@Skyscraper; Damn, that sucks! I hope they are working. I've seen drives survive surprising impacts before so you might be lucky. Good luck man.

brostenen wrote:

Have you tried one of those Zalman GPU coolers? Remember customers wanted the blue one with heatpipes n' stuff, back in 2005.
I was really impressed by how well it cooled card's like Radeon9800's and Geforce6800's.
They are to be found on ebay as of now. Not cheap, nor expensive I would say. And they come with a fan too.

Unfortunately, it is the CPU which heats up and I cannot find a better heatsink anywhere. I am sure one will show up eventually. Meanwhile I should be able to half-ass it until I get something better, I'll just wire the fan into the -5V rail for a bit of extra speed and hope it lasts until a replacement comes up - worst case scenario there is I break a cheap 60mm fan, which I have spares of anyway. I should be in Maplin soon as I need headphones and I remember them having a large Socket A cooler that they could never sell, they went from shelves ages ago but in the past I've asked about things and they have had them in the back of the store covered in dust, it is worth a shot as the worst that can happen is that they do not have one.

The GPU heatsink is awesome, ASUS even covered the RAM - a lot of vendors did not, including nVidia themselves - though they did not use paste or pads, so I added that when I put it back together.

I found a 40GB Seagate I did not know I owned - think it is from my NEC so I will image it. I think this will make a good drive for this machine. I can add others later if I need to. I do not fancy using my ExcelStor 80GB ever again, even though it is faster and larger it has a lot of miles and was sounding akin to a hammer drill last time I had it running.

HighTreason wrote:

Unfortunately, it is the CPU which heats up

Sorry... 😁 That slipped my attention. Read wrong, aparently still got hangovers from our one year engagement night-out.
Anyway. For a better cooler. Hmmm. I have this cooler, wich is really nice. (it can cool 775's too)

Skyscraper wrote:

Im testing old HDDs

I had a big box with ~30 old HDDs that took a 1 meter* nose dive down on my parquet floor half a year ago. Im now testing to see which drives survived and which drives sounds like waste shredders. The HDDs were just stacked in the box without any protection at all so I will be happy if half of them survived...

* 1 meter is ~3 feet in the medieval system of measurements.

I dropped a 40gb WD drive with a fresh/fully updated install of XP Pro from 6 feet last month on edge and it has not even hiccuped. I bet around 3/4 survive just fine! 😊

I'm starting to archive a variety of online FTP sites that contain drivers/programs/BIOS/firmware/etc. I went to a site that had a ton of Packard Bell stuff last year (motherboard schematics, etc), and now its not online. I'm sure its mirrored somewhere else, but I figured with my business internet connection with no bandwidth restriction at 100mbps, I'll start my own archive of FTP sites. I can see a massive archive being wiped out with no mirror on accident at some point. Storage is cheap, and might as well have my own archive.

Oh...and I went through all of my old ISA cards that I will never ever use. This forum needs a "JUST PAY FOR SHIPPING" thread...since I'll never use an ISA hercules card or a variety of EGA/CGA cards, or some weird microchannel ethernet card.

HighTreason wrote:

Guess what I think I fixed... […]
Show full quote

Guess what I think I fixed...

Awesome_Ti.jpg

The GeForce 4 Ti 4600! F*** yeah! My overkill DX8 card. Oh, the Ti 4800 SE seems to work too. Now I can build that Athlon box to replace my old, beaten and dysfunctional Athlon box... Oh... Hang on... No, it keeps overheating and that heatsink is the biggest one I have. I cannot find one anywhere either, so I am stuck, my only option is to drop to the 1500+ as I cannot cool this 2600+ with any heatsink I own. The only comparable heatsink is on another 2600+ but they are two very different models of CPU and I tried it before, it is a much smaller heatsink and cannot dissipate the heat at all.

I also cannot afford to damage the CPU as this version is very hard to find; http://www.cpu-world.com/CPUs/K7/AMD-Athlon%2 … A2600DKV3C.html I do not think I will be so lucky as to find one at the glorious price of 1.99 again. Only thing I can think of is to get a fan adapter, but it would not fit in the case. Perhaps I can find a server-grade 60mm fan instead as noise is not a problem, I could even build a temp control circuit (existing fan actually has one) if it were a problem.

