If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#11
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
J. P. Gilliver (John) wrote:
No, as the article explains, if you only have one drive, and it was connected to the middle connector, you'd have an unterminated stub of cable, which isn't good electrically (reflections and so on). Apparently for the 40 as opposed to 80 cables that _was_ the case, as they just left out line 28 to the second connector (i. e. master was on the middle connector). I like to draw pictures for people. For a single IDE drive, it *always* goes on the end, like this. Mobo X --------------+--------+ | | Master Cable_Select (if 80 wire, CS is allowed) When you add a second drive, it can be like this. Or, you can run CS on both drives, if you are using an 80 wire cable (with that twist in it). I didn't want to junk up the diagram, by adding CS to the table for both drives. Mobo X --------------+--------+ | | Slave Master Master with Slave (some brands have a distinction on the jumpers) Do *not* do this, as the end of the cable constitutes a stub and causes excess reflections and corrupted data. Mobo X --------------+--------+ | | Oopsy ULooz Even in the best of circumstances, the signals on that cable look horrible. The signals look more horrible in that last case. One of the things that SATA does, is banish those bad design ideas... to the pit. With point to point SATA, there are no more simulation nightmares for engineers to look at. Someone (of course) can still make a SATA cable out-of-spec, but the field reports seem to be pretty good. Almost as if most of the rolls of raw cables come from one cable plant, and that helps keep the process "honest". You should not bend a SATA cable until it kinks, as that causes unpredictable results to your data. You could get away with it, or not. Don't crush the excess SATA cable and tightly wrap duct tape around it. Bad. If you have too much SATA cable, buy a shorter one from the store and try again. HTH, Paul |
#12
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
On Sun, 17 Dec 2017 13:01:49 +0000, "J. P. Gilliver (John)"
wrote: In message , writes: On Sun, 17 Dec 2017 06:57:37 +0000, "J. P. Gilliver (John)" wrote: [] Yep, it looks like the old drive was set to SLAVE. However, I just changed the new one to CS and copied a bunch of stuff to it from my first drive. Then I deleted some stuff and copied a whole bunch of small clipart pics to it, and then deleted some of them, and after that I copied a huge ISO file to it, which is almost 1gb in size. After all of that, I defragged that drive with no problem. It appears that it needs to be set to CS. Maybe that was the whole problem. I'll copy more stuff to it and delete other stuff and see if it keeps working properly now. So far, so good! I hesitate to ask, but when you say "the old drive" above, do you mean the CD drive that failed years ago, or do you mean the HD-that-was-G/H/I whose failure started this whole saga? If the latter, I wonder if setting that to CS might have cured the original problem )-:! There is no CD drive involved in any of this. In fact the CD drive in this computer died years ago. I really dont have any need for one on my Win98 machine. I have the Win98 install files right on the HDD. I dont play games or anything that needs a CD. I do have a CD player on my XP machine, but rarely use it. But I would need it to reinstall XP, and once and awhile I copy a music CD and turn it into MP3 songs for my MP3 player. If I want to listen to CDs, I have a regular CD player on my stereo. Anyhow, I was referring to my old 2nd drive / Slave (G: H: I. I ma tempted to try the actual Master and Slave settings with the jumpers and see if that works. I dont know if one way is better than the other, or not? Does anyone know which jumper setting is the best? Does anyone know whether using master/slave jumpering with a cable on which CS works might cause problems? I never understood how that CS works, or why it's even an option. Older [] If the cable truly has the same connections on all three connectors, then I can't see how it's selecting either. I know floppy drive cables had a twist in the cable. Yep, floppy cables do have a twist, but not these IDE Hard drive cables. So how that CS works is beyond my comprehension. I do know that for Maybe there's an internal break in one line - so the cable from the mobo to the first connector is 80 way, but between them is 79 or 78 way? (I take it there's nothing obvious like one of the connectors having one of its holes blanked.) This is a brand new cable. Of course anything can be defective. Ah, I've just looked: https://en.wikipedia.org/wiki/Parallel_ATA#Cable_select says it is done using pin 28 - often just by omitting the contact