canthearu wrote on 2020-05-25, 12:38:

... like who wants to listen to old quantum fireball drives 😀

Me. You don't like the sound of old fireballs? Those drives were great. I wish somebody would invent an HDD sound emulator for SD cards. I don't like the spindle noise much, but seek noises are good.

Anonymous Coward wrote on 2020-05-26, 10:05:

canthearu wrote on 2020-05-25, 12:38:

... like who wants to listen to old quantum fireball drives 😀

Me. You don't like the sound of old fireballs? Those drives were great. I wish somebody would invent an HDD sound emulator for SD cards. I don't like the spindle noise much, but seek noises are good.

Yeah, seek noises ARE great. But the idle whine (spindle noise) on the old fireballs is a whole lot of "NO THANKS!"

Intel486dx33 wrote on 2020-05-25, 14:49:

starcube wrote on 2020-05-25, 14:00:

Reading your own benchmark results really, really helps: […]
Show full quote

Intel486dx33 wrote on 2020-05-25, 13:50:

As you can it is very close to your score.
So there was not much improvement in going with a PCI motherboard.

Reading your own benchmark results really, really helps:

1. My system memory bandwidth is 244% higher than yours.
2. My system memory and cache memory throughputs are 35% higher than yours.
3. My video memory bandwidth is 257% higher than yours (PCI > VLB).
4. My TopBench score is 26% higher than yours (237 > 199).
5. My disk transfer rates are twice of yours.
6. Even my CPU score is higher than yours even though we have the same processor.

I am highly amused at your continuing attempts to put my system down. It's becoming rather silly.

Yes, but your CPU score is only 2% higher than mine. That is the point the CPU is topped out at about 40 points in Speedsys CPU test
I am using an Intel Over drive 486dx4-100 CPU NOT the Standard 486dx4-100 CPU.

His Intel DX4 is (as I assume) set to WB instead of WT. Perhaps yours is set to WT, which could be a bit slower. Otoh, my memory is a bit dusty on whether overdrives could actually be re-set in this way (or if they were stuck using WB) as typically overdrives lacked some of the basic flexibility functionalities that the ordinary CPUs would have back then.

With 486s, there were a lot of other things that could determine the system's performance as there were also many different chipsets around with a ridiculous amount of different busses, all connected to each other in different ways with a lot more tweaking options then we have today. And today, age has also become a factor and modern parts like SD cards were also not around back then.

I find it always interesting to see 2 vastly different systems of the same generation compared to each other. Even if they were compared to mine I would still enjoy learning about it as it can perhaps teach me a few new things. I never see it as something personal as someone else will probably have build his own retro rig with a different purpose.
I personally never cared whether someone elses rig of the same platform is vastly faster then mine (chances are they usually were) but when it comes to retro rigs, wanting more speed is easiest done by simply building around a more recent platform. It's also why I'm usually not a fan of (wildly) overclocking old hardware and especially if it will kill parts that are already harder to come by.
But I do find the rough data interesting regardless of whether the faster system was build by me or not.

I've always build my own systems for stability and out of curiosity for something and I've had only very few systems that somehow ended up being very crachy with the vast majority basically doing what they were supposed to do.

Anonymous Coward wrote on 2020-05-26, 10:05:

canthearu wrote on 2020-05-25, 12:38:

... like who wants to listen to old quantum fireball drives 😀

Me. You don't like the sound of old fireballs? Those drives were great. I wish somebody would invent an HDD sound emulator for SD cards. I don't like the spindle noise much, but seek noises are good.

The higher pitched continuous whining I find very annoying, but the coffeegrinder noises when they are reading/writing are fine in my book 😜
I've forgotten how many of those older drives sounded like and the laptop drives I ended up using in my latest bacthes of retro rigs were awefully quiet 😜

PARKE wrote on 2020-05-25, 16:00:

starcube wrote on 2020-05-25, 14:48:

I just looked on eBay, there are a few for sale, but the prices are not very appealing, especially for a CPU with an easily cracked core.

They are in general prohibitively expensive and maybe only worth it if you are seriously after the fastest s370 coppermine for an in-spec 440bx system.
Anyway, for general interest here an image of the failed Intel attempt to market a >1000 Mhz cpu in 2000:1113Mhz2.jpg

I can't remember ever having seen this design slot 1 HSF, but it's clearly based on the same designs that Intel's later s370 stock HSFs were based on (but before their slightly later s370 Tualatin-era stock HSFs that often used those (now often very brittle) green plastic clamps).