Have not tested GPU in Direct3D yet, but before the inductor was broken off it worked fine, without the inductor it caused a reboot as soon as the system displayed graphics for a few seconds. It no longer does this so I assume it will work, I can't see anything else just randomly dying because an obscure component went open circuit - it is not outside the realms of possibility I guess, but it seems unlikely. Thank you, dead XGL card... I hate butchering things like that, but it really was dead, very dead, the GPU was actually melted when I got it and of course it did not work... Has sat in the drawer since, though it donated its heatsink to something around a year ago.

I sure hope this thing works. I think I can get away with running it for now as it idles at 55°C and will go up to around 70° when loaded hard... Apparently they take 85° but I DO NOT like to run the machine above 60° EVER. In fact, I try to keep machines below 45°C at all times, am I fighting a losing battle here though? I mean, this CPU clearly runs much hotter than the 200MHz versions (This is a 133MHz chip), perhaps it always will? I still will try to improve cooling ASAP though and I will still set the fail-safe, I know it is responsive on this board and just cuts power unceremoniously, which is good.

EDIT: Oh, I also wanted to add. Something funny happens when you plug a U5S-Super33 into a Micronics M4Pi (A very finicky motherboard) - it detects a "486SX Operating at 100MHz" according to the POST screen. Obviously it is not, but I found it funny. Intel should have made a 486SX4 because that board reports this for other CPUs such as the 486DX4 (All makes) and the Cx5x86. It will not POST with an Am5x86 and is the only board I know of to have this problem, especially with it running the Cyrix. Very strange board and I may not own it much longer, so it is good to learn a bit more about its quirks. It actually isn't a bad board paired with a 486DX2 and an S3 video card.

266 FSB Athlon chips don't run that hot.. I would say check your TIM install and check the default voltage on cpu world and make sure your motherboard isn't over-volting it. Also check for capacitors. I have a Soyo Dragon motherboard for AthlonXP that has some swelling capacitors near the cpu socket and it makes all chips put in it run +10c hotter than in other motherboards.

Also, I have a 266 FSB AthlonXP chip in my MS-DOS machine, that even takes a voltage bump and a slight overclock of +350 mhz and runs at 350 FSB and still only idles at 35c, load is like 51c-55c. And it's nothing special, just a generic socket 462 all-aluminum heatsink with a 80mm fan mount.

On the other hand however, my 400-FSB Barton-core AthlonXP chip in my win98se machine that runs with a little overclock @ 2560 mhz uses a thermaltake volcano all-copper heatsink and a TT Tornado fan on it (with a fan controller) idles at 50c-55c and gets up to 65c-68c under load. -THOSE- late model AthlonXP chips are the ones that run really hot. I'm seriously considering water cooling it some time in 2016.

dogchainx wrote:

I'm starting to archive a variety of online FTP sites that contain drivers/programs/BIOS/firmware/etc. I went to a site that had a ton of Packard Bell stuff last year (motherboard schematics, etc), and now its not online. I'm sure its mirrored somewhere else, but I figured with my business internet connection with no bandwidth restriction at 100mbps, I'll start my own archive of FTP sites. I can see a massive archive being wiped out with no mirror on accident at some point. Storage is cheap, and might as well have my own archive.

Oh...and I went through all of my old ISA cards that I will never ever use. This forum needs a "JUST PAY FOR SHIPPING" thread...since I'll never use an ISA hercules card or a variety of EGA/CGA cards, or some weird microchannel ethernet card.

If you want stuff to archive, check out http://www.dreamlandbbs.com/index.html An old BBS turned in to an online file repository, some interesting old software and game demos and stuff here.

ODwilly wrote:

Skyscraper wrote:

Im testing old HDDs

I had a big box with ~30 old HDDs that took a 1 meter* nose dive down on my parquet floor half a year ago. Im now testing to see which drives survived and which drives sounds like waste shredders. The HDDs were just stacked in the box without any protection at all so I will be happy if half of them survived...

* 1 meter is ~3 feet in the medieval system of measurements.

I dropped a 40gb WD drive with a fresh/fully updated install of XP Pro from 6 feet last month on edge and it has not even hiccuped. I bet around 3/4 survive just fine! 😊

In my experience, don't knock external HDDs off the desk when they're plugged in. Expensive accident 😉

dogchainx wrote:

I'm starting to archive a variety of online FTP sites that contain drivers/programs/BIOS/firmware/etc. I went to a site that had a ton of Packard Bell stuff last year (motherboard schematics, etc), and now its not online. I'm sure its mirrored somewhere else, but I figured with my business internet connection with no bandwidth restriction at 100mbps, I'll start my own archive of FTP sites. I can see a massive archive being wiped out with no mirror on accident at some point. Storage is cheap, and might as well have my own archive.

