debs3759 wrote on 2021-01-18, 19:10:

nathanieltolbert wrote on 2021-01-17, 19:32:

So what is the difference between 486 and 486DX?

The only difference is the name.

So was it a naming nomenclature that they came up with after the fact when they decided to release lower end, cost reduced chips, like the Celerons? I was curious because I couldn't find any information. But I vaguely remember when the 486 came out it was just called the 486 and then a bit later the SX and DX came out and I didn't know what kind of difference there was -edit- aside from the fact that the SX doesn't have an activated FPU -edit-. And trying to find information I can find The 486 labeled chips have the ident information of 486DX on the printing at the bottom of the chip. Thank you for the clarification. I appreciate it.

nathanieltolbert wrote on 2021-01-18, 20:56:

debs3759 wrote on 2021-01-18, 19:10:

nathanieltolbert wrote on 2021-01-17, 19:32:

So what is the difference between 486 and 486DX?

The only difference is the name.

So was it a naming nomenclature that they came up with after the fact when they decided to release lower end, cost reduced chips, like the Celerons? I was curious because I couldn't find any information. But I vaguely remember when the 486 came out it was just called the 486 and then a bit later the SX and DX came out and I didn't know what kind of difference there was -edit- aside from the fact that the SX doesn't have an activated FPU -edit-. And trying to find information I can find The 486 labeled chips have the ident information of 486DX on the printing at the bottom of the chip. Thank you for the clarification. I appreciate it.

No, when Intel removed or disabled the onchip FPU.on their 486s. Calling those 486SXs.

Celerons are fine as well. They just had less cache than the "Pentiums".

Caluser2000 wrote on 2021-01-18, 21:03:

No, when Intel removed or disabled the onchip FPU.on their 486s. Calling those 486SXs.

Celerons are fine as well. They just had less cache than the "Pentiums".

That's kind of what I meant. That the SX specifically notes the removal of the FPU feature, like the Celeron which were usually gimped with part of the L1 cache. And as such, they were more affordable chips to purchase. Apologies for wording my thought poorly. What is interesting to me is that I remember hearing about the 486 back in the day and it having a built in FPU making it significantly more powerful than the 386DX it was replacing. And then later hearing about SX and DX variants. I was curious what was difference between the setting on my motherboard because it has 486, 486SX and then 487SX/486DX. And I was under the impression that there was no different between them, but I wasn't sure. I am trying to remember passing information from when I was 10 or 11 and so I cannot say how accurate my memory is. Thank you for the information.

-Edit- Sorry I can see how what I said didn't make sense. I know that the difference between the SX and the DX models are, the SX has no FPU and the DX does. But I was curious what the difference was, if any between a straight chip labeled 486, and the SX and DX variants. Seems from what I have been told is that the 486 and the DX chip from intel were the same, I think? I wonder if the guy on CPU galaxy has an intel CPU that is just labeled as 486 without a DX and SX delineation?

So I have the machine running well and it's now running fast, aside from the fact that I cannot seem to identify the l1 cache, which is fine. But I have another issue. My BTC 16x CD-ROM drive won't read all of my cd-roms. Not just burned ones, but it is having problems with my World of Xeen CD-ROMs where it just errors out and fails. I has seen several videos where people open up their CD-ROM drives and there is a pot that the twist to increase the strength of the laser, but I cannot seem to find it on mine. I did take some pictures of the laser assembly since that was where it was located on other cd rom drives but I can't see it. There are three points that look like pots, but they don't look they have any method of being able to modify them? Hopefully I can figure out how to post this picture properly.

Well, the 486dx is a 33mhz bus.

Even the original 486-20 had a DX in the long string on the bottom of the chip (from what I can remember)

Cant find a pic of a survivor from 1989

nathanieltolbert wrote on 2021-01-18, 23:07:

-Edit- Sorry I can see how what I said didn't make sense. I know that the difference between the SX and the DX models are, the SX has no FPU and the DX does. But I was curious what the difference was, if any between a straight chip labeled 486, and the SX and DX variants. Seems from what I have been told is that the 486 and the DX chip from intel were the same, I think? I wonder if the guy on CPU galaxy has an intel CPU that is just labeled as 486 without a DX and SX delineation?

More than likely a early Intel 486 ccpu wih functioning fpu as mentioned above- ie no difference at all.

Intel486dx33 wrote on 2021-01-18, 23:28:

Well, the 486dx is a 33mhz bus.

25, 33 and 5o buddy.

