I have the DX version, not the DX2 version. I know it needs a fan on its standard 50mhz setting, but can I remove the fan and just run it like a regular DX-33?
The fan I have on it doesn't work, and I don't have a replacement.
Yes, it can run it without fan and heat sink at 33. It should run cooler than a normal 33 since it should use a smaller die process.
"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium
Ok Thanks. I dont know much about DX-50s, I was worried it was an experimental version of a 486 that just runs hot all the time.
clownwolf wrote on 2023-06-14, 03:13:Ok Thanks. I dont know much about DX-50s, I was worried it was an experimental version of a 486 that just runs hot all the time.
It like all 33mhz+ 486s do run hot all the time but they are also perfectly stable, a heatsync makes any antique chip last longer even if it doesn’t “need it”
Good Luck
clownwolf wrote on 2023-06-14, 01:29:I have the DX version, not the DX2 version. I know it needs a fan on its standard 50mhz setting, but can I remove the fan and just run it like a regular DX-33?
The fan I have on it doesn't work, and I don't have a replacement.
Basically DX-33: No cooler required
DX-40 and DX2-50: Heatsink required
DX-50 and DX2-66: Heatsink and fan required
So you may run it with 40 MHz instead of 50 MHz as long as you keep the heatsink.
To the question in the topic name: The DX-50 includes JTAG (hardware based test mode). Nothing you should worry about.
DX2/4 3V parts do not need a fan, a cooler will suffice.
Retronautics: A digital gallery of my retro computers, hardware and projects.
Disruptor wrote on 2023-06-14, 04:02:To the question in the topic name: The DX-50 includes JTAG (hardware based test mode). Nothing you should worry about.
Don't most 486 have JTAG? I thought even some 386 had it
See my graphics card database at www.gpuzoo.com
Constantly being worked on. Feel free to message me with any corrections or details of cards you would like me to research and add.
Irony thing is, I had a Am486DX 40 back in the day. This thing wouldn't stay running even at 25MHz, as it get hot. So had to go and buy heatsink with fan. After that, attempted overclocking to 50MHz and that stayed this way for years playing games.
Cheers,
Great Northern aka Canada.
pentiumspeed wrote on 2023-06-14, 19:40:Irony thing is, I had a Am486DX 40 back in the day. This thing wouldn't stay running even at 25MHz, as it get hot. So had to go and buy heatsink with fan. After that, attempted overclocking to 50MHz and that stayed this way for years playing games.
Cheers,
I had several 486sx25’s overclocked to 40 or 50 mhz , usually it appeared something other than the cpu was the limit, many boards had a 50mhz FSB but it really didn’t want to work
Ah well
Anonymous Coward wrote on 2023-06-14, 02:50:Yes, it can run it without fan and heat sink at 33. It should run cooler than a normal 33 since it should use a smaller die process.
Are you sure about that?
https://www.cpu-world.com/CPUs/80486/Intel-A80486DX-50.html states 1.0 or 0.8 process
https://www.cpu-world.com/CPUs/80486/Intel-A80486DX-33.html apparently same goes for the DX-33?
Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀
If your board has voltage regulator, you can decrease CPU TDP further by setting voltage somewhere in 3.4-4.0v range.
I must be some kind of standard: the anonymous gangbanger of the 21st century.
I use my DX-50 with just a heatsink, no fan (albeit a chunky one, not your usual tiny 486 heatsink). Sure, it gets quite warm after a while but nothing out of the ordinary. I don't think any DX-50s used fans back when they first came out, relying on case airflow instead. CPU fans really weren't a common sight in PCs of the early '90s.
A small heatsink should be enough for a DX running at 33MHz, lots of people used to run those completely bare back then!
I do not recall cpu fans before Pentium. Note that even Intel's own 100MHz Overdeive CPUs or Cyrix DX2/4s come with just a heatsink. A fan on DX4 and beyond is a good idea but not necessary.
Retronautics: A digital gallery of my retro computers, hardware and projects.
H3nrik V! wrote on 2023-06-15, 05:30:Are you sure about that? […]
Anonymous Coward wrote on 2023-06-14, 02:50:Yes, it can run it without fan and heat sink at 33. It should run cooler than a normal 33 since it should use a smaller die process.