from the middle (slave, grey) connector, so you'd have to look extremely hard to see it! It also says line 28 is only used so the drives know which they are, not for control by the mobo, so if the drives are jumpered as master and slave anyway, it is ignored (and that doesn't have to be master at the end). So you can try it if you want. When the controller says "master drive, please respond", both drives receive the command, but only one of them responds - either because it is jumpered as master, or because both are jumpered as CS and one of them knows it is master. (Apparently also "drive 0" and "drive 1" - apparently "master" and "slave" don't actually appear in the specification.) Which does suggest that having one drive "hard jumpered" and the other as CS _could_ cause problems, depending on position on the cable. Ok, that explains it.... awhile I had the Master drive on the first connector and Slave on the last connector. THAT IS WRONG, but it was that way for a year or more and worked fine. Maybe it dont much matter which cable comes first, but according to several articles, the last connector goes to the first drive (which seems backwards). No, as the article explains, if you only have one drive, and it was connected to the middle connector, you'd have an unterminated stub of cable, which isn't good electrically (reflections and so on). Apparently for the 40 as opposed to 80 cables that _was_ the case, as they just left out line 28 to the second connector (i. e. master was on the middle connector). I better understand this now. I know I have the cable right now, and since I changed that jumper to CS, it looks like everything works fine now. (At least so far). I have copied and deleted files and defragged and scandisked, and ran Norton Dick Doctor. I even ran scandisk from Dos. Everything checks out ok. I sort of am wondering if the problem on my old slave drive may have been caused by the jumpers being incorrectly set, but I had them drives that way for at least 2 years. I'd think that would have shown up a lot sooner. |
#13
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
On Sun, 17 Dec 2017 14:02:20 -0500, Paul wrote:
J. P. Gilliver (John) wrote: No, as the article explains, if you only have one drive, and it was connected to the middle connector, you'd have an unterminated stub of cable, which isn't good electrically (reflections and so on). Apparently for the 40 as opposed to 80 cables that _was_ the case, as they just left out line 28 to the second connector (i. e. master was on the middle connector). I like to draw pictures for people. For a single IDE drive, it *always* goes on the end, like this. Mobo X --------------+--------+ | | Master Cable_Select (if 80 wire, CS is allowed) When you add a second drive, it can be like this. Or, you can run CS on both drives, if you are using an 80 wire cable (with that twist in it). I didn't want to junk up the diagram, by adding CS to the table for both drives. Mobo X --------------+--------+ | | Slave Master Master with Slave (some brands have a distinction on the jumpers) Do *not* do this, as the end of the cable constitutes a stub and causes excess reflections and corrupted data. Mobo X --------------+--------+ | | Oopsy ULooz Even in the best of circumstances, the signals on that cable look horrible. The signals look more horrible in that last case. One of the things that SATA does, is banish those bad design ideas... to the pit. With point to point SATA, there are no more simulation nightmares for engineers to look at. Someone (of course) can still make a SATA cable out-of-spec, but the field reports seem to be pretty good. Almost as if most of the rolls of raw cables come from one cable plant, and that helps keep the process "honest". You should not bend a SATA cable until it kinks, as that causes unpredictable results to your data. You could get away with it, or not. Don't crush the excess SATA cable and tightly wrap duct tape around it. Bad. If you have too much SATA cable, buy a shorter one from the store and try again. HTH, Paul I have no problem with SATA drives. I dont think they would work for Win98 though. One thing about them, I dont understand why the data cable has so few wires compared to the IDE cables, and even more puzzling why the power connectors have all those pins, when there is still only 5V 12V and a copule grounds needed (4 wires). Why do they have all them pins? Why didnt they just use the common 4 pin connecters they have used for years. All that did is make power supplies more complicated and the need to buy adapters to use older power supplies. That "dd" thing sounds too much like Linux command line **** to me. I dont touch that ****.... I'd rather run scandisk, chkdsk, or Norton Disk Doctor (NDD). Ndd runs faster than scandisk, so I normally run that. Scandisk took 13 hours to scan a 40gb drive, Ndd takes 4 or 5 hours to do the same. I do question how much drives are abused by running all this stuff. Like, how much life is taken away from drives by beating the crap out of them with these sector by sector tests? This is not like normal use, this is extreme abuse. Some of those tests are made so they can be run 10x or even more. Not only would that make my computer unusable for several days, but probably eliminates 50% of the drive's lifespan. |