Wasn't there also an unstable non-doubled 40MHz as well at one point?

nathanieltolbert wrote on 2021-01-19, 05:14:

Wasn't there also an unstable non-doubled 40MHz as well at one point?

Not by Intel.
Intel's try was the single speed 50MHz variant.
AMD and Cyrix had 40MHz models. Their stability depends on how the motherboard and (VLB, sometimes PCI) cards handle the 40MHz bus.
And yes, 50MHz is an even more serious problem with VLB (though I could run it at 60MHz with a few of my VGAs, with additional wait state of course).

Edit. Oh I forgot UMC!

Caluser2000 wrote on 2021-01-19, 01:26:

Intel486dx33 wrote on 2021-01-18, 23:28:

Well, the 486dx is a 33mhz bus.

25, 33 and 5o buddy.

And 40 (Cyrix, AMD and UMC)

alvaro84 wrote on 2021-01-19, 07:39:

Not by Intel. Intel's try was the single speed 50MHz variant. AMD and Cyrix had 40MHz models. Their stability depends on how the […]
Show full quote

Not by Intel.
Intel's try was the single speed 50MHz variant.
AMD and Cyrix had 40MHz models. Their stability depends on how the motherboard and (VLB, sometimes PCI) cards handle the 40MHz bus.
And yes, 50MHz is an even more serious problem with VLB (though I could run it at 60MHz with a few of my VGAs, with additional wait state of course).

On some PCI boards, the Intel DX 50 can even run at 66 MHz. Admittedly, my sample size of one single DX 50 isn't very meaningful 😀

nathanieltolbert wrote on 2021-01-17, 19:22:

It's just so strange that Cachechk doesn't identify that there is an L1 cache at all.

If I have understood correctly how cachechk works, it derives the cache size and cache levels from the speed it calculates in different ranges.
When your cache performs similarly up to 256kb it won't have a way to detect levels in that range.

frudi wrote on 2021-01-19, 10:17:

On some PCI boards, the Intel DX 50 can even run at 66 MHz. Admittedly, my sample size of one single DX 50 isn't very meaningful 😀

No problem, even one data point can prove that something's possible. It can't tell you how common it is, but it's certainly possible.
Heck, I had to get 3 pcs of 300A Celerons to find one running flawlessly at 450MHz stock voltage. I was quite surprised 😁
And still zero 100MHz-stable POD83s.
But I should give a "486DX-60" a try. 66...? I'm not sure. Because VLB and all.

alvaro84 wrote on 2021-01-19, 12:35:

But I should give a "486DX-60" a try. 66...? I'm not sure. Because VLB and all.

That's why he said PCI motherboards.

appiah4 wrote on 2021-01-19, 09:54:

Caluser2000 wrote on 2021-01-19, 01:26:

Intel486dx33 wrote on 2021-01-18, 23:28:

Well, the 486dx is a 33mhz bus.

25, 33 and 5o buddy.

And 40 (Cyrix, AMD and UMC)

Thanks for adding those.

I was specificly refering to Intel but more the merrier. I dought that 486DXguy will take that in at all though.

nathanieltolbert wrote on 2021-01-19, 05:14:

Wasn't there also an unstable non-doubled 40MHz as well at one point?

Sadly Intel insisted on this holy terror

Thank you for the clarification on the DX-40 non-clock doubled CPUs. I did not realize that Intel did a 50MHz variant. But going back to my other weird thing. The CD rom drive. Were the pictures I provided not good enough? I think I see three different things that could be potentiometers to move to increase the laser strength. Are those not what I need? Am I just out of luck on getting this CD-ROM drive to read all of my legit CD's let alone Burned CDs so I can transfer drivers more easily. Sadly I don't have a working ISA 10-BaseT card. -Edit- If there is more that is needed regarding the CD-ROM drive, please let me know and I will do my best to get the information for you.

rmay635703 wrote on 2021-01-19, 16:31:

Sadly Intel insisted on this holy terror

CF39475D-45CD-4568-B19A-39CB9AA5DF3E.png

What a beautiful and pointless chip, definitely missing from my collection...

alvaro84 wrote on 2021-01-19, 18:14:

rmay635703 wrote on 2021-01-19, 16:31:

Sadly Intel insisted on this holy terror

CF39475D-45CD-4568-B19A-39CB9AA5DF3E.png

What a beautiful and pointless chip, definitely missing from my collection...

This one is more pointless, IMHO, ecpecially considering the 1991 release date .

https://en.wikichip.org/wiki/intel/80486/486sx-16