Also security patches often also makes things slower. The variant of Heartbeat that was vulnerable to heartbleed was _definitely_ faster then the patched variant. Because the patch was a extra check.

@roycsinclair

10 ай бұрын

And the checking for viruses, while new viruses keep getting created the virus software still has to check for ALL of the older viruses just in case someone loads an ancient floppy disk containing one of them again.

@animepussy8356

10 ай бұрын

I disabled the patch like a chad to gain 3 extra FPS i am smart

@BillAnt

10 ай бұрын

My 2012 Asus ROG laptop running Win7 is running just as fast if not faster than on day one. Simply because I maintain it carefully and don't install every piece of crap software on it. No more official security patches by Microsoft, but no problem, I get free critical micro-patches from 0patch. ;)

@benbaselet2026

10 ай бұрын

Indeed, new microcode for CPUs or BIOS and OS level workarounds for vulnerabilities or disabling features that are vulnerable can cause really bad loss of performance. If you have stuff that's not online.. maybe it's better not to pach or update things if everything is fine.

@TheRedneckAtheist

10 ай бұрын

@@benbaselet2026 Remember XP service packs? Pretty much everybody with even the most modest comfort level in command line figured out how to slipstream the packs into new install media. The service packs were great on a clean install but slow and/or flaky as fuck if you just let windows update take care of it.

In my opinion Wirth's / Andy and Bill's law also plays a major role. When I look at the code that my colleagues write performance seems to be the last thing on their mind because modern hardware can often compensate for bad code.

@bobthemagicmoose

10 ай бұрын

Though I agree with the cause, I disagree with the characterization. The code now is more abstract and generalized which makes it easier to create, iterate, and modify. True, you could write it all in assembly and have insane performance gains but it would cost a lot more and would be abandoned the moment a new system were released. So it may not be performant, but it is more readable, reliable, and adaptable.

@Fiyaaaahh

10 ай бұрын

@@bobthemagicmoose I'm all for abstractions and high level languages, but _within_ that high level language you should make a basic effort to not write code that sucks ass. E.g. it blows my mind how commonly linear search is used in modern applications.

@warrenpuckett4203

10 ай бұрын

I finally figure out a formula. Get 3 times as processor than you need to run all the software. Twice as much memory. 4 times as much hard drive. Then under clock and under volt. When it gets too slow. set it back to factory specs. Plus a bit more overclock You will have two years to find the nextest, greatest and bestest PC build after that. Maybe.

I find even web browsing gradually gets slower as the browsers are updated and the web technology changes. It really does feel like we are not really getting any more for the extra processing that is required. In fact, my view is we are getting less functionality.

@roycsinclair

10 ай бұрын

We are getting new functionality, the problem is that it's not NEEDED, it's just bloat now. A reason to make new releases. Microsoft Word for example was a perfected bit of software when they added the real time spell check, every change since then has been to excuse the release of a new version and not to add a useful new feature that the whole world can use.

@MitzaMaxwell

10 ай бұрын

If you're using Microsoft Edge, that's probably true. Otherwise, it is because of more and more advertisements and cookies that spy on your shopping habits and other things. Which one do you use?

@roycsinclair

10 ай бұрын

@@MitzaMaxwell Edge is just Chrome with MS spying added to the already excessive Google spying so I stay away from anything based on Chrome. That's causing problems sometimes because lazy web developers who write to Chrome only instead of standards. Again, there's not much that HTML does today that it didn't do well enough ten years ago but they keep changing the standard and yet have never fixed the things that make spying easy.

@MitzaMaxwell

10 ай бұрын

@roycsinclair Thanks for your comment. Which browser do you recommend? Currently I use Firefox with some privacy add-ons.

@ThePortuguesePlayer

10 ай бұрын

I have a 2 GB Intel Atom machine where I usually could open 5 chromium tabs in, then later it got to 4, then 3 and now it's at 2. Nothing changed on the computer or its use, the pages just seem to use a lot more RAM now and browsing in general is getting slower.

In the Windows XP days, my solution to the problem was simply wiping out the harddrive about once every year or 2 and simply reinstalling windows. That always worked like a charm, fresh and clean. I still do that now with my windows 10 and 11 machines, works like a charm. Linux I dont have to worry about because, well...Linux

@Patrick-857

10 ай бұрын

Windows is the world's most successful computer virus.

Windows 95/98 suffered from what some folks called "Registry Rot"/ The more software you installed and uninstalled, the more unstable and slow Windows became. And why does windows volume shoot up to 100% every time you install new audio hardware??

Intel has had some security problems that required patches that have been devastating to their speed. Downfall~Meltdown~Rowhammer AMD has a bunch too. The patched do slow down your computer by a good amount too. 30%?

@MitzaMaxwell

10 ай бұрын

Are you absolutely sure that it is Intel and AMD's fault. I am more inclined to think that it is Microsoft that has a vulnerable operating system. Please explain why it is a CPU problem and not the operating system? Thanks in advance.