#14
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
wrote:
I have no problem with SATA drives. I dont think they would work for Win98 though. One thing about them, I dont understand why the data cable has so few wires compared to the IDE cables, and even more puzzling why the power connectors have all those pins, when there is still only 5V 12V and a copule grounds needed (4 wires). Why do they have all them pins? Why didnt they just use the common 4 pin connecters they have used for years. All that did is make power supplies more complicated and the need to buy adapters to use older power supplies. The SATA 7 pin data uses TX+,TX-,RX+,RX-, and those are differential serial connections. The data travels serially, a bit at a time, like a modem. Only it happens at 6Gbit/sec, which is "faster than your microwave oven". It's a signal at microwave frequencies. So that's how they squeezed down the data cable, by going serial. USB uses this approach too. USB3 uses TX+,TX-,RX+,RX- . This document is 3.6MB and it has a picture of what the SATA data signal looks like, at 6Gbit/sec. And it isn't even an eye diagram picture - the picture is standalone ones and zeros. Page 4 has the picture. http://download.tek.com/datasheet/4HW_19377_15_0.pdf ******* The SATA connector is designed for SATA backplanes. It's usage in desktop computers is an afterthought. The hard drive was supposed to slide into a hole in a chassis, and the back of the drive mates with a backplane connector that "sticks out" of the backplane PCB board. And via hotswap, on a server you could add or remove drives while the server remained powered and running. ******* The 15 pin power is 5 groups of 3 pins each. A pin carries 1 ampere of current. Three pins carry 3 amps. And 3 amps is just enough for the +12V source, to run the hard drive motor. At one time, some hard drives would draw 3 amps for the first ten seconds, until the spindle was up to speed. So the contact count for power, was made generous enough to run existing hard drives. Actual current flow measurements, show drives now being "all over the place" with regard to the level of current flow at startup. I don't think I found any samples I tested, drawing the whole 3 amps. The groups on the power connector are 3.3V, 5V, 12V, GND, GND The expectation is, a design might use two of three power sources, so only two ground groups are needed. A conventional disk drive uses 5V, 12V, GND, GND and so there are just enough grounds to match the current flow level on the supply pins. Paul |
#15
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
|
#16
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
|
#17
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
On Mon, 18 Dec 2017 21:15:58 -0500, Paul wrote:
The reason I worry about this sort of stuff, is I was actually sent on a plane trip from work, to root cause why we had excessive disk failures at a certain site in the US. When I saw what they were doing, my jaw dropped :-) And I haven't been the same since :-) It's not only the UPS driver who has evil in his heart. Paul I would absolutely run a RAM test, may be nothing to do with the drive Any PC you pull from storage, should have some basic health tests done on it. There's no harm in trying that. I did that on my new machine only a couple days ago... just in case. Paul How do you test RAM? Hopefully its a Windows program, not linux.... |
#18
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
Il giorno Sat 16 Dec 2017 08:26:40p, ** ha inviato su
microsoft.public.windowsxp.general il messaggio . Vediamo cosa ha scritto: This is an old IBM brand computer from about 2001 an almost seventeen-years-old computer is like a Ford Model T you can't ask the hardware to be fully functional after so much time it's your mistake to pretend to work safely with it imho -- /-\ /\/\ /\/\ /-\ /\/\ /\/\ /-\ T /-\ -=- -=- -=- -=- -=- -=- -=- -=- - -=- http://www.bb2002.it ............ [ al lavoro ] ........... |
#19
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
On Tue, 19 Dec 2017 07:40:50 +0000 (UTC), Ammammata
