Quake hits the FPU, but a FPU specific bench might show more of a difference.

In order to do a truly final test, you'd have to take several AMD processors and several Intels, benchmark them, and average the results. I could easily see those tiny differences arising between, for instance, two seemingly identical Intel chips from different batches.

That's in that 3% margin of error, so I consider it a tie. Interesting though, that when there is a difference, it's always the Intel that is on top. 😀

keenmaster486 wrote:

In order to do a truly final test, you'd have to take several AMD processors and several Intels, benchmark them, and average the results. I could easily see those tiny differences arising between, for instance, two seemingly identical Intel chips from different batches.

It's the motherboard that can cause a difference because of varying clock speeds. But I used the same and tested the CPUs within minutes 😀

clueless1 wrote:

That's in that 3% margin of error, so I consider it a tie. Interesting though, that when there is a difference, it's always the Intel that is on top. 😀

I know right?

I basically got the exact same results. Same most of the time but when there is a difference the Intel is very slightly ahead.

Cyrix dx2 66mhz varies. on a few tests is was slightly ahead but usually it was lagging behind a little bit like in DOOM and Quake benchmarks.

Cool. I don't have any other chips, would have been nice to test them also. It gets a bit confusing with the many versions. There are also 16 KB Cache WB versions, I found an Intel model locally, so that will be interesting to compare.

I just want to confirm, there was no sound-card involved? I would leave it out, to get a better idea of core processing ability, but I would also run it with something extremely common, such as an SB16 to get a real-world benchmark in games.

Logistics wrote:

I just want to confirm, there was no sound-card involved? I would leave it out, to get a better idea of core processing ability, but I would also run it with something extremely common, such as an SB16 to get a real-world benchmark in games.

No sound card was used.

So... a "reverse engineered" Intel 486 DX2, is as fast as the original chip.
Or did AMD had a license to made 486 clones ?

As far as I know 486 AMD chips are intel clones - same architecture, same design, so basically identical to their intel counterparts - witch can be seen in your results as the margin is way too low to take into consideration. Generally a intel or AMD 486 cpu with the same specs will perform identically.

What I'm trying to say is that you will very likely see the same discrepancy between two different intel DX2-66 chips, or two different AMD DX2-66 chips.

Cyrix CPUs are another matter. They are an original in-house design, as such they will perform differently from same clocked intel/amd counterparts. The cyrix 486 series is a little slower, while a 100MHz cyrix 586 is quite a bit faster then both AMD and Intel 100Mhz DX4's - a gap that widens even more when enabling 686 enhancements.

Good job on the comparison!

The difference is a slight bit bigger when you get to the DX4's, as Intel apparently made some small enhancements that AMD didn't manage to copy. Maybe the difference is already there between write-back versions of the DX2?

firage wrote:

Good job on the comparison!

The difference is a slight bit bigger when you get to the DX4's, as Intel apparently made some small enhancements that AMD didn't manage to copy. Maybe the difference is already there between write-back versions of the DX2?

What strikes me about intel's DX4 chips is how poorly they overclock - if at all. A 66MHz AMD DX2 will often OC to 100MHz, but an intel version rarely will - same for the 100MHz parts - all AMD 100MHz chips i've come across OC do 120 stable - as for intel DX4 chips, I only have one that does 120MHz, but it requires 3.8 volts to run windows, 4v to run quake.

Also, if I recall corectly, AMD launched their 100MHz chip (8kb model) in 1993 while intel launched theirs in late spring the following year. This is an understandable move, since boosting the 486's performance would eat into intel's upcoming Pentium processor sales, so releasing a faster model would practically mean shooting themselves in the foot. This is why intel's fastest 486 is the 100MHz DX4. There were rumors of a 120Mhz chip, but they were quicly silenced a few months before the pentium's launch.

what I see is that your constrained by all your benchmarks. since they all score identical, they all have more headroom than the benchmark allows for.

i did some tests between intel and amd dx4-100, both are 16kb WB cache versions, and the conclusion is totally different:
intel is significantly faster(by 10-15%) in integer benchmarks, while amd is noticeably faster(by 3-5%) in floatpoint benchmarks.

clueless1 wrote:

That's in that 3% margin of error, so I consider it a tie. Interesting though, that when there is a difference, it's always the Intel that is on top. 😀

Not surprising, since AMD basically just reverse-engineered the 386 and 486 designs from Intel. They're pretty much a carbon-copy. They didn't actually design an x86 CPU of their own until the K5.
The Am486DX2-66 is probably an early one. They later modified the design slightly with more/faster caches and such. Then again, so did Intel. So later 486s will have some more variation than early ones.

kanecvr wrote:

What strikes me about intel's DX4 chips is how poorly they overclock - if at all. A 66MHz AMD DX2 will often OC to 100MHz

Oh really?
That's more of a 'luck of the draw', or at least having a very late model, built on a mature process, more than anything.
I had an early Am486DX2-66 back in the day, and it wouldn't overclock at all... in fact, it needed a heatsink and fan just to reach 66 MHz. I couldn't get it to 80 at all.
The thing eventually burnt out (at its stock 66 MHz speed no less).
I then replaced it with an Intel 486DX2-66, which was an OverDrive model with a heatsink glued on. It could run 66 MHz fanless, and easily did 80 MHz when I put the fan from the AMD one on. That CPU is still running today... at 80 MHz.

And I can also vouch for the benchmark results... In fact, in my case I didn't see any difference *at all*.
The only difference I ever noticed is that there's a bug in the Am486DX2-66 which prevented me from running OS/2, while the Intel one ran OS/2 fine in the same system.

agent_x007 wrote:

So... a "reverse engineered" Intel 486 DX2, is as fast as the original chip.
Or did AMD had a license to made 486 clones ?

Nope, AMD never had a license for using Intel's 386 or 486 designs.
AMD claimed they did, and just cloned the CPUs. Intel sued them, and AMD was found guilty of using Intel's copyrighted microcode.
However, because of the economic importance of the x86 architecture, the court decided that Intel had to license the instructionset to third parties. Which is how the AMD-Intel x86 cross-license first came into place. After rewriting the microcode, AMD could finally start selling their 386 clone, and the 486 followed shortly after.

The differences seen here are probably because of slightly different microcode only.

No, not quite. The cross license was originally in place as a condition of IBM using Intel technology in the PC. Due to a squabble, AMD did not have access to the 386 and 486 design from Intel as they had with previous designs. When they came out with their own versions based on reverse engineering, Intel charged them with taking shortcuts (they did violate the clean room to reverse engineer the microcode).

The court accepted AMD's settlement offer, I believe at least partly due to the fact that the original cross licensing agreement would have allowed AMD to pay Intel to use their technology until 1996 had Intel not dissolved it. The settlement preserved that status quo.

gdjacobs wrote:

No, not quite. The cross license was originally in place as a condition of IBM using Intel technology in the PC.

That's not a cross-license.
That was the original license for production of second-source x86 CPUs. It was about chip designs, which is why it had to be renewed for every new CPU.
The cross-license is a completely different thing, and deals with the x86 on a more abstract level (intellectual property), such as instruction sets and patents required to implement an x86-compatible CPU.

That is exactly what I meant by saying "AMD never had a license for using Intel's 386 or 486 designs".
The original license was a condition from IBM for the original PC, for 8088, and later extended to 286 for the AT.
By the time the 386 arrived, there were plenty of clone manufacturers, and Intel didn't solely need to rely on IBM. As such, Intel no longer cared about IBM's conditions, and the license was never extended beyond 286 CPUs, and production of subsequent x86 CPUs from Intel was never outsourced. IBM probably didn't care either, since Intel had become much larger and more powerful, and at the same time IBM had become a smaller player in the PC market.

AMD used this original license to claim that the terms allowed them to manufacture any x86 CPU from Intel. However, Intel never handed them the designs for anything newer than the 286, so AMD had to reverse-engineer the CPUs.
And as I said, AMD's claim to the original license was invalidated by the court. Which instated the x86 cross-license as we still know it today (although the terms have changed somewhat in recent years. Originally, AMD was only allowed to outsource 20% of production. Under those terms, Global Foundries would never have been possible).

Edit: This looks like a good overview: http://jolt.law.harvard.edu/digest/patent/int … gal-perspective