Suppose we all have our 'thing' in this game - mine just happens to be dual boards (must have a dozen or so now). Although not all in working systems I'll happily use a dual system for whatever I need, whether that benefits from dual cpus or not. Guess I just care more about the tech side of it than the perceived practicalities. Also, there are still some ridiculous deals to be had on these boards as I suspect sellers think they won't be able to so easily shift old workstation / server gear - I'm not complaining 😀

In running period correct games there is no advantage whatsoever in having a dual Pentium 3 versus a single one. However in multitasking there is a huge and visible advantage! Also when using Windows XP there is a tangible performance benefit (actually enormous) in running a dual processor system; using Windows 2000 does not feels like having such a performance boost when going dual.
Case in point:
System 1:
Pentium 3 1000 Coppermine FSB133
3*512mb SDRAM PC133
Abit VH6T
Abit geforce4 ti4200 128mb
Windows XP
System 2:
Pentium 3 1000 Coppermine FSB133 * 2
3*512mb SDRAM PC133
Abit VP6
Abit geforce4 ti4200 128mb
Windows XP
System 2 feels snappier and more responsive with XP; the boot times are lower and also the shutdown times; everything runs better including PCMARK 2002 and 2004; 3dmark scores are identical so in games there is no difference but in everyday tasks the dual processor system is running far more smoothly; you can have winamp, Nero burning a DVD, Windows update running and a game with no visible slowdowns!

As of last year I set myself a new rule of only getting systems with 2 sockets or above.

On the plus side you can play around with high end hardware.
and if your just gaming it's not going to benefit you, but not really hurt either.

But if your purely a gamer, makes no sense whatsoever.

My main PC back in the early 2000s was a 2nd hand dual P2 Xeon system. It ran Windows 2000. When I added the 2nd CPU I did notice it booted a lot faster. It was odd that the taskbar icons would load in different orders each time because of inconsistency in how each was getting dispatched to either CPU during the boot process.
There was a period of about a year when I used to run Norton, and it had a bug that would sometimes consume 100% CPU time in a single process. Since I had 2 CPUs I didn't notice it until looking in the task manager. Other people on the internet were screaming about how that bug caused their computer to slow to a crawl.
I don't remember if it was Norton or Microsoft Outlook that did that.

Although there were some fringe benefits, I think the value for typical desktop use was somewhere between marginal to moderate at best. It was a cheap upgrade (I got the 2nd CPU for like $10 or something) so it was easily worth it, but if I was building a system with brand new parts then it probably wouldn't have been. I remember considering building a dual AthlonMP machine around that time and I think I was right to not do it.

The greatest use I ever had for the dual CPUs was when I started encoding MPEG-2 videos on that (dual P2) computer. Video encoding scales very well with dual CPUs and it did run twice as fast when set to use 2 threads. But that didn't come up until such a late date that normal people were using Athlons and P4s, so I wasn't exactly blowing anybody's doors off with a dual P2.

As I started collecting used computer hardware, I upgraded to a single overclocked Coppermine P3 on an Asus P2B-F. When I landed a dual capable P2B-D, I had the SMP fascination out of my system so I decided to sell that board for money instead of using it. If it could actually clock the same as the single CPU board then I might have been interested in keeping it.

..
Much more recently, like a few years ago, I built a dual P3-933MHz system on a VIA motherboard with the 694DX (or whatever that chipset is called). I had seen some attempts at running Skyrim on a Pentium 3 and wanted to try my hand at it. This is not a practical application, obviously. I didn't get far enough with that to experiment with single vs dual CPU, but I'm just assuming it helps with games from 2011. Not at all with games from 2001 though.

System 2 feels snappier and more responsive with XP; the boot times are lower and also the shutdown times

Adding SSD with negate such difference.

There's no sensible reason for dual CPU's unless you're actually running the applications they were made for - 90's desktop productivity. It's not about performance, it's about having two CPU's! It's just... neato.

cyclone3d wrote on 2020-06-30, 20:14:

Well, if you run an OS that supports more than one CPU core, then you do gain some performance even with single threaded games because there is a second core to handle background stuff instead of the single CPU having to task switch constantly.

Super easy to do a comparison as you can just remove a single CPU so you are testing on the same exact system and not having to think about performance differences between different hardware combinations.

A dual socket system will most likely be faster with two CPU's than one, but high performance single socket systems beat it either way (while also supporting a larger range of CPU's and overclocking). Comparatively low volume industrial/work station boards aren't optimized for speed, especially not real time graphics performance. For a single threaded application, you are losing performance going with a dual socket vs. a good single socket.

The Serpent Rider wrote on 2020-07-01, 11:21:

System 2 feels snappier and more responsive with XP; the boot times are lower and also the shutdown times

Adding SSD with negate such difference.