@SeanLavery

10 ай бұрын

@@MitzaMaxwell It's hardware level patches that are fixing actual hardware security issues. The OS is secure, but the hardware has issues that can be exploited to allow software to either steal information that should be inaccessible, or to cause something to allow remote code execution. Many of the CPU problems are fixed by turning off certain accelerators, which means the same exact code running on the CPU will now run slower, but it's safer. And yes, both AMD and intel have said its their fault and released driver/bios/CPU fixes that adjust how the CPU is allowed to work. These bugs exist on the hardware level, and no amount of changing your OS can fix them. Exploitable on windows/Linux/others.

@SeanBZA

10 ай бұрын

@@SeanLavery Yes hardware faults, and no way to fix at all short of redesigning the chips themselves, which is why you have different step levels, each one fixing some of the hardware bugs, and sometimes with the only fix being "do not use this function", and the microcode being updated to disable it entirely. most relate to how speculation works, as the processor will use spare clock cycles at the fastest speed to guess the result of any branch, flag check or arithmetic function, based on past results, and start to execute the next instructions, so that if the result is right, it already has done the work. Thus you can detect wrong results from the time it takes to do this, as a wrong result means the whole cache now has to roll back the core to the now proved correct result, and dump all the execution that took place, and this time is easy to monitor from context switches, so you can know fairly confidently the result of any branch or flag. The same for RAM, cells adjacent bleed charge to each other, and you can use this to reliably flip a particular bit by thrashing the bits around it to flip it, allowing you to change your program priority and owner, so allowing it to read memory it should be segfaulted out of. Or use the hammering to flip your memory bit, to reflect the values of the adjacent cell, to get that info out.

@MitzaMaxwell

10 ай бұрын

​@@SeanLavery It is not good. The other day, a Danish web provider lost everything, including the customers' data. And the backup is also not secure because the malicious software has also been stored on the backup. How can you secure your systems and data?

@sputukgmail

10 ай бұрын

It is perhaps worth noting - the side channel attacks like Spectre and Meltdown are probably irrelevant on personal computers in practice, and it is a reasonable choice to not include the firmware changes to ‘fix’ then and hence avoid the loss of performance from the loss of some of the speculative execution features. The reason - to exploit these vulnerabilities, and attacker would already need to have significantly compromised a device in some way in order to perform the exploit - and if they can do that, there will usually be much easier and direct ways to leverage that access to gain complete control of a system. (The Browser is a specific case where this may not be true, but most browsers now include deliberate noise in timing functions to render such attacks within their environment, ineffective). Such side channel attacks on the CPU are much more of an issue in cloud computing environments where they could be used to leak information from one tenant to another, or from the host into a tenant environment, potentially allowing an attacker to break out of the virtual sandbox and compromise the cloud host, or just leak information from other tenants. But home users, might consider not applying the firmware mitigations for these attacks. That said - since Windows and other OS including Linux and MacOS have all been patched to include updated firmware for the processors, and also re-written to include ‘defensive coding’ against side channel attacks (with various degrees of success), the degree to which a user could ‘opt out’ of these mitigations impacting performance is probably not that practical now I think about it…

OSs are not supposed to be giant octopuses, with every tentacle wrapping around your computer, like accessing the disc every second. Everybody wants to use all the resources to show off with huge graphical interfaces! I remember In the nineties, mouse drivers used to be a few Kb in size, now they are tens of megabytes. And don’t get me started on all the massive printer installations I’ve used over the years. Computers don’t get slower, OSs get more rubbish. It’s all designed to make you buy more hardware.😃

@pharoah327

10 ай бұрын

​@@baraka629with Linux you have different problems, not less problems, just different problems.

@MitzaMaxwell

10 ай бұрын

@@pharoah327 Yes it's true. But there are no recurring problems, once they're fixed they're fixed. I got tired of Windows because it was the same recurring problems over and over again, and when you got a newer version of Windows, the old "Fixes" didn't work anymore, So when you have worked professionally with these "fixes" on Windows in over twenty years, there comes a day when you just don't want to Fix it anymore. The problem is that it is difficult to buy a computer with Linux on (at a generous price). Here I am writing this on my old computer instead of using my new Windows 11 laptop. I turned it on and said oh no not that again. Several updates and reboots over and over again, before I could try the "new wonder". Windows is what it is, Linux is what you want it to be.

I wrote some ATA and SCSI drivers. Sector read performance becomes terrible once the firmware starts using spare sectors. Also I've noticed tons of lazy programmers don't cleanup temp files when they are done, which slows things down. I wish windows had a "cleared on boot" temp folder I could use.

Too many comments to check if someone mentioned it, but some reasons for gradual decrease in performance are the ever increasing size of log files that don't get rotated, software updates which include new features thus taking longer to load, and in some cases disk fragmentation. In most cases the issue is related to increased disk I/O as that is noticeable, but installing more and more programs over time which keep some background processes running can also slow down the computer enough to get a sluggish user experience but it's quite difficult to end up with 100% CPU load that way so it's usually virus scans, search indexing, and poorly written software that end up maxing out the CPU.

You should look into issues with SSDs too - with multi level storage of bits - where when empty, the storage uses 1 bit per ‘storage cell’ and can read and write quickly, but once the drive reaches half capacity, it starts storing multiple bits in the same cell (not a simple on/off but varying voltages to correspond to 4 different binary values), and that makes reading and writing those bits slower. (Recent episode of Security Now Steve Gibson dives into this) We used to have similar issues with spinning disks getting slower over time - sectors failing and needing multiple read/write attempts (below the OS level) as well as fragmentation leading to slower overall performance. There are actually quite a few ways in which the hardware itself can lose performance - another, you could have RAM that fails, and instead of having all the memory available, the PC might only ‘see’ some of it and be having to cope with it. Yes, usually this would be detected, but RAM can fail in ways where the PC simply doesn’t see it plugged in, so fails silently to the user.

