OT Hints on replacing computer CMOS battery?

[bob_villa]

No sign of the OP...but we can sure BS on this stuff...

[Mayayana]

| If the Cmos battery is DEAD the settings are already either totally | lost or compromized.

Yes, but often people are replacing a failing battery, indicated by something like not keeping accurate time. I didn't see any reason to assume the battery is completely dead.

[trader_4]

I know what CMOS RAM is, it's 40 year old technology. That's why for a long time, the settings have been stored in Flash memory now, because it doesn't require battery backup. As I said, AFAIK, the only thing that requires a battery is the clock.

[bob_villa]

As per usual, you contradict yourself, "...for a long time"-"AFAIK"!

[bob_villa]

I know you will nitpick this to death...so I offer-up this from Tom's Hardware:

[Terry Coombs]

I have a (dead now) Tosh laptop that had a dead CMOS battery . The only thing that was ever affected is the clock if I let it drain the (main) battery .

[trader_4]

No, as usual, you're making things up. Nothing I said there is contradictory. Think about this, Flash has been available for over two decades, EEPROM even longer. They are non-volatile inexpensive and don't require a battery. Flash is even used for the Bios, so it would be very easy to use a tiny section of that to also store the setup data, it would essentially be free.

[Mayayana]

| > Actually, I'd be curious, too. It surprised me | > that the defaults didn't work. And it was an | > OEM PC (Dell), not home made. But I don't | > remember the details now. | | How can this be...considering the computer guru that you are? Was this before the IT expertise? | :) I seem to learn a new bit every day. Yesterday I was trying to update DownloadHelper for Firefox. It turned out I had to update Firefox itself. Now I have to figure out how to get back my beloved Cape Neddick lighthouse "throbber". (Apparently the Mozilla people decided that page-load indicators are no longer fashionable.) And it turns out the new DownloadHelper is poorly designed, anyway. I'm also finding weird odds and ends. Geo.enabled in about:config. What the?!! It turns out a function to tell websites my location is enabled by default! Should I go back to the older, less annoying FF 24 that I had? Whatever happens, it will have been at least 2 or 3 hours wasted to deal with all of these unnecessary changes to already overproduced software.

Computers and OSs are so complex. On top of that, the tech industry depends on constant, often frivolous obsolescence and manufactured abstruseness. It never ends. (I read yesterday that Microsoft's phone business is dangerously close to dying... which could threaten their idea for Metro apps sold through their own store to compete with Apple... which could be ruinous for the few developers who were sucker enough to put all of their eggs in the Metro basket: A few years of re-education in a new technology that may end up entirely useless before it gets going. So now they'll need to retrain yet again to become expert in the language du jour, which may or may not be around next year. And actually, it's not even as bad as it looks because there was never any money in phone apps to begin with. Apps are the goldrush. Google, Apple and Microsoft have figured out how to cut their risks by selling shovels cheap and taking a 30% cut of any gold found.)

The Dell belonged to my elderly father. He called Dell and apparently got suckerpunched by the salesman. A 90 year old man who does email ended up with an expensive and powerful, dual-CPU computer. Something like 6,600 GHz. He's since stopped having his own computer, but at the time he was a daring explorer. I don't know how he got into the BIOS, but he didn't hesitate to fiddle around in there.

Personally I'm ambivalent about Dell. I have a couple that were given to me. They seem OK, and the support for drivers is good. But I would never buy one myself. The drivers are all custom packaged. One has to get them from Dell. And the hardware is often customized. The first time I ran into a Dell was in salvaging from a Win95 box that had an unusually good graphics card, which I decided to save. But when I opened it up, it turned out there wasn't a card. There was a sort of stick, with a square plug that went into a motherboard socket I'd never seen before. It was useless for reuse in any normal Windows computer. Emachines was the same way. The hardware was custom. The power supplies were too weak, but they were so small there was no way to fit a normal one in the case. Things like that can result in unexpected costs. But with eMachines at least the product was dirt cheap.

So it wouldn't seem *really* odd to me that an OEM box doesn't have default BIOS settings. But with Dell nothing surprises me. They built their business by catering to business needs, being very responsive with support to people who were happy to pay through the nose and almost certainly would never open the case. Hardware lock-in is a good strategy for Dell. And once the IT dept learns the Dell way of doing things they have no reason to complain.

[trader_4]

sumption,3066-10.html

You have Toms, but how about American Megatrends, (AMI) which is a major supplier of BIOS technology to the PC industry. They should know where the BIOS setup is actually stored today:

Historically, the BIOS and its settings were stored into CMOS (complementar y metal-oxide semiconductor) and was commonly referred to as the CMOS Setup . The CMOS and Real-Time Clock (RTC) required an electric charge to maintai n their settings. This was typically performed by an on-board battery. As t he battery aged, the electrical charge that maintained the CMOS settings di minished. The BIOS and RTC would then revert to its default settings, resul ting in, "Press F1 to enter the CMOS setup."

Note: The terms CMOS setup and BIOS setup were frequently used interchangea bly in the 1990s and far into the 2000s.

