxe Paint II disks onto the heads of my that disk drive. No disks would read completely without errors after that in the same drive. I kinda expected that and knew it had to be cleaned very carefully.

And I'm very lazy with such things, and basically only got around to doing that tonight!

I kinda wish I had taken a picture of the lower head before I cleaned it. There was definitely very visible dirt all over it! Anyway, it took a while, multiple attempts reading/writing disks, cleaning, then attempting again, etc ... but now that same drive is reading and writing disks just fine once more. Hooray! 😀

I put together a 286/8 and played around with it.

Then I formatted a disk!

Sort of:

So GSETUP'd the BIOS and indicated that the b:\ floppy was 720k (it is), yet under DOS 3.3 and 4.01 it force formats them to 179k. It can read the directory listing of a 720k diskette, but it will throw a General Failure error when it tries to run anything. The same 179K format is readable in my 386.

Issuing these commands under 3.3 and 4.01 give me an "invalid parameter" - Format b: /f:720 or Format b: /n:9 /t:80 (or whatever the parameters are for a 720K diskette)

I tried changing the BIOS settings to 1.44 Mb, 360k, both still format 179k . Changing it to a 1.2Mb gives me a "track 0 disk unusable" error.

Can it be that this computer just doesn't support 3.5 inch floppies?

*Edit and this is a BIOS limitation, it looks like it won't support 3.5" drives at all. Oddly, if you put a 1.44 Mb and register it as a 1.2Mb, you can read (but not write or format) 720k diskettes.

I always hated BIOS limits. Some 486 boards had HDD limits up to 512MB (or 504MB formatted) and required either a card that has higher HDD support, XT-IDE BIOS or card, or Overlay software.

Got a CD drive connected externally by feeding an IDE cable through the rear of the Packard Bell Pack-Mate 28 Plus's modem panel (removed said modem) with electrical tape wrapped around it and it works.

Well, sort of. When I put in the Megarace CD, once in a while, it'll throw an access denied error message, but it works. Also reinstalled Terra Nova Strike Force Centauri with the CD drive and it runs real smooth on an Evergreen 586 (AMD Am5x86WB-P75).

The drive's power is being used with the USB HDD power supply (for SATA and 40-pin IDE drives) so that the drive will have power going to the drive (might make a Molex power connector or by connecting a Molex connector internally through the blank slot for the COM port and get power from the PSU instead of using the other supply I have connected to the drive at this time).

Added a new games console to my collection -

This arrived today and I currently only have the 1 game for it, so just played a bit of that. I have more games and an official controller on the way, plus an Everdrive cartridge on it's way from China.

I never owned a SNES myself back in the day and only ever played it at friends' houses.

bjwil1991 wrote on 2020-04-22, 04:53:

I always hated BIOS limits. Some 486 boards had HDD limits up to 512MB (or 504MB formatted) and required either a card that has higher HDD support, XT-IDE BIOS or card, or Overlay software.

This BIOS only goes up to Type 23 for hard drives (which is fine as I have a Type 1) so I'm OK with that. I wonder if there's a floppy overlay software somewhere out there... I'll have to search.

I also used that "pass the IDE cable out the back" method - it works, it's not pretty, but it works 😉

Finished setting up my "secret" retro rig. I have been playing on this alot recently but it was all temporary with wires everywhere and meaning I could not use the closet for its actual purpose 😀

The only con (or is it a pro?) Is that I must stand to play.

Oh, the joys of SCSI...

All I wanted to do was test a single 1GB SCSI-2 SE drive for basic functionality and bad sectors. So I got my more-or-less trusty Dawicontrol DC-974 card, stuck it in my Knoppix test system, connected the drive via a 50p flatcable with active terminator on the end and...
- system didn't get past POST. After much head-scratching turned out to be the bootROM on the SCSI card. Removed and re-seated it, then POST finised, ROM initialized and...
- Knoppix got to the penguin (initializing kernel) then hung. Damned Linux kernel decided that its disk order was different to that of BIOS or grub and that its root filesystem should be on the SCSI disk. Could probably fix with parameters in grub, but couldn't be bothered. Hooked up my Gotek and booted from that instead...
- despite SCSI BIOS initializing the drive as D:, no D:, no drive recognized by fdisk or gdisk either. Also drive making repetitive noises. Dead drive?
- if so all SCSI drives dead as a 24MB SCSI-1 drive and a 1.8GB SCA drive via SCA-50p adapter also failed. Dead cable?
- maybe, but if so my other 50p cable and other terminator were also dead as same behaviour. Dead controller?
- nope, with 50p-68p adapter and 68p cable to SCA adapter it did work fine.

So it looks like I have two dead cables and/or terminators here, and I still don't know if the drive works 🙁

Unfortunately my other SCSI controllers are a bit exotic (one EISA, the other a combi SCSI+IDE controller that hangs any system with other IDE controller). This evening I wanted to work on an EISA system anyway (that's why I wanted a known-good drive...), so if that works will see what that controller thinks of the situation...

A few hours ago I received an RGB Scart cable for my SNES. It was a cheap cable and had the expected poor reviews about it not working on a PAL SNES console. Sure enough it didn't work on my SNES, with the screen fading to near black after a couple of seconds.

Thankfully though the fix was pretty easy, just open up the Scart plug, snip off 4 capacitors and re-solder the cables to the pins. 10 minutes later, a fully working RGB Scart cable without having to pay crazy prices for "PAL compatible" ones.

Got the Amiga 500 & 1084S-D2 monitor booted up on my test bench. Some functionality is there at least. I am going to make a Gotek external floppy with a switch select so it can function as a boot floppy, and then I have to fix the monitor to get the green cathode ray functional again. Should just be a broken solder joint or a failed transistor. I have the repair manual, so it will just be a matter of working through the schematic. I currently don't have time to do that because I'm at work more than I'm home, so it will have to wait.