@minecraftsabotage

10 ай бұрын

I have also had issues with junky SSD controller slowing down as the drive wears out. At least that is what I think happened. I never confirmed the root cause of the issue, but the SSD had become ridiculously slow (with very low IOPs), and replacing it made the problem disappear.

As someone that lived through the dark days before SSD's I rarely complain about slow speed. If someone told you they cried with joy the first time they ever booted up their machine with an ssd they're probably not joking. When you turned on your computer was usually the time you went to get yourself some coffee or use the restroom. There was debates on if it was okay to just leave the computers running but honestly the debate didn't matter because we'd just leave them on so we wouldn't have to wait turning them back on again.

If you need episode ideas, I'm curious how basic is turned in to machine code or just "stuff that works". It's related to the early computers you have, it all seems like magic how a small machine runs it. When the memory is kilobytes how do you interpret in to useful code.

This is a very simplified explanation. Honestly we all know that before NTFS became the standard, a defrag could make a huge difference, not least on HDDs of the time, as opposed to the SDDs that are the norm today. That's the hardware side... software side is also guilty. I do microcontroller programming, and today SoCs have vast amounts of resources compared to my first PC, an Amiga 500. Having an environment where you don't need to count memory in kb and clocks in kHz makes a programmer lazy. Those that still do the counting are usually hobbyists that challenge themselves. Back in the day, a programmer that could save 1 out of 3 bytes of memory, doing clever bitwise operations, was king on a system with 256kb of RAM. Today it's more like "what does 100mb matter if we can save 30% on development time and 50% on maintaining the code?" This slack approach to memory, storage and CPU cycles does have an impact on whole systems over time. Remember when it was possible to squeeze an emergency/recovery linux onto a single 1.44mb floppy? So you could try to repair a broken partition table, or remove a faulty entry in the fstab file. The kernel alone can't fit on a floppy today.

As computers become more powerful, later software versions become slower to use the extra power. It's the Parkinson's Law of computers: Software expands to consume the available cycles.

@GoingtoHecq

10 ай бұрын

It basically follows the same rules as cars and roads.

@shadowbenshorebatten2542

10 ай бұрын

What a fucked up name for a computer law 😂

@macethorns1168

10 ай бұрын

Correction: As computers become more powerful, later WINDOWS software versions become slower. FTFY

@DaveHoskinsCG

10 ай бұрын

@@macethorns1168 don’t forget Apple computers

@DaveHoskinsCG

10 ай бұрын

Human waste expands to fill the space provided for it.

Disk retries increase over time, mostly due to physical wear of the disk coating, or flash cells but partially due to aging of the circuits that deal with the extremely low voltages. NIC retransmits increase over time due to signal quality loss from aging of the power components. Moisture ingress into the PCB will mess up impedance matching, causing echos of data values to increasingly reflect around, which can cause PCIe retries.

@asm_nop

10 ай бұрын

Imagine all the older folks confused and complaining because the no-name QLC DRAM-less SATA SSD they bought on Amazon 3-5 years ago just isn't cutting it as a boot disk anymore.

@perwestermark8920

10 ай бұрын

I have not seen any indications of any disk read retries or increased NIC retransmissions as hardware gets older. The counters for resends are normally supervised if running supervision software so it's quickly flagged. For networks, lots of people end up with issues from cheap routers what has a heatsink fitted with dual-sided tape. The heatsink comes lose or slides slowly to the side. This makes the router overheat. And that can make the router either hang or result in network retransmissions. This is also a common reason to ask "have you tried to power off/on your router". But both HDD and SSD are affected by fragmentation even if it's a much smaller issue for SSD. And SSD are also affected by huge slowdowns if almost full if the drive doesn't have a pool of spare sectors in an overprovisioning buffer. Then it needs to copy data from flash, erase flash blocks, wait for the erase to finish - just to be able to write back any changes. For an SSD with overprovisioning, even a full SSD can take a pre-erased flash block from the overprovisioning pool. So no need to wait for the slow erase cycle. The just released block can have erase started and then be moved to the overprovisioning pool.

@perwestermark8920

10 ай бұрын

​@@asm_nopI fail to see how "older folks" is relevant. Much hardware you buy are designed by "older folks" - often called engineers. And there are huge amounts of young people who understands zilch about how computers actually works. Because they got born around the equipment instead of reading technical publications about the new and magic inventions soon to show up around us - and including actual technical descriptions.

@douggale5962

10 ай бұрын

@@perwestermark8920 The drive is a blatant liar about them, if it is not an enterprise drive. Enterprise drives admit how bad they are immediately. Consumer drives desperately rely on ECC, it pushes its luck all the way, to try to make it to the end of the warranty without running out of spares. The OS can't tell a retry is occurring, it just appears to take longer to complete a particular command. Some drives have downright malicious firmware, such that they will never relocate a block unless a write occurs to said block. This can cause a stable hardly readable block to take down the whole system, eventually.