You may want to work with archiveteam's project in case you discover FTP sites they don't yet archive.
http://www.archiveteam.org/index.php?title=FTP
https://archive.org/details/ftpsites
Past VOGONS discussion: Preservation of historical ftp site content

Stiletto wrote:

ODwilly wrote:

Skyscraper wrote:

Im testing old HDDs

I had a big box with ~30 old HDDs that took a 1 meter* nose dive down on my parquet floor half a year ago. Im now testing to see which drives survived and which drives sounds like waste shredders. The HDDs were just stacked in the box without any protection at all so I will be happy if half of them survived...

* 1 meter is ~3 feet in the medieval system of measurements.

I dropped a 40gb WD drive with a fresh/fully updated install of XP Pro from 6 feet last month on edge and it has not even hiccuped. I bet around 3/4 survive just fine! 😊

In my experience, don't knock external HDDs off the desk when they're plugged in. Expensive accident 😉

I have a Seagate 500GB external 2.5" HDD that has seen alot of abuse but so far I have been lucky 😀. I guess a more heavy 3.5" disk would be more sensitive to being pushed off the table while copying stuff 😵, I have one of those aswell but its only used for backups. The 3.5" disks probably also have less shock protection as they are not used in portable computers.

Funny you should mention that, because I have a 160gb external 2.5 WD drive that the enclosure is broken that has gone through a washer and dryer in my pants pocket before and gets thrown around all the time. Tough little bugger! Laptop drives are pretty amazing.

HighTreason wrote:

Guess what I think I fixed... […]
Show full quote

Guess what I think I fixed...

Awesome_Ti.jpg

The GeForce 4 Ti 4600! F*** yeah! My overkill DX8 card. Oh, the Ti 4800 SE seems to work too. Now I can build that Athlon box to replace my old, beaten and dysfunctional Athlon box... Oh... Hang on... No, it keeps overheating and that heatsink is the biggest one I have. I cannot find one anywhere either, so I am stuck, my only option is to drop to the 1500+ as I cannot cool this 2600+ with any heatsink I own. The only comparable heatsink is on another 2600+ but they are two very different models of CPU and I tried it before, it is a much smaller heatsink and cannot dissipate the heat at all.

I also cannot afford to damage the CPU as this version is very hard to find; http://www.cpu-world.com/CPUs/K7/AMD-Athlon%2 … A2600DKV3C.html I do not think I will be so lucky as to find one at the glorious price of 1.99 again. Only thing I can think of is to get a fan adapter, but it would not fit in the case. Perhaps I can find a server-grade 60mm fan instead as noise is not a problem, I could even build a temp control circuit (existing fan actually has one) if it were a problem.

Have not tested GPU in Direct3D yet, but before the inductor was broken off it worked fine, without the inductor it caused a reboot as soon as the system displayed graphics for a few seconds. It no longer does this so I assume it will work, I can't see anything else just randomly dying because an obscure component went open circuit - it is not outside the realms of possibility I guess, but it seems unlikely. Thank you, dead XGL card... I hate butchering things like that, but it really was dead, very dead, the GPU was actually melted when I got it and of course it did not work... Has sat in the drawer since, though it donated its heatsink to something around a year ago.

I sure hope this thing works. I think I can get away with running it for now as it idles at 55°C and will go up to around 70° when loaded hard... Apparently they take 85° but I DO NOT like to run the machine above 60° EVER. In fact, I try to keep machines below 45°C at all times, am I fighting a losing battle here though? I mean, this CPU clearly runs much hotter than the 200MHz versions (This is a 133MHz chip), perhaps it always will? I still will try to improve cooling ASAP though and I will still set the fail-safe, I know it is responsive on this board and just cuts power unceremoniously, which is good.

EDIT: Oh, I also wanted to add. Something funny happens when you plug a U5S-Super33 into a Micronics M4Pi (A very finicky motherboard) - it detects a "486SX Operating at 100MHz" according to the POST screen. Obviously it is not, but I found it funny. Intel should have made a 486SX4 because that board reports this for other CPUs such as the 486DX4 (All makes) and the Cx5x86. It will not POST with an Am5x86 and is the only board I know of to have this problem, especially with it running the Cyrix. Very strange board and I may not own it much longer, so it is good to learn a bit more about its quirks. It actually isn't a bad board paired with a 486DX2 and an S3 video card.