Kingston 120gb over SATA to IDE adapter in both systems!

maxtherabbit wrote on 2020-07-01, 02:24:

my dual coppermine setup runs XP on a Tyan Tiger 100 BX board with AGP (Radeon 9700 Pro) and ISA (using an ALS007 sound card)

interesting, most of the dual pentium3 boards i have seen come without isa slot.
Do you know any good PCI soundcard that works with DOS games and sounds exactly the same as a soundblaster16 ?

As the self proclaimed king of multi cpu systems here, there is really only one point. I like them.

The WHOLE 'hobby' of screwing around with vintage hardware is empirically a silly thing. Why program a game for DOS when you can do the same work for a new game for new systems? Why bother gaming on old hardware when you can emulate it?

Once you get into the era where games can even SEE the 2nd cpu, you can just run those games on a new system (in most cases).

The only advantage to a dual cpu setup with single core software is, cpu2 can do background tasks while cpu1 can run the game at full speed. (requires a multi cpu capable OS still)

*EVERYONE* has a fetish.

See, this is why I thought the whole dual core thing was stupid. Practically all software mainstream users had was single threaded. So we got a bunch of benchmarks and hype over what, initially, had no effect.

Of course I felt that way about a lot of things. Consider first generation SATA drives, which in some cases were slower than their ATA counterparts due to controller woes. Or the first PCIE cards - more bandwidth, but why, there was no card that could use it. In time it all made sense. But the dual core thing, damn, it was years and years before that had a material impact on what you were doing.

Same thing with 64 bit. I got a 64 bit processor in 2003. Absolutely no use for it, XP-64 was renowned for its incompatibilities and driver woes, that's just what AMD was selling at the time. So it sounded cool but it was a completely academic capability.

luckybob wrote on 2020-07-04, 07:45:

As the self proclaimed king of multi cpu systems here, there is really only one point. I like them. […]
Show full quote

As the self proclaimed king of multi cpu systems here, there is really only one point. I like them.

The WHOLE 'hobby' of screwing around with vintage hardware is empirically a silly thing. Why program a game for DOS when you can do the same work for a new game for new systems? Why bother gaming on old hardware when you can emulate it?

Once you get into the era where games can even SEE the 2nd cpu, you can just run those games on a new system (in most cases).

The only advantage to a dual cpu setup with single core software is, cpu2 can do background tasks while cpu1 can run the game at full speed. (requires a multi cpu capable OS still)

*EVERYONE* has a fetish.

Hear hear!

I fully 2nd luckybob

I had a dual P3-450 system back in the day, and while it was good for desktop use (with windows 2000) it was fast and responsive, for gaming, it made zero difference

hwh wrote on 2020-07-04, 08:23:

See, this is why I thought the whole dual core thing was stupid. Practically all software mainstream users had was single threaded. So we got a bunch of benchmarks and hype over what, initially, had no effect.

Wrong, as soon as you were multitasking, even with purely single-threaded stuff, you reaped instant and big benefits. If all you do is use the PC as a game console, running one single-threaded game only, then indeed, no benefit. But if you were running a word processor, an MP3 player (in P3-days, Winamp was a non-negligible load) or - perhaps most telling - wanted play a game while you waited for a CD to burn, the benefit was immediate. With single CPU, that CD burn failed if the game ever pushed the CPU near 100%. With dual CPU, the CD burnt correctly.

For retro gaming the multitasking use case is less relevant, but in the day it was the reason why these things sold at all outside of the 3DStudioMax/Autocad community. But the time the first dualcores came along a few years later, things like Photoshop has also gone multithreaded and the utility increased - but don't overestimate how much stuff is still single-threaded even today. The vast majority of programs, even games, still only uses one thread. Thing is you're running hundreds of them concurrently. More cores mean faster servicing of processes waiting for CPU. If you don't think that matters I challenge you to limit yourself to single-threaded programs and disable all but the first core on whatever CPU you most modern machine uses. You will have a measureable but more importantly a noticeable slowdown.

Of course I felt that way about a lot of things. Consider first generation SATA drives, which in some cases were slower than their ATA counterparts due to controller woes. Or the first PCIE cards - more bandwidth, but why, there was no card that could use it. In time it all made sense. But the dual core thing, damn, it was years and years before that had a material impact on what you were doing.

I clearly remember upgrading from an So939 Athlon64 to an Athlon64 X2 (the first mainstream dualcore) somewhere in late 2005 and being amazed by the increased usability of the system, despite clock speed & cache remaining the same (3200+ > X2 3800+)

Take a look at the multitasking benchmarks here. The very first dualcore in-depth review shows massive improvements in performance running two (single-threaded) applications at the same time.

Same thing with 64 bit. I got a 64 bit processor in 2003. Absolutely no use for it, XP-64 was renowned for its incompatibilities and driver woes, that's just what AMD was selling at the time. So it sounded cool but it was a completely academic capability.