Are you sure about that?
https://www.cpu-world.com/CPUs/80486/Intel-A80486DX-50.html states 1.0 or 0.8 process
https://www.cpu-world.com/CPUs/80486/Intel-A80486DX-33.html apparently same goes for the DX-33?
I'm going from memory here, but this is my understanding. The majority of the 33MHz parts are 1 micron. When the DX-50 was introduced, the 1 micron process was used for a short time, but there were heat related issues and it was switched to the .8 micron process when the DX/2 was introduced. Probably at least some 33MHz parts made after 1992 were also made on .8 micron.
I can't remember if the 1 micron 50MHz parts were recalled or not, but I don't think I've ever seen one in the wild.
"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium
Anonymous Coward wrote on 2023-06-15, 07:00:H3nrik V! wrote on 2023-06-15, 05:30:Are you sure about that? […]
Anonymous Coward wrote on 2023-06-14, 02:50:Yes, it can run it without fan and heat sink at 33. It should run cooler than a normal 33 since it should use a smaller die process.
Are you sure about that?
https://www.cpu-world.com/CPUs/80486/Intel-A80486DX-50.html states 1.0 or 0.8 process
https://www.cpu-world.com/CPUs/80486/Intel-A80486DX-33.html apparently same goes for the DX-33?I'm going from memory here, but this is my understanding. The majority of the 33MHz parts are 1 micron. When the DX-50 was introduced, the 1 micron process was used for a short time, but there were heat related issues and it was switched to the .8 micron process when the DX/2 was introduced. Probably at least some 33MHz parts made after 1992 were also made on .8 micron.
I can't remember if the 1 micron 50MHz parts were recalled or not, but I don't think I've ever seen one in the wild.
That could actually make sense 😀
Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀
debs3759 wrote on 2023-06-14, 18:29:Don't most 486 have JTAG? I thought even some 386 had it
Not the first 486DX's (CHMOS IV = 1 µm ?)
Not the first PGA 486SX's (CHMOS IV = 1 µm ?)
Not the first 487SX's (CHMOS IV = 1 µm ?)
Not the first low power 486DX's (CHMOS IV = 1 µm)
Not the first low power 486SX's (CHMOS IV = 1 µm)
But the first PQFP 486SX's had TDI, TDO, TMS, TCK. (CHMOS V = 0,8 µm ?)
Same with the first 486 DX 50. (CHMOS V = 0,8 µm ?)
The power supply current [mA] comparison of the first SX's is:
MHz / CHMOS IV / CHMOS V Percent / 487SX
16 / 525 / 450 (85,7 %) / 625
20 / 600 / 500 (83,3 %) / 775
25 / 700 / 560 (80 %) / 950
It seems like the first PGA 486SX's were produced in CHMOS IV (1 µm).
It seems like the first PQFP 486SX's were produced in CHMOS V (0,8 µm).
It seems like the first 487SX's were produced in CHMOS IV (1 µm).
With a 487SX present the PQFP 486SX introduces a Power Down Mode
MHz CHMOS IV CHMOS V
16 / 400 / 50
20 / 500 / 50
25 / 600 / 50
So, in fact, at 25 MHz 486SX / 487SX combo can draw 1550 mA, compared to a 486 DX with 700 mA.
The power supply current table of the first 486 DX's is interesting.
MHz Icc
25 700
33 900
50 1000
The input capacitance of the 25/33 MHz ones is 20 pF, whereas the 50 MHz variant has 13-17 pF.
Conclusion: It seems like the first 486DX's were produced in CHMOS IV (1 µm), but the first 50 MHz ones in CHMOS V (0,8 µm).
For the first 487SX's the input capacitance is also 20 pF.
The imput capacitance of the first PGA 486SX's is also 20 pF, whereas the first PQFP 486SX's have 6-10 pF.
Conclusion: It seems like the first PGA 486SX's were produced in CHMOS IV (1 µm) and have a disabled FPU, but the first PQFP 486SX's are in CHMOS V (0,8 µm) and have no FPU at all.
All values are taken from the book "Intel, Microprocessors, Volume I, 1992"