If you don't especially want that particualar cpu you can just buy another one, preferable a mobile and just pin-mod the socket, very easy.

Guide: http://www.ocinside.de/workshop_en/amd_pinmod/

Best socket A cooler I've found (and I've been looking) is the Thermaltake Big Typhoon. You need the 4 mounting holes around the socket though. The best cooler that uses the socket clips could probably be the Thermalright slk-900a or maybe Swiftech MCX462-V, but I'm note sure.

I have the mounting holes, but the only Big Typhoon cooler I can find is some heatpipe ridden monstrosity that would not fit in the case. The seller also does not know what it fits onto.

I'll think of something. I am not wiling to cut back on CPU performance, editing images that are over 12000 pixels (Yes, five digits!) each way is hard enough already and I don't need to make it worse. 24 frames per second, an average of at least 4 layers and sequences running over a minute long... All I need is slower processors. I'm already mad at having to move back to Windows 98 all these years on.

Swapped CTs around. My Magnavox 386SX now has a CT2290 with 4X Mitsumi IDE CD-ROM connected to it. 😀

Also installed Windows 98 on my IBM NetVista. The 40GB IDE HDD I found had Windows 2000 on it and would not boot. I used diskpart.exe to remove the partitions. Impossible to do with Windows 98 (Windows 2000 is NTFS, not FAT16/32). 😵

Played some Pirates Gold! Awesome game.

Then played some MID files: including Wacht am Rhein, Das Boot, Air Wolf, Zapfenstreich, Deutschlandlied, Panzerlied. Great tunes: excellent renditions on the SC-55ST. 🤣

That MO drive is still cranky. Doesn't like cold starts but once it is behaving it's a dream machine. Even got it to boot off an OD finally. Slow, but this was someone's dream. 🤣

Skyscraper wrote:

brostenen wrote:

Skyscraper wrote:

Im testing old HDDs

I had a big box with ~30 old HDDs that took a 1 meter* nose dive down on my parquet floor half a year ago. Im now testing to see which drives survived and which drives sounds like waste shredders. The HDDs were just stacked in the box without any protection at all so I will be happy if half of them survived...

* 1 meter is ~3 feet in the medieval system of measurements.

Hope the damage is minimal.

So far 5 out of 5 drives have been OK, 30KB or so in damaged sectors on a 540MB drive but those could very well have been there before the "accident". I have never fully tested all the drives before, just made sure that they were spinning up without nasty sounds, were recognized by the BIOS and that I could write a test file to them. This time Im deleting the partitions, creating new ones and formating them.

I need to test the rest of the drives in another computer as the Intel 430FX chipset board Im using now wont recognize disks larger than ~2.1GB.

I'm super interested on how you test the drives. 😁
I've never been able to fully test non LBA-compatible drives, so as to be sure of their condition. 😒

Just when I thought I finally had my floppy drives perfectly aligned...

this happens.

edit:
Just discovered, if I remove the top circuit board from drive A, it reads just fine. As soon as I move it back, the top head fails to read over half the sectors in the inner tracks. Can't see any difference in signal level on the scope with or without PCB. What the heck is going on here?!?!

GeorgeMan wrote:

Skyscraper wrote:

Skyscraper wrote:

Im testing old HDDs

I had a big box with ~30 old HDDs that took a 1 meter* nose dive down on my parquet floor half a year ago. Im now testing to see which drives survived and which drives sounds like waste shredders. The HDDs were just stacked in the box without any protection at all so I will be happy if half of them survived...

* 1 meter is ~3 feet in the medieval system of measurements.

So far 5 out of 5 drives have been OK, 30KB or so in damaged sectors on a 540MB drive but those could very well have been there before the "accident". I have never fully tested all the drives before, just made sure that they were spinning up without nasty sounds, were recognized by the BIOS and that I could write a test file to them. This time Im deleting the partitions, creating new ones and formating them.

I need to test the rest of the drives in another computer as the Intel 430FX chipset board Im using now wont recognize disks larger than ~2.1GB.