Even though I like the idea of clamping a fan assembly onto the metal HS in this way, those fans were so annoying for me to listen to that I usually opted to remove the fanassemblies from the metal heatsink and attached some other fan(s) to it 🤣

But still interesting to see this heatsink design used in this way.

Here are some speedsys scores using the same PCI type of motherboard but with AMD 5x86-133 CPU’s over clocked to 160mhz.
The Average CPU score is 60-points compared to the 40-points on the 486dx4-100
So there is a 33.3% improvement in CPU performance. ( just for comparison ).
The CPU/Motherboard combo of this 486dx4-100 build just did not make sense to me but it looks like it
Did improve performance in other areas than the CPU.

Intel486dx33 wrote on 2020-05-26, 11:07:

Here are some speedsys scores using the same PCI type of motherboard but with AMD 5x86-133 CPU’s over clocked to 160mhz. The Ave […]
Show full quote

Here are some speedsys scores using the same PCI type of motherboard but with AMD 5x86-133 CPU’s over clocked to 160mhz.
The Average CPU score is 60-points compared to the 40-points on the 486dx4-100
So there is a 33.3% improvement in CPU performance. ( just for comparison ).
The CPU/Motherboard combo of this 486dx4-100 build just did not make sense to me but it looks like it
Did improve performance in other areas than the CPU.

Of course the CPU score is faster than a 486DX4 - YOU CHANGED THE CPU!

Your posts really don't make any sense to me.

canthearu wrote on 2020-05-26, 10:13:

Anonymous Coward wrote on 2020-05-26, 10:05:

canthearu wrote on 2020-05-25, 12:38:

... like who wants to listen to old quantum fireball drives 😀

Me. You don't like the sound of old fireballs? Those drives were great. I wish somebody would invent an HDD sound emulator for SD cards. I don't like the spindle noise much, but seek noises are good.

Yeah, seek noises ARE great. But the idle whine (spindle noise) on the old fireballs is a whole lot of "NO THANKS!"

Lords, I hated the high-pitched spindle whine of early hard drives, it's like something's drilling into your skull. They got a LITTLE quieter around 1999, not as ear-piercing but still annoying.

When fluid-dynamic bearings became a thing in 2003, I was praising the gods of storage for finally granting my prayers for silence.

All my rigs currently use drives with FDBs because like you I love the seek noises but hate the spindle whine. And it just doesn't feel fully retro without a mechanical hard drive.

(Full disclosure: My 486 uses a quieter revision of the ball-bearing 8.4GB WD Caviar, and the location of the drive in the case makes it less audible to the outside.)

starcube wrote on 2020-05-26, 12:28:

Intel486dx33 wrote on 2020-05-26, 11:07:

Here are some speedsys scores using the same PCI type of motherboard but with AMD 5x86-133 CPU’s over clocked to 160mhz. The Ave […]
Show full quote

Here are some speedsys scores using the same PCI type of motherboard but with AMD 5x86-133 CPU’s over clocked to 160mhz.
The Average CPU score is 60-points compared to the 40-points on the 486dx4-100
So there is a 33.3% improvement in CPU performance. ( just for comparison ).
The CPU/Motherboard combo of this 486dx4-100 build just did not make sense to me but it looks like it
Did improve performance in other areas than the CPU.

Of course the CPU score is faster than a 486DX4 - YOU CHANGED THE CPU!

Your posts really don't make any sense to me.

Never mind. I was just confused about what you were trying to do with your 486dx4-100 build. I am mainly focused on best game performance per computer build. And after the 486dx2-66 DOS game performance does not improve enough to play modern DOS games in good quality. You need to upgrade to a 1st gen Pentium for that. But you already know that.

Tetrium wrote on 2020-05-26, 10:50:

With 486s, there were a lot of other things that could determine the system's performance as there were also many different chipsets around with a ridiculous amount of different busses, all connected to each other in different ways with a lot more tweaking options then we have today.

This is the point I was trying to make to him this entire time.

The motherboard is half the equation for system performance but he just keeps going on about how the CPU scores are close.

Of course they are - they're the same CPU!

