286 vs. 386SX
dfarq.homeip.net
dfarq.homeip.net
TBH that kind of a short shelf life wasn't just a 386 thing. Clocks speeds and architectures were advancing quickly, and all chips had a really short shelf life.
For example, in the span of 5 years (say '91-96) you could upgrade from a 386SX 16MHz to a 486DX 50MHz to a Pentium 90MHz, each time paying about the same amount of money but getting a 3x speed-up. And other components like video cards were improving just as quickly.
People were upgrading every couple of years because the difference between older and newer models was night and day. Imagine if in 2015 you bought an Intel i3 3GHz and this year you could buy an i7 15GHz with 8x the RAM for the same price.
A 2 year old PC felt woefully underpowered in that era, and a 4 year old PC was almost useless if you wanted to use any "current" software. You'd be out of drive space, unable to run a lot of programs/games, and limping along.
Now, my 8 year old Macbook Air is still more or less as functional and useful as it was when I got it.
If one can’t discern issues with a four-five yo computer, I would very humbly suggest it also says something about the demands of the owner stultifying to a degree.
I've seen people load up a 200 machine cluster on AWS doing quite mundane tasks.
Information complexity is the only physical quantity that doesn't obey conservation laws, so this kind of thing really isn't impressive or interesting.
Sure, I'll admit it may not be fair to compare, considering the AWS host type, copying the data out and in, and all the other overhead.
That said, the project sure looked like it was resume driven 'big data'.
Then, show me the equivalent of a 2015 game on par on the gap to a game from 2021 as a 486 game like Doom compared to a Pentium III game like Max Payne.
> eh, bullshit
Precisely what's bullshit? You're telling me that your workloads on modern computers are stressing them out and you simultaneously don't see a difference between a 6th gen platform and an 11th? No, you don't see a substantial difference in modern platforms because you don't have a need for the performance gains.
My point stands.
For those whose workloads aren't multicore-intensive (or GPU-intensive, which is pretty much the same thing), it's been a dry, boring 10 years in the PC business.
Right? I remember the enormous performance jumps from 286 to 386 to 486DX2 to Pentium to P6S...
Today, I'm still using a late 2013 MBP. Other than the lame 128GB of disk space, it is still my primary machine and is fine for everything I do. Most of my work is done in the cloud anyway so its essentially a 1970's dumb terminal.
This is my frustration with Apple's policy of dropping support for hardware in MacOS. It made sense in the 90's to upgrade every 2-3 years because you got 1.5-3x more performance each time. So 6 year old hardware was almost an order of magnitude less capable.
Fast-forwarding to today, a "legacy" 10 year old ("Mid-2011") MacBook Pro supports just as much memory (16GB) as Apple's current M1 offerings. The M1 does put up some very impressive numbers on the single-thread CPU front, but that's because we've gotten used to such small progress every year--it's only about 2x the speed of the 2011 MBP for single thread tasks.
You still get that level of performance increase. It's just that it's to the point of being imperceptible, or invisible, to the vast majority of users. The only ones who really notice are the ones really pushing the machines with things like 3d rendering, 12k video production, and high end audio (I can run 80 plugins now where I could only run 70 before!).
Texting? Doesn't matter how fast your machine gets, you won't notice. It will perceivably perform just as good as a phone that texted 20 years ago. You can run benchmarks and see, oh yeah, this is faster, but it's not noticeably faster to your human inputs.
In the days of 300 baud modems, they were so slow you could actually read text as it was sent over the line, as if someone was just a very fast typist and you were watching them. Now, you get full multiple pages of text with videos that pop up almost instantaneously. Once we got to 1200 baud, it was too fast to read as it was coming over the line. You couldn't catch up. Ever since that threshold was crossed, it doesn't matter how fast speeds get, they're all faster than the human brain can absorb so they are basically the same in our minds.
Our 9-year-old Air went to live with friends who suddenly needed another computer (for their kid) at the beginning of COVID. It still works fine, though some sites are slow.
Sure, APPLE isn't releasing updates for it anymore, but that doesn't actually affect whether or not Word runs, or whatever.
I'll also say that biggest and most dramatic upgrade I ever had was moving from a 1988 AT clone to a 1991 386/33 (Gateway 2000, baby). We'd pick a directory with loads of files in it and do a dir just to watch it scroll by so insanely fast. Simpler times for sure.