wrote: This is an old IBM brand computer from about 2001 an almost seventeen-years-old computer is like a Ford Model T you can't ask the hardware to be fully functional after so much time it's your mistake to pretend to work safely with it imho Bull****. Unless it's dead, it works just fine, and I have used it almost daily for the past 17 years. I have radios and power tools from the 1950s and 60s that still work fine too. Then again, I'm almost 70 years old, and I still work pretty well too. You sound like one of todays spoiled rotten youth who cant stand to have anything more than 2 years old, and waste a lot of your parents hard earnings on expensive toys, because your old toy (which is one year old) no longer gives you any thrills). I cant wait to see all you spoiled youth crash when the economy goes to ****. Maybe you'll learn the value of a dollar that YOU had to earn yourself, and you'll have to wear an old coat you've had for 4 years and be forced to use a 5 year old cellphone or computer, or do without them completely. Your generation makes me want to puke. Not only are you spoiled so rotten that you stink, but you are the worst polluters to ever inhabit this earth, with all the waste you create (because it's old). |
#20
|
|||
|
|||
New HDD, has corrupted Data - AGAIN
wrote:
On Mon, 18 Dec 2017 21:15:58 -0500, Paul wrote: The reason I worry about this sort of stuff, is I was actually sent on a plane trip from work, to root cause why we had excessive disk failures at a certain site in the US. When I saw what they were doing, my jaw dropped :-) And I haven't been the same since :-) It's not only the UPS driver who has evil in his heart. Paul I would absolutely run a RAM test, may be nothing to do with the drive Any PC you pull from storage, should have some basic health tests done on it. There's no harm in trying that. I did that on my new machine only a couple days ago... just in case. Paul How do you test RAM? Hopefully its a Windows program, not linux.... Windows does have a memory diagnostic. On WinXP, you get a copy by downloading it. On more modern Windows OSes, it's included on C: (but you have to figure out where it's located of course). https://web.archive.org/web/20070102...en/windiag.asp (640KB - shirely a joke) https://web.archive.org/web/20070102.../en/mtinst.exe The virustotal scan suggests that might be installing itself as a multiboot. I need to know this, to know whether it's a good idea for me to click this or not :-) What it might be doing, is adding an item to the boot menu, for a boot-time memory test. https://www.virustotal.com/#/file/ce...0ffbe/behavior In any case, that's one way to do it. Try it out, tell me what happened or something :-) I *have* run the Windows memory tester on a later Windows, but I cannot recollect right now what menu I saw it in. It's possible it was testing while Windows was running, and it didn't test every memory location (the 300MB of locations Windows is using). ******* This one, you combine it with a blank floppy. It tests every memory location, except "BIOS reserved" locations totaling around 1MB or so. If it were to write to BIOS locations, some BIOS call might crash later. There is actually a table the BIOS presents, of locations "you must not touch". http://www.memtest.org The downloads are half way down that web page. For example, the floppy in front of me, is this one. 274,506 bytes. http://www.memtest.org/download/4.10....10.floppy.zip The contents of the ZIP a memtestp.bin install64.bat install.bat ---- insert blank floppy, run this one in Command Prompt README.txt dd.exe rawrite.exe What the file set does, is a sector-by-sector transfer of "memtestp.bin" to the sectors of the floppy. When finished, if you try to list the floppy "there is nothing on it". There is no file system on the floppy. You insert the floppy later if you want and boot from it. Once the 640x480 screen appears, the floppy contents are stored in memory, so you can pop the floppy out and put it away somewhere. The "memtestp.bin" is like an OS and when the BIOS hands off control, that program runs the whole machine and does the memory test. After it has completed one pass, press the esc key to exit and boot the OS again. You can stop the test at any time by pressing that key. If it finds errors, they're printed in the middle of the screen. If there are too many errors, the error list will scroll. When I had one completely dead memory chip, it scrolled... a lot. Paul |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Corrupted COMCTL32.DLL | Herbert Chan | General | 2 | January 15th 06 04:11 AM |
Corrupted Winsock | jfs7 | Software & Applications | 1 | October 4th 04 06:16 AM |
Corrupted directories | lanman | General | 4 | September 7th 04 05:56 PM |
corrupted files ?? | bkelran | General | 2 | July 7th 04 07:25 AM |
corrupted file | [email protected] | Disk Drives | 1 | July 1st 04 12:07 AM |