@perwestermark8920

10 ай бұрын

@@douggale5962 Good supervision software tends to see such retries. And they really are very uncommon. Not really something that makes normal users suffer slowdowns relevant to this video scope. Having a HDD perform regular SMART scans is relevant - because that's when the HDD gets a chance to decide if a sector is weak and should be refreshed - and possibly relocated. If never running SMART long tests, the people can suffer issues because it may take years since a sector was last checked. And the disk can't know if it needs to rely on ECC for correct recover (which normally still is instant unless trigging the disks rewrite rule to refresh the sector, i.e. forcing additional rotations) unless it's finally is asked to read that specific sector. And that can push it from a hardware ECC recovered issue into a offline uncorrectable issue. Which is a sad moment for people without backup. Quite some years ago I noted I had accumulated more than 1000 years of powered HDD time - yes, from quite a lot of disks. Multiple SMART metrics relates to wear. Like parameter 1 - raw read error rate. Or parameter 7 - seek error rate. And 197 - hardware ECC recovered. I have only seen one single HDD - out of hundreds - where that hardware ECC counter value actually started to speed up ticking. I tend to replace HDD after about 5-7 years power-on time. Not because of wear-out but because of power consumption and max number of spindles - better to replace with fewer but larger disks, unless the machine is fine with just a single disk. Buy then that's a usage case normally better handled by a silent and power-efficient SSD. One note - I make sure my HDD doesn't do silly amounts of load cycles. I.e. parking of the heads. Some HDD firmware have notorious bad default configuration where a disk can pass the expected SMART lifetime wear limit for load cycles in just months. The load cycle timeout must be set in relation to expected disk access frequency so the disk doesn't do 1000+ load cycles/day. I had one Green aka eco drive consume 40% of lifetime wear in about 3 weeks. Because the machine was brand new and I didn't had time to finish up everything on it directly from day one. So it parked the head just before every Linux disc sync operation. Very silly oops from the HDD manufacturer.

Also, your device might be underclocking to stretch the battery life long (Apple...), updates and software are tested on the most modern components, and they do not typically optimize for older systems (they had to balance developer time against the computing resources of the client devices). Some components (like hard disks, as another commentator explained) do wear out - the computer navigating around these issues can also slow down the computer. The biggest issue though is that software isn't made targeting your device.

@jannikheidemann3805

10 ай бұрын

I notice that with the KZread app for Android, it's bloated and slow.

Pretty much the first "short" that i didn't immediately stop. Way to recondition me, Dave

Hey Dave! I was just thinking.. You are pretty awesome! Have a good one.

There is a psychological aspect though to. Like using an old pc back when it was new felt fast. But as tech got better it felt slower and slower. We used to turn pcs on and leave the room as they started up. Or click a game then leave the room. And those pcs we thought were fast at the time. Today you would never put up with that so it now feels crazy slow

Thermal compound dose tend to get worse with age too.

The problem is only Windows related. My now old Linux laptop has not slowed down despite several upgrades from Linux Mint version 18 to Linux Mint 20 Cinnamon. It is now so worn that I have bought a new one not because this one is slow, but to have one ready when this one falls apart. It is already held together with duct tape. The new one is made of aluminum because it requires solid hardware to be able to last as long as the Linux software. It was the other way round as I was a Windows fan. I am no longer particularly impressed with Microsoft products and it is no longer true that Linux is for experts. Sad to have to say this because I have been a Microsoft user since MS-DOS 3.3 and am also a Microsoft product specialist in Windows NT with several diplomas.

@macethorns1168

10 ай бұрын

You are exactly and 100% correct. Every newer version of Windows is slower than the last, usually goes unnoticed because hardware is refreshed at the same time.

A problem pointed out by IBM some decades ago. OS/2 initially spread files around on disk, leaving gaps around file. As old files expanded, they could expand into empty space nearby. In Windows, files were allocated adjacent to each others. It's not clear to me that defragmenting the disk was good. You got a big lump of fully allocated space and then a big lump of unallocated space. When an existing file expanded, the next space was quite some distance away. Not a problem with SSDs I expect. I don't think software updates or new features or even new software caused much slowdown. Firefox (and probably Thunderbird) long had memory leaks which my use pattern exacerbated, but the Mozilla crowd wasn't very interested. That could bring a Pentium III to its knees.

as long as my computer can solve for prime numbers 2 billion times faster than an altair, i'm happy

Hey Dave, as someone who knows "how the sausage is made", can you explain for those of us who just see the results why it is that Windows has seemingly constant updates that take so long to apply? Even on a fairly recent (ryzen 5950x) recent system with plenty of RAM (64 gigs) and a gen4 nvme Windows takes forever to update while my Linux installs (3 separate ones, xubuntu, popos, and an arch based one) take like 5-10 minutes. It is slower in every step, from download to install to reboot to finishing install to pressuring me to use a Microsoft account. I realize you're not a current dev, but maybe you could shed some light (LED even) on what's going on

@KKLD

10 ай бұрын

Windows Update is not great at utilising more than 1 CPU core concurrently, so it can only use 1/16 of your CPU's cores.

