Installed and purchased Lenny Loosejocks in Space by Ezone, and I did get this game for free back in 2007 when Ezone had the points system when you get in first place, you get a free game by the end of either the week or month, which was nice.

I repaired a Canon MD-5501 1.2MB 5.25" Floppy drive today!

I bought a lot of three "as-is" 5.25" drives on eBay last year for a good price and the genius seller did nothing but wrap them in one layer of newspaper (yes WRAP), and throw them in a box. Needless to say, they weren't kind to each other. The TEAC FD-55GFR was damaged badly and the Canon had a 6 pin connector nearly ripped clean off the PCB. The black 360K Panasonic drive however, worked fine and was in decent shape... thank goodness! I used that drive extensively for testing old disks. But, I still needed a 1.2MB drive!

I really wanted the TEAC to work, but when I tried fixing it last year I got nowhere. I think the PCB was actually cracked.

I've been trying to hoe-out my superfluous PC parts lately, and decided to take another look at these old drives rather than just leave them sitting here for the rest of eternity. The connector on this board is the type that holds a ribbon cable with 6 wide-spaced tinned wires that just fit in with no clips or pins. It hit me today that the easiest way to fix this would be to just desolder the damaged connector (and all of the severed pins) and put the wires directly through the holes. The issue with that was soldering the wires in place, since the board has to be mounted on the drive solder-side-down and only then are the wires in the right place. After looking at my busted TEAC drive, I noticed it had a similar setup with the wires soldered directly to the board... but they were soldered from the bottom up! So, after getting all of the old solder and busted pins out of the holes (a pain in the butt... had to use a vacuum desoldering station, pliers and a thumb tack...) I put the drive back together with the ribbon cable wires poking through their corresponding holes, I bent them over, soldered them in place and... voila! The drive works flawlessly! I booted from a 28 year old DOS 3.3 disk as well as a bootable MS Flight Simulator disk with absolutely no errors. This drive works incredibly well. I'm so glad I took the time to fix this rather than throw it in the trash. 😀

(Ignore the messed up graphics in Flight Sim... its an incredibly old version and doesn't seem to like running on a VGA card, even one as old as an ET4000AX)

Got 256MB of PC133 in the mail, and found that my first computer's motherboard HATES it. Then I dug up a stick of 128MB from a bag of RAM I found in a box in the basement, and it works fine. Lovely.

I did some tests on that Cached Promise controller yesterday (Need help on this Promise ISA caching controller), stacking it up against a standard Winbond ISA controller. For the test, I used Speedsys and 270mb 3600rpm Quantum Maverick Prodrive. Keeping the RPM's low, should in theory and in my mind, make the Cached controller somehow boost the drive. Ohh boy, was I wrong. I got the exact same kind of results when using a 3600rpm drive, as well as I got with a 4200rpm, 5400rpm and 7200rpm drives. As well as a CF card and a DOM module. Namely that the controller degrades the performance of storage, compared to a non cached standard ISA IDE controller.

The question is why? And as I have no means of testing out on WindowsNT 4.0 or 3.51. I have no freaking clue, if the Promise DC99 Cached ISA controller in reality is meant for NT or what the true purpouse of this card is.

I have included 3 pictures:

#1: Standard Winbond NON-Cached.
#2: Promise DC99 Cached Controller with 2.5mb cache (512k onboard + 2mb 30pin modules)
#3: For the laughs of it, Winbond Controller with 512mb Transcend CF card.

The computer setup, that I used for it. Was the following:

- ISA Only Socket 3 motherboard, with 256kb 15ns Cache
- Intel 486dx33 CPU
- 16mb 60ns 72pin EDO Memory
- Tseng ET4000
- Winbond and Promise DC99 Controllers
- 270mb 3600rpm Quantum Maverick ProDrive

As you all can see. The results for the Cached Controller is grimm.... I would have liked to see better results.
So I will keep to the CF card on the Winbond controller, for my dx33 build. A shame. The promise card just looks awesomme.
Yet with the fact that it does not provide compatibility with the Goldstar CD-Rom drive. I see no use for the Promise.

Can anyone give any example of were exactly this controller was introduced? What type of Computer is it meant for?

Speedsys Results:

quite frankly - imho no write tests so You just omitted the one thing this controller should be good at; also this thing is most effective when You do reads and writes alternating in quick succession (windows 95/nt mostly i guess); the max speed of all disks on this controller results from how fast the chip on its pcb can do the advanced processing required (more complex and time consuming than with the winbond) - also the promise has a big slow cache to check while the winbond has a very small and fast cache that it needs to access;
btw your disk performance numbers with that winbond are nice imho, i believe my 5x86 was slower;

mrau wrote:

quite frankly - imho no write tests so You just omitted the one thing this controller should be good at; also this thing is most effective when You do reads and writes alternating in quick succession (windows 95/nt mostly i guess); the max speed of all disks on this controller results from how fast the chip on its pcb can do the advanced processing required (more complex and time consuming than with the winbond) - also the promise has a big slow cache to check while the winbond has a very small and fast cache that it needs to access;
btw your disk performance numbers with that winbond are nice imho, i believe my 5x86 was slower;

Hmmm.... Well. I need to use another tool for testing, I did use a full hdd test in all three cases, using Speedsys.
If I use norton sysinfo, it will report that some tests can not be used, because of an intelligent controller.