It was all about memory, or rather address space. In 2003, 1GB was a lot of RAM and a normal user wasn't going to hit any limits, but 2003 was when the first mainstream memory controllers were able to support 4GB RAM. It therefore made sense to have a CPU that could also address all of it natively, and to have an OS that could support that. Of course it was a niche application back then, unless you were going to hit your 32b 4GB barrier you were better off with 32b, particularly given the bad state of OS support (Linux 64b libraries and 32b compatibility were also a mess at the time), but it added a lot of value to those users who actually needed the >3GB RAM.

By 2008 4GB of RAM was mainstream and OS support had improved to a generally usable level (as far as Vista could ever really be considered usable).

But back to retro stuff: the argument that SMP makes no sense if you only play games is more than valid, but that it made no sense for all-round systems back in the day: no way.

SMP was taken mainstream because CPU manufacturers ran into a performance wall and couldn't get much more out of a single CPU core anymore. It also made a nice new subject to get people more interested in new CPUs. SMP programming is difficult and not feasible for many applications, so for mainstream usage it's almost always preferable to have a single, faster CPU. But that stopped being a viable roadmap when the improvement of single cores was decelerating.

A big hindrance to the adoption of AMD64 features was that WinNT is a restrictive environment. Back in the days of DOS, any program could use 386/486/etc 32-bit instructions and DOS didn't do anything to interfere. If you locked the system, DOS didn't care. But with Windows, if WinXP32 didn't understand AMD64 then no program running under it could take advantage.

SETBLASTER wrote on 2020-06-30, 17:41:

But what is the purpose of building such machines? just cosmetics to show something that is not common for home use?
or is there a real benefit of building that for playing DOS, Windows3.1 and windows98 games?

Several have mentioned this but I think it's worth reiterating. Not only is there no benefit on DOS or Win9x, there is zero benefit. The second CPU will be ignored.
Writing an operating system for multiple CPUs greatly complicates things. There is a lot of locking/unlocking to do with things in memory so that multiple CPUs don't stomp on each other. Here is some high level tech info about it: https://en.wikipedia.org/wiki/Giant_lock
It's kind of amazing to me that there were some SMP versions of Mac OS 9. I can't imagine attempting SMP without even having memory protection. MS never attempted this with Win9x.

If you have some interest in messing with NT or Linux, go for it... otherwise unless the second CPU makes the board oddball enough to get a low price or anything, it doesn't accomplish anything.

dionb wrote on 2020-07-04, 12:53:

But back to retro stuff: the argument that SMP makes no sense if you only play games is more than valid, but that it made no sense for all-round systems back in the day: no way.

Wrong, your unsourced assertion that average users not only ran multiple programs at once but that those programs by themselves on average used 100% of the processor makes no sense. Even now, we are limited as people to doing one thing. If the computer needs time, we can switch to a different thing. Of course if your software is multithreaded you can tie it all up.

The dual core chip is not twice as fast as the single chip; that's the point. You need to have the need to run multithreaded software or the habit of running multiple programs at once, which, back in the day, was a foolish thing as your computer would crash and your CDs would fail to write and generally you got poor performance when you tried to multitask.

hwh wrote on 2020-07-06, 08:20:

dionb wrote on 2020-07-04, 12:53:

But back to retro stuff: the argument that SMP makes no sense if you only play games is more than valid, but that it made no sense for all-round systems back in the day: no way.

Wrong, your unsourced assertion that average users not only ran multiple programs at once but that those programs by themselves on average used 100% of the processor makes no sense. Even now, we are limited as people to doing one thing. If the computer needs time, we can switch to a different thing. Of course if your software is multithreaded you can tie it all up.

Run any CPU-limited game and you hit 100% on one CPU. Run anything else at the same time and something is going to have to wait. That's not rocket science.

But you want a source? How about this:

Use case: you're using MS Office and WinZip and your antivirus program kicks in in the background. That's not an exotic edge-case, it's something that still happens to us daily, and did 20 years ago as well. Instead of "MS Office and WinZip", you can read any game you happen to be running, or indeed burning that CD.

The dual core chip is not twice as fast as the single chip; that's the point. You need to have the need to run multithreaded software or the habit of running multiple programs at once, which, back in the day, was a foolish thing as your computer would crash and your CDs would fail to write and generally you got poor performance when you tried to multitask.

It was only foolish if your hardware couldn't handle it. In fact you're proving my point here: if you wanted to multitask, you had a valid use case for multiple CPUs/cores.

With my dual p3 system it allowed me to not only host a Quake 3 or original counter strike server but play the games on the system at the same time without issues. I set the server to use cpu 1 and the game itself to use cpu 2 and it worked beautifully. When I would try that on a single cpu system it would become a jittery mess.

So in other words, yeah if you want to run two heavy duty programs at the same time for reasons, having a separate CPU is nice. I hope that's not surprising and isn't worthy of starting an argument over.

Hyperthreading seemed to be pretty satisfying for some people before dual cores too. I wonder if that would smooth out the above game + server scenario's jitters.