@0LoneTech

10 ай бұрын

The short of it is: Because Microsoft don't care. They'll brag for days about how they do extensive testing, try to adjust code for every eventuality, etc, but atomic switching of udpates has been well known technology for many decades and they just don't care to do it. Compare e.g. NixOS which can absolutely have an update that takes ages but never interrupts the user. Microsoft just refuse to do that even on their consoles where the software platform is deliberately controlled. It also isn't that their update procedure is so unstructured; you can capture all of that and build a transaction, as demonstrated by STM. It's a less efficient way to do it, requiring name spaces and snapshots, but that's also old hat - and they've built those for functions like driver rollbacks (system restore points) and running multiple browser versions (by VMs). FreeBSD made file system changes in general do this with the soft-update system, again decades ago. System updating is typically I/O bound, not CPU or RAM bound (though Microsoft's malware service sure tends to eat a lot of those resources too).

Not sure ... seems like an intended feature to me. Nothing prevents better register management.

Security (and other) patches also come with a performance penalty

@MitzaMaxwell

10 ай бұрын

Nor does it go any faster when the system is so vulnerable that it is necessary to have a virus scanner constantly running in the background to protect the system.

@lucidmoses

10 ай бұрын

@@MitzaMaxwellAre you suggesting that if you put in the security patches that you don't need an anti virus program?

@MitzaMaxwell

10 ай бұрын

@@lucidmoses Do not run without an antivirus program if you are using Windows, it is not advisable.

@lucidmoses

10 ай бұрын

@@MitzaMaxwell Ok, then I'm not sure what the point of your previous comment was. AV runs either way.

@MitzaMaxwell

10 ай бұрын

@@lucidmosesThe point is AV slows down the system but it must be present. So it makes a difference which antivirus you use. some slow down less than others. It's been a long time since I've researched different AVs, so unfortunately I can't recommend any today. I myself use Microsoft Defender, it has become reliable and safe enough for me. But my new computer recommends Norton, does that mean two AVs are running at the same time? I can't stand Windows anymore. That is the point. I use Linux now, but that doesn't mean you can be careless, programs, files and e-mail must be safe on Linux too. But you only scan when necessary, so the system runs without slowing down the rest of the time.

Thank you.

There's nothing like a fresh Windows install every now and then. An easy job when the libraries with documents, video's pictures, etc are stored on a drive other than C:\

@krathoon2338

10 ай бұрын

Sometimes, you just have to blast away your Windows.

@macethorns1168

10 ай бұрын

...or when you do everything in VMs and your host OS has no extra software installed.

@MitzaMaxwell

10 ай бұрын

Yes For ever! Nowadays, fortunately, we have better alternatives than Windows. My computer is so old that it can no longer be purchased. The brand Medion no longer exists. The Computer is faster now even as when it was new (15" Intel i5 laptop). The secret is that it has been upgraded from Windows7 to Linux Mint.

@SeanBZA

10 ай бұрын

@@MitzaMaxwell Mine came with Vista approved stickers on. But DDR2 is getting scarce, so an upgrade is in the future. Also want to move from the core 2 Duo as well, only 4 cores....... and 2 are virtual, using spare processor cycles.

@UmVtCg

10 ай бұрын

@@MitzaMaxwell Sure, I've got a laptop with Ubuntu, but try gaming on that. It's all about using the right tool for the right job. My Nvidia card needs Windows and so does my gamepass subscription

Also, software itself changes, through updates that bring new features, and thus, might require more resources.

A spinning rust drive CAN feel slower when it gets fragmented or fills to close its capacity since the data might not stored in a fast area due to the spinning platters. I still use spinning drives when throwing together a test box for one of those "what does rm -rf /" really look like: scenarios. I know, I should be on Azure or digitalocean spinning up virtuals but sometimes having a box sitting in front of me is more fun. Ya know, cuz I'm old school.

Hardware that relies on FLASH storage definitely can slow down with age. Flash is written in blocks/pages of a few bytes - for example 256 bytes. Therefore, even if you only want to change one byte in a block, all 256 bytes must be written. Flash can only be written to a finite number of times (unlike SRAM, DRAM, or FRAM), so sooner or later a write to a flash memory page will fail. In this case, a different page must be selected. Wear levelling hardware internal to flash components reduce the amount of wear by swapping different pages in and out of the memory space, but sooner or later, you begin to run our of good pages, and thus you can spend longer selecting a good memory page as it takes to write to the memory itself. This is a particular problem is the memory is full of data as it means there are less free pages to select from. This only applies to flash devices though - it's one of the reasons your phones and tablets get slower with age. I'm not sure about SSD drives - I can't remember if they're built with flash memory or not.

@macethorns1168

10 ай бұрын

You're not exactly wrong, but for some strange reason it seems that Linux never has this problem. Just Windows.

@0LoneTech

10 ай бұрын

​@@macethorns1168It's a real problem but doesn't really affect Windows in particular, because it's relegated to the storage controllers inside the SSDs or hard drives (yes, they have them too, ever since SCSI and ATA). Some Linux systems expose this, e.g. when using UBIFS it takes time to load remappings by scanning the entire memory (if you can't reliably erase them, you may need to compare all to find the latest).

I found the fragmentation of the MFT and Reg in the XP days. Also, I found that one way to keep 98 speedy was to automate a full and updated copy, blasted to a reformated partition worked wonders.

if you're going to make changes to a PC it can be worth benchmarking before and after to help you keep track, used this myself many times.

