Updating BIOS, speed up your PC?

[pendrachken]

Supercooling with liquid nitrogen would be more effective.

Till you hit coldbugs and can't boot / degrade performance due to electron flow restrictions in the logic gates ( usually happens somewhere in the 0 to -20C range ). You have to really push the CPU to put out enough heat to keep everything all nice and happy in the center of the CPU while still taking the heat away with your LN2 chiller.

[bester69]

This thread has drift into the bushes but I'll add my two cents nonetheless.

Updating BIOS, in my experience, usually makes sense in the therm of getting more stability from the system or adding functionalities to it, like different model CPU's or RAM. Updating the BIOS, per se, should not bring any significant boost to the system but I can come up with an exception: I owned an ASUS motherboard that received several BIOS updates in a very short time and besides bug solving, there was nothing of great relevance listed until there came an update than review the BUS protocol, opening more bandwidth for the system to function.

At the time I was a Window$ user and the thing did really had some degree of impact on the system. It took less time to boot up and ran smoother. Before the update, I used to have the system choke just because. But then again, the last Time I used Window$ the damn thing choked on itself just because... Never mind that last remark.

Want real hard evidence? Benchmark it!

Here It's.

Many poeple talk just for talking, they really dont know much; As you can see here, we can finally conclude updating BIOS can sometimes boost up speed.
I said before im consider myseld kind of good in psychophysics, and i could see Windows 10, loggin session faster once i updated my bios.

[bester69]

I found an instructional video for you, bester: http://9gag.com/gag/aYeLQ9w/how-to-clean-your-laptop

Yes Stev, but she forgot to cover screen with a bag or something, furthermore you need something like a dryer program washdiscs, if not there is a risk electronic circuits screw themselves, perhpas if she used an hair dryier could be more secure.


thanks anyway.

[phenest]

As you can see here, we can finally conclude updating BIOS can sometimes boost up speed.

No. That is your conclusion. And with no facts to support your latest BIOS update because your haven't mentioned the contents of the changelog where it shows that was the point of the update.

I said before im consider myseld kind of good in psychophysics, and i could see Windows 10, loggin session faster once i updated my bios.

Mostly a psychological and subjective observation. Unless, or course, your observations are coming from the timestamps in a log somewhere and not through perception.

Updating BIOS, in my experience, usually makes sense in the therm of getting more stability from the system or adding functionalities to it, like different model CPU's or RAM.

Remember though, the BIOS is only there so the computer can POST. If you have a modern OS, the BIOS is ignored mostly as the kernel communicates with the hardware directly. And BIOS updates never add functionality. It can't because it's only software. Hardware does what hardware does and no software can change that. Having said that, the BIOS can restrict hardware, and a BIOS update simply reveals it, but it was always there.

I'm hoping I can hack the BIOS on my Pentium 4 system to reveal the Virtualization Technology (VT-x) of the CPU.

[bester69]

Mostly a psychological and subjective observation. Unless, or course, your observations are coming from the timestamps in a log somewhere and not through perception.

I think i can trust in my perception, it that wasn't the case, i didnt feel any difference. I also can do it with most of the kernels, i see if they're performing better, worse or around the same (I suppose if you have powerfull computer it's not same easy to do it).

So then, why am i going to kill my brains with testing logs?

[qyron]

Updating BIOS, in my experience, usually makes sense in the therm of getting more stability from the system or adding functionalities to it, like different model CPU's or RAM.

Okay, I'll admit to badly phrasing my message. Can I get a 10% off for not being English-speaking native?

Remember though, the BIOS is only there so the computer can POST. If you have a modern OS, the BIOS is ignored mostly as the kernel communicates with the hardware directly. And BIOS updates never add functionality. It can't because it's only software. Hardware does what hardware does and no software can change that. Having said that, the BIOS can restrict hardware, and a BIOS update simply reveals it, but it was always there.