I'm super interested on how you test the drives. 😁
I've never been able to fully test non LBA-compatible drives, so as to be sure of their condition. 😒

I will be testing them using the same unscientific method I used for the smaller drives only with DOS 7, FAT32 and a motherboard with better support for disks larger than 2.1GB.

1. See if the disk spins up without making nasty sounds
2. See if the system can find the drive.
3. Remove all partitions with fdisk
4. Create a new one using 100% of the availible space
5. Format the partition and see if any faulty sectors are found or if I hear any grinding noises.
6. Pretend that the disk is totally fine.

This should work for all disks as the largest disk in the box is 40GB, fdisk handles disk sizes up to at least 128GB even if it fails to report the available space before creating the partition for disks larger than 32GB.

I acquired an Acer 386SX/20N last week. First thing I did was open it up and have look (I like to inspect old hardware before running it). Couldn't find the CMOS battery initially as I was expecting a barrel or a CR2032 and instead found a Dallas RTC. This was very exciting as I'd never come across one but had read about them on this excellent site.

After intending to employ the hacky method I decided to get a new replacement off eBay instead as they were only £3. Still had to get my soldering iron out to do some repair on the Conner CP3000 logic board.

Everything works, Win 3.1 booted right up - I've got my childhood computer back! 😊 Now I need an Adlib.

Skyscraper wrote:

I will be testing them using the same unscientific method I used for the smaller drives only with DOS 7, FAT32 and a motherboard […]
Show full quote

I will be testing them using the same unscientific method I used for the smaller drives only with DOS 7, FAT32 and a motherboard with better support for disks larger than 2.1GB.

1. See if the disk spins up without making nasty sounds
2. See if the system can find the drive.
3. Remove all partitions with fdisk
4. Create a new one using 100% of the availible space
5. Format the partition and see if any faulty sectors are found or if I hear any grinding noises.
6. Pretend that the disk is totally fine.

This should work for all disks as the largest disk in the box is 40GB, fdisk handles disk sizes up to at least 128GB even if it fails to report the available space before creating the partition for disks larger than 32GB.

Don't forget you can also run ms-dos scandisk and do a surface scan, sector by sector. Slow but this is usually what I do on smaller disks. Although sometimes I'll just connect em to a newer system with IDE like an AM2/AM3/775 system and just run a hdtune surface scan in windows. Should work, even trying to check out a 200 MB IDE drive. The IDE format hasn't changed since it's incarnation, any very old IDE drive will work in any later system if it's functional.

kithylin wrote:

Skyscraper wrote:

I will be testing them using the same unscientific method I used for the smaller drives only with DOS 7, FAT32 and a motherboard […]
Show full quote

I will be testing them using the same unscientific method I used for the smaller drives only with DOS 7, FAT32 and a motherboard with better support for disks larger than 2.1GB.

1. See if the disk spins up without making nasty sounds
2. See if the system can find the drive.
3. Remove all partitions with fdisk
4. Create a new one using 100% of the availible space
5. Format the partition and see if any faulty sectors are found or if I hear any grinding noises.
6. Pretend that the disk is totally fine.

This should work for all disks as the largest disk in the box is 40GB, fdisk handles disk sizes up to at least 128GB even if it fails to report the available space before creating the partition for disks larger than 32GB.

Don't forget you can also run ms-dos scandisk and do a surface scan, sector by sector. Slow but this is usually what I do on smaller disks. Although sometimes I'll just connect em to a newer system with IDE like an AM2/AM3/775 system and just run a hdtune surface scan in windows. Should work, even trying to check out a 200 MB IDE drive. The IDE format hasn't changed since it's incarnation, any very old IDE drive will work in any later system if it's functional.

I do not have the patience for surface scans 😁, perhaps with the smallest disk,in the box (170MB) but with larger disks a surface scan takes ages 😜

These disks will not be used for anything critical but I still want to know which ones I can throw away.

Skyscraper wrote:

I do not have the patience for surface scans 😁, perhaps with the smallest disk,in the box (170MB) but with larger disks a surface scan takes ages 😜

These disks will not be used for anything critical but I still want to know which ones I can throw away.

It's your life... any disk with even one bad sector will lead to corruption and data loss.. and should be discarded. Personally I won't use any mechanical drive with a bad sector on it, IDE or sata. Along those notes, I have an older 80 MB IBM SCSI drive somewhere in a box that still has no bad sectors and spins up fine.. 😀