The next most impressive upgrade I ever had was about 10 years ago. I had a 2010-era Macbook Pro (early Intel) that shipped with a spinning drive, back before everything was SSD. At some point in its life -- it was ultimately stolen in 2012, so call it 2011 -- I swapped the spinner for an SSD and OH MY GOD THE DIFFERENCE.
No other machine upgrade in my 30 years of computing has come close to the "holy shit!" moments of these two. I wonder if my ultimate shift to Apple Silicon will bring some of that -- I hear good things, so I hope so.
FWIW, that difference isn't nearly as noticeable under Linux. It's so lightweight on RAM and so good at caching frequently-accessed data that it can be incredibly snappy even when using spinning rust. SSDs mostly speed up your boot process and the rare IO-heavy workload. Though it wasn't until the late 2000s-early 2010s that RAM began to be truly abundant on out-of-the-box configs, and that was the same timeframe as the switch to SSD's.
If I recall, Ubuntu 16.04 still suffered from this. And Android Studio could very reliably trigger this situation on my older/upgraded early i7/16GB RAM/SSD laptop. Hard power cycle to recover.
When I was preparing to emigrate in 2014, I put an SSD and a big HDD in my desktop-replacement laptop, and the transformation was a revelation. Ubuntu 14.04 went from booting in only a minute or so to booting in a single-digit number of seconds. It was astonishing.
And was considerably less expensive than a very top end ($2500-3500 in 1992-1994 dollars) desktop built with something like a Pentium 60 or 66 MHz.
Inflation calculator tells me that a $2500 desktop PC in 1993 would be the same as about $4700 today. For 4700 you could build a real beast of a machine.
I remember when the best price/performance was a 300 Mhz Celeron A that you could overclock to 450 Mhz on an inexpensive A-bit BH6 motherboard. Paired with a 3dfx Banshee, and I remember being able to build a respectable gaming rig for under $600.
These days, $700 will barely get you a GPU, even at MSRP.
I'm referring to the whole time frame before the Cyrix 5x86 and similar were even a thing... There were plenty of AMD 286 and 386 CPUs sold in the early 1990s.
AMD has had moments where it really stuck it to intel but always fell back to 2nd, I hope this time it sticks.
x86 vendors duking it out while Apple keeps both honest isn't a bad market for a buyer.
Now with magnitudes more speed and memory available, loading Word is...somewhere in between.
Opening it is a joy... what was a slow, lardy app when it was new is now lean, mean and fast.
Something similar happened with graphics cards, each new generation made stuff look that much better.
These days I can still use a 2013 Macbook to play MineCraft, doesn't feel any different. Compiling code probably is different, but most everyday things would not be much different.
Oh and of course an obvious question to go along with the whole 90s CPU story:
https://www.maketecheasier.com/why-cpu-clock-speed-isnt-incr...
Programs start much faster, quit much faster, web pages load much, much, much faster. Compiling is anywhere from 4x to 10x faster, as is interpreter startup time for scripts.
Now, granted, it's a jump from one architecture to another and a time span of 7 years, but this upgrade felt magical the way the mid-90s hardware upgrades felt.
Anyway, my best friend and I ran the DRG game lab (developer relations group) - where we (intel) paid millions to gaming companies to optimize their games to the intel arch and things like SIMD instruction sets... we game tested (subjectively) games running on intel vs AMD machines... it was also the lab where we were able to prove that a subjectively performant PC could cost less than $1,000 === The Celeron Processor
I was even the person who first sent an email (1997) to engineering asking why we couldnt stack multiple processors on top of one another....
I later learned on a hike with a head of marketing that in the proc labs Intel had a 64-core test fab. (This was fucking 1998 when that was revealed to me under NDA etc...)
---
"looking closely at the demo, we can see - as you can see on the screenshots - that the SSE version is less detailed than the non-SSE version (see the floor). Intel would it try to roll journalists in the flour?"
Two other examples from that link are SSE "optimized" Rage Software Expendables running slower than original, and "Pentium III optimized" PowerSlide by GT Interactive with zero fps difference. At least those two never released, unlike "Designed for Intel MMX" POD from 1997 with one optional audio filter added and 1/6 of the box covered by Intel marketing.
One has to wonder how much was Intel bribing for such blatant lies and manipulation back then.
The original author missed that the 386SX had the memory addressing models that the 386 had in addition to the backward compatible 286 modes. So you could access your "Lotus eXtended Memory" much more quickly than you could on a 286 based DOS machine (aka the PC/AT). A lot of businesses that ran on large Lotus spreadsheets used the 'SX for just that reason.
I don't think we get the same kind of performance gap in desktop pcs anymore.