Tetrium wrote on 2020-05-26, 10:50:

His Intel DX4 is (as I assume) set to WB instead of WT. Perhaps yours is set to WT, which could be a bit slower. Otoh, my memory is a bit dusty on whether overdrives could actually be re-set in this way (or if they were stuck using WB) as typically overdrives lacked some of the basic flexibility functionalities that the ordinary CPUs would have back then.

Initially 486DX2 and 486DX4 only supported WT cache.

Later revisions supported both WB and WT. You can distinguish them by the stepping number.

My DX4 is indeed the WB variant.

starcube wrote on 2020-05-26, 13:05:

Initially 486DX2 and 486DX4 only supported WT cache. […]
Show full quote

Tetrium wrote on 2020-05-26, 10:50:

His Intel DX4 is (as I assume) set to WB instead of WT. Perhaps yours is set to WT, which could be a bit slower. Otoh, my memory is a bit dusty on whether overdrives could actually be re-set in this way (or if they were stuck using WB) as typically overdrives lacked some of the basic flexibility functionalities that the ordinary CPUs would have back then.

Initially 486DX2 and 486DX4 only supported WT cache.

Later revisions supported both WB and WT. You can distinguish them by the stepping number.

My DX4 is indeed the WB variant.

The vast majority of Intel DX2 and DX4 chips that I found in the wild (as in dumpsterdiving) were WT. I think I ended up with very few WB Intel DX2/DX4 chips because of that.
There were also several AMD variants of these chips. Some with 8KB cache and some with 16KB and then there were the chips of the other manufacturers. The 486 era was exciting for sure!

I don't know why but I could never manage to match or have faster 486 performance with pci video cards when comparing to my vlb system. If I put my 486 33Mhz cpu to LS-486E motherboard with pci video cards, it always gets worse performance than when I am using the cpu in my normal vlb system. Tried many pci video cards from S3 trio64 and S3 Virge/GX to a pci voodoo 3 card. Many people say that PCI is better but I can't confirm that myself.

Baoran wrote on 2020-05-27, 00:12:

I don't know why but I could never manage to match or have faster 486 performance with pci video cards when comparing to my vlb system. If I put my 486 33Mhz cpu to LS-486E motherboard with pci video cards, it always gets worse performance than when I am using the cpu in my normal vlb system. Tried many pci video cards from S3 trio64 and S3 Virge/GX to a pci voodoo 3 card. Many people say that PCI is better but I can't confirm that myself.

I think your 33MHz processor is the real culprit here, bottlenecking the PCI graphics cards. You'd see a difference at 100MHz and above.

Tetrium wrote on 2020-05-26, 23:49:

The vast majority of Intel DX2 and DX4 chips that I found in the wild (as in dumpsterdiving) were WT. I think I ended up with very few WB Intel DX2/DX4 chips because of that.
There were also several AMD variants of these chips. Some with 8KB cache and some with 16KB and then there were the chips of the other manufacturers. The 486 era was exciting for sure!

I think it's because by the time the WB variants were released in October 1994, Pentium machines were already mainstream.

starcube wrote on 2020-05-26, 12:39:

This is the point I was trying to make to him this entire time. […]
Show full quote

Tetrium wrote on 2020-05-26, 10:50:

With 486s, there were a lot of other things that could determine the system's performance as there were also many different chipsets around with a ridiculous amount of different busses, all connected to each other in different ways with a lot more tweaking options then we have today.

This is the point I was trying to make to him this entire time.

The motherboard is half the equation for system performance but he just keeps going on about how the CPU scores are close.

Of course they are - they're the same CPU!

It’s like a kit car. Where we both have the same engine. But my kit car is basic model and yours is a sports car model. It really does not make the engine run faster. It just looks after.

In 486 world PCI is not in general faster than VLB. Good VLB boards are equal to good PCI boards at same clock. And measuring the Bus-Speed with Speedsys is not the best method to find out the true bus performance.

In my experiences with comparing L1 cache strategies (using intel/AMD DX2 and DX4 with SIS and UMC chipsets), L1WB makes nearly zero to L1WT in Application-Benchmarks (Doom, Quake, PCP-Bench, Winstone), if you have 256kB or more L2 in WB. Surprisingly, in case of intel DX2 the L1WB/L2WB was a little bit slower than L1WT/L2WB. The 486 WB CPUs from intel where intended for low cost systems without L2 cache.

