Antinomy wrote:

All AMD 5x86 166MHz happen to be fake. And most 150MHz as well (some are real, but there's not to much of those). As for Cyrix - I've heard about 120 but 133 for real? Can you confirm it wasn't a fake?

Sorry, typo. I wanted to type 160 instead of 166 -_-

I can't comfirm it was a fake, but as we had a group buy on cpu-world.com of these Cyrix 133 chips and all the other buyers are expert cpu collectors, I doubt they are fake. But I don't have the chip at my home yet though, I'll still need it to be shipped to my place, along with a whole other bunch of cpu's. I only do this every couple months or so to save money on shipping costs.

We bought this lot from Ebay (includes a picture of the chips) and I got 1 of them because everyone wanted one!
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem … %253D&viewitem=

Wow, very nice lot that you've got! If you see any more around, let me know. But I still think they might be remarked samples.

Might be (btw, I'm only getting 1 of these chips). When I can test it, I'll know for sure if it can run @ 133Mhz or not.
I wasn't the one who found it though, I just said I wanted one (Well, actually 2, but since we had so many wanting 1 or more, all of us got 1 😁 ).

Another one to think about - what's the bus and multiplier of the 133? 😉 Don't you mention something strange?

Antinomy wrote:

Another one to think about - what's the bus and multiplier of the 133? 😉 Don't you mention something strange?

Isn't that just 33Mhz and x3?

33*3 = 100, my friend 😀
It's the frequency equation for Cyrix 5x86 100. For 120 it's 40*3. For 133... 😉

In case someone wondered why 33x3 isn't 99, it's because the frequency is actually 33.33333333... or 66.666666666 or 99.99999999....

33*3 = 100, my friend Happy
It's the frequency equation for Cyrix 5x86 100. For 120 it's 40*3. For 133... Wink

I'm sure that he thought about 4x33MHz = 133MHz.

486 motherboard specification does not provide 4x external multiplier. To overcome this limitation, CPU producers created internal multiplier. In this scenario you have to set CPU multiplier to 2x and it will work as 4x multiplier.

udam_u wrote:

33*3 = 100, my friend Happy
It's the frequency equation for Cyrix 5x86 100. For 120 it's 40*3. For 133... Wink

I'm sure that he thought about 4x33MHz = 133MHz.

486 motherboard specification does not provide 4x external multiplier. To overcome this limitation, CPU producers created internal multiplier. In this scenario you have to set CPU multiplier to 2x and it will work as 4x multiplier.

Haha, you're right! I meant 4x33 of course!

heh...I'll just blame this €2 keyboard 😜

Well, 4x multiplier makes sense. But I don't remember it being hardware strap available. The last sentence should be read as "there's no way to set jumpers to start it with a 4x multi". Cyrix 5x86 has only 2x and 3x available by h/w straps while 1x and 4x can be set only by software. Meaning a BIOS should know what this is. And can you find a single board that would know?

I didn't read the datasheet close enough, but I was checking the multipliers.

The last sentence should be read as "there's no way to set jumpers to start it with a 2x multiplier because in this case it works always as 4x multiplier". (:

Really? That turns things the other way.

Can anyone suggest a 486 board with PCI, good memory performance?

It's an excellent idea to use a variety of different video cards for testing. I use 4. Virge 325, Diamond Viper V330, Dell Nitro, and Trio64v+. They all seem to behave slightly differently, depending on what type of overclocking you are doing. For each speed configuration test, try all 4 video cards.

Thank a lot! I will use this advice (I have added S3 virge PCI and cirruslogic ISA to my test kit). (:

Currently my AMD DX4-120 is running at 133MHz (2x66MHz) on ECS 8810PAIO so I'm sure that my two Shuttle HOT-433 are not able to work with such high FSB. (: Shuttle has one big adventage - 4 PCI slots but this is not the best platform for overclocking.

Is there a complete list anywhere of 486 boards with this chipset?

You ´could´ try TH99, or download the full version and do a search for the exact chipset from Windows.

It´s the best `full` collection of 486 data that I know of.

Ehr, one problem - there's quite a number of boards which chipset is marked as unknown there. Guess, we need to stack our power and make a better database 😉

Tetrium wrote:

You ´could´ try TH99, or download the full version and do a search for the exact chipset from Windows.

It´s the best `full` collection of 486 data that I know of.

That collection isn't searchable and chipsets aren't mentioned for the vast majority of the boards so even searching manually isn't likely to produce much in the way of results. I just thought someone would have documented the boards that have this chipset somewhere.

Didn't see such a thing in the Web. Yet again, Tetrium know quite a number of oldschool lovers, so we can make our pretty own database. Some info can be taken from the Web and some - from our own hardware.

There is a better way -> just put this into Google search (:
site:http://www.artofhacking.com/th99/ "Chip Set"+"UMC"+"32-bit PCI"

sliderider wrote:

Tetrium wrote:

You ´could´ try TH99, or download the full version and do a search for the exact chipset from Windows.

It´s the best `full` collection of 486 data that I know of.

That collection isn't searchable and chipsets aren't mentioned for the vast majority of the boards so even searching manually isn't likely to produce much in the way of results. I just thought someone would have documented the boards that have this chipset somewhere.

Just download the full version and search using Windows search. Works for me, it's how I found the boards with the "magic jumper" from within TH99.

You basically search TH99 as it's on your own harddrive 😉

But still, you are correct, many of the chipsets are "unknown", but to tell you the truth, I know of no single place at all that has all these boards in 1 place. Afaik it simply doesn't exist.