Yeah... The speed of the Winbond controller alone, surprised me as well. Back in 1995/96 I had a Iwill SideJR Pro VLB
controller, and with an 800mb Quantum Trailblazer, it gave some 800 to 1000kb/s transfer rate, as far as I remember.
The Winbond controller that I used, was the one in the picture below. I have detached the HDD cable for a better shot.

Regarding the Cached controller's CPU. Then it is an 80186 running at 12mhz.

Managed to reach the 49fps "realm" in Doom Shareware with a Pentium @ 66mhz, clean boot'd on a CF card - this puts it quasi in league with the DX4 120s.

I just exploded my first tantalum capacitor!

I installed a near-mint condition Sceptre Trident TVGA8900B ISA card in my tester board, covered my face (because tantalums) and sure enough, this one exploded, left a long black scorch mark across the board and filled my office with a plume of smoke. Finally! 🤣

Over the past year I've covered my face and shut my eyes every time I powered up an old tantalum-equipped board for the first time. It finally payed off.

Too bad this card has 5 large (one of then blown), and half a dozen smaller tantalums. If I want to repair this card at some point, its going to take some time and I'll have to find lots of suitable caps. I think you can replace a tantalum with a suitable elecrolytic in many situations, but I don't know much about it personally.

@brostenen

Cache only helps when you are making lots of random reads and writes. It will never help a modern CF card. they are just too quick. They still are cool!

luckybob wrote:

@brostenen

Cache only helps when you are making lots of random reads and writes. It will never help a modern CF card. they are just too quick. They still are cool!

I am beginning to think, that this card is well suited for old 286/386 class file servers.

Got to have a nice afternoon of vintage computing today. Had to occupy my wife's little brother for a while, so we tinkered with my P3-450 test bench system trying to get it to properly play a "music.exe" file I found on some misc floppy disk from 1984. It was a 2KB file with 5 music tracks that play through the PC speaker. Playing them on the P3 450 resulted in a split second "BLURB" coming from the speaker for each song. Disabling cache made it a couple seconds of hyper-fast music. Inserting a P2-266 and running it at 200Mhz with disabled cache allowed for slightly slower music. Putting the program on a disk and running it on my IBM 5150 with the 7Mhz 286 upgrade running resulted in actual music, but it was much too fast still. Once I flipped the switch on the back of the computer to switch back to the 4.7Mhz 8088, the music.exe programmed worked perfectly and the music was surprisingly well done. I was quite impressed. To make it even better, one of the songs turned out to be a chunk of the song "The Entertainer", which is a song my brother in law really likes. So this was awesome.

After this, we installed Star Trek - The Next Generation - The Transinium Challenge from 1989 on my IBM. That is one of the most boring but hilariously weird games I've ever played. It worked great and we had some great laughs. Strangely, the game has a program that converts all of the graphics to EGA 16 color from CGA. It says it needs a ton of space though (an additional 1.5MB, on top of the 2.5MB free on my 20MB hard drive). So, I'll have to save the super high quality 16 color graphics after deleting some other games... 🤣

I also got my CT1350B + CMS working on my test bench to play some music while my daughter is playing the original EGA version of Mixed Up Mother Goose on my IBM (for the 5th time through). Fun times. 😀

EDIT: I've attached the music program in case anyone wants to try it out. You'll most likely need a 4.7Mhz 8088 for it to sound right, since its from 1984 and played too fast even on a 7.16Mhz 286.

I went through a lot of my vintage computer boxes before winter hits, making sure stuff in storage is okay. Holy crap, I have a lot of....crap. Next year, once I have my house built, I'm going to do a weekly blog, and maybe even make a few youtube videos of "opening vintage computer hoard boxes".

Oh, and I found a massive collection of Commodore 64 stuff for dirt cheap. I'll be picking it up Tuesday, so I'll post next week about my find.

Managed to repair this nice 80286 Mainboard with an old Chips P82C201 Chipset.
- There was a loose contact in the CPU socket, so I needed to solder a new one. Its another type but nobody will notice. Can you still buy these LCC sockets? Didnt find this one anywhere

- Then it complained about memory errors. I gave it up and installed a set of known working memory chips

- Then it still didnt power up. I noticed 2 40 Pin sockets were empty. One for the 80287..and the other one? Right, no keyboard controller. I found one in my spare parts and luckily its working

- board powered up now, but no build-in BIOS setup? Memory count still not correct, 128K? Was not able to boot DOS 6.22 and tried 3.33. It worked and I was able to run setup.