When I lead from-scratch projects; whenever possible, I insist provisioning the dev team with workstations powered approximately the same as our production target. Too many times, I've seen well-intentioned leads and directors order top-of-the-line gaming rigs. Teams develop and test on those, only to deploy to average home users running on Cerelon or Athlon chips. Only to later scratch their heads in frustration at negative reviews and/or elusive crash reports.

I run Linux, not a problem for me... In fact older computers when loaded with Linux appear faster. Installing user space apps like games shouldn't introduce Windows rot, that is a bug, conveniently left there to ensure you buy a new computer and windows license every n years...

I feel like we're missing the impact of a mechincal drive on performance. For those who still use them, they are the number one reason why a computer is slow. Just transferring the image to solid state media creates a night and day difference.

@macethorns1168

10 ай бұрын

W10 still runs like a dog on hardware that W7 flew on.

Brief and Blunt: Bloating... thank you, Dave!

I'm pretty sure in many cases it's hard drives slowly deteriorating.

Dave's attempt of a shorts. So glad it didn't happen otherwise I'll have to scroll all the way down my notifications again 😅

Also if your storage is close to full, that does have an impact for similar reasons explained in the vid of when you have too little ram. This is true for desktop and mobile

Never had the problem on my Amiga.

Failing hard drives are a big performance hit…

When I was younger I regularly formatted the HDD:s of my computer reinstalling everything from scratch. Perhaps a little extreme method to unbloat your computer, but back in the late 90:s early 00:s it worked.

Can we talk about Side by Side bloat? Windows update bloat? Code bloat in general? (Old man screaming at clouds voice)

As an IT tec I confidentially say, a lot of is computer illiteracy. People panic when the file 'disappears' after it downloaded only to download it again rather than going to the downloads folder. You can forget about deleting old files because they are clueless on how too. Right clicking confueses the hell out of people, and we are talking about people who use a computer for living; teachers and those who work for corporations alike.

DMA all the way baby! Nice post.

I've been told in college (basically) that it's a good idea for the average user to back up their files then reinstall windows at least 1 time a year, because of the scenario you point out here, but in relation to the system registry specifically... granted, I have no idea if that's still a good idea with windows 10/11, considering that when I got this advice I was able to get such advanced CD keys, for windows Vista.(and of course my college was a Microsoft partner, so any CD keys were free to technology students. Does Microsoft still have that program, and does it really need it these days with free upgrades?) Basically, what I have been told is this, when you install and uninstall applications and other software, the system registry can get bloated and fragmented over time, and the only real easy way to combat that at the time anyway, was to reinstall Windows if you have done a lot of say testing with it or reinstallation of drivers and things over and over again or uninstalling one game and installing another then uninstalling that one and so on and so on and so forth... Another thing that could cause the issues, that I haven't talked about is your desktop screen in particular, your users desktop on Windows xp, which back then, if you stored a lot of large program folders and ISOs and things, the increasing size of the items on the desktop would then slow down the boot and login process exponentially over time... if course, that also likely only applies up to windows xp, but still, if anyone needed or wanted to know that, then there you go...

@a4d9

10 ай бұрын

It's always good to have a backup of the data that you care about. When using spinning disks, reinstalling the operation system gave you a defrag for free. With reinstall you probably got the latest drivers, and didn't install unwanted programs. Having large or many files on your desktop is mostly a problem if you use roaming profiles and it has to copy the files to and from the network every time you login/logout.

@northwiebesick7136

10 ай бұрын

@@a4d9 while everything you say is true, I'm talking from experience with the cluttered desktop slowdown issue, and THAT was using a home computer NOT a business machine... as in, a normal account, with a few gigs or more of data sitting on the desktop... once I moved past windows xp that "problem" went away for me

When I switched to MacOS this problem seemed to diminish a lot, not gone mind you but diminished to the point where you can tell performance differences only between the processors (Apple M1/M2 vs Intel Macs)

In many cases it's also using newer software on older machines. Newer software may expect a different level of hardware than the machine it's actually running on. Time marches forward and if you're not upgrading your hardware, the software will get to be too much for it eventually.

@CrazyTobster

10 ай бұрын

I don't know, look a Linux, unless your paying for support the drivers can be generic and Linux can be super fast old hardware... Look at operating systems like Xubuntu which is made for older hardware. (Not be confused with Ubuntu) As an IT guy who gets moan at all the time for their PC being slow, it normally because their desktop is a horror movies wil all their files on it or/and they have never deleted anything after it been downloaded. After two years of not tidying up down loading a couple of time every day what do people expect. I don't really understand how keeping everything on the desktop is easier to find; it a lot more difficult. People don't understand the more you put on your desktop the longer it will take to boot up your computer. I got a more than average amount of applications on my desktop which can be a struggle at times but they are applications I use every day and need to access first thing in the morning; it is different from having seemingly an infinite amount of Word, Project and Excel documents on your desktop that should be in your documents folder.

I have an old Compaq cq57 laptop 2.4GHz Dual core Celero running Windows 10. I replaced its HDD with an SSD, upgraded it from 2GB to 8GB Ram. I am surprised how decently it behaves.