Three years later, I bought a 386/33 because the 286 was, by comparison, dog slow. And it cost less than the AT had.
Computer Shopper was such an amazing magazine to go through, and I would always try and find the best deal from the systems advertised, not that I had any real money. Usually it was from the ads in the back, not the big pretty Gateway ads in the front.
Some prebuilts, especially from Dell, HP, Acer and Lenovo will have custom motherboards, custom power supplies and custom cases that can't be re-used later on. Those usually have the very crappiest cooling solutions too. So they will end up costing you more money down the road, because you can't incrementally upgrade the pieces as much as you could with a self-built PC.
I really miss that type of advertising. I don't want advertising intruding on unrelated activities, ruining tools and destroying my neighbourhood with billboards.
But there are times I want to be sold to. I want 100 firms to show off what they have and extol their virtues in an easy to browse format.
But in the 90's we went from 286 and 386 class machines to the scorching-fast 750MHz AMD Athlon. It was very depressing to sink a few thousand dollars into a new mid-range system, only to see it worth about half that a year later. And a doorstop 3 years later.
Now we're back to the point where almost any computer you buy will do a good job for common tasks (i.e. not gaming and crypto mining) for at least 5 years or more. My current machine is a 7-year-old laptop and runs all modern software just fine. Unthinkable just a couple decades ago!
The 386 was still being used up until Window 95 came out. It could run on a 4MB 386 but anyone who did probably quickly upgraded to at the very least a 486/66, which still struggled even with 8MB. You really needed a Pentium, or a 100MHz 486 clone.
A 7-year old processor is somewhere around a haswell or broadwell, right? An i7 of that gen had a passmark score of like 950, whereas an i7 of current gen is about 1700. Even ignoring improvements in mobile graphics, power consumption, displays, and memory capacity, a modern laptop provides a completely different experience and level of capability than a 7-year old laptop.
As I said in another comment that got voted down to hell and even flagged (for goodness sake) these comments about the adequacy of very old hardware probably tell us more about the owners' unchanging software tooling and work habits more than they convince us that technical advancement has come to a near standstill.
WFH in 2020 required many people to add frequent VC on top of their normal workload. A 7 year old laptop would fully show its age if that normal workload was already pressing the laptop's boundaries. God help you if you wanted to add OBS into the mix. I can play an acceptable form of Assassin's Creed on my Tiger Lake laptop. I couldn't comfortably do that with my Skylake laptop. Examples and scenarios are legion. I'll use any resources I can get in a 13-14" mobile platform and then some.
I'm going to do mostly the same thing, although I'll certainly acknowledge hardware has gotten faster and more power efficient. It's not nearly the same growth rate as before, of course.
But, as a grumpy old person, I don't see a whole lot of change in what people are doing with their computers. Yeah, a new one would be more powerful and often use less energy, but would it change what I can do? Newer games are a real thing, but can I spreadsheet faster? Does it make chatting better or just enable more bloat? I used to send animated gifs to friends with AIM in the late 90s on a pentium 75, and the gifs were really animated gifs. Somehow, a modern chat client needs a quad core machine.
Maybe more/different VC stuff, but I usually run VC on a haswell chromebook with 2 cores, so that doesn't seem hard.
Check out where the broadwell and haswell procs are on this plot relative to a 10th gen processor.
You probably couldn't spreadsheet faster on an 11th gen, but you could have more workbooks open, drive a couple large hi-res monitors, do VC with OBS in the middle, while screen sharing, while running several other apps, and add in a few VMs/containers and an IDE on an 11th gen platform (e.g. XPS 9310), all of which you absolutely could not do on a Broadwell or even Skylake platform - I've tried. It's all about what your workload/requirement is, right?
However, the leaps in the 90's were great enough that you wouldn't have been able to run even the latest OS (or apps written for it) on mid-range 7-year-old hardware, at least not with any kind of comfort. I would have been almost outlandish to even try. Windows 95 should have theoretically run on a 386 from the early 90's, but I'm not sure it would have really been usable. In the late 80's you might have got a 286 -- not much luck running Windows 3.11 four or five years later on that. If you got a Pentium circa 1995, good luck running Windows XP or even Windows 2000 on it.
Looking at the charts in the Phoronix article, the overall performance (geometric mean) approximately doubled from Broadwell to 10th gen. That's more than I would have expected based on how incremental the generation-to-generation changes appear, and it's certainly substantial if your workload involves any kind of heavy lifting. However, in the 90's, performance would have doubled every couple of years or so. Within a time frame similar to Broadwell to 10th gen, it would probably have more than quadrupled.