But the board still does not seem to detect more than 128K? Strange, should be more.

Unfortunately I didnt find any documentation, the board is labeled as "MS-0010-2" VER 4.0".

I am done with it, for me its running.

Predator99 wrote:

But the board still does not seem to detect more than 128K?

I am having similar challenges with an XT system due to lack of documentation. From the photo it looks like your board has 512KB. 256kBx16 chips (32KB each) the other two chips are parity. Have you tried different combinations with that red dip switch next to the memory banks? My XT had a fully loaded 640K on-board and a previous owner had incorrectly set the dip switches so it only used 512K. May be the same on your board. I'm still dealing with a weird memory card expansion memory count oddity. The fun of these early PCs is the Wild West designs and layouts and without the documentation you're left pulling levers and pushing buttons hoping you don't hit the eject seat button.

My previous post of the fun I'm having.

What retro activity did you get up to today?

Finally finished installing my DOS/98SE box w. XP dualboot:
Specs:
Intel Pentium 4 3.0 HT 800 FSB Prescott S478
Intel D865GBF i865G mainboard -> hoping to replace it with i865PE
4x512MB Infineon DDR400 CL3 matched (I have a workeround in Win98SE to hardcap it at 1GB, 2GB is useful for XP though)
GeForce 4 Ti4200 GPU (LeadTek WinFast A250LE TDH MyVivo)
WD400 40GB PATA IDE for 98SE, 80GB SATA Maxtor for XP

Ran 3dm2001SE yesterday - 10039pts on 800x600 and 8869 in 1024x768. OC'ing the card gives another ~500 points in 1024x768.

The other day I cleaned + retrobrighted this Optiplex GX1 which I am still awaiting parts for. https://imgur.com/a/lKORu

Only had 4 hours of sun so I will try again today. It came out pretty good already but I want to see how much better it can get.

oeuvre wrote:

The other day I cleaned + retrobrighted this Optiplex GX1 which I am still awaiting parts for. https://imgur.com/a/lKORu

Only had 4 hours of sun so I will try again today. It came out pretty good already but I want to see how much better it can get.

That is a rather convenient floppy drive placement I gotta say. (if the machine is run placed on the ground that is.)

Last night I started getting a pretty bad flicker on my 486 PC. I verified it wasn't the monitor, the particular game, or anything transient (shut down/restarted, checked the VGA cable), which sunk my stomach as that meant my only VLB graphics card might be going bad. So I pulled the card and gave it a good visual exam. Couldn't see any damage:

I went ahead and cleaned all the contacts and put it back in. Upon firing it back up, flickering was gone. So maybe the card got a little loose in the VLB slot or the contacts were dirty. Who knows?

liqmat wrote:

I am having similar challenges with an XT system due to lack of documentation. From the photo it looks like your board has 512KB […]
Show full quote

Predator99 wrote:

But the board still does not seem to detect more than 128K?

I am having similar challenges with an XT system due to lack of documentation. From the photo it looks like your board has 512KB. 256kBx16 chips (32KB each) the other two chips are parity. Have you tried different combinations with that red dip switch next to the memory banks? My XT had a fully loaded 640K on-board and a previous owner had incorrectly set the dip switches so it only used 512K. May be the same on your board. I'm still dealing with a weird memory card expansion memory count oddity. The fun of these early PCs is the Wild West designs and layouts and without the documentation you're left pulling levers and pushing buttons hoping you don't hit the eject seat button.

My previous post of the fun I'm having.

What retro activity did you get up to today?

Indeed, it should be 512k. Yes, I tried some dip combinations but nothing changed. I set 512 manually in the CMOS and I get still this error, it only disappears when setting to 128K. I dont think its a hardware problem and I am not going to use this board. Its just for my collection therefore I will not care any more about this.

Does your XT have a CMOS (would be quite unusual)? Maybe you have to set the correct memory size there first?

clueless1 wrote:

Last night I started getting a pretty bad flicker on my 486 PC. I verified it wasn't the monitor, the particular game, or anyth […]
Show full quote

Last night I started getting a pretty bad flicker on my 486 PC. I verified it wasn't the monitor, the particular game, or anything transient (shut down/restarted, checked the VGA cable), which sunk my stomach as that meant my only VLB graphics card might be going bad. So I pulled the card and gave it a good visual exam. Couldn't see any damage:

CL-GD5428VL-1.jpg

CL-GD5428VL-2.jpg

I went ahead and cleaned all the contacts and put it back in. Upon firing it back up, flickering was gone. So maybe the card got a little loose in the VLB slot or the contacts were dirty. Who knows?

That's all part of the fun that VLB brings.... 😉