@starcube
Its a beautiful and well designed 486 system! But why you chose a 486 ALI chipset for a dream system? 😉

Intel486dx33 wrote on 2020-05-27, 05:53:

starcube wrote on 2020-05-26, 12:39:

This is the point I was trying to make to him this entire time. […]
Show full quote

Tetrium wrote on 2020-05-26, 10:50:

With 486s, there were a lot of other things that could determine the system's performance as there were also many different chipsets around with a ridiculous amount of different busses, all connected to each other in different ways with a lot more tweaking options then we have today.

This is the point I was trying to make to him this entire time.

The motherboard is half the equation for system performance but he just keeps going on about how the CPU scores are close.

Of course they are - they're the same CPU!

It’s like a kit car. Where we both have the same engine. But my kit car is basic model and yours is a sports car model. It really does not make the engine run faster. It just looks after.

I don't think this comparison is helping your argument.
Drivetrain, suspension, tires and brakes make a distinct difference when you know how to drive your car and you're actually trying to push it. And that is assuming they're using the same chassis. Different chassis introduce a huge difference. Even on the daily commute an optimised gearbox and tires with low rolling friction will make a relevant difference in fuel consumption if you're willing to use those features.
Plus modern engines behave very differently depending on the software they're running even if all the physical parts are identical.
In any case the engine is only part of the equasion. I think the same applies to CPUs in computers. 😀

starcube wrote on 2020-05-27, 01:12:

Baoran wrote on 2020-05-27, 00:12:

I don't know why but I could never manage to match or have faster 486 performance with pci video cards when comparing to my vlb system. If I put my 486 33Mhz cpu to LS-486E motherboard with pci video cards, it always gets worse performance than when I am using the cpu in my normal vlb system. Tried many pci video cards from S3 trio64 and S3 Virge/GX to a pci voodoo 3 card. Many people say that PCI is better but I can't confirm that myself.

I think your 33MHz processor is the real culprit here, bottlenecking the PCI graphics cards. You'd see a difference at 100MHz and above.

Does the cpu affect vidspeed that much? I am getting up to 50% higher numbers in vidspeed with vlb card compared to my pci video cards.
I am getting numbers of over 33k with my vlb card while pci cards are somewhere between 20k and 25k in vidspeed.

PC-Engineer wrote on 2020-05-27, 05:55:

In 486 world PCI is not in general faster than VLB. Good VLB boards are equal to good PCI boards at same clock. And measuring the Bus-Speed with Speedsys is not the best method to find out the true bus performance.

I would be very interested in some actual numbers backing this up. Sounds rather implausible given how faster, more capable graphics and disk controller chipsets only exist in PCI form. We're not discussing bus speeds here, but whole-system performance.

PC-Engineer wrote on 2020-05-27, 05:55:

In my experiences with comparing L1 cache strategies (using intel/AMD DX2 and DX4 with SIS and UMC chipsets), L1WB makes nearly zero to L1WT in Application-Benchmarks (Doom, Quake, PCP-Bench, Winstone), if you have 256kB or more L2 in WB. Surprisingly, in case of intel DX2 the L1WB/L2WB was a little bit slower than L1WT/L2WB.

I think you shouldn't be mixing Intel and AMD chips in your tests as AMD chips are significantly slower at floating point math than Intel chips at the same clockspeed.

WB L1 on an Intel chip gives you up to a 15% boost in FPU-intensive workloads, so I am actually surprised at your results. Quake especially, since it's so FPU-heavy.

PC-Engineer wrote on 2020-05-27, 05:55:

The 486 WB CPUs from intel where intended for low cost systems without L2 cache.

Where did you read that? That really makes no sense to me. A system with no L2 would be dog-slow regardless of the L1 mode and by the time the DX4 WB appeared (October 1994) cache was cheap.

Having a DX4 WB in a L2-less system would literally make as much sense as having a Ferrari engine in a Fiat 126.

PC-Engineer wrote on 2020-05-27, 05:55:

@starcube
Its a beautiful and well designed 486 system! But why you chose a 486 ALI chipset for a dream system? 😉

Because if you read the datasheets, only the ALi chipset is designed to actually take advantage of EDO DRAM. While late SiS and UMC will boot with EDO DRAM (early revisions will not), they don't actually support EDO data rates and just run it as FPM. 😉