I have an old work laptop (6years) and a much newer Linux home laptop that is faster. I notice and it’s nice to have the speed but Im not suffering with my old machine.
We’re a long way from watching a spreadsheet recalculate (AppleWorks on the Apple 2 Im thinking of you)..
Sure for gaming/ music or video work or high end development tasks it makes a difference. It probably matters more than people think but they aren’t suffering with older hardware.
Some development platforms are looking to induce upgrades however (electron apps for example)
That's precisely my point. Having low requirements for hardware is a poor basis upon which to judge the advancement of hardware. If you don't notice a difference between a 6th gen and 11th(!) gen proc and associated platform (for which we have comparative benchmarks, btw), your demands of the hardware have just remained too low for the improvements to register. For a basic home user or low-intensity office user, an XPS 9310 with 32GB of RAM would be a waste today. For people that can use it, it's a godsend compared to a Haswell/Broadwell/Skylake laptop.
Or just a smaller gap. Doom (1993) vs Quake II (1997).
Night and day.
By comparison, kid, today you should be able to play a FULLY raytraced new release of AC or Watch Dogs with crazy photo-realist graphics (superior to even Forza 5) at least at 1080p.
Because that was the gap between a 1993 PC and a "Multimedia PC" with a Voodoo and MMX Pentium, if not a Pentium II which would curb-stomp the 486.
Oh, and we are emulating the PSX with VGS from Connectix, so add PS4 emulation on top of that with a mid-range PC from today.
BTW, the first computer I personally owned was an 8086, so I was there, Gandalf. It's what I first learned to really program on. The timeframe you're talking about, the progression of tech through the nineties, I'm intimately familiar with.
I feel dissapointed.
I remember the excitement when the SB meant my computer could play, e.g., human speech in my games—I forget what the games of that day were; one of the early Ultimas?—rather than just the beeps and boops of the PC speaker (although some people could do amazing things with that speaker!).
Good memories :-)
Thanks for sparking a fond memory ^_^
CONFIG.SYS be damned!
Eventually IDE/ATAPI drives became the norm, and I believe the SB16 dropped the proprietary CD-ROM.
GUS was definitely a game changer for me also, I can still remember how much cleaner and yet at the same time warmer the sounds outputted by GUS were when used correctly.
At that time these chips and cards still had their signature sounds, kinda missing this period in a way.
They certainly could do amazing things with that speaker. I remember being blown away when I loaded up a PCPlus magazine "superdisk" cover disk some time in the early nineties IIRC and speech came out of the PC speaker -- "Welcome to SuperDisk 61" (or whatever number it was).
Dumb question (maybe I've been out of Windows world too long) but what does this mean?
It was a HUGE upgrade.
First, 12Mhz to 40Mhz was an amazing upgrade, but that was definitely not the reason why I upgraded.
The real reason was *compatibilityé.
The 286 was compatible with close to nothing by the early 90's. Doom? Sure, runs slow on the 386sx, but on the 286? Does not run at all. And the same goes for all games and programs using one of: 1) 32 bit DOS extenders 2) EMS 3) any amount of XMS more than my 286 could handle 4) >600KB conventional memory, because there's only that much HIMEM.SYS can do 5) Windows 3.0 in 386-enhanced mode
The 386 40 was my first "homebuilt" before I had ever heard the term.
Wolf3D came out in 1992, it was the perfect companion for a 286. Same goes for Duke Nukem II in 1993.
EMS also works fine, heck it even works fine on an 8088. What doesn't work are things like EMM386 that emulate EMS using extended memory.
The 286 supported up to 16MB of XMS and if you could afford that much RAM, you probably could afford a 386 or 486 CPU..
This didn't work on the 386 and it was the end of the CPU industry in USSR/Russia.
The 286 technically also protected mode but it really sucked.
A fair-few DOS applications (via DPMI) also implemented virtual memory schemes like a "real OS" - I remember POVRay 2.0 officially requiring 2 MB of RAM, but was able to start on my 1MB 386SX machine (albeit with 1-2 mins of HDD-thrashing...).
It barely ran the Doom demo but it did run. I ran OS/2 2.1 For Windows on it too at one point but eventually I build a home built Cyrix 486 DX40 that ran OS/2 Warp and Dos games like Doom at blistering speed.
You couldn't run OS/2 2.1 on a 286 at all back then.