If you have a really old HDD that also can slow down the PC quite a bit, and that is a real hardware problem.

@macethorns1168

10 ай бұрын

...if you're running Winows. Ubuntu on spinning rust is just as fast as Windows on SSDs.

I have on a couple of occasions over the years found that creating a new user and switching to it solved the dreaded gradual slowing of windows with time. I thought I had made a great discovery but it doesn't always seem to "work". For this reason I always install software for "All Users". I just copy the users files over and all seems OK.

So many factors. (1) Fragged files, (2) unnecessary processes that slow startup, (3) "helpers" that are more like cripplers, (4) Chrome tabs that steal gigs of memory/100% a core, (5) running too much at once/too many tabs --> infinite swapping.

The Windows Registry is evil, there is no way to clean it up - in that way linux and macos is way way easier to deal with.

You forgot to mention that software sometimes "outgrows" the host machine. I mean, sometimes tge min specs required change with each updatr.

Also OS upgrades some times make it slower because the new OS was designed to run on modern hardware.

Is there a way to bring the task manager to the top through a keyboard command when its not set up that way before hand?

I'm dying to know how you set up your UDM and whether or not you upgraded to the UDM SE since that video is quite old. I recall a video where you started to talk about aggregation but missed if you ever uploaded the promised video on that. I got the UDM SE set up and running (all on that box and a patch panel) but I'm considering the $269 aggregation switch and a unifi switch next. Hope you don't mind me asking. Thanks!

Dave. Dave! DAVE! Now that I have your attention. Could you do a video on pagefile size. I am reading about VMs and see that the 4KB page size is based on the processor type and architecture, but why adjust it? Why can't I find where to change this in Windows? If I have lots of unused RAM because I pulled out my credit card and bought all the things at Microcenter, could I get an advantage from increasing the default page size? Please Davesplain this me. I am referring to the minimum size of memory transferred back and forth between the disk and RAM. Why 4KB? Thank you, Dave. Please upvote to ensure Dave sees.

I think timing crystals get less consistent with time, and that causes more checksum fails in communications. At least that's what I was told.

@SeanBZA

10 ай бұрын

Not true, spread spectrum clocks also jitter the clock to make noise more broad band, to cheat emissions tests that sweep across a band, so the average noise becomes lower, and the clock can jitter by 5% easily. Critical timing functions use separate crystals, though all timing on a CPU and computer is designed with tolerance in mind, as nothing outside the CPU cores is too critical timing wise, so long as edges of data and clocks line up with the correct relation to each other. The most critical timing is not in PC's, but in cellular use, where your timings even compensate for the distance from the base station to the mobile device, as the shorter the dead time between switching between each mobile handset you have, the more data you can transmit to the particular handset per frame, and thus you get distance compensation, as well as dynamic power control, so that the receive side sees a near constant signal from all handsets at the base station , so it does not have to degrade data rate, and the transmit power to each handset in the antenna field varies wildly, so they get a near constant level as well, The other handsets are muting the receivers while waiting, so they do not get desensitised by high signal levels, and the base station will dynamically allocate the handset position per cycle, to not have high power signals during the potential window time. That needs an accurate clock, and no base station runs off a regular crystal, they have normally heated ovens that keep the crystal temperature stable, and this is in turn fed from a Rubidium or Cesium atomic clock, that in turn is synchronised with a GPS receiver per base station, to get a clock accurate to within a clock cycle across the entire network, and giving free time to each mobile, as they also need an accurate clock, though the base station will compensate for the drift in the phone clock as well, in the telemetry channel each phone receives.

@plumbr13

10 ай бұрын

I understood every word of that. Just not the sentences.

A new subscriber question... have you previously covered the Y2K Calendar/Clock issue???

hey dave ... why is it, that windows still strictly needs a registry instead the linux aproach of storing (config)informations in actual files?

So how do we fix the software bloating? Is there a routine maintenance to reduce bloating to restore performance that is comparable to when it was new? Im having trouble finding this answer online.

@shallex5744

10 ай бұрын

the best thing you can do is not use Windows

I'm temporarily using a 15 year old PC built for Windows Vista. I put in an SSD and 4GB of RAM and installed Linux Mint. It's definitely slow at loading webpages with its aging Phenom X4 9650 and GeForce 8200.

It's Windows software being bogged down overtime.

I think if I use my PC in a way that builds up bloat, then I am using it wrong, and I would like to learn to avoid bloating the computer when cycling software.

HDDs could also be an exception, luckily quite easy to replace.

The reason is that the code we all write nowadays is total bs. Lets be honest. Thats the only reason why we get performance downgrade every year. Coz of software updates.

How about this: after factory reset, ancient phones stay unusably slow even if you keep them offline. I reproduced this on three different Samsung Galaxy S4's.

There are 2 more factors in PCs getting slower over time; HDD and SSDs. They gradually fail and may mark sectors as dead. That means it has to find a new place to write the data, that process can take time, and may cause more fragmentation on HDDs, which in turn increases the seek/read time.

Planned obsolescence?

So what is the solution? Uninstall/Reinstall?