Beginning in the late 1990s, main board manufacturers started to store the BIOS into flash memory. There are two benefits for doing this.

*Flash memory does not require a charge to maintain its settings. *The size of the BIOS can be increased.

Note: Although the BIOS settings are stored in Flash memory, a battery is r equired to maintain the main board's RTC settings."

Pay close attention to that last sentence. It's *exactly* what I said. Now, to be sure, I can't say every MB implementation is done that way. Which is why I said "AFAIK". And to be fair, from Googling I saw other sources saying the setup is still stored in CMOS. A big part of the problem is that it's all referred to, based on past history, as CMOS setup, regardless of where it's actually stored.

Finally, we have the post Mr. Combs just made, where he says he has a Toshiba laptop with a dead battery and the only thing it ever lost was the clock. If the BIOS setup is in CMOS, how do you explain that?

[bob_villa]

So now you're talking hypothetically, "...so it would be very easy...", but not factual! And you're saying flash, when in fact they use NVRAM...try not to guess at things!

[trader_4]

Take a look at my post, where I quoted AMI which is an actual major supplier of BIOS and similar technology to the PC industry. They make the software that stores the BIOS data, they certainly know where it's stored. That's about as "factual" as you can get. Here, they directly address the issue:

"Note: Although the BIOS settings are stored in Flash memory, a battery is required to maintain the main board's RTC settings."

Note that is consistent with what Mr. Coombs just posted. He has a laptop with a dead CMOS battery and he says it works just fine, the only thing he loses is the date/time.

[bob_villa]

The 2nd time you referred to your cohort to back-up your blathering...his CMOS could have been marginal, holding settings but not RTC.

[trader_4]

es battery backup.

", but not factual! And you're saying flash, when in fact they use NVRAM... try not to guess at things!

is required to maintain the main board's RTC settings."

CMOS could have been marginal, holding settings but not RTC.

You act like Mr. Coombs is all that I posted and keep avoiding the 12ft elephant in the room:

"Note: Although the BIOS settings are stored in Flash memory, a battery is required to maintain the main board's RTC settings."

That's from AMI, a major supplier of BIOS's to the PC industry. They supply the software that manages the BIOS settings. I guess they're my "cohort" too....

Or how about these:

"Motherboard battery

Note that desktop and laptop PCs now store the EFI settings in non-volatile permanent memory that retains information when the PC is switched off, whi ch means that the EFI does not require to make use of a battery supplied by the PC's motherboard to retain its data. However, motherboards still have a battery because the Real Time Clock (RTC) that keeps date and time inform ation has to be powered. "

Nonvolatile BIOS memory

From Wikipedia, the free encyclopedia

"Nonvolatile BIOS memory refers to a small memory on PC motherboards that i s used to store BIOS settings. It was traditionally called CMOS RAM because it used a volatile, low-power complementary metal-oxide-semiconductor (CMO S) SRAM (such as the Motorola MC146818 or similar) powered by a small batte ry when system power was off (called the CMOS battery).

The term remains in wide use but it has grown into a misnomer: nonvolatile storage in contemporary computers is often in EEPROM or flash memory (like the BIOS code itself); the remaining usage for the battery is then to keep the real-time clock (RTC) going."

[Scott Lurndal]

While this is not inaccurate, the CMOS is still used when booting legacy operating systems (e.g. win2k) on newer model desktop computers.

Note that this is also specific to intel/amd x86 systems. ARM64 servers will store all configuration information using SPI or eMMC non-volatile memory and don't have the Intel CMOS RTC.

The default values in his CMOS memory are presumably sufficient for operation. He may not be getting optimal performance from his system, however.

[Scott Lurndal]

Easy - the default values are sufficient for his level of operations.

[bob_villa]

e storage in contemporary computers is often in EEPROM or flash memory (lik e the BIOS code itself); the remaining usage for the battery is then to kee p the real-time clock (RTC) going."

The 12' elephant in the room is your ego...the point you are totally ignori ng is your use of the word flash instead of NVRAM...get with the "program"!

[trader_4]

I'm fine with all except the last part. After what you posted, isn't the alternative and more likely situation that Mr. Coombs PC has the BIOS setup actually stored in Flash or EEPROM and hence it doesn't lose them even though it has a dead battery?

[bob_villa]

Keep grasping as stars...that's what dreamers do.

[trader_4]

ile storage in contemporary computers is often in EEPROM or flash memory (l ike the BIOS code itself); the remaining usage for the battery is then to k eep the real-time clock (RTC) going."

ring is your use of the word flash instead of NVRAM...get with the "program "!

Apparently you can't read. AMI clearly says EEPROM or Flash is often used. Neither of those are NVRAM. I said Flash. You claimed it was stored only in CMOS RAM and required a battery or the setup would be lost. Clearly AMI says otherwise. Feel free to dig yourself deeper. You're in way over your head. head.

[trader_4]

Keep lying instead of admitting you're wrong. Again, from AMI, a leading supplier of BIOSs:

"Note: Although the BIOS settings are stored in Flash memory, a battery is required to maintain the main board's RTC settings."

Exactly what I said.....