I think the biggest issue of the 386SX was the data bus but I enjoyed my time with it. It is a bear to get modern storage to work with my Victor where as I see people with 286s on YT adding all sorts of modern goodies custom made for classic computers. It makes no sense why my Radioshack motherboard has problems with new modern tech and the 286s that I have seen don't but it may be due to bios issues. Atleast I had upgraded my 107mb hard disk in the Victor to a Maxtor 540mb drive and it worked with the software Maxtor had iirc though I lost the ability to play games if I was using MSDos at the time because it took too much HiMem.
Those were the days. I really was stretched trying to get things running on that 386SX other than productivity software for Highschool...
PS... My Highschool had IBM PS/2 Model 30's and 40's with Microchannel buses. So we couldn't sneak upgrading our PC lab computers with Joystick boards. At least we could learn Turbo Pascal.
The Harris 286 chips overclocked handily. The difference between 25Mhz and 40Mhz on a 286 was noticeable, too; except when anything hit the ISA bus which din't push much at all.
There were a few systems in that era with SRAM instead of DRAM; I always regretted not catching one. Helped someone track one down for their week long spreadsheet runs.
Not much was left for documents, so I would "garbage collect" old stuff into floppies.
Still was quite useful for about 5 years, however being an SX meant that eventually I could not keep up with my favourite flight sims as they started asking for 386DX as minimum.
Was still on time to edit it.
It eventually returned after MS paid out.
"Few people ever upgraded their 386SXs that far" is a bit of an understatement. In 1990 2MB of ram cost about the same as a 386DX CPU. By 1992 ram had dropped to about $50/MB, but the Am386DX and the 486SX, both of which blew the i386SX out of the water were generally available at this point and cost less than 4MB of ram.
One thing TFA doesn't mention is that the Am386SX (and SXL) had usage in battery powered applications for some time after this (not as common as today) due to their very thrifty power usage (with a fully static core, there was no lower limit on the clock-speed it could run at and it was lower power usage than Intel's SL when running at full speed).
There was something weird where windows wouldn't run in 386 protected mode. I was sure it was DX and SX, but this article has the assertion that protected mode worked with SX. I know my dad replaced the system, I assumed he updated the CPU, but maybe it was a bad mobo or something.
DIY computing to me will always be RLL/MFM hard disks with insane ribbon cables and IRQ toggles on ISA slots.
God, I am old.
It was also a bit of an art back then to stuff as much of the needed TSRs and device drivers into the memory above 640K, to leave as much room as possible for applications.
These days I'm dealing with an SoC company that has a dozen variants of the same processor, which have different combinations of four letters in the suffix in various combinations.
I worked at a computer consulting company 1990-1991 and started my own company after that. Got very good at puzzling out how to jam as many ISA cards as my customers thought they needed into their computers. But I really think of that as the bad old days.
I remember getting computers on a network running SHGEN-1 and SHGEN-1 with Novell Netware 2.15 on 360k floppies on a 286. It took an hour for a single computer! I wouldn’t want to go back to that for anything.
I paid $3,500 for a 486DX2 66MHz around 1992/93 I can't imagine what a 486 would have cost in 1991.
I saved up for the math coprocessor, and that thing was a game changer. It cut the processing times by at least half. I can really, really appreciate what a small, modern day cell phone is capable of in terms of compute and graphics power. Even smart watches are more powerful than my first computer.
It's intresting that for some reason before ~2002 it was possible to run Linux(up to 2.0?) and NetBSD (for sure 1.6.1) at just 2MB of RAM.
For some reason most modern operating systems requires 4MB of ram - even if kernel size is far less (like 700kb), as there is SOMETHING hardcoded in the kernel to prevent boot under 4MB of RAM. I think this could be due the "large memory pages" switch from 4kb to 4MB - but I never found anyone knowledge enough to confirm it :(
Anyone?
https://en.wikipedia.org/wiki/Chips_and_Technologies and
https://en.wikipedia.org/wiki/Tseng_Labs_ET4000 was really fast.
For instance Smalltalk/V 286 ran very well on it. Compared to a 386sx@25Mhz with more RAM.
I imagine that our hand held devices will at some point replace our desktops alltogether, allowing plug and play wireless connections to our monitors, etc and letting us run the most advanced software & games directly on the same devices we currently use to text and make phone calls. But we're not there yet
I'm sure this was done for marketing in some cases but I think Intel also worked out how to reclaim chips were the FPU was damaged as well.
A Packard Bell 486DX2 66Mhz.