True. But as you stated, updates can unlock or improve the output of the hardware present and if BIOS is locking the functions of the hardware, I severely doubt the kernel will be able to superimpose the motherboard software. For some reason you have to configure (at least I do but I'm not a tech wizard or guru) at BIOS level some functions to enable OS to monitor such functions.

Simple examples that come to my mind is hardware temperature (motherboard, CPU and GPU) and cooler fan speed (CPU, GPU and case) and energy management, to cite those I have personal experience with. RAID modes may also be another, but I am yet to experiment with this.
My humble motherboard, out of the box, with a Debian vanilla install, negated such probing - the system itself warned me I had to activate such functions at BIOS level - and kept the CPU at full power, all the time. After a little tweaking and messing around, I had the machine telling me what I wanted to know and running according to system load requirements.

But this isn't me coming to the line defending bester; to actually measure system response, you have to benchmark the thing. Before and after BIOS update. And from this I won't budge.

[phenest]

Mostly a psychological and subjective observation. Unless, or course, your observations are coming from the timestamps in a log somewhere and not through perception.

I think i can trust in my perception, it that wasn't the case, i didnt feel any difference. I also can do it with most of the kernels, i see if they're performing better, worse or around the same (I suppose if you have powerfull computer it's not same easy to do it).

So then, why am i going to kill my brains with testing logs?

And we all know you hate reading.

[phenest]

Updating BIOS, in my experience, usually makes sense in the therm of getting more stability from the system or adding functionalities to it, like different model CPU's or RAM.

Okay, I'll admit to badly phrasing my message. Can I get a 10% off for not being English-speaking native?

Have 10% off.

Remember though, the BIOS is only there so the computer can POST. If you have a modern OS, the BIOS is ignored mostly as the kernel communicates with the hardware directly. And BIOS updates never add functionality. It can't because it's only software. Hardware does what hardware does and no software can change that. Having said that, the BIOS can restrict hardware, and a BIOS update simply reveals it, but it was always there.

True. But as you stated, updates can unlock or improve the output of the hardware present and if BIOS is locking the functions of the hardware, I severely doubt the kernel will be able to superimpose the motherboard software. For some reason you have to configure (at least I do but I'm not a tech wizard or guru) at BIOS level some functions to enable OS to monitor such functions.

When I said unlock, I just meant that a setting is made available to you, but that doesn't mean an extra feature has been added. Before it was made available, it was still there, but maybe fixed to a default setting or maybe disabled. But remember, your BIOS update may give you new settings for example, but if the hardware is not capable, then that setting will do nothing.

Simple examples that come to my mind is hardware temperature (motherboard, CPU and GPU) and cooler fan speed (CPU, GPU and case) and energy management, to cite those I have personal experience with. RAID modes may also be another, but I am yet to experiment with this.
My humble motherboard, out of the box, with a Debian vanilla install, negated such probing - the system itself warned me I had to activate such functions at BIOS level - and kept the CPU at full power, all the time. After a little tweaking and messing around, I had the machine telling me what I wanted to know and running according to system load requirements.

But this isn't me coming to the line defending bester; to actually measure system response, you have to benchmark the thing. Before and after BIOS update. And from this I won't budge.

The BIOS can switch features on and off, that is true, and the kernel can do nothing about it. But providing a feature is enabled, the kernel can communicate with it directly. Imagine that your motherboard has no further BIOS updates available. If the BIOS has bugs, then the OS can be potentially unstable. The kernel communicates directly where it can to help prevent instabilities. I have SSD's connected to an old tower that the BIOS cannot detect and so they don't appear in the BIOS. But Debian can see them just fine and use them without any problem.

[stevepusser]

In short, an upgrade may or may not provide a speed boost. It might fix a bug that affected you, or may introduce new ones that affect you when the old version had no issues for you. In the worst case scenario, it can convert your machine into a doorstop.

[qyron]

Have 10% off.

Thanks!