@UmVtCg

10 ай бұрын

Reinstall

@MitzaMaxwell

10 ай бұрын

Uninstall! and for God's sake don't reinstall. If you want your old computer to feel fast? Install the latest version of Linux based on your taste. There are many KZreadrs who demo the most popular Linux flavors. It is also possible to test a Linux version on a USB key without installing it. An old version of Linux is not necessary if the computer is only 5 years old. You can have both the latest software version and a faster computer. Today there are Linux versions that look and feel like Windows and you can run Windows programs on Linux with Wine. I have grown fond of Linux Mint.

Dave, is new software coded badly and inefficient compared to the good old days?

My experience is budget gaming ram, especially stuff that runs high XMP / EXPO clocks and will degrade over time. If you sit and do a memory validation, sweep once every 3 months what was once solid might be ever so slightly unstable the small random bit Flips are corrected by the system kernel on the fly but not without a performance hit. ECC memory will solve this problem, but it runs a good bit slower and is far more stable than desktop gaming memory as most server memory follows jdec / iec spec very tightly

@roycsinclair

10 ай бұрын

My home system has 80GB of ECC memory though I usually don't use more then 40GB the extra memory is automatically used as cache which is even faster than that nice SSD. I have disliked non-ECC memory from it's introduction so I usually build systems that use ECC memory.

@ZoeyR86

10 ай бұрын

@roycsinclair I understand my workstation has 256GB of ECC memory I frequently run out. I ordered more today, actually. But my 5995wx + RTX 4090 (has ecc memory) has been an ai power house

@MitzaMaxwell

10 ай бұрын

Wish that was our only problem. The problem for non-gamers is that the operating system is weighed down more and more every time Microsoft patches the vulnerable Windows operating system. For example, I had a computer that ran fine until service pack 2, more RAM help with that problem. Then came service pack 3 and the computers became snail slow and there was nothing that could be done about it. so both me, my brother and my father had to go out and buy three new computers.

Don't forget general software bloat. Not that a guy from Microsoft would know anything about that.

@paulmichaelfreedman8334

10 ай бұрын

Bloatware. Shareware versions of crap bogging down system right off the shelf, and mostly absolutely useless.

@SianaGearz

10 ай бұрын

@@paulmichaelfreedman8334 "Shareware" is an interesting term to use today, given it stems from the time when independent software developers couldn't span up their own marketing and distribution for the lack of universal Internet access, and relied on people who like the software copying it to their friends and the local BBSes they frequent. So Shareware that you would encounter (unless you bought one of those Shareware collection CD-ROMs) was usually very frugal and high quality software.

@katrinabryce

10 ай бұрын

To an extent, yes. But if you run Windows 98 on modern hardware, it will be a lot faster than running Windows 98 on period-appropriate hardware, but still a lot slower than running a modern operating system that supports the full range of additional things you can do on the modern hardware.

@mariemccann5895

10 ай бұрын

@@katrinabryce You're on fire!

Because windows.

I always assumed a lot of those "updates" are there just to make your PC run slower so you get annoyed and buy a new one lol.

This why you have some people who re-image their computers after a while. It gets rid of the bloat.

@mariemccann5895

10 ай бұрын

A re-image wont work, it'll wipe most of your data.

I did a test on a very old laptop recently. It was completely unusable. I installed from the original disk. Performance was acceptable. No 3rd party software was installed. I only then installed all the Microsoft SPs. Back to being unusable.

Yeah, but I install a fresh copy of windows 95 on my Asus P5A motherboard with an AMD K62 and it just seems 1 million times slower than it did in 1999. I think perception plays a large role..

Don't buy used video cards expecting them to run at full performance, especially if they've been mining

Shorts must be a 60 seconds or less, Dave

Or if you have an iPhone, the chipset will throttle itself based on the health of the battery.

@mariemccann5895

10 ай бұрын

It's not just iPhones that do this...

this applies only to operating systems they want you to pay for, which is bad

Look out fornlow storage space, I think that is the main culprit. Also too many startup programs enabled

Dave, as one old guy to another... swapping memory in the age of NVMe SSDs on PCIe 4 busses isn't the performance hit it used to be...

The longer you've had your computer, the more software you've installed. When new, there may only be a few 10's of services running. A few years later, there may be hundreds of services running, which will certainly slow down your system. You could get a new system, or just re-image your old one to bring it back to factory new condition.

It’s true for windows but not for osx or Linux

Barring some physical failure, computers do _not_ get slower. By the same token, they don't get faster either. So if you keep installing newer and more CPU cycle-needing software on the same computer, it's not going to compensate for the added load.

My AOL is slowing my 17.5k modem to a stop.(27) 24hour days to download windows ME.

Registry corruption/fragmentation. Accumulated TSRs. Primarily Windows problem.

Slightly off topic which slow your pc virus malware how can we track them down and DELETE them

You forgot about updates, they almost never improve performance, they almost pretty much always destroy performance, unless the initial release was so pathetically lazy and buggy and unoptimized, that that it was already as bad as it gets and there was nowhere to go but up. "Security" updates can affect it too, like how the "patches" for Spectre/Meltdown killed up to 20% of performance.

This was meant as a short lol

just note my screen is 16:9 sized ▀̿▀̿ not ▮ shaped ?

👍

Fresh install every two years or GTFOH. Seriously. It forces you to have proper backups too ;)

Some knowledge mankind was not supposed to have

Are you trying to slim down using portrait mode? 🤣

Because they are just like their creators.