When I said unlock, I just meant that a setting is made available to you, but that doesn't mean an extra feature has been added. Before it was made available, it was still there, but maybe fixed to a default setting or maybe disabled. But remember, your BIOS update may give you new settings for example, but if the hardware is not capable, then that setting will do nothing.

The hardware must comply to a defined function, absolutely, but what can we say if a hardware is listed to support USB1.0 and after an update the hardware responded to USB2.0?
I had this happening in a machine (my very first computer) where the motherboard was listed as USB1.0 only capable and after an update it became 2.0 compliant. The hardware was there, definitely, but the BIOS software upgrade putting down the orders in it set it free. Could we consider this an added feature? An average user would because hardware issues tend to slip unnoticed. Comes down to the recurring joke of downloading more RAM....

The BIOS can switch features on and off, that is true, and the kernel can do nothing about it. But providing a feature is enabled, the kernel can communicate with it directly. Imagine that your motherboard has no further BIOS updates available. If the BIOS has bugs, then the OS can be potentially unstable. The kernel communicates directly where it can to help prevent instabilities. I have SSD's connected to an old tower that the BIOS cannot detect and so they don't appear in the BIOS. But Debian can see them just fine and use them without any problem.

When something like that happens, I prefer to say its the system overcoming BIOS handicaps.

The most similar thing I can recall happening to me was having a motherboard that required drivers installation for the network card to work. Under XP it was a nightmare to get that cursed motherboard to finish an installation without having a fit. First time I got it running Linux I was bracing myself to get an awful lot of trouble to configure the network but the kernel found it and loaded it. Again, it was a later BIOS update that solved that issue but the board kicked the bucket shortly after.

/end

I've had enough of this but it's always good to get feedback in these forums; I'm always bound to learn something new.

[phenest]

The hardware must comply to a defined function, absolutely, but what can we say if a hardware is listed to support USB1.0 and after an update the hardware responded to USB2.0?
I had this happening in a machine (my very first computer) where the motherboard was listed as USB1.0 only capable and after an update it became 2.0 compliant. The hardware was there, definitely, but the BIOS software upgrade putting down the orders in it set it free. Could we consider this an added feature? An average user would because hardware issues tend to slip unnoticed. Comes down to the recurring joke of downloading more RAM....

Did you get USB 2.0 speeds? Something being compliant is different to something actually giving USB 2.0 speeds. What you're suggesting isn't possible. Unless it was USB 2.0 in the first place.

The BIOS can switch features on and off, that is true, and the kernel can do nothing about it. But providing a feature is enabled, the kernel can communicate with it directly. Imagine that your motherboard has no further BIOS updates available. If the BIOS has bugs, then the OS can be potentially unstable. The kernel communicates directly where it can to help prevent instabilities. I have SSD's connected to an old tower that the BIOS cannot detect and so they don't appear in the BIOS. But Debian can see them just fine and use them without any problem.

When something like that happens, I prefer to say its the system overcoming BIOS handicaps.

That would not be possible to implement workarounds for buggy BIOSes. There are too many for kernel developers to take in to account, let alone the ones they don't know about. And how would you know if it's buggy? That would require extensive, time consuming testing. It is easier to simply circumvent it altogether and communicate with the hardware directly. It removes an unnecessary layer.

[qyron]

I got. The hardware was just "muffled" by the software lack.

The motherboard was an ASUS that I later read some reviews stating it had been one of those models that came to the market too soon, to meet the demands of the new arriving CPU's (I was running a P4 Prescott series then) and was loaded with bugs at BIOS level.

Even BUS speed (marketed ate 800MHz) was originally much lower and the machine "choked" very often; I got my fair share of BSOD and cold reboots with that machine. Every single upgrade to the BIOS was a nightmare to do (it was so buggy that even the flash BIOS function was prone to crashes) and it received so many it became almost a chore checking for it.

After a great number of updates, it reached a maturity state